From 1ba5bf993c6a3142e18e68ea6452b347f9cb5635 Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Tue, 5 Jul 2016 10:18:08 +0200 Subject: [PATCH 001/271] xfrm: fix crash in XFRM_MSG_GETSA netlink handler If we hit any of the error conditions inside xfrm_dump_sa(), then xfrm_state_walk_init() never gets called. However, we still call xfrm_state_walk_done() from xfrm_dump_sa_done(), which will crash because the state walk was never initialized properly. We can fix this by setting cb->args[0] only after we've processed the first element and checking this before calling xfrm_state_walk_done(). Fixes: d3623099d3 ("ipsec: add support of limited SA dump") Cc: Nicolas Dichtel Cc: Steffen Klassert Signed-off-by: Vegard Nossum Acked-by: Nicolas Dichtel Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index d516845e16e3..4fb04ced5867 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -896,7 +896,8 @@ static int xfrm_dump_sa_done(struct netlink_callback *cb) struct sock *sk = cb->skb->sk; struct net *net = sock_net(sk); - xfrm_state_walk_done(walk, net); + if (cb->args[0]) + xfrm_state_walk_done(walk, net); return 0; } @@ -921,8 +922,6 @@ static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb) u8 proto = 0; int err; - cb->args[0] = 1; - err = nlmsg_parse(cb->nlh, 0, attrs, XFRMA_MAX, xfrma_policy); if (err < 0) @@ -939,6 +938,7 @@ static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb) proto = nla_get_u8(attrs[XFRMA_PROTO]); xfrm_state_walk_init(walk, proto, filter); + cb->args[0] = 1; } (void) xfrm_state_walk(net, walk, dump_one_state, &info); From 73efc3245fd3edb3632d82a3a9c5d5d975a02efc Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Wed, 27 Jul 2016 08:03:18 +0200 Subject: [PATCH 002/271] xfrm: get rid of incorrect WARN AFAICT this message is just printed whenever input validation fails. This is a normal failure and we shouldn't be dumping the stack over it. Looks like it was originally a printk that was maybe incorrectly upgraded to a WARN: commit 62db5cfd70b1ef53aa21f144a806fe3b78c84fab Author: stephen hemminger Date: Wed May 12 06:37:06 2010 +0000 xfrm: add severity to printk Cc: Stephen Hemminger Cc: Steffen Klassert Signed-off-by: Vegard Nossum Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 4fb04ced5867..1a4f142dd50a 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2117,7 +2117,7 @@ static int xfrm_add_acquire(struct sk_buff *skb, struct nlmsghdr *nlh, err = verify_newpolicy_info(&ua->policy); if (err) - goto bad_policy; + goto free_state; /* build an XP */ xp = xfrm_policy_construct(net, &ua->policy, attrs, &err); @@ -2149,8 +2149,6 @@ static int xfrm_add_acquire(struct sk_buff *skb, struct nlmsghdr *nlh, return 0; -bad_policy: - WARN(1, "BAD policy passed\n"); free_state: kfree(x); nomem: From 7677c7560c3e80fde08a7e710d378dedabf950c3 Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Wed, 27 Jul 2016 08:44:15 +0200 Subject: [PATCH 003/271] xfrm: get rid of another incorrect WARN During fuzzing I regularly run into this WARN(). According to Herbert Xu, this "certainly shouldn't be a WARN, it probably shouldn't print anything either". Cc: Stephen Hemminger Cc: Steffen Klassert Cc: Herbert Xu Signed-off-by: Vegard Nossum Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 1a4f142dd50a..cb65d916a345 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2051,9 +2051,6 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, if (up->hard) { xfrm_policy_delete(xp, p->dir); xfrm_audit_policy_delete(xp, 1, true); - } else { - // reset the timers here? - WARN(1, "Don't know what to do with soft policy expire\n"); } km_policy_expired(xp, p->dir, up->hard, nlh->nlmsg_pid); From 6916fb3b10b3cbe3b1f9f5b680675f53e4e299eb Mon Sep 17 00:00:00 2001 From: Tobias Brunner Date: Fri, 29 Jul 2016 09:57:32 +0200 Subject: [PATCH 004/271] xfrm: Ignore socket policies when rebuilding hash tables Whenever thresholds are changed the hash tables are rebuilt. This is done by enumerating all policies and hashing and inserting them into the right table according to the thresholds and direction. Because socket policies are also contained in net->xfrm.policy_all but no hash tables are defined for their direction (dir + XFRM_POLICY_MAX) this causes a NULL or invalid pointer dereference after returning from policy_hash_bysel() if the rebuild is done while any socket policies are installed. Since the rebuild after changing thresholds is scheduled this crash could even occur if the userland sets thresholds seemingly before installing any socket policies. Fixes: 53c2e285f970 ("xfrm: Do not hash socket policies") Signed-off-by: Tobias Brunner Acked-by: Herbert Xu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index b5e665b3cfb0..45f9cf97ea25 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -626,6 +626,10 @@ static void xfrm_hash_rebuild(struct work_struct *work) /* re-insert all policies by order of creation */ list_for_each_entry_reverse(policy, &net->xfrm.policy_all, walk.all) { + if (xfrm_policy_id2dir(policy->index) >= XFRM_POLICY_MAX) { + /* skip socket policies */ + continue; + } newpos = NULL; chain = policy_hash_bysel(net, &policy->selector, policy->family, From 4d0bd46a4d55383f7b925e6cf7865a77e0f0e020 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 8 Aug 2016 08:45:33 +0200 Subject: [PATCH 005/271] Revert "wext: Fix 32 bit iwpriv compatibility issue with 64 bit Kernel" This reverts commit 3d5fdff46c4b2b9534fa2f9fc78e90a48e0ff724. Ben Hutchings pointed out that the commit isn't safe since it assumes that the structure used by the driver is iw_point, when in fact there's no way to know about that. Fortunately, the only driver in the tree that ever runs this code path is the wilc1000 staging driver, so it doesn't really matter. Clearly I should have investigated this better before applying, sorry. Reported-by: Ben Hutchings Cc: stable@vger.kernel.org [though I guess it doesn't matter much] Fixes: 3d5fdff46c4b ("wext: Fix 32 bit iwpriv compatibility issue with 64 bit Kernel") Signed-off-by: Johannes Berg --- net/wireless/wext-core.c | 25 ++----------------------- 1 file changed, 2 insertions(+), 23 deletions(-) diff --git a/net/wireless/wext-core.c b/net/wireless/wext-core.c index dbb2738e356a..6250b1cfcde5 100644 --- a/net/wireless/wext-core.c +++ b/net/wireless/wext-core.c @@ -958,29 +958,8 @@ static int wireless_process_ioctl(struct net *net, struct ifreq *ifr, return private(dev, iwr, cmd, info, handler); } /* Old driver API : call driver ioctl handler */ - if (dev->netdev_ops->ndo_do_ioctl) { -#ifdef CONFIG_COMPAT - if (info->flags & IW_REQUEST_FLAG_COMPAT) { - int ret = 0; - struct iwreq iwr_lcl; - struct compat_iw_point *iwp_compat = (void *) &iwr->u.data; - - memcpy(&iwr_lcl, iwr, sizeof(struct iwreq)); - iwr_lcl.u.data.pointer = compat_ptr(iwp_compat->pointer); - iwr_lcl.u.data.length = iwp_compat->length; - iwr_lcl.u.data.flags = iwp_compat->flags; - - ret = dev->netdev_ops->ndo_do_ioctl(dev, (void *) &iwr_lcl, cmd); - - iwp_compat->pointer = ptr_to_compat(iwr_lcl.u.data.pointer); - iwp_compat->length = iwr_lcl.u.data.length; - iwp_compat->flags = iwr_lcl.u.data.flags; - - return ret; - } else -#endif - return dev->netdev_ops->ndo_do_ioctl(dev, ifr, cmd); - } + if (dev->netdev_ops->ndo_do_ioctl) + return dev->netdev_ops->ndo_do_ioctl(dev, ifr, cmd); return -EOPNOTSUPP; } From 7e196aa1a011da35a04cb1f161e186563aa8d8db Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 6 Jul 2016 12:18:34 +0000 Subject: [PATCH 006/271] clk: sunxi: pll2: Fix return value check in sun4i_pll2_setup() In case of error, the functions clk_register_composite() and clk_register_divider() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Signed-off-by: Wei Yongjun Signed-off-by: Maxime Ripard --- drivers/clk/sunxi/clk-a10-pll2.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/clk/sunxi/clk-a10-pll2.c b/drivers/clk/sunxi/clk-a10-pll2.c index 0ee1f363e4be..d8eab90ae661 100644 --- a/drivers/clk/sunxi/clk-a10-pll2.c +++ b/drivers/clk/sunxi/clk-a10-pll2.c @@ -73,7 +73,7 @@ static void __init sun4i_pll2_setup(struct device_node *node, SUN4I_PLL2_PRE_DIV_WIDTH, CLK_DIVIDER_ONE_BASED | CLK_DIVIDER_ALLOW_ZERO, &sun4i_a10_pll2_lock); - if (!prediv_clk) { + if (IS_ERR(prediv_clk)) { pr_err("Couldn't register the prediv clock\n"); goto err_free_array; } @@ -106,7 +106,7 @@ static void __init sun4i_pll2_setup(struct device_node *node, &mult->hw, &clk_multiplier_ops, &gate->hw, &clk_gate_ops, CLK_SET_RATE_PARENT); - if (!base_clk) { + if (IS_ERR(base_clk)) { pr_err("Couldn't register the base multiplier clock\n"); goto err_free_multiplier; } From fbe359f12ce40e3126bea959b135656e4a305897 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 6 Jul 2016 12:21:47 +0000 Subject: [PATCH 007/271] clk: sunxi: Fix return value check in sun8i_a23_mbus_setup() In case of error, the function of_io_request_and_map() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Signed-off-by: Wei Yongjun Signed-off-by: Maxime Ripard --- drivers/clk/sunxi/clk-sun8i-mbus.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/clk/sunxi/clk-sun8i-mbus.c b/drivers/clk/sunxi/clk-sun8i-mbus.c index 411d3033a96e..b200ebf159ee 100644 --- a/drivers/clk/sunxi/clk-sun8i-mbus.c +++ b/drivers/clk/sunxi/clk-sun8i-mbus.c @@ -48,7 +48,7 @@ static void __init sun8i_a23_mbus_setup(struct device_node *node) return; reg = of_io_request_and_map(node, 0, of_node_full_name(node)); - if (!reg) { + if (IS_ERR(reg)) { pr_err("Could not get registers for sun8i-mbus-clk\n"); goto err_free_parents; } From 156ad0d7eae4825a0d94a4bf68571e97302f2501 Mon Sep 17 00:00:00 2001 From: Chen-Yu Tsai Date: Tue, 26 Jul 2016 15:04:23 +0800 Subject: [PATCH 008/271] clk: sunxi-ng: Fix inverted test condition in ccu_helper_wait_for_lock The condition passed to read*_poll_timeout() is the break condition, i.e. wait for this condition to happen and return success. The original code assumed the opposite, resulting in a warning when the PLL clock rate was changed but never lost it's lock as far as the readout indicated. This was verified by checking the read out register value. Fixes: 1d80c14248d6 ("clk: sunxi-ng: Add common infrastructure") Signed-off-by: Chen-Yu Tsai Acked-by: Maxime Ripard Signed-off-by: Maxime Ripard --- drivers/clk/sunxi-ng/ccu_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/clk/sunxi-ng/ccu_common.c b/drivers/clk/sunxi-ng/ccu_common.c index fc17b5295e16..51d4bac97ab3 100644 --- a/drivers/clk/sunxi-ng/ccu_common.c +++ b/drivers/clk/sunxi-ng/ccu_common.c @@ -31,7 +31,7 @@ void ccu_helper_wait_for_lock(struct ccu_common *common, u32 lock) return; WARN_ON(readl_relaxed_poll_timeout(common->base + common->reg, reg, - !(reg & lock), 100, 70000)); + reg & lock, 100, 70000)); } int sunxi_ccu_probe(struct device_node *node, void __iomem *reg, From 06421a7821f9040467e0db2702309b72aa2b7af4 Mon Sep 17 00:00:00 2001 From: Chen-Yu Tsai Date: Tue, 26 Jul 2016 15:04:24 +0800 Subject: [PATCH 009/271] clk: sunxi-ng: nk: Make ccu_nk_find_best static make C=2 reports: CHECK drivers/clk/sunxi-ng/ccu_nk.c drivers/clk/sunxi-ng/ccu_nk.c:17:6: warning: symbol 'ccu_nk_find_best' was not declared. Should it be static? ccu_nk_find_best is only used within ccu_nk.c. So make it static to get rid of this warning. Fixes: adbfb0056e03 ("clk: sunxi-ng: Add N-K-factor clock support") Signed-off-by: Chen-Yu Tsai Signed-off-by: Maxime Ripard --- drivers/clk/sunxi-ng/ccu_nk.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/clk/sunxi-ng/ccu_nk.c b/drivers/clk/sunxi-ng/ccu_nk.c index 4470ffc8cf0d..d6fafb397489 100644 --- a/drivers/clk/sunxi-ng/ccu_nk.c +++ b/drivers/clk/sunxi-ng/ccu_nk.c @@ -14,9 +14,9 @@ #include "ccu_gate.h" #include "ccu_nk.h" -void ccu_nk_find_best(unsigned long parent, unsigned long rate, - unsigned int max_n, unsigned int max_k, - unsigned int *n, unsigned int *k) +static void ccu_nk_find_best(unsigned long parent, unsigned long rate, + unsigned int max_n, unsigned int max_k, + unsigned int *n, unsigned int *k) { unsigned long best_rate = 0; unsigned int best_k = 0, best_n = 0; From 1625f4529957738be7d87cf157e107b8fb9d23b9 Mon Sep 17 00:00:00 2001 From: Alexey Kodanev Date: Wed, 10 Aug 2016 13:54:57 +0300 Subject: [PATCH 010/271] net/xfrm_input: fix possible NULL deref of tunnel.ip6->parms.i_key Running LTP 'icmp-uni-basic.sh -6 -p ipcomp -m tunnel' test over openvswitch + veth can trigger kernel panic: BUG: unable to handle kernel NULL pointer dereference at 00000000000000e0 IP: [] xfrm_input+0x82/0x750 ... [] xfrm6_rcv_spi+0x1e/0x20 [] xfrm6_tunnel_rcv+0x42/0x50 [xfrm6_tunnel] [] tunnel6_rcv+0x3e/0x8c [tunnel6] [] ip6_input_finish+0xd5/0x430 [] ip6_input+0x33/0x90 [] ip6_rcv_finish+0xa5/0xb0 ... It seems that tunnel.ip6 can have garbage values and also dereferenced without a proper check, only tunnel.ip4 is being verified. Fix it by adding one more if block for AF_INET6 and initialize tunnel.ip6 with NULL inside xfrm6_rcv_spi() (which is similar to xfrm4_rcv_spi()). Fixes: 049f8e2 ("xfrm: Override skb->mark with tunnel->parm.i_key in xfrm_input") Signed-off-by: Alexey Kodanev Signed-off-by: Steffen Klassert --- net/ipv6/xfrm6_input.c | 1 + net/xfrm/xfrm_input.c | 14 +++++++------- 2 files changed, 8 insertions(+), 7 deletions(-) diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c index 0eaab1fa6be5..00a2d40677d6 100644 --- a/net/ipv6/xfrm6_input.c +++ b/net/ipv6/xfrm6_input.c @@ -23,6 +23,7 @@ int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb) int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi) { + XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6 = NULL; XFRM_SPI_SKB_CB(skb)->family = AF_INET6; XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr); return xfrm_input(skb, nexthdr, spi, 0); diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 1c4ad477ce93..6e3f0254d8a1 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -207,15 +207,15 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) family = XFRM_SPI_SKB_CB(skb)->family; /* if tunnel is present override skb->mark value with tunnel i_key */ - if (XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4) { - switch (family) { - case AF_INET: + switch (family) { + case AF_INET: + if (XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4) mark = be32_to_cpu(XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4->parms.i_key); - break; - case AF_INET6: + break; + case AF_INET6: + if (XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6) mark = be32_to_cpu(XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6->parms.i_key); - break; - } + break; } /* Allocate new secpath or COW existing one. */ From c81396f3da22aa8f1e8fbf7943616a0839c4d63d Mon Sep 17 00:00:00 2001 From: Cathy Luo Date: Thu, 21 Jul 2016 16:30:24 +0530 Subject: [PATCH 011/271] mwifiex: fix large amsdu packets causing firmware hang Sometimes host prepares and downloads a large amsdu packet to firmware which leads to a memory corruption in firmware. The reason is __dev_alloc_skb() may allocate larger buffer than required size. This patch solves the problem by checking "adapter->tx_buf_size" instead of relying on skb_tailroom(). Signed-off-by: Cathy Luo Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo --- drivers/net/wireless/marvell/mwifiex/11n_aggr.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/marvell/mwifiex/11n_aggr.c b/drivers/net/wireless/marvell/mwifiex/11n_aggr.c index dc49c3de1f25..c47d6366875d 100644 --- a/drivers/net/wireless/marvell/mwifiex/11n_aggr.c +++ b/drivers/net/wireless/marvell/mwifiex/11n_aggr.c @@ -205,7 +205,8 @@ mwifiex_11n_aggregate_pkt(struct mwifiex_private *priv, do { /* Check if AMSDU can accommodate this MSDU */ - if (skb_tailroom(skb_aggr) < (skb_src->len + LLC_SNAP_LEN)) + if ((skb_aggr->len + skb_src->len + LLC_SNAP_LEN) > + adapter->tx_buf_size) break; skb_src = skb_dequeue(&pra_list->skb_head); From 11d7a0bb95eaaba1741bb24a7c3c169c82f09c7b Mon Sep 17 00:00:00 2001 From: David Ahern Date: Sun, 14 Aug 2016 19:52:56 -0700 Subject: [PATCH 012/271] xfrm: Only add l3mdev oif to dst lookups Subash reported that commit 42a7b32b73d6 ("xfrm: Add oif to dst lookups") broke a wifi use case that uses fib rules and xfrms. The intent of 42a7b32b73d6 was driven by VRFs with IPsec. As a compromise relax the use of oif in xfrm lookups to L3 master devices only (ie., oif is either an L3 master device or is enslaved to a master device). Fixes: 42a7b32b73d6 ("xfrm: Add oif to dst lookups") Reported-by: Subash Abhinov Kasiviswanathan Signed-off-by: David Ahern Signed-off-by: Steffen Klassert --- net/ipv4/xfrm4_policy.c | 2 +- net/ipv6/xfrm6_policy.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 7b0edb37a115..e07ed8b1deb3 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -29,7 +29,7 @@ static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4, memset(fl4, 0, sizeof(*fl4)); fl4->daddr = daddr->a4; fl4->flowi4_tos = tos; - fl4->flowi4_oif = oif; + fl4->flowi4_oif = l3mdev_master_ifindex_by_index(net, oif); if (saddr) fl4->saddr = saddr->a4; diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index c074771a10f7..dd84ecd1221b 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -36,7 +36,7 @@ static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos, int oif, int err; memset(&fl6, 0, sizeof(fl6)); - fl6.flowi6_oif = oif; + fl6.flowi6_oif = l3mdev_master_ifindex_by_index(net, oif); fl6.flowi6_flags = FLOWI_FLAG_SKIP_NH_OIF; memcpy(&fl6.daddr, daddr, sizeof(fl6.daddr)); if (saddr) From 05860bed491b114a9f2d7a4f6e09fb02c0b69056 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Fri, 19 Aug 2016 13:37:38 +0300 Subject: [PATCH 013/271] ath9k: fix client mode beacon configuration For pure station mode, iter_data.primary_beacon_vif was used and passed to ath_beacon_config, but not set to the station vif. This was causing the following warning: [ 100.310919] ------------[ cut here ]------------ [ 100.315683] WARNING: CPU: 0 PID: 7 at compat-wireless-2016-06-20/drivers/net/wireless/ath/ath9k/beacon.c:642 ath9k_calculate_summary_state+0x250/0x60c [ath9k]() [ 100.402028] CPU: 0 PID: 7 Comm: kworker/u2:1 Tainted: G W 4.4.15 #5 [ 100.409676] Workqueue: phy0 ieee80211_ibss_leave [mac80211] [ 100.415351] Stack : 8736e98c 870b4b20 87a25b54 800a6800 8782a080 80400d63 8039b96c 00000007 [ 100.415351] 803c5edc 87875914 80400000 800a47cc 87a25b54 800a6800 803a0fd8 80400000 [ 100.415351] 00000003 87875914 80400000 80094ae0 87a25b54 8787594c 00000000 801ef308 [ 100.415351] 803ffe70 801ef300 87193d58 87b3a400 87b3ad00 70687930 00000000 00000000 [ 100.415351] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 [ 100.415351] ... [ 100.451703] Call Trace: [ 100.454235] [<800a6800>] vprintk_default+0x24/0x30 [ 100.459110] [<800a47cc>] printk+0x2c/0x38 [ 100.463190] [<800a6800>] vprintk_default+0x24/0x30 [ 100.468072] [<80094ae0>] print_worker_info+0x148/0x174 [ 100.473378] [<801ef308>] serial8250_console_putchar+0x0/0x44 [ 100.479122] [<801ef300>] wait_for_xmitr+0xc4/0xcc [ 100.484014] [<87193d58>] ieee80211_ibss_leave+0xb90/0x1900 [mac80211] [ 100.490590] [<80081604>] warn_slowpath_common+0xa0/0xd0 [ 100.495922] [<801a359c>] dump_stack+0x14/0x28 [ 100.500350] [<80071a00>] show_stack+0x50/0x84 [ 100.504784] [<80081604>] warn_slowpath_common+0xa0/0xd0 [ 100.510106] [<87024c60>] ath9k_calculate_summary_state+0x250/0x60c [ath9k] [ 100.517105] [<800816b8>] warn_slowpath_null+0x18/0x24 [ 100.522256] [<87024c60>] ath9k_calculate_summary_state+0x250/0x60c [ath9k] [ 100.529273] [<87025418>] ath9k_set_txpower+0x148/0x498 [ath9k] [ 100.535302] [<871d2c64>] cleanup_module+0xa74/0xd4c [mac80211] [ 100.541237] [<801ef308>] serial8250_console_putchar+0x0/0x44 [ 100.547042] [<800a5d18>] wake_up_klogd+0x54/0x68 [ 100.551730] [<800a6650>] vprintk_emit+0x404/0x43c [ 100.556623] [<871b9db8>] ieee80211_sta_rx_notify+0x258/0x32c [mac80211] [ 100.563475] [<871ba6a4>] ieee80211_sta_rx_queued_mgmt+0x63c/0x734 [mac80211] [ 100.570693] [<871aa49c>] ieee80211_tx_prepare_skb+0x210/0x230 [mac80211] [ 100.577609] [<800af5d4>] mod_timer+0x15c/0x190 [ 100.582220] [<871ba8b8>] ieee80211_sta_work+0xfc/0xe1c [mac80211] [ 100.588539] [<871940b4>] ieee80211_ibss_leave+0xeec/0x1900 [mac80211] [ 100.595122] [<8009ec84>] dequeue_task_fair+0x44/0x130 [ 100.600281] [<80092a34>] process_one_work+0x1f8/0x334 [ 100.605454] [<80093830>] worker_thread+0x2b4/0x408 [ 100.610317] [<8009357c>] worker_thread+0x0/0x408 [ 100.615019] [<8009357c>] worker_thread+0x0/0x408 [ 100.619705] [<80097b68>] kthread+0xdc/0xe8 [ 100.623886] [<80097a8c>] kthread+0x0/0xe8 [ 100.627961] [<80060878>] ret_from_kernel_thread+0x14/0x1c [ 100.633448] [ 100.634956] ---[ end trace aafbe57e9ae6862f ]--- Fixes: cfda2d8e2314 ("ath9k: Fix beacon configuration for addition/removal of interfaces") Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath9k/main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index a394622c9022..098745dc683b 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -1154,6 +1154,7 @@ void ath9k_calculate_summary_state(struct ath_softc *sc, bool changed = (iter_data.primary_sta != ctx->primary_sta); if (iter_data.primary_sta) { + iter_data.primary_beacon_vif = iter_data.primary_sta; iter_data.beacons = true; ath9k_set_assoc_state(sc, iter_data.primary_sta, changed); From 7711aaf08ad3fc4d0e937eec1de0a63620444ce7 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Fri, 19 Aug 2016 13:37:46 +0300 Subject: [PATCH 014/271] ath9k: fix using sta->drv_priv before initializing it A station pointer can be passed to the driver on tx, before it has been marked as associated. Since ath9k_sta_state was initializing the entry too late, it resulted in some spurious crashes. Fixes: df3c6eb34da5 ("ath9k: Use sta_state() callback") Cc: stable@vger.kernel.org Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath9k/main.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index 098745dc683b..6ca4337a98c4 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -1564,13 +1564,13 @@ static int ath9k_sta_state(struct ieee80211_hw *hw, struct ath_common *common = ath9k_hw_common(sc->sc_ah); int ret = 0; - if (old_state == IEEE80211_STA_AUTH && - new_state == IEEE80211_STA_ASSOC) { + if (old_state == IEEE80211_STA_NOTEXIST && + new_state == IEEE80211_STA_NONE) { ret = ath9k_sta_add(hw, vif, sta); ath_dbg(common, CONFIG, "Add station: %pM\n", sta->addr); - } else if (old_state == IEEE80211_STA_ASSOC && - new_state == IEEE80211_STA_AUTH) { + } else if (old_state == IEEE80211_STA_NONE && + new_state == IEEE80211_STA_NOTEXIST) { ret = ath9k_sta_remove(hw, vif, sta); ath_dbg(common, CONFIG, "Remove station: %pM\n", sta->addr); From 15dacf880e49ce3ecee05eb1a0c6b8e363dbacdc Mon Sep 17 00:00:00 2001 From: "mhiramat@kernel.org" Date: Mon, 15 Aug 2016 18:40:57 +0900 Subject: [PATCH 015/271] brcmfmac: Check rtnl_lock is locked when removing interface MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Check rtnl_lock is locked in brcmf_p2p_ifp_removed() by passing rtnl_locked flag. Actually the caller brcmf_del_if() checks whether the rtnl_lock is locked, but doesn't pass it to brcmf_p2p_ifp_removed(). Without this fix, wpa_supplicant goes softlockup with rtnl_lock holding (this means all other process using netlink are locked up too) e.g. [ 4495.876627] INFO: task wpa_supplicant:7307 blocked for more than 10 seconds. [ 4495.876632] Tainted: G W 4.8.0-rc1+ #8 [ 4495.876635] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [ 4495.876638] wpa_supplicant D ffff974c647b39a0 0 7307 1 0x00000000 [ 4495.876644] ffff974c647b39a0 0000000000000000 ffff974c00000000 ffff974c7dc59c58 [ 4495.876651] ffff974c6b7417c0 ffff974c645017c0 ffff974c647b4000 ffffffff86f16c08 [ 4495.876657] ffff974c645017c0 0000000000000246 00000000ffffffff ffff974c647b39b8 [ 4495.876664] Call Trace: [ 4495.876671] [] schedule+0x3c/0x90 [ 4495.876676] [] schedule_preempt_disabled+0x15/0x20 [ 4495.876682] [] mutex_lock_nested+0x176/0x3b0 [ 4495.876686] [] ? rtnl_lock+0x17/0x20 [ 4495.876690] [] rtnl_lock+0x17/0x20 [ 4495.876720] [] brcmf_p2p_ifp_removed+0x4d/0x70 [brcmfmac] [ 4495.876741] [] brcmf_remove_interface+0x196/0x1b0 [brcmfmac] [ 4495.876760] [] brcmf_p2p_del_vif+0x111/0x220 [brcmfmac] [ 4495.876777] [] brcmf_cfg80211_del_iface+0x21b/0x270 [brcmfmac] [ 4495.876820] [] nl80211_del_interface+0xfe/0x3a0 [cfg80211] [ 4495.876825] [] genl_family_rcv_msg+0x1b5/0x370 [ 4495.876832] [] ? trace_hardirqs_on+0xd/0x10 [ 4495.876836] [] genl_rcv_msg+0x7d/0xb0 [ 4495.876839] [] ? genl_family_rcv_msg+0x370/0x370 [ 4495.876846] [] netlink_rcv_skb+0x97/0xb0 [ 4495.876849] [] genl_rcv+0x28/0x40 [ 4495.876854] [] netlink_unicast+0x1d3/0x2f0 [ 4495.876860] [] ? netlink_unicast+0x14b/0x2f0 [ 4495.876866] [] netlink_sendmsg+0x2eb/0x3a0 [ 4495.876870] [] sock_sendmsg+0x38/0x50 [ 4495.876874] [] ___sys_sendmsg+0x27f/0x290 [ 4495.876882] [] ? mntput_no_expire+0x5/0x3f0 [ 4495.876888] [] ? mntput_no_expire+0x8e/0x3f0 [ 4495.876894] [] ? mntput_no_expire+0x5/0x3f0 [ 4495.876899] [] ? mntput+0x24/0x40 [ 4495.876904] [] ? __fput+0x190/0x200 [ 4495.876909] [] __sys_sendmsg+0x45/0x80 [ 4495.876914] [] SyS_sendmsg+0x12/0x20 [ 4495.876918] [] entry_SYSCALL_64_fastpath+0x23/0xc1 [ 4495.876924] [] ? trace_hardirqs_off_caller+0x1f/0xc0 Signed-off-by: Masami Hiramatsu Acked-by: Rafał Miłecki Signed-off-by: Kalle Valo --- drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c | 2 +- drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c | 8 +++++--- drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.h | 2 +- 3 files changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c index 8d16f0204985..65e8c8766441 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c @@ -743,7 +743,7 @@ static void brcmf_del_if(struct brcmf_pub *drvr, s32 bsscfgidx, * serious troublesome side effects. The p2p module will clean * up the ifp if needed. */ - brcmf_p2p_ifp_removed(ifp); + brcmf_p2p_ifp_removed(ifp, rtnl_locked); kfree(ifp); } } diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c index 66f942f7448e..de19c7c92bc6 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c @@ -2297,7 +2297,7 @@ int brcmf_p2p_del_vif(struct wiphy *wiphy, struct wireless_dev *wdev) return err; } -void brcmf_p2p_ifp_removed(struct brcmf_if *ifp) +void brcmf_p2p_ifp_removed(struct brcmf_if *ifp, bool rtnl_locked) { struct brcmf_cfg80211_info *cfg; struct brcmf_cfg80211_vif *vif; @@ -2306,9 +2306,11 @@ void brcmf_p2p_ifp_removed(struct brcmf_if *ifp) vif = ifp->vif; cfg = wdev_to_cfg(&vif->wdev); cfg->p2p.bss_idx[P2PAPI_BSSCFG_DEVICE].vif = NULL; - rtnl_lock(); + if (!rtnl_locked) + rtnl_lock(); cfg80211_unregister_wdev(&vif->wdev); - rtnl_unlock(); + if (!rtnl_locked) + rtnl_unlock(); brcmf_free_vif(vif); } diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.h index a3bd18c2360b..8ce9447533ef 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.h @@ -155,7 +155,7 @@ struct wireless_dev *brcmf_p2p_add_vif(struct wiphy *wiphy, const char *name, int brcmf_p2p_del_vif(struct wiphy *wiphy, struct wireless_dev *wdev); int brcmf_p2p_ifchange(struct brcmf_cfg80211_info *cfg, enum brcmf_fil_p2p_if_types if_type); -void brcmf_p2p_ifp_removed(struct brcmf_if *ifp); +void brcmf_p2p_ifp_removed(struct brcmf_if *ifp, bool rtnl_locked); int brcmf_p2p_start_device(struct wiphy *wiphy, struct wireless_dev *wdev); void brcmf_p2p_stop_device(struct wiphy *wiphy, struct wireless_dev *wdev); int brcmf_p2p_scan_prep(struct wiphy *wiphy, From b64abcb7dae6060c67ab0e548da3ef923c49641d Mon Sep 17 00:00:00 2001 From: "mhiramat@kernel.org" Date: Mon, 15 Aug 2016 18:41:12 +0900 Subject: [PATCH 016/271] brcmfmac: Change vif_event_lock to spinlock Change vif_event_lock to spinlock from mutex, since this lock is used in wait_event_timeout() via vif_event_equals(). This caused a warning report as below. As far as I can see, this lock protects regions where updating structure members, not function calls. Also, since those regions are not called from interrupt handlers (of course, it was a mutex), spin_lock is used instead of spin_lock_irqsave. [ 186.678550] ------------[ cut here ]------------ [ 186.678556] WARNING: CPU: 2 PID: 7140 at /home/mhiramat/ksrc/linux/kernel/sched/core.c:7545 __might_sleep+0x7c/0x80 [ 186.678560] do not call blocking ops when !TASK_RUNNING; state=2 set at [] prepare_to_wait_event+0x60/0x100 [ 186.678560] Modules linked in: brcmfmac xt_CHECKSUM rfcomm ipt_MASQUERADE nf_nat_masquerade_ipv4 xt_addrtype br_netfilter xt_tcpudp ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ip_set nfnetlink ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_raw ip6table_security ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_filter ip6_tables iptable_raw iptable_security iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_filter ip_tables x_tables bnep nls_iso8859_1 i2c_designware_platform i2c_designware_core snd_hda_codec_hdmi snd_hda_codec_realtek dcdbas snd_hda_codec_generic snd_hda_intel snd_hda_codec intel_rapl snd_hda_core x86_pkg_temp_thermal intel_powerclamp coretemp [ 186.678594] snd_pcm crct10dif_pclmul crc32_pclmul aesni_intel aes_x86_64 joydev glue_helper snd_hwdep lrw gf128mul uvcvideo ablk_helper snd_seq_midi cryptd snd_seq_midi_event snd_rawmidi videobuf2_vmalloc videobuf2_memops snd_seq input_leds videobuf2_v4l2 cfg80211 videobuf2_core snd_timer videodev serio_raw btusb snd_seq_device media btrtl rtsx_pci_ms snd mei_me memstick hid_multitouch mei soundcore brcmutil idma64 virt_dma intel_lpss_pci processor_thermal_device intel_soc_dts_iosf hci_uart btbcm btqca btintel bluetooth int3403_thermal dell_smo8800 intel_lpss_acpi intel_lpss int3402_thermal int340x_thermal_zone intel_hid mac_hid int3400_thermal shpchp sparse_keymap acpi_pad acpi_thermal_rel acpi_als kfifo_buf industrialio kvm_intel kvm irqbypass parport_pc ppdev lp parport autofs4 btrfs xor raid6_pq [ 186.678631] usbhid nouveau ttm i915 rtsx_pci_sdmmc mxm_wmi i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops psmouse drm ahci rtsx_pci nvme nvme_core libahci i2c_hid hid pinctrl_sunrisepoint video wmi pinctrl_intel fjes [last unloaded: brcmfmac] [ 186.678646] CPU: 2 PID: 7140 Comm: wpa_supplicant Not tainted 4.8.0-rc1+ #8 [ 186.678647] Hardware name: Dell Inc. XPS 15 9550/0N7TVV, BIOS 01.02.00 04/07/2016 [ 186.678648] 0000000000000000 ffff9d8c64b5b900 ffffffff98442f23 ffff9d8c64b5b950 [ 186.678651] 0000000000000000 ffff9d8c64b5b940 ffffffff9808b22b 00001d790000000d [ 186.678653] ffffffff98c75e78 000000000000026c 0000000000000000 ffff9d8c2706d058 [ 186.678655] Call Trace: [ 186.678659] [] dump_stack+0x85/0xc2 [ 186.678666] [] __warn+0xcb/0xf0 [ 186.678668] [] warn_slowpath_fmt+0x4f/0x60 [ 186.678671] [] ? prepare_to_wait_event+0x60/0x100 [ 186.678672] [] ? prepare_to_wait_event+0x60/0x100 [ 186.678674] [] __might_sleep+0x7c/0x80 [ 186.678680] [] mutex_lock_nested+0x33/0x3b0 [ 186.678682] [] ? trace_hardirqs_on+0xd/0x10 [ 186.678689] [] brcmf_cfg80211_wait_vif_event+0xcd/0x130 [brcmfmac] [ 186.678691] [] ? wake_atomic_t_function+0x60/0x60 [ 186.678697] [] brcmf_p2p_del_vif+0xf9/0x220 [brcmfmac] [ 186.678702] [] brcmf_cfg80211_del_iface+0x21b/0x270 [brcmfmac] [ 186.678716] [] nl80211_del_interface+0xfe/0x3a0 [cfg80211] [ 186.678718] [] genl_family_rcv_msg+0x1b5/0x370 [ 186.678720] [] ? trace_hardirqs_on+0xd/0x10 [ 186.678721] [] genl_rcv_msg+0x7d/0xb0 [ 186.678722] [] ? genl_family_rcv_msg+0x370/0x370 [ 186.678724] [] netlink_rcv_skb+0x97/0xb0 [ 186.678726] [] genl_rcv+0x28/0x40 [ 186.678727] [] netlink_unicast+0x1d3/0x2f0 [ 186.678729] [] ? netlink_unicast+0x14b/0x2f0 [ 186.678731] [] netlink_sendmsg+0x2eb/0x3a0 [ 186.678733] [] sock_sendmsg+0x38/0x50 [ 186.678734] [] ___sys_sendmsg+0x27f/0x290 [ 186.678737] [] ? mntput_no_expire+0x5/0x3f0 [ 186.678739] [] ? mntput_no_expire+0x8e/0x3f0 [ 186.678741] [] ? mntput_no_expire+0x5/0x3f0 [ 186.678743] [] ? mntput+0x24/0x40 [ 186.678744] [] ? __fput+0x190/0x200 [ 186.678746] [] __sys_sendmsg+0x45/0x80 [ 186.678748] [] SyS_sendmsg+0x12/0x20 [ 186.678749] [] entry_SYSCALL_64_fastpath+0x23/0xc1 [ 186.678751] [] ? trace_hardirqs_off_caller+0x1f/0xc0 [ 186.678752] ---[ end trace e224d66c5d8408b5 ]--- Signed-off-by: Masami Hiramatsu Acked-by: Arend van Spriel Signed-off-by: Kalle Valo --- .../broadcom/brcm80211/brcmfmac/cfg80211.c | 26 +++++++++---------- .../broadcom/brcm80211/brcmfmac/cfg80211.h | 2 +- 2 files changed, 14 insertions(+), 14 deletions(-) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 2628d5e12c64..5db56a74e7df 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -5635,7 +5635,7 @@ static s32 brcmf_notify_vif_event(struct brcmf_if *ifp, ifevent->action, ifevent->flags, ifevent->ifidx, ifevent->bsscfgidx); - mutex_lock(&event->vif_event_lock); + spin_lock(&event->vif_event_lock); event->action = ifevent->action; vif = event->vif; @@ -5643,7 +5643,7 @@ static s32 brcmf_notify_vif_event(struct brcmf_if *ifp, case BRCMF_E_IF_ADD: /* waiting process may have timed out */ if (!cfg->vif_event.vif) { - mutex_unlock(&event->vif_event_lock); + spin_unlock(&event->vif_event_lock); return -EBADF; } @@ -5654,24 +5654,24 @@ static s32 brcmf_notify_vif_event(struct brcmf_if *ifp, ifp->ndev->ieee80211_ptr = &vif->wdev; SET_NETDEV_DEV(ifp->ndev, wiphy_dev(cfg->wiphy)); } - mutex_unlock(&event->vif_event_lock); + spin_unlock(&event->vif_event_lock); wake_up(&event->vif_wq); return 0; case BRCMF_E_IF_DEL: - mutex_unlock(&event->vif_event_lock); + spin_unlock(&event->vif_event_lock); /* event may not be upon user request */ if (brcmf_cfg80211_vif_event_armed(cfg)) wake_up(&event->vif_wq); return 0; case BRCMF_E_IF_CHANGE: - mutex_unlock(&event->vif_event_lock); + spin_unlock(&event->vif_event_lock); wake_up(&event->vif_wq); return 0; default: - mutex_unlock(&event->vif_event_lock); + spin_unlock(&event->vif_event_lock); break; } return -EINVAL; @@ -5792,7 +5792,7 @@ static void wl_deinit_priv(struct brcmf_cfg80211_info *cfg) static void init_vif_event(struct brcmf_cfg80211_vif_event *event) { init_waitqueue_head(&event->vif_wq); - mutex_init(&event->vif_event_lock); + spin_lock_init(&event->vif_event_lock); } static s32 brcmf_dongle_roam(struct brcmf_if *ifp) @@ -6691,9 +6691,9 @@ static inline bool vif_event_equals(struct brcmf_cfg80211_vif_event *event, { u8 evt_action; - mutex_lock(&event->vif_event_lock); + spin_lock(&event->vif_event_lock); evt_action = event->action; - mutex_unlock(&event->vif_event_lock); + spin_unlock(&event->vif_event_lock); return evt_action == action; } @@ -6702,10 +6702,10 @@ void brcmf_cfg80211_arm_vif_event(struct brcmf_cfg80211_info *cfg, { struct brcmf_cfg80211_vif_event *event = &cfg->vif_event; - mutex_lock(&event->vif_event_lock); + spin_lock(&event->vif_event_lock); event->vif = vif; event->action = 0; - mutex_unlock(&event->vif_event_lock); + spin_unlock(&event->vif_event_lock); } bool brcmf_cfg80211_vif_event_armed(struct brcmf_cfg80211_info *cfg) @@ -6713,9 +6713,9 @@ bool brcmf_cfg80211_vif_event_armed(struct brcmf_cfg80211_info *cfg) struct brcmf_cfg80211_vif_event *event = &cfg->vif_event; bool armed; - mutex_lock(&event->vif_event_lock); + spin_lock(&event->vif_event_lock); armed = event->vif != NULL; - mutex_unlock(&event->vif_event_lock); + spin_unlock(&event->vif_event_lock); return armed; } diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h index 7d77f869b7f1..8889832c17e0 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h @@ -227,7 +227,7 @@ struct escan_info { */ struct brcmf_cfg80211_vif_event { wait_queue_head_t vif_wq; - struct mutex vif_event_lock; + spinlock_t vif_event_lock; u8 action; struct brcmf_cfg80211_vif *vif; }; From 89e1f6d2b956649fbe0704d543a90b8e0cf872b0 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 01:02:18 +0800 Subject: [PATCH 017/271] netfilter: nft_reject: restrict to INPUT/FORWARD/OUTPUT After I add the nft rule "nft add rule filter prerouting reject with tcp reset", kernel panic happened on my system: NULL pointer dereference at ... IP: [] nf_send_reset+0xaf/0x400 Call Trace: [] ? nf_reject_ip_tcphdr_get+0x160/0x160 [] nft_reject_ipv4_eval+0x61/0xb0 [nft_reject_ipv4] [] nft_do_chain+0x1fa/0x890 [nf_tables] [] ? __nft_trace_packet+0x170/0x170 [nf_tables] [] ? nf_ct_invert_tuple+0xb0/0xc0 [nf_conntrack] [] ? nf_nat_setup_info+0x5d4/0x650 [nf_nat] [...] Because in the PREROUTING chain, routing information is not exist, then we will dereference the NULL pointer and oops happen. So we restrict reject expression to INPUT, FORWARD and OUTPUT chain. This is consistent with iptables REJECT target. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nft_reject.h | 4 ++++ net/ipv4/netfilter/nft_reject_ipv4.c | 1 + net/ipv6/netfilter/nft_reject_ipv6.c | 1 + net/netfilter/nft_reject.c | 16 ++++++++++++++++ net/netfilter/nft_reject_inet.c | 7 ++++++- 5 files changed, 28 insertions(+), 1 deletion(-) diff --git a/include/net/netfilter/nft_reject.h b/include/net/netfilter/nft_reject.h index 60fa1530006b..02e28c529b29 100644 --- a/include/net/netfilter/nft_reject.h +++ b/include/net/netfilter/nft_reject.h @@ -8,6 +8,10 @@ struct nft_reject { extern const struct nla_policy nft_reject_policy[]; +int nft_reject_validate(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nft_data **data); + int nft_reject_init(const struct nft_ctx *ctx, const struct nft_expr *expr, const struct nlattr * const tb[]); diff --git a/net/ipv4/netfilter/nft_reject_ipv4.c b/net/ipv4/netfilter/nft_reject_ipv4.c index c24f41c816b3..2c2553b9026c 100644 --- a/net/ipv4/netfilter/nft_reject_ipv4.c +++ b/net/ipv4/netfilter/nft_reject_ipv4.c @@ -46,6 +46,7 @@ static const struct nft_expr_ops nft_reject_ipv4_ops = { .eval = nft_reject_ipv4_eval, .init = nft_reject_init, .dump = nft_reject_dump, + .validate = nft_reject_validate, }; static struct nft_expr_type nft_reject_ipv4_type __read_mostly = { diff --git a/net/ipv6/netfilter/nft_reject_ipv6.c b/net/ipv6/netfilter/nft_reject_ipv6.c index 533cd5719c59..92bda9908bb9 100644 --- a/net/ipv6/netfilter/nft_reject_ipv6.c +++ b/net/ipv6/netfilter/nft_reject_ipv6.c @@ -47,6 +47,7 @@ static const struct nft_expr_ops nft_reject_ipv6_ops = { .eval = nft_reject_ipv6_eval, .init = nft_reject_init, .dump = nft_reject_dump, + .validate = nft_reject_validate, }; static struct nft_expr_type nft_reject_ipv6_type __read_mostly = { diff --git a/net/netfilter/nft_reject.c b/net/netfilter/nft_reject.c index 0522fc9bfb0a..c64de3f7379d 100644 --- a/net/netfilter/nft_reject.c +++ b/net/netfilter/nft_reject.c @@ -26,11 +26,27 @@ const struct nla_policy nft_reject_policy[NFTA_REJECT_MAX + 1] = { }; EXPORT_SYMBOL_GPL(nft_reject_policy); +int nft_reject_validate(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nft_data **data) +{ + return nft_chain_validate_hooks(ctx->chain, + (1 << NF_INET_LOCAL_IN) | + (1 << NF_INET_FORWARD) | + (1 << NF_INET_LOCAL_OUT)); +} +EXPORT_SYMBOL_GPL(nft_reject_validate); + int nft_reject_init(const struct nft_ctx *ctx, const struct nft_expr *expr, const struct nlattr * const tb[]) { struct nft_reject *priv = nft_expr_priv(expr); + int err; + + err = nft_reject_validate(ctx, expr, NULL); + if (err < 0) + return err; if (tb[NFTA_REJECT_TYPE] == NULL) return -EINVAL; diff --git a/net/netfilter/nft_reject_inet.c b/net/netfilter/nft_reject_inet.c index 759ca5248a3d..e79d9ca2ffee 100644 --- a/net/netfilter/nft_reject_inet.c +++ b/net/netfilter/nft_reject_inet.c @@ -66,7 +66,11 @@ static int nft_reject_inet_init(const struct nft_ctx *ctx, const struct nlattr * const tb[]) { struct nft_reject *priv = nft_expr_priv(expr); - int icmp_code; + int icmp_code, err; + + err = nft_reject_validate(ctx, expr, NULL); + if (err < 0) + return err; if (tb[NFTA_REJECT_TYPE] == NULL) return -EINVAL; @@ -124,6 +128,7 @@ static const struct nft_expr_ops nft_reject_inet_ops = { .eval = nft_reject_inet_eval, .init = nft_reject_inet_init, .dump = nft_reject_inet_dump, + .validate = nft_reject_validate, }; static struct nft_expr_type nft_reject_inet_type __read_mostly = { From 93fac10b99d78eb2c50a739cba2e590c7332d539 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 21:58:16 +0800 Subject: [PATCH 018/271] netfilter: nfnetlink: use list_for_each_entry_safe to delete all objects cttimeout and acct objects are deleted from the list while traversing it, so use list_for_each_entry is unsafe here. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nfnetlink_acct.c | 6 +++--- net/netfilter/nfnetlink_cttimeout.c | 5 +++-- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c index 70eb2f6a3b01..d44d89b56127 100644 --- a/net/netfilter/nfnetlink_acct.c +++ b/net/netfilter/nfnetlink_acct.c @@ -343,12 +343,12 @@ static int nfnl_acct_del(struct net *net, struct sock *nfnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const tb[]) { - char *acct_name; - struct nf_acct *cur; + struct nf_acct *cur, *tmp; int ret = -ENOENT; + char *acct_name; if (!tb[NFACCT_NAME]) { - list_for_each_entry(cur, &net->nfnl_acct_list, head) + list_for_each_entry_safe(cur, tmp, &net->nfnl_acct_list, head) nfnl_acct_try_del(cur); return 0; diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index 68216cdc7083..f74fee1e2d0a 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -350,12 +350,13 @@ static int cttimeout_del_timeout(struct net *net, struct sock *ctnl, const struct nlmsghdr *nlh, const struct nlattr * const cda[]) { - struct ctnl_timeout *cur; + struct ctnl_timeout *cur, *tmp; int ret = -ENOENT; char *name; if (!cda[CTA_TIMEOUT_NAME]) { - list_for_each_entry(cur, &net->nfct_timeout_list, head) + list_for_each_entry_safe(cur, tmp, &net->nfct_timeout_list, + head) ctnl_timeout_try_del(net, cur); return 0; From 23aaba5ad55547db62bada5066c8fb6412d5b1c2 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 21:58:17 +0800 Subject: [PATCH 019/271] netfilter: cttimeout: put back l4proto when replacing timeout policy We forget to call nf_ct_l4proto_put when replacing the existing timeout policy. Acctually, there's no need to get ct l4proto before doing replace, so we can move it to a later position. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nfnetlink_cttimeout.c | 31 +++++++++++++---------------- 1 file changed, 14 insertions(+), 17 deletions(-) diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index f74fee1e2d0a..6844c7af0b8f 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -98,31 +98,28 @@ static int cttimeout_new_timeout(struct net *net, struct sock *ctnl, break; } - l4proto = nf_ct_l4proto_find_get(l3num, l4num); - - /* This protocol is not supportted, skip. */ - if (l4proto->l4proto != l4num) { - ret = -EOPNOTSUPP; - goto err_proto_put; - } - if (matching) { if (nlh->nlmsg_flags & NLM_F_REPLACE) { /* You cannot replace one timeout policy by another of * different kind, sorry. */ if (matching->l3num != l3num || - matching->l4proto->l4proto != l4num) { - ret = -EINVAL; - goto err_proto_put; - } + matching->l4proto->l4proto != l4num) + return -EINVAL; - ret = ctnl_timeout_parse_policy(&matching->data, - l4proto, net, - cda[CTA_TIMEOUT_DATA]); - return ret; + return ctnl_timeout_parse_policy(&matching->data, + matching->l4proto, net, + cda[CTA_TIMEOUT_DATA]); } - ret = -EBUSY; + + return -EBUSY; + } + + l4proto = nf_ct_l4proto_find_get(l3num, l4num); + + /* This protocol is not supportted, skip. */ + if (l4proto->l4proto != l4num) { + ret = -EOPNOTSUPP; goto err_proto_put; } From 533e33009897c7dd1b0424c0d4b3331b222d5681 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 21:58:18 +0800 Subject: [PATCH 020/271] netfilter: cttimeout: unlink timeout objs in the unconfirmed ct lists KASAN reported this bug: BUG: KASAN: use-after-free in icmp_packet+0x25/0x50 [nf_conntrack_ipv4] at addr ffff880002db08c8 Read of size 4 by task lt-nf-queue/19041 Call Trace: [] dump_stack+0x63/0x88 [] kasan_report_error+0x528/0x560 [] kasan_report+0x58/0x60 [] ? icmp_packet+0x25/0x50 [nf_conntrack_ipv4] [] __asan_load4+0x61/0x80 [] icmp_packet+0x25/0x50 [nf_conntrack_ipv4] [] nf_conntrack_in+0x550/0x980 [nf_conntrack] [] ? __nf_conntrack_confirm+0xb10/0xb10 [nf_conntrack] [ ... ] The main reason is that we missed to unlink the timeout objects in the unconfirmed ct lists, so we will access the timeout objects that have already been freed. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nfnetlink_cttimeout.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index 6844c7af0b8f..139e0867e56e 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -302,7 +302,16 @@ static void ctnl_untimeout(struct net *net, struct ctnl_timeout *timeout) const struct hlist_nulls_node *nn; unsigned int last_hsize; spinlock_t *lock; - int i; + int i, cpu; + + for_each_possible_cpu(cpu) { + struct ct_pcpu *pcpu = per_cpu_ptr(net->ct.pcpu_lists, cpu); + + spin_lock_bh(&pcpu->lock); + hlist_nulls_for_each_entry(h, nn, &pcpu->unconfirmed, hnnode) + untimeout(h, timeout); + spin_unlock_bh(&pcpu->lock); + } local_bh_disable(); restart: From 960fa72f67f1be6891d63a5518860d1ae4e14b88 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 22:57:56 +0800 Subject: [PATCH 021/271] netfilter: nft_meta: improve the validity check of pkttype set expr "meta pkttype set" is only supported on prerouting chain with bridge family and ingress chain with netdev family. But the validate check is incomplete, and the user can add the nft rules on input chain with bridge family, for example: # nft add table bridge filter # nft add chain bridge filter input {type filter hook input \ priority 0 \;} # nft add chain bridge filter test # nft add rule bridge filter test meta pkttype set unicast # nft add rule bridge filter input jump test This patch fixes the problem. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nft_meta.h | 4 ++++ net/bridge/netfilter/nft_meta_bridge.c | 1 + net/netfilter/nft_meta.c | 17 +++++++++++++---- 3 files changed, 18 insertions(+), 4 deletions(-) diff --git a/include/net/netfilter/nft_meta.h b/include/net/netfilter/nft_meta.h index d27588c8dbd9..1139cde0fdc5 100644 --- a/include/net/netfilter/nft_meta.h +++ b/include/net/netfilter/nft_meta.h @@ -36,4 +36,8 @@ void nft_meta_set_eval(const struct nft_expr *expr, void nft_meta_set_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr); +int nft_meta_set_validate(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nft_data **data); + #endif diff --git a/net/bridge/netfilter/nft_meta_bridge.c b/net/bridge/netfilter/nft_meta_bridge.c index 4b901d9f2e7c..ad47a921b701 100644 --- a/net/bridge/netfilter/nft_meta_bridge.c +++ b/net/bridge/netfilter/nft_meta_bridge.c @@ -86,6 +86,7 @@ static const struct nft_expr_ops nft_meta_bridge_set_ops = { .init = nft_meta_set_init, .destroy = nft_meta_set_destroy, .dump = nft_meta_set_dump, + .validate = nft_meta_set_validate, }; static const struct nft_expr_ops * diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 2863f3493038..8a6bc7630912 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -291,10 +291,16 @@ int nft_meta_get_init(const struct nft_ctx *ctx, } EXPORT_SYMBOL_GPL(nft_meta_get_init); -static int nft_meta_set_init_pkttype(const struct nft_ctx *ctx) +int nft_meta_set_validate(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nft_data **data) { + struct nft_meta *priv = nft_expr_priv(expr); unsigned int hooks; + if (priv->key != NFT_META_PKTTYPE) + return 0; + switch (ctx->afi->family) { case NFPROTO_BRIDGE: hooks = 1 << NF_BR_PRE_ROUTING; @@ -308,6 +314,7 @@ static int nft_meta_set_init_pkttype(const struct nft_ctx *ctx) return nft_chain_validate_hooks(ctx->chain, hooks); } +EXPORT_SYMBOL_GPL(nft_meta_set_validate); int nft_meta_set_init(const struct nft_ctx *ctx, const struct nft_expr *expr, @@ -327,15 +334,16 @@ int nft_meta_set_init(const struct nft_ctx *ctx, len = sizeof(u8); break; case NFT_META_PKTTYPE: - err = nft_meta_set_init_pkttype(ctx); - if (err) - return err; len = sizeof(u8); break; default: return -EOPNOTSUPP; } + err = nft_meta_set_validate(ctx, expr, NULL); + if (err < 0) + return err; + priv->sreg = nft_parse_register(tb[NFTA_META_SREG]); err = nft_validate_register_load(priv->sreg, len); if (err < 0) @@ -407,6 +415,7 @@ static const struct nft_expr_ops nft_meta_set_ops = { .init = nft_meta_set_init, .destroy = nft_meta_set_destroy, .dump = nft_meta_set_dump, + .validate = nft_meta_set_validate, }; static const struct nft_expr_ops * From 4249fc1f023a2106170bbf715e2e1a0ebc2d5b1f Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 23 Aug 2016 10:20:31 +0200 Subject: [PATCH 022/271] netfilter: ebtables: put module reference when an incorrect extension is found commit bcf493428840 ("netfilter: ebtables: Fix extension lookup with identical name") added a second lookup in case the extension that was found during the first lookup matched another extension with the same name, but didn't release the reference on the incorrect module. Fixes: bcf493428840 ("netfilter: ebtables: Fix extension lookup with identical name") Signed-off-by: Sabrina Dubroca Acked-by: Phil Sutter Signed-off-by: Pablo Neira Ayuso --- net/bridge/netfilter/ebtables.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index cceac5bb658f..0833c251aef7 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -368,6 +368,8 @@ ebt_check_match(struct ebt_entry_match *m, struct xt_mtchk_param *par, match = xt_find_match(NFPROTO_BRIDGE, m->u.name, 0); if (IS_ERR(match) || match->family != NFPROTO_BRIDGE) { + if (!IS_ERR(match)) + module_put(match->me); request_module("ebt_%s", m->u.name); match = xt_find_match(NFPROTO_BRIDGE, m->u.name, 0); } From a7d4b8f2565ad0dfdff9a222d1d87990c73b36e8 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Tue, 16 Aug 2016 14:38:24 +0200 Subject: [PATCH 023/271] KVM: s390: don't use current->thread.fpu.* when accessing registers As the meaning of these variables and pointers seems to change more frequently, let's directly access our save area, instead of going via current->thread. Right now, this is broken for set/get_fpu. They simply overwrite the host registers, as the pointers to the current save area were turned into the static host save area. Cc: stable@vger.kernel.org # 4.7 Fixes: 3f6813b9a5e0 ("s390/fpu: allocate 'struct fpu' with the task_struct") Reported-by: Hao QingFeng Signed-off-by: David Hildenbrand Signed-off-by: Christian Borntraeger --- arch/s390/kvm/kvm-s390.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index f142215ed30d..607ec91966c7 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -2231,9 +2231,10 @@ int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) return -EINVAL; current->thread.fpu.fpc = fpu->fpc; if (MACHINE_HAS_VX) - convert_fp_to_vx(current->thread.fpu.vxrs, (freg_t *)fpu->fprs); + convert_fp_to_vx((__vector128 *) vcpu->run->s.regs.vrs, + (freg_t *) fpu->fprs); else - memcpy(current->thread.fpu.fprs, &fpu->fprs, sizeof(fpu->fprs)); + memcpy(vcpu->run->s.regs.fprs, &fpu->fprs, sizeof(fpu->fprs)); return 0; } @@ -2242,9 +2243,10 @@ int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) /* make sure we have the latest values */ save_fpu_regs(); if (MACHINE_HAS_VX) - convert_vx_to_fp((freg_t *)fpu->fprs, current->thread.fpu.vxrs); + convert_vx_to_fp((freg_t *) fpu->fprs, + (__vector128 *) vcpu->run->s.regs.vrs); else - memcpy(fpu->fprs, current->thread.fpu.fprs, sizeof(fpu->fprs)); + memcpy(fpu->fprs, vcpu->run->s.regs.fprs, sizeof(fpu->fprs)); fpu->fpc = current->thread.fpu.fpc; return 0; } From 29617e1cfc2aa869154f5ed2580b756ec2c3cb28 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Thu, 25 Aug 2016 10:56:48 -0400 Subject: [PATCH 024/271] MAINTAINERS: Add myself as reviewer for Samsung Exynos support I've been helping reviewing and testing Exynos SoC support patches for the last couple of years. But it would be easier for me if I'm cc'ed for patches, so I'm adding myself as reviewer for this entry. Signed-off-by: Javier Martinez Canillas Acked-by: Kukjin Kim Acked-by: Sylwester Nawrocki Signed-off-by: Krzysztof Kozlowski --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index c9cd8d34b753..8a8a48512f4b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1614,6 +1614,7 @@ N: rockchip ARM/SAMSUNG EXYNOS ARM ARCHITECTURES M: Kukjin Kim M: Krzysztof Kozlowski +R: Javier Martinez Canillas L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) L: linux-samsung-soc@vger.kernel.org (moderated for non-subscribers) S: Maintained From 6654674cb7b5953ac04fc9d7f5f511676ae97e29 Mon Sep 17 00:00:00 2001 From: Jorik Jonker Date: Sat, 27 Aug 2016 21:04:33 +0200 Subject: [PATCH 025/271] clk: sunxi-ng: Fix wrong reset register offsets The reset register offsets for UART*, I2C* and SCR were off by a few bytes. Signed-off-by: Jorik Jonker Signed-off-by: Maxime Ripard --- drivers/clk/sunxi-ng/ccu-sun8i-h3.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/clk/sunxi-ng/ccu-sun8i-h3.c b/drivers/clk/sunxi-ng/ccu-sun8i-h3.c index 9af359544110..267f99523fbe 100644 --- a/drivers/clk/sunxi-ng/ccu-sun8i-h3.c +++ b/drivers/clk/sunxi-ng/ccu-sun8i-h3.c @@ -783,14 +783,14 @@ static struct ccu_reset_map sun8i_h3_ccu_resets[] = { [RST_BUS_I2S1] = { 0x2d0, BIT(13) }, [RST_BUS_I2S2] = { 0x2d0, BIT(14) }, - [RST_BUS_I2C0] = { 0x2d4, BIT(0) }, - [RST_BUS_I2C1] = { 0x2d4, BIT(1) }, - [RST_BUS_I2C2] = { 0x2d4, BIT(2) }, - [RST_BUS_UART0] = { 0x2d4, BIT(16) }, - [RST_BUS_UART1] = { 0x2d4, BIT(17) }, - [RST_BUS_UART2] = { 0x2d4, BIT(18) }, - [RST_BUS_UART3] = { 0x2d4, BIT(19) }, - [RST_BUS_SCR] = { 0x2d4, BIT(20) }, + [RST_BUS_I2C0] = { 0x2d8, BIT(0) }, + [RST_BUS_I2C1] = { 0x2d8, BIT(1) }, + [RST_BUS_I2C2] = { 0x2d8, BIT(2) }, + [RST_BUS_UART0] = { 0x2d8, BIT(16) }, + [RST_BUS_UART1] = { 0x2d8, BIT(17) }, + [RST_BUS_UART2] = { 0x2d8, BIT(18) }, + [RST_BUS_UART3] = { 0x2d8, BIT(19) }, + [RST_BUS_SCR] = { 0x2d8, BIT(20) }, }; static const struct sunxi_ccu_desc sun8i_h3_ccu_desc = { From 8a07fed44b126f48020f122b9e6bf05d8c48f281 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Tue, 23 Aug 2016 10:25:58 +0100 Subject: [PATCH 026/271] drm/i915/dvo: Remove dangling call to drm_encoder_cleanup() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If we hit the error path, we have never called drm_encoder_init() and so have nothing to cleanup. Doing so hits a null dereference: [ 10.066261] BUG: unable to handle kernel NULL pointer dereference at 00000104 [ 10.066273] IP: [] mutex_lock+0xa/0x15 [ 10.066287] *pde = 00000000 [ 10.066295] Oops: 0002 [#1] [ 10.066302] Modules linked in: i915(+) video i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm iTCO_wdt iTCO_vendor_support ppdev evdev snd_intel8x0 snd_ac97_codec ac97_bus psmouse snd_pcm snd_timer snd pcspkr uhci_hcd ehci_pci soundcore sr_mod ehci_hcd serio_raw i2c_i801 usbcore i2c_smbus cdrom lpc_ich mfd_core rng_core e100 mii floppy parport_pc parport acpi_cpufreq button processor usb_common eeprom lm85 hwmon_vid autofs4 [ 10.066378] CPU: 0 PID: 132 Comm: systemd-udevd Not tainted 4.8.0-rc3-00013-gef0e1ea #34 [ 10.066389] Hardware name: MicroLink /D865GLC , BIOS BF86510A.86A.0077.P25.0508040031 08/04/2005 [ 10.066401] task: f62db800 task.stack: f5970000 [ 10.066409] EIP: 0060:[] EFLAGS: 00010286 CPU: 0 [ 10.066417] EIP is at mutex_lock+0xa/0x15 [ 10.066424] EAX: 00000104 EBX: 00000104 ECX: 00000000 EDX: 80000000 [ 10.066432] ESI: 00000000 EDI: 00000104 EBP: f5be8000 ESP: f5971b58 [ 10.066439] DS: 007b ES: 007b FS: 0000 GS: 00e0 SS: 0068 [ 10.066446] CR0: 80050033 CR2: 00000104 CR3: 35945000 CR4: 000006d0 [ 10.066453] Stack: [ 10.066459] f503d740 f824dddf 00000000 f61170c0 f61170c0 f82371ae f850f40e 00000001 [ 10.066476] f61170c0 f5971bcc f5be8000 f9c2d401 00000001 f8236fcc 00000001 00000000 [ 10.066491] f5144014 f5be8104 00000008 f9c5267c 00000007 f61170c0 f5144400 f9c4ff00 [ 10.066507] Call Trace: [ 10.066526] [] ? drm_modeset_lock_all+0x27/0xb3 [drm] [ 10.066545] [] ? drm_encoder_cleanup+0x1a/0x132 [drm] [ 10.066559] [] ? drm_atomic_helper_connector_reset+0x3f/0x5c [drm_kms_helper] [ 10.066644] [] ? intel_dvo_init+0x569/0x788 [i915] [ 10.066663] [] ? drm_encoder_init+0x43/0x20b [drm] [ 10.066734] [] ? intel_modeset_init+0x1436/0x17dd [i915] [ 10.066791] [] ? i915_driver_load+0x85a/0x15d3 [i915] [ 10.066846] [] ? i915_driver_open+0x5/0x5 [i915] [ 10.066857] [] ? firmware_map_add_entry.part.2+0xc/0xc [ 10.066868] [] ? pci_device_probe+0x8e/0x11c [ 10.066878] [] ? driver_probe_device+0x1db/0x62e [ 10.066888] [] ? kernfs_new_node+0x29/0x9c [ 10.066897] [] ? pci_match_device+0xd9/0x161 [ 10.066905] [] ? kernfs_create_dir_ns+0x42/0x88 [ 10.066914] [] ? __driver_attach+0xe6/0x11b [ 10.066924] [] ? kobject_add_internal+0x1bb/0x44f [ 10.066933] [] ? driver_probe_device+0x62e/0x62e [ 10.066941] [] ? bus_for_each_dev+0x46/0x7f [ 10.066950] [] ? driver_attach+0x1a/0x34 [ 10.066958] [] ? driver_probe_device+0x62e/0x62e [ 10.066966] [] ? bus_add_driver+0x217/0x32a [ 10.066975] [] ? 0xf8403000 [ 10.066982] [] ? driver_register+0x5f/0x108 [ 10.066991] [] ? do_one_initcall+0x49/0x1f6 [ 10.067000] [] ? pick_next_task_fair+0x14b/0x2a3 [ 10.067008] [] ? __schedule+0x15c/0x4fe [ 10.067016] [] ? preempt_schedule_common+0x19/0x3c [ 10.067027] [] ? do_init_module+0x17/0x230 [ 10.067035] [] ? _cond_resched+0x12/0x1a [ 10.067044] [] ? kmem_cache_alloc+0x8f/0x11f [ 10.067052] [] ? do_init_module+0x17/0x230 [ 10.067060] [] ? kfree+0x137/0x203 [ 10.067068] [] ? do_init_module+0x76/0x230 [ 10.067078] [] ? load_module+0x2a39/0x333f [ 10.067087] [] ? SyS_finit_module+0x96/0xd5 [ 10.067096] [] ? vm_mmap_pgoff+0x79/0xa0 [ 10.067105] [] ? do_fast_syscall_32+0xb5/0x1b0 [ 10.067114] [] ? sysenter_past_esp+0x47/0x75 [ 10.067121] Code: c8 f7 76 c1 e8 8e cc d2 ff e9 45 fe ff ff 66 90 66 90 66 90 66 90 90 ff 00 7f 05 e8 4e 0c 00 00 c3 53 89 c3 e8 75 ec ff ff 89 d8 08 79 05 e8 fa 0a 00 00 5b c3 53 89 c3 85 c0 74 1b 8b 03 83 [ 10.067180] EIP: [] mutex_lock+0xa/0x15 SS:ESP 0068:f5971b58 [ 10.067190] CR2: 0000000000000104 [ 10.067222] ---[ end trace 049f1f09da45a856 ]--- Reported-by: Meelis Roos Fixes: 580d8ed522e0 ("drm/i915: Give encoders useful names") Reviewed-by: David Weinehall Signed-off-by: Chris Wilson Cc: Ville Syrjälä Cc: drm-intel-fixes@lists.freedesktop.org Signed-off-by: Jani Nikula Link: http://patchwork.freedesktop.org/patch/msgid/20160823092558.14931-1-chris@chris-wilson.co.uk (cherry picked from commit 8f76aa0ebe0b7787afe768d9df80031e832d2520) --- drivers/gpu/drm/i915/intel_dvo.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c index 47bdf9dad0d3..b9e5a63a7c9e 100644 --- a/drivers/gpu/drm/i915/intel_dvo.c +++ b/drivers/gpu/drm/i915/intel_dvo.c @@ -554,7 +554,6 @@ void intel_dvo_init(struct drm_device *dev) return; } - drm_encoder_cleanup(&intel_encoder->base); kfree(intel_dvo); kfree(intel_connector); } From 360647959323ec066cc55d82133df73e30530824 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 11 Jul 2016 09:25:45 +0300 Subject: [PATCH 027/271] iwlwifi: mvm: consider P2p device type for firmware dump triggers When the user really wanted a dump on P2P Client, he coudln't get it because we checked vif->type but didn't take vif->p2p into account. Fix that. Signed-off-by: Emmanuel Grumbach Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.h index f7dff7612c9c..e9f1be9da7d4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.h @@ -105,7 +105,8 @@ iwl_fw_dbg_trigger_vif_match(struct iwl_fw_dbg_trigger_tlv *trig, { u32 trig_vif = le32_to_cpu(trig->vif_type); - return trig_vif == IWL_FW_DBG_CONF_VIF_ANY || vif->type == trig_vif; + return trig_vif == IWL_FW_DBG_CONF_VIF_ANY || + ieee80211_vif_type_p2p(vif) == trig_vif; } static inline bool From bdc98b13716eafc454457e068d6c8aaffaaa71e4 Mon Sep 17 00:00:00 2001 From: Sharon Dvir Date: Mon, 25 Jul 2016 16:11:05 +0300 Subject: [PATCH 028/271] iwlwifi: mvm: check if vif is NULL before using it wdev_to_ieee80211_vif() might return NULL. Check that vif != NULL before dereferencing it. Signed-off-by: Sharon Dvir Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index b4fc86d5d7ef..6a615bb73042 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -467,6 +467,8 @@ struct iwl_mvm_vif { static inline struct iwl_mvm_vif * iwl_mvm_vif_from_mac80211(struct ieee80211_vif *vif) { + if (!vif) + return NULL; return (void *)vif->drv_priv; } From ff6e58e648ed5f3cc43891767811d5c3c88bbd41 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Wed, 3 Aug 2016 22:06:43 +0300 Subject: [PATCH 029/271] iwlwifi: mvm: don't use ret when not initialised fw-dbg code return ret but that variable was either 0 or not initialised. Return 0 always. Signed-off-by: Emmanuel Grumbach Fixes: 6a95126763fb ("iwlwifi: mvm: send dbg config hcmds to fw if set in tlv") Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c index 1abcabb9b6cd..46b52bf705fb 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c @@ -960,5 +960,6 @@ int iwl_mvm_start_fw_dbg_conf(struct iwl_mvm *mvm, u8 conf_id) } mvm->fw_dbg_conf = conf_id; - return ret; + + return 0; } From a904a08b5fee5317ff0f7b8212aa5d0776795a52 Mon Sep 17 00:00:00 2001 From: "Peer, Ilan" Date: Wed, 10 Aug 2016 13:48:35 +0300 Subject: [PATCH 030/271] iwlwifi: mvm: Advertise support for AP channel width change The iwlmvm driver supports channel width change in AP mode. Add the proper flag. Signed-off-by: Ilan Peer Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index 6d6064534d59..5dd77e336617 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -624,6 +624,7 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm) hw->wiphy->features |= NL80211_FEATURE_P2P_GO_CTWIN | NL80211_FEATURE_LOW_PRIORITY_SCAN | NL80211_FEATURE_P2P_GO_OPPPS | + NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE | NL80211_FEATURE_DYNAMIC_SMPS | NL80211_FEATURE_STATIC_SMPS | NL80211_FEATURE_SUPPORTS_WMM_ADMISSION; From 554d072e7bc3e56de5893c8181110a547b2062c9 Mon Sep 17 00:00:00 2001 From: Arik Nemtsov Date: Mon, 29 Aug 2016 12:37:35 +0300 Subject: [PATCH 031/271] mac80211: TDLS: don't require beaconing for AP BW Stop downgrading TDLS chandef when reaching the AP BW. The AP provides the necessary regulatory protection in this case. This fixes https://bugzilla.kernel.org/show_bug.cgi?id=153961, which reported an infinite loop here. Reported-by: Kamil Toman Signed-off-by: Arik Nemtsov Signed-off-by: Johannes Berg --- net/mac80211/tdls.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/net/mac80211/tdls.c b/net/mac80211/tdls.c index b5d28f14b9cf..afca7d103684 100644 --- a/net/mac80211/tdls.c +++ b/net/mac80211/tdls.c @@ -333,10 +333,11 @@ ieee80211_tdls_chandef_vht_upgrade(struct ieee80211_sub_if_data *sdata, if (!uc.center_freq1) return; - /* proceed to downgrade the chandef until usable or the same */ + /* proceed to downgrade the chandef until usable or the same as AP BW */ while (uc.width > max_width || - !cfg80211_reg_can_beacon_relax(sdata->local->hw.wiphy, &uc, - sdata->wdev.iftype)) + (uc.width > sta->tdls_chandef.width && + !cfg80211_reg_can_beacon_relax(sdata->local->hw.wiphy, &uc, + sdata->wdev.iftype))) ieee80211_chandef_downgrade(&uc); if (!cfg80211_chandef_identical(&uc, &sta->tdls_chandef)) { From 61aaa0e8c1c15d9e045f0577f046be50f2f571ab Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Linus=20L=C3=BCssing?= Date: Fri, 19 Aug 2016 22:02:48 +0200 Subject: [PATCH 032/271] cfg80211: Add stub for cfg80211_get_station() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This allows modules using this function (currently: batman-adv) to compile even if cfg80211 is not built at all, thus relaxing dependencies. Signed-off-by: Linus Lüssing Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 9c23f4d33e06..beb7610d64e9 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1102,6 +1102,7 @@ struct station_info { struct cfg80211_tid_stats pertid[IEEE80211_NUM_TIDS + 1]; }; +#if IS_ENABLED(CONFIG_CFG80211) /** * cfg80211_get_station - retrieve information about a given station * @dev: the device where the station is supposed to be connected to @@ -1114,6 +1115,14 @@ struct station_info { */ int cfg80211_get_station(struct net_device *dev, const u8 *mac_addr, struct station_info *sinfo); +#else +static inline int cfg80211_get_station(struct net_device *dev, + const u8 *mac_addr, + struct station_info *sinfo) +{ + return -ENOENT; +} +#endif /** * enum monitor_flags - monitor flags From b030485220caf862c71db6fb8b8ad016ce7f7565 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Sun, 21 Aug 2016 03:27:45 -0400 Subject: [PATCH 033/271] ARM: EXYNOS: Clear OF_POPULATED flag from PMU node in IRQ init callback The Exynos PMU node is an interrupt, clock and PMU (Power Management Unit) controller, and these functionalities are supported by different drivers that matches the same compatible strings. Since commit 15cc2ed6dcf9 ("of/irq: Mark initialised interrupt controllers as populated") the OF core flags interrupt controllers registered with the IRQCHIP_DECLARE() macro as OF_POPULATED, so platform devices with the same compatible string as the interrupt controller will not be registered. This prevents the PMU platform device to be registered so the Exynos PMU driver is never probed. This breaks (among other things) Suspend-to-RAM. Fix this by clearing the OF_POPULATED flag in the PMU IRQ init callback, to allow the Exynos PMU platform driver to be probed. The patch is based on Philipp Zabel's "ARM: imx6: mark GPC node as not populated after irq init to probe pm domain driver". Fixes: 15cc2ed6dcf9 ("of/irq: Mark initialised interrupt controllers as populated") Signed-off-by: Javier Martinez Canillas Signed-off-by: Krzysztof Kozlowski --- arch/arm/mach-exynos/suspend.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/arm/mach-exynos/suspend.c b/arch/arm/mach-exynos/suspend.c index 3750575c73c5..06332f626565 100644 --- a/arch/arm/mach-exynos/suspend.c +++ b/arch/arm/mach-exynos/suspend.c @@ -255,6 +255,12 @@ static int __init exynos_pmu_irq_init(struct device_node *node, return -ENOMEM; } + /* + * Clear the OF_POPULATED flag set in of_irq_init so that + * later the Exynos PMU platform device won't be skipped. + */ + of_node_clear_flag(node, OF_POPULATED); + return 0; } From c73c2484901139c28383b58eabcbf4d613e91518 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sun, 28 Aug 2016 16:59:52 +0800 Subject: [PATCH 034/271] netfilter: nf_tables_netdev: remove redundant ip_hdr assignment We have already use skb_header_pointer to get the ip header pointer, so there's no need to use ip_hdr again. Moreover, in NETDEV INGRESS hook, ip header maybe not linear, so use ip_hdr is not appropriate, remove it. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_netdev.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/netfilter/nf_tables_netdev.c b/net/netfilter/nf_tables_netdev.c index 5eefe4a355c6..75d696f11045 100644 --- a/net/netfilter/nf_tables_netdev.c +++ b/net/netfilter/nf_tables_netdev.c @@ -30,7 +30,6 @@ nft_netdev_set_pktinfo_ipv4(struct nft_pktinfo *pkt, if (!iph) return; - iph = ip_hdr(skb); if (iph->ihl < 5 || iph->version != 4) return; From 313a61d30761217ce4383018de1cc0d5d503a376 Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Tue, 30 Aug 2016 13:57:38 -0700 Subject: [PATCH 035/271] drm/vc4: Allow some more signals to be packed with uniform resets. The intent was to make sure people don't sneak in a small immediate or something to change the interpretation of the uniform update args, but these signals are just fine. Fixes a validation failure in the current X server on some Render operation. Signed-off-by: Eric Anholt --- drivers/gpu/drm/vc4/vc4_validate_shaders.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vc4/vc4_validate_shaders.c b/drivers/gpu/drm/vc4/vc4_validate_shaders.c index 46527e989ce3..2543cf5b8b51 100644 --- a/drivers/gpu/drm/vc4/vc4_validate_shaders.c +++ b/drivers/gpu/drm/vc4/vc4_validate_shaders.c @@ -309,8 +309,14 @@ validate_uniform_address_write(struct vc4_validated_shader_info *validated_shade * of uniforms on each side. However, this scheme is easy to * validate so it's all we allow for now. */ - - if (QPU_GET_FIELD(inst, QPU_SIG) != QPU_SIG_NONE) { + switch (QPU_GET_FIELD(inst, QPU_SIG)) { + case QPU_SIG_NONE: + case QPU_SIG_SCOREBOARD_UNLOCK: + case QPU_SIG_COLOR_LOAD: + case QPU_SIG_LOAD_TMU0: + case QPU_SIG_LOAD_TMU1: + break; + default: DRM_ERROR("uniforms address change must be " "normal math\n"); return false; From 237e15dfd5d651868726111c3a9d828bec700490 Mon Sep 17 00:00:00 2001 From: Ashok Raj Nagarajan Date: Fri, 19 Aug 2016 13:37:37 +0300 Subject: [PATCH 036/271] ath10k: fix get rx_status from htt context On handling amsdu on rx path, get the rx_status from htt context. Without this fix, we are seeing warnings when running DBDC traffic like this. WARNING: CPU: 0 PID: 0 at net/mac80211/rx.c:4105 ieee80211_rx_napi+0x88/0x7d8 [mac80211]() [ 1715.878248] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G W 3.18.21 #1 [ 1715.878273] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [ 1715.878293] [] (show_stack) from [] (dump_stack+0x70/0xbc) [ 1715.878315] [] (dump_stack) from [] (warn_slowpath_common+0x64/0x88) [ 1715.878339] [] (warn_slowpath_common) from [] (warn_slowpath_null+0x18/0x20) [ 1715.878395] [] (warn_slowpath_null) from [] (ieee80211_rx_napi+0x88/0x7d8 [mac80211]) [ 1715.878474] [] (ieee80211_rx_napi [mac80211]) from [] (ath10k_htt_t2h_msg_handler+0xb48/0xbfc [ath10k_core]) [ 1715.878535] [] (ath10k_htt_t2h_msg_handler [ath10k_core]) from [] (ath10k_htt_t2h_msg_handler+0xbf8/0xbfc [ath10k_core]) [ 1715.878597] [] (ath10k_htt_t2h_msg_handler [ath10k_core]) from [] (ath10k_htt_txrx_compl_task+0xa54/0x1170 [ath10k_core]) [ 1715.878639] [] (ath10k_htt_txrx_compl_task [ath10k_core]) from [] (tasklet_action+0xb4/0x130) [ 1715.878659] [] (tasklet_action) from [] (__do_softirq+0xe0/0x210) [ 1715.878678] [] (__do_softirq) from [] (irq_exit+0x84/0xe0) [ 1715.878700] [] (irq_exit) from [] (__handle_domain_irq+0x98/0xd0) [ 1715.878722] [] (__handle_domain_irq) from [] (gic_handle_irq+0x38/0x5c) [ 1715.878741] [] (gic_handle_irq) from [] (__irq_svc+0x40/0x74) [ 1715.878753] Exception stack(0xc05f9f50 to 0xc05f9f98) [ 1715.878767] 9f40: ffffffed 00000000 00399e1e c000a220 [ 1715.878786] 9f60: 00000000 c05f6780 c05f8000 00000000 c05f5db8 ffffffed c05f8000 c04d1980 [ 1715.878802] 9f80: 00000000 c05f9f98 c0018110 c0018114 60000013 ffffffff [ 1715.878822] [] (__irq_svc) from [] (arch_cpu_idle+0x2c/0x50) [ 1715.878844] [] (arch_cpu_idle) from [] (cpu_startup_entry+0x108/0x234) [ 1715.878866] [] (cpu_startup_entry) from [] (start_kernel+0x33c/0x3b8) [ 1715.878879] ---[ end trace 6d5e1cc0fef8ed6a ]--- [ 1715.878899] ------------[ cut here ]------------ Fixes: 18235664e7f9 ("ath10k: cleanup amsdu processing for rx indication") Signed-off-by: Ashok Raj Nagarajan Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath10k/htt_rx.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c index 78db5d679f19..24c8d65bcf34 100644 --- a/drivers/net/wireless/ath/ath10k/htt_rx.c +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c @@ -1525,7 +1525,7 @@ static void ath10k_htt_rx_h_filter(struct ath10k *ar, static int ath10k_htt_rx_handle_amsdu(struct ath10k_htt *htt) { struct ath10k *ar = htt->ar; - static struct ieee80211_rx_status rx_status; + struct ieee80211_rx_status *rx_status = &htt->rx_status; struct sk_buff_head amsdu; int ret; @@ -1549,11 +1549,11 @@ static int ath10k_htt_rx_handle_amsdu(struct ath10k_htt *htt) return ret; } - ath10k_htt_rx_h_ppdu(ar, &amsdu, &rx_status, 0xffff); + ath10k_htt_rx_h_ppdu(ar, &amsdu, rx_status, 0xffff); ath10k_htt_rx_h_unchain(ar, &amsdu, ret > 0); - ath10k_htt_rx_h_filter(ar, &amsdu, &rx_status); - ath10k_htt_rx_h_mpdu(ar, &amsdu, &rx_status); - ath10k_htt_rx_h_deliver(ar, &amsdu, &rx_status); + ath10k_htt_rx_h_filter(ar, &amsdu, rx_status); + ath10k_htt_rx_h_mpdu(ar, &amsdu, rx_status); + ath10k_htt_rx_h_deliver(ar, &amsdu, rx_status); return 0; } From 4e80ffab0e722911842b59299dbf8033709673e5 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Fri, 19 Aug 2016 13:37:37 +0300 Subject: [PATCH 037/271] ath10k: Remove driver log suggesting QCA9887 support is experimental Support for QCA9887 is no longer experimental and if there are any issues we need to address them Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath10k/pci.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index 9a22c478dd1b..07933c51a850 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -3162,7 +3162,6 @@ static int ath10k_pci_probe(struct pci_dev *pdev, pci_hard_reset = ath10k_pci_qca988x_chip_reset; break; case QCA9887_1_0_DEVICE_ID: - dev_warn(&pdev->dev, "QCA9887 support is still experimental, there are likely bugs. You have been warned.\n"); hw_rev = ATH10K_HW_QCA9887; pci_ps = false; pci_soft_reset = ath10k_pci_warm_reset; From 8e018c21da3febb558586b48c8db0d6d66cb6593 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 25 Aug 2016 10:09:39 -0700 Subject: [PATCH 038/271] raid5-cache: fix a deadlock in superblock write There is a potential deadlock in superblock write. Discard could zero data, so before discard we must make sure superblock is updated to new log tail. Updating superblock (either directly call md_update_sb() or depend on md thread) must hold reconfig mutex. On the other hand, raid5_quiesce is called with reconfig_mutex hold. The first step of raid5_quiesce() is waitting for all IO finish, hence waitting for reclaim thread, while reclaim thread is calling this function and waitting for reconfig mutex. So there is a deadlock. We workaround this issue with a trylock. The downside of the solution is we could miss discard if we can't take reconfig mutex. But this should happen rarely (mainly in raid array stop), so miss discard shouldn't be a big problem. Cc: NeilBrown Signed-off-by: Shaohua Li --- drivers/md/raid5-cache.c | 46 +++++++++++++--------------------------- 1 file changed, 15 insertions(+), 31 deletions(-) diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index 51f76ddbe265..1b1ab4a1d132 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -96,7 +96,6 @@ struct r5l_log { spinlock_t no_space_stripes_lock; bool need_cache_flush; - bool in_teardown; }; /* @@ -704,31 +703,22 @@ static void r5l_write_super_and_discard_space(struct r5l_log *log, mddev = log->rdev->mddev; /* - * This is to avoid a deadlock. r5l_quiesce holds reconfig_mutex and - * wait for this thread to finish. This thread waits for - * MD_CHANGE_PENDING clear, which is supposed to be done in - * md_check_recovery(). md_check_recovery() tries to get - * reconfig_mutex. Since r5l_quiesce already holds the mutex, - * md_check_recovery() fails, so the PENDING never get cleared. The - * in_teardown check workaround this issue. + * Discard could zero data, so before discard we must make sure + * superblock is updated to new log tail. Updating superblock (either + * directly call md_update_sb() or depend on md thread) must hold + * reconfig mutex. On the other hand, raid5_quiesce is called with + * reconfig_mutex hold. The first step of raid5_quiesce() is waitting + * for all IO finish, hence waitting for reclaim thread, while reclaim + * thread is calling this function and waitting for reconfig mutex. So + * there is a deadlock. We workaround this issue with a trylock. + * FIXME: we could miss discard if we can't take reconfig mutex */ - if (!log->in_teardown) { - set_mask_bits(&mddev->flags, 0, - BIT(MD_CHANGE_DEVS) | BIT(MD_CHANGE_PENDING)); - md_wakeup_thread(mddev->thread); - wait_event(mddev->sb_wait, - !test_bit(MD_CHANGE_PENDING, &mddev->flags) || - log->in_teardown); - /* - * r5l_quiesce could run after in_teardown check and hold - * mutex first. Superblock might get updated twice. - */ - if (log->in_teardown) - md_update_sb(mddev, 1); - } else { - WARN_ON(!mddev_is_locked(mddev)); - md_update_sb(mddev, 1); - } + set_mask_bits(&mddev->flags, 0, + BIT(MD_CHANGE_DEVS) | BIT(MD_CHANGE_PENDING)); + if (!mddev_trylock(mddev)) + return; + md_update_sb(mddev, 1); + mddev_unlock(mddev); /* discard IO error really doesn't matter, ignore it */ if (log->last_checkpoint < end) { @@ -827,7 +817,6 @@ void r5l_quiesce(struct r5l_log *log, int state) if (!log || state == 2) return; if (state == 0) { - log->in_teardown = 0; /* * This is a special case for hotadd. In suspend, the array has * no journal. In resume, journal is initialized as well as the @@ -838,11 +827,6 @@ void r5l_quiesce(struct r5l_log *log, int state) log->reclaim_thread = md_register_thread(r5l_reclaim_thread, log->rdev->mddev, "reclaim"); } else if (state == 1) { - /* - * at this point all stripes are finished, so io_unit is at - * least in STRIPE_END state - */ - log->in_teardown = 1; /* make sure r5l_write_super_and_discard_space exits */ mddev = log->rdev->mddev; wake_up(&mddev->sb_wait); From ad5b0f7685dbfc4730987cd16af3c5ebe8133f10 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 30 Aug 2016 10:29:33 -0700 Subject: [PATCH 039/271] raid5: guarantee enough stripes to avoid reshape hang If there aren't enough stripes, reshape will hang. We have a check for this in new reshape, but miss it for reshape resume, hence we could see hang in reshape resume. This patch forces enough stripes existed if reshape resumes. Reviewed-by: NeilBrown Signed-off-by: Shaohua Li --- drivers/md/raid5.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index da583bb43c84..b95c54c0e596 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -6639,6 +6639,16 @@ static struct r5conf *setup_conf(struct mddev *mddev) } conf->min_nr_stripes = NR_STRIPES; + if (mddev->reshape_position != MaxSector) { + int stripes = max_t(int, + ((mddev->chunk_sectors << 9) / STRIPE_SIZE) * 4, + ((mddev->new_chunk_sectors << 9) / STRIPE_SIZE) * 4); + conf->min_nr_stripes = max(NR_STRIPES, stripes); + if (conf->min_nr_stripes != NR_STRIPES) + printk(KERN_INFO + "md/raid:%s: force stripe size %d for reshape\n", + mdname(mddev), conf->min_nr_stripes); + } memory = conf->min_nr_stripes * (sizeof(struct stripe_head) + max_disks * ((sizeof(struct bio) + PAGE_SIZE))) / 1024; atomic_set(&conf->empty_inactive_list_nr, NR_STRIPE_HASH_LOCKS); From 9264251ee2a55bce8fb93826b3f581fb9eb7e2c2 Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Wed, 31 Aug 2016 14:16:44 +0200 Subject: [PATCH 040/271] bridge: re-introduce 'fix parsing of MLDv2 reports' commit bc8c20acaea1 ("bridge: multicast: treat igmpv3 report with INCLUDE and no sources as a leave") seems to have accidentally reverted commit 47cc84ce0c2f ("bridge: fix parsing of MLDv2 reports"). This commit brings back a change to br_ip6_multicast_mld2_report() where parsing of MLDv2 reports stops when the first group is successfully added to the MDB cache. Fixes: bc8c20acaea1 ("bridge: multicast: treat igmpv3 report with INCLUDE and no sources as a leave") Signed-off-by: Davide Caratti Acked-by: Nikolay Aleksandrov Acked-by: Thadeu Lima de Souza Cascardo Signed-off-by: David S. Miller --- net/bridge/br_multicast.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index a5423a1eec05..c5fea9393946 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -1138,7 +1138,7 @@ static int br_ip6_multicast_mld2_report(struct net_bridge *br, } else { err = br_ip6_multicast_add_group(br, port, &grec->grec_mca, vid); - if (!err) + if (err) break; } } From c6f1dc4d9ce275c1bd9d90b5630f36ca6abccc97 Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:27 +0800 Subject: [PATCH 041/271] net: ethernet: mediatek: fix fails from TX housekeeping due to incorrect port setup which net device the SKB is complete for depends on the forward port on txd4 on the corresponding TX descriptor, but the information isn't set up well in case of SKB fragments that would lead to watchdog timeout from the upper layer, so fix it up. Signed-off-by: Sean Wang Acked-by: John Crispin Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index f1609542adf1..7fc2ff0c5bf0 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -588,14 +588,15 @@ static int mtk_tx_map(struct sk_buff *skb, struct net_device *dev, dma_addr_t mapped_addr; unsigned int nr_frags; int i, n_desc = 1; - u32 txd4 = 0; + u32 txd4 = 0, fport; itxd = ring->next_free; if (itxd == ring->last_free) return -ENOMEM; /* set the forward port */ - txd4 |= (mac->id + 1) << TX_DMA_FPORT_SHIFT; + fport = (mac->id + 1) << TX_DMA_FPORT_SHIFT; + txd4 |= fport; tx_buf = mtk_desc_to_tx_buf(ring, itxd); memset(tx_buf, 0, sizeof(*tx_buf)); @@ -653,7 +654,7 @@ static int mtk_tx_map(struct sk_buff *skb, struct net_device *dev, WRITE_ONCE(txd->txd3, (TX_DMA_SWC | TX_DMA_PLEN0(frag_map_size) | last_frag * TX_DMA_LS0)); - WRITE_ONCE(txd->txd4, 0); + WRITE_ONCE(txd->txd4, fport); tx_buf->skb = (struct sk_buff *)MTK_DMA_DUMMY_DESC; tx_buf = mtk_desc_to_tx_buf(ring, txd); From 549e5495467ac2a76489f1304e2b4a7ad85e1644 Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:28 +0800 Subject: [PATCH 042/271] net: ethernet: mediatek: fix incorrect return value of devm_clk_get with EPROBE_DEFER 1) If the return value of devm_clk_get is EPROBE_DEFER, we should defer probing the driver. The change is verified and works based on 4.8-rc1 staying with the latest clk-next code for MT7623. 2) Changing with the usage of loops to work out if all clocks required are fine Signed-off-by: Sean Wang Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 39 ++++++++++++--------- drivers/net/ethernet/mediatek/mtk_eth_soc.h | 22 +++++++----- 2 files changed, 36 insertions(+), 25 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 7fc2ff0c5bf0..a5dcf5741495 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -50,6 +50,10 @@ static const struct mtk_ethtool_stats { MTK_ETHTOOL_STAT(rx_flow_control_packets), }; +static const char * const mtk_clks_source_name[] = { + "ethif", "esw", "gp1", "gp2" +}; + void mtk_w32(struct mtk_eth *eth, u32 val, unsigned reg) { __raw_writel(val, eth->base + reg); @@ -1814,6 +1818,7 @@ static int mtk_probe(struct platform_device *pdev) if (!eth) return -ENOMEM; + eth->dev = &pdev->dev; eth->base = devm_ioremap_resource(&pdev->dev, res); if (IS_ERR(eth->base)) return PTR_ERR(eth->base); @@ -1848,21 +1853,21 @@ static int mtk_probe(struct platform_device *pdev) return -ENXIO; } } + for (i = 0; i < ARRAY_SIZE(eth->clks); i++) { + eth->clks[i] = devm_clk_get(eth->dev, + mtk_clks_source_name[i]); + if (IS_ERR(eth->clks[i])) { + if (PTR_ERR(eth->clks[i]) == -EPROBE_DEFER) + return -EPROBE_DEFER; + return -ENODEV; + } + } - eth->clk_ethif = devm_clk_get(&pdev->dev, "ethif"); - eth->clk_esw = devm_clk_get(&pdev->dev, "esw"); - eth->clk_gp1 = devm_clk_get(&pdev->dev, "gp1"); - eth->clk_gp2 = devm_clk_get(&pdev->dev, "gp2"); - if (IS_ERR(eth->clk_esw) || IS_ERR(eth->clk_gp1) || - IS_ERR(eth->clk_gp2) || IS_ERR(eth->clk_ethif)) - return -ENODEV; + clk_prepare_enable(eth->clks[MTK_CLK_ETHIF]); + clk_prepare_enable(eth->clks[MTK_CLK_ESW]); + clk_prepare_enable(eth->clks[MTK_CLK_GP1]); + clk_prepare_enable(eth->clks[MTK_CLK_GP2]); - clk_prepare_enable(eth->clk_ethif); - clk_prepare_enable(eth->clk_esw); - clk_prepare_enable(eth->clk_gp1); - clk_prepare_enable(eth->clk_gp2); - - eth->dev = &pdev->dev; eth->msg_enable = netif_msg_init(mtk_msg_level, MTK_DEFAULT_MSG_ENABLE); INIT_WORK(ð->pending_work, mtk_pending_work); @@ -1905,10 +1910,10 @@ static int mtk_remove(struct platform_device *pdev) { struct mtk_eth *eth = platform_get_drvdata(pdev); - clk_disable_unprepare(eth->clk_ethif); - clk_disable_unprepare(eth->clk_esw); - clk_disable_unprepare(eth->clk_gp1); - clk_disable_unprepare(eth->clk_gp2); + clk_disable_unprepare(eth->clks[MTK_CLK_ETHIF]); + clk_disable_unprepare(eth->clks[MTK_CLK_ESW]); + clk_disable_unprepare(eth->clks[MTK_CLK_GP1]); + clk_disable_unprepare(eth->clks[MTK_CLK_GP2]); netif_napi_del(ð->tx_napi); netif_napi_del(ð->rx_napi); diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h index f82e3acb947b..6e1ade7a25c5 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h @@ -290,6 +290,17 @@ enum mtk_tx_flags { MTK_TX_FLAGS_PAGE0 = 0x02, }; +/* This enum allows us to identify how the clock is defined on the array of the + * clock in the order + */ +enum mtk_clks_map { + MTK_CLK_ETHIF, + MTK_CLK_ESW, + MTK_CLK_GP1, + MTK_CLK_GP2, + MTK_CLK_MAX +}; + /* struct mtk_tx_buf - This struct holds the pointers to the memory pointed at * by the TX descriptor s * @skb: The SKB pointer of the packet being sent @@ -370,10 +381,7 @@ struct mtk_rx_ring { * @scratch_ring: Newer SoCs need memory for a second HW managed TX ring * @phy_scratch_ring: physical address of scratch_ring * @scratch_head: The scratch memory that scratch_ring points to. - * @clk_ethif: The ethif clock - * @clk_esw: The switch clock - * @clk_gp1: The gmac1 clock - * @clk_gp2: The gmac2 clock + * @clks: clock array for all clocks required * @mii_bus: If there is a bus we need to create an instance for it * @pending_work: The workqueue used to reset the dma ring */ @@ -400,10 +408,8 @@ struct mtk_eth { struct mtk_tx_dma *scratch_ring; dma_addr_t phy_scratch_ring; void *scratch_head; - struct clk *clk_ethif; - struct clk *clk_esw; - struct clk *clk_gp1; - struct clk *clk_gp2; + struct clk *clks[MTK_CLK_MAX]; + struct mii_bus *mii_bus; struct work_struct pending_work; }; From 1b43079906367d4d6e1ad00f04fff525a32e26c4 Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:29 +0800 Subject: [PATCH 043/271] net: ethernet: mediatek: fix API usage with skb_free_frag use skb_free_frag() instead of legacy put_page() Signed-off-by: Sean Wang Acked-by: John Crispin Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index a5dcf5741495..c9e25a753524 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -870,7 +870,7 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget, /* receive data */ skb = build_skb(data, ring->frag_size); if (unlikely(!skb)) { - put_page(virt_to_head_page(new_data)); + skb_free_frag(new_data); netdev->stats.rx_dropped++; goto release_desc; } From d3bd1ce4db8e843dce421e2f8f123e5251a9c7d3 Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:30 +0800 Subject: [PATCH 044/271] net: ethernet: mediatek: remove redundant free_irq for devm_request_irq allocated irq these irqs are not used for shared irq and disabled during ethernet stops. irq requested by devm_request_irq is safe to be freed automatically on driver detach. Signed-off-by: Sean Wang Acked-by: John Crispin Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index c9e25a753524..1ffde919174d 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -1513,8 +1513,6 @@ static void mtk_uninit(struct net_device *dev) phy_disconnect(mac->phy_dev); mtk_mdio_cleanup(eth); mtk_irq_disable(eth, ~0); - free_irq(eth->irq[1], dev); - free_irq(eth->irq[2], dev); } static int mtk_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) From 7c6b0d76fa02213393815e3b6d5e4a415bf3f0e2 Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:31 +0800 Subject: [PATCH 045/271] net: ethernet: mediatek: fix logic unbalance between probe and remove original mdio_cleanup is not in the symmetric place against where mdio_init is, so relocate mdio_cleanup to the right one. Signed-off-by: Sean Wang Acked-by: John Crispin Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 1ffde919174d..bf5b7e1d9450 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -1511,7 +1511,6 @@ static void mtk_uninit(struct net_device *dev) struct mtk_eth *eth = mac->hw; phy_disconnect(mac->phy_dev); - mtk_mdio_cleanup(eth); mtk_irq_disable(eth, ~0); } @@ -1916,6 +1915,7 @@ static int mtk_remove(struct platform_device *pdev) netif_napi_del(ð->tx_napi); netif_napi_del(ð->rx_napi); mtk_cleanup(eth); + mtk_mdio_cleanup(eth); platform_set_drvdata(pdev, NULL); return 0; From 79e9a41438527a4cc2b426a2dfb92cd0825d7d29 Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:32 +0800 Subject: [PATCH 046/271] net: ethernet: mediatek: fix issue of driver removal with interface is up mtk_stop() must be called to stop for freeing DMA resources acquired and restoring state changed by mtk_open() firstly when module removal. Signed-off-by: Sean Wang Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index bf5b7e1d9450..556951e66978 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -1906,6 +1906,14 @@ err_free_dev: static int mtk_remove(struct platform_device *pdev) { struct mtk_eth *eth = platform_get_drvdata(pdev); + int i; + + /* stop all devices to make sure that dma is properly shut down */ + for (i = 0; i < MTK_MAC_COUNT; i++) { + if (!eth->netdev[i]) + continue; + mtk_stop(eth->netdev[i]); + } clk_disable_unprepare(eth->clks[MTK_CLK_ETHIF]); clk_disable_unprepare(eth->clks[MTK_CLK_ESW]); From b5776f01bf1f36a6c89c6dc407dda6509002e0d7 Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:33 +0800 Subject: [PATCH 047/271] net: ethernet: mediatek: fix the missing of_node_put() after node is used done inside mtk_mdio_init This patch adds the missing of_node_put() after finishing the usage of of_get_child_by_name. Signed-off-by: Sean Wang Acked-by: John Crispin Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 556951e66978..409efcf9b5c8 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -324,6 +324,7 @@ static int mtk_mdio_init(struct mtk_eth *eth) err = of_mdiobus_register(eth->mii_bus, mii_np); if (err) goto err_free_bus; + of_node_put(mii_np); return 0; From 1e515b7fdbaeef29b1025c09b50ebbb132a39208 Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:34 +0800 Subject: [PATCH 048/271] net: ethernet: mediatek: use devm_mdiobus_alloc instead of mdiobus_alloc inside mtk_mdio_init a lot of parts in the driver uses devm_* APIs to gain benefits from the device resource management, so devm_mdiobus_alloc is also used instead of mdiobus_alloc to have more elegant code flow. Using common code provided by the devm_* helps to 1) have simplified the code flow as [1] says 2) decrease the risk of incorrect error handling by human 3) only a few drivers used it since it was proposed on linux 3.16, so just hope to promote for this. Ref: [1] https://patchwork.ozlabs.org/patch/344093/ Signed-off-by: Sean Wang Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 23 ++++++--------------- 1 file changed, 6 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 409efcf9b5c8..0367f510f81a 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -295,7 +295,7 @@ err_phy: static int mtk_mdio_init(struct mtk_eth *eth) { struct device_node *mii_np; - int err; + int ret; mii_np = of_get_child_by_name(eth->dev->of_node, "mdio-bus"); if (!mii_np) { @@ -304,13 +304,13 @@ static int mtk_mdio_init(struct mtk_eth *eth) } if (!of_device_is_available(mii_np)) { - err = 0; + ret = 0; goto err_put_node; } - eth->mii_bus = mdiobus_alloc(); + eth->mii_bus = devm_mdiobus_alloc(eth->dev); if (!eth->mii_bus) { - err = -ENOMEM; + ret = -ENOMEM; goto err_put_node; } @@ -321,20 +321,11 @@ static int mtk_mdio_init(struct mtk_eth *eth) eth->mii_bus->parent = eth->dev; snprintf(eth->mii_bus->id, MII_BUS_ID_SIZE, "%s", mii_np->name); - err = of_mdiobus_register(eth->mii_bus, mii_np); - if (err) - goto err_free_bus; - of_node_put(mii_np); - - return 0; - -err_free_bus: - mdiobus_free(eth->mii_bus); + ret = of_mdiobus_register(eth->mii_bus, mii_np); err_put_node: of_node_put(mii_np); - eth->mii_bus = NULL; - return err; + return ret; } static void mtk_mdio_cleanup(struct mtk_eth *eth) @@ -343,8 +334,6 @@ static void mtk_mdio_cleanup(struct mtk_eth *eth) return; mdiobus_unregister(eth->mii_bus); - of_node_put(eth->mii_bus->dev.of_node); - mdiobus_free(eth->mii_bus); } static inline void mtk_irq_disable(struct mtk_eth *eth, u32 mask) From aa6e8a54f653daaf34ff7fdceb2d992d714eea7c Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 1 Sep 2016 10:47:35 +0800 Subject: [PATCH 049/271] net: ethernet: mediatek: fix error handling inside mtk_mdio_init Return -ENODEV if the MDIO bus is disabled in the device tree. Signed-off-by: Sean Wang Acked-by: John Crispin Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 0367f510f81a..d9199151a83e 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -304,7 +304,7 @@ static int mtk_mdio_init(struct mtk_eth *eth) } if (!of_device_is_available(mii_np)) { - ret = 0; + ret = -ENODEV; goto err_put_node; } From c0338aff2260ea6c092806312dbb154cec07a242 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Sun, 28 Aug 2016 21:28:26 -0700 Subject: [PATCH 050/271] kcm: fix a socket double free Dmitry reported a double free on kcm socket, which could be easily reproduced by: #include #include int main() { int fd = syscall(SYS_socket, 0x29ul, 0x5ul, 0x0ul, 0, 0, 0); syscall(SYS_ioctl, fd, 0x89e2ul, 0x20a98000ul, 0, 0, 0); return 0; } This is because on the error path, after we install the new socket file, we call sock_release() to clean up the socket, which leaves the fd pointing to a freed socket. Fix this by calling sys_close() on that fd directly. Fixes: ab7ac4eb9832 ("kcm: Kernel Connection Multiplexor module") Reported-by: Dmitry Vyukov Cc: Tom Herbert Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- net/kcm/kcmsock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c index cb39e05b166c..411693288648 100644 --- a/net/kcm/kcmsock.c +++ b/net/kcm/kcmsock.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -2029,7 +2030,7 @@ static int kcm_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) if (copy_to_user((void __user *)arg, &info, sizeof(info))) { err = -EFAULT; - sock_release(newsock); + sys_close(info.fd); } } From 5fe118c9fd6b8e129788c88a10ac0af82b255f81 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Mon, 29 Aug 2016 08:29:52 -0400 Subject: [PATCH 051/271] qed*: Disallow dcbx configuration for VF interfaces. Dcbx configuration is not supported for VF interfaces. Hence don't populate the callbacks for VFs and also fail the dcbx-query for VFs. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dcbx.c | 4 ++++ drivers/net/ethernet/qlogic/qede/qede_main.c | 3 ++- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index 226cb08cc055..7ad1667ee354 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -19,6 +19,7 @@ #include "qed_dcbx.h" #include "qed_hsi.h" #include "qed_sp.h" +#include "qed_sriov.h" #ifdef CONFIG_DCB #include #endif @@ -945,6 +946,9 @@ static int qed_dcbx_query_params(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt; int rc; + if (IS_VF(p_hwfn->cdev)) + return -EINVAL; + p_ptt = qed_ptt_acquire(p_hwfn); if (!p_ptt) return -EBUSY; diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index a6eb6af8cbe8..9544e4c41359 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -2520,7 +2520,8 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, edev->ops->register_ops(cdev, &qede_ll_ops, edev); #ifdef CONFIG_DCB - qede_set_dcbnl_ops(edev->ndev); + if (!IS_VF(edev)) + qede_set_dcbnl_ops(edev->ndev); #endif INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); From 5ec5dfa4d29448002bb05018ba12620e7fafffd1 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Mon, 29 Aug 2016 08:29:53 -0400 Subject: [PATCH 052/271] qed: Set selection-field while configuring the app entry in ieee mode. Management firmware requires the selection-field (SF) to be set for configuring the application/protocol entry in IEEE mode. Without this setting, the app entry will be configured incorrectly in MFW. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dcbx.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index 7ad1667ee354..192a886edfe9 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -1063,23 +1063,31 @@ qed_dcbx_set_app_data(struct qed_hwfn *p_hwfn, for (i = 0; i < DCBX_MAX_APP_PROTOCOL; i++) { entry = &p_app->app_pri_tbl[i].entry; if (ieee) { - *entry &= ~DCBX_APP_SF_IEEE_MASK; + *entry &= ~(DCBX_APP_SF_IEEE_MASK | DCBX_APP_SF_MASK); switch (p_params->app_entry[i].sf_ieee) { case QED_DCBX_SF_IEEE_ETHTYPE: *entry |= ((u32)DCBX_APP_SF_IEEE_ETHTYPE << DCBX_APP_SF_IEEE_SHIFT); + *entry |= ((u32)DCBX_APP_SF_ETHTYPE << + DCBX_APP_SF_SHIFT); break; case QED_DCBX_SF_IEEE_TCP_PORT: *entry |= ((u32)DCBX_APP_SF_IEEE_TCP_PORT << DCBX_APP_SF_IEEE_SHIFT); + *entry |= ((u32)DCBX_APP_SF_PORT << + DCBX_APP_SF_SHIFT); break; case QED_DCBX_SF_IEEE_UDP_PORT: *entry |= ((u32)DCBX_APP_SF_IEEE_UDP_PORT << DCBX_APP_SF_IEEE_SHIFT); + *entry |= ((u32)DCBX_APP_SF_PORT << + DCBX_APP_SF_SHIFT); break; case QED_DCBX_SF_IEEE_TCP_UDP_PORT: *entry |= ((u32)DCBX_APP_SF_IEEE_TCP_UDP_PORT << DCBX_APP_SF_IEEE_SHIFT); + *entry |= ((u32)DCBX_APP_SF_PORT << + DCBX_APP_SF_SHIFT); break; } } else { From c5e801dac4daff972013613ae3edae6668b2aa7c Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Mon, 29 Aug 2016 08:29:54 -0400 Subject: [PATCH 053/271] qed: Clear dcbx memory buffers before the usage. This patch takes care of clearing the uninitialized buffer before using it. 1. pfc pri-enable bitmap need to be cleared before setting the requested enable bits. Without this, the un-touched values will be merged with requested values and sent to MFW. 2. The data in app-entry field need to be cleared before using it. 3. Clear the output data buffer used in qed_dcbx_query_params(). Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dcbx.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index 192a886edfe9..ad023fe7d156 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -988,6 +988,7 @@ qed_dcbx_set_pfc_data(struct qed_hwfn *p_hwfn, if (p_params->pfc.prio[i]) pfc_map |= BIT(i); + *pfc &= ~DCBX_PFC_PRI_EN_BITMAP_MASK; *pfc |= (pfc_map << DCBX_PFC_PRI_EN_BITMAP_SHIFT); DP_VERBOSE(p_hwfn, QED_MSG_DCB, "pfc = 0x%x\n", *pfc); @@ -1062,6 +1063,7 @@ qed_dcbx_set_app_data(struct qed_hwfn *p_hwfn, for (i = 0; i < DCBX_MAX_APP_PROTOCOL; i++) { entry = &p_app->app_pri_tbl[i].entry; + *entry = 0; if (ieee) { *entry &= ~(DCBX_APP_SF_IEEE_MASK | DCBX_APP_SF_MASK); switch (p_params->app_entry[i].sf_ieee) { @@ -1193,6 +1195,7 @@ int qed_dcbx_get_config_params(struct qed_hwfn *p_hwfn, return -ENOMEM; } + memset(dcbx_info, 0, sizeof(*dcbx_info)); rc = qed_dcbx_query_params(p_hwfn, dcbx_info, QED_DCBX_OPERATIONAL_MIB); if (rc) { kfree(dcbx_info); @@ -1230,6 +1233,7 @@ static struct qed_dcbx_get *qed_dcbnl_get_dcbx(struct qed_hwfn *hwfn, return NULL; } + memset(dcbx_info, 0, sizeof(*dcbx_info)); if (qed_dcbx_query_params(hwfn, dcbx_info, type)) { kfree(dcbx_info); return NULL; From 6af7e4f77259ee946103387372cb159f2e99a6d4 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 1 Sep 2016 08:52:29 -0500 Subject: [PATCH 054/271] PCI: Mark Haswell Power Control Unit as having non-compliant BARs The Haswell Power Control Unit has a non-PCI register (CONFIG_TDP_NOMINAL) where BAR 0 is supposed to be. This is erratum HSE43 in the spec update referenced below: The PCIe* Base Specification indicates that Configuration Space Headers have a base address register at offset 0x10. Due to this erratum, the Power Control Unit's CONFIG_TDP_NOMINAL CSR (Bus 1; Device 30; Function 3; Offset 0x10) is located where a base register is expected. Mark the PCU as having non-compliant BARs so we don't try to probe any of them. There are no other BARs on this device. Rename the quirk so it's not Broadwell-specific. Link: http://www.intel.com/content/www/us/en/processors/xeon/xeon-e5-v3-spec-update.html Link: http://www.intel.com/content/www/us/en/processors/xeon/xeon-e5-v3-datasheet-vol-2.html (section 5.4, Device 30 Function 3) Link: https://bugzilla.kernel.org/show_bug.cgi?id=153881 Reported-by: Paul Menzel Tested-by: Prarit Bhargava Signed-off-by: Bjorn Helgaas Acked-by: Myron Stowe --- arch/x86/pci/fixup.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/arch/x86/pci/fixup.c b/arch/x86/pci/fixup.c index 837ea36a837d..6d52b94f4bb9 100644 --- a/arch/x86/pci/fixup.c +++ b/arch/x86/pci/fixup.c @@ -553,15 +553,21 @@ static void twinhead_reserve_killing_zone(struct pci_dev *dev) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x27B9, twinhead_reserve_killing_zone); /* - * Broadwell EP Home Agent BARs erroneously return non-zero values when read. + * Device [8086:2fc0] + * Erratum HSE43 + * CONFIG_TDP_NOMINAL CSR Implemented at Incorrect Offset + * http://www.intel.com/content/www/us/en/processors/xeon/xeon-e5-v3-spec-update.html * - * See http://www.intel.com/content/www/us/en/processors/xeon/xeon-e5-v4-spec-update.html - * entry BDF2. + * Devices [8086:6f60,6fa0,6fc0] + * Erratum BDF2 + * PCI BARs in the Home Agent Will Return Non-Zero Values During Enumeration + * http://www.intel.com/content/www/us/en/processors/xeon/xeon-e5-v4-spec-update.html */ -static void pci_bdwep_bar(struct pci_dev *dev) +static void pci_invalid_bar(struct pci_dev *dev) { dev->non_compliant_bars = 1; } -DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x6f60, pci_bdwep_bar); -DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x6fa0, pci_bdwep_bar); -DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x6fc0, pci_bdwep_bar); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x2fc0, pci_invalid_bar); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x6f60, pci_invalid_bar); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x6fa0, pci_invalid_bar); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x6fc0, pci_invalid_bar); From 5b004412e2b7894105ea6043d380b7ab21f244ef Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 1 Sep 2016 10:37:40 +0200 Subject: [PATCH 055/271] mlxsw: spectrum_router: Fix failure caused by double fib removal from HW In mlxsw we squash tables 254 and 255 together into HW. Kernel adds/dels /32 ip to/from both 254 and 255. On del path, that causes the same prefix being removed twice. Fix this by introducing reference counting for private mlxsw fib entries. That required a bit of code reshuffle. Also put dev into fib entry key so the same prefix could be represented once per every router interface. Fixes: 61c503f976b5 ("mlxsw: spectrum_router: Implement fib4 add/del switchdev obj ops") Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- .../ethernet/mellanox/mlxsw/spectrum_router.c | 144 ++++++++++++------ 1 file changed, 101 insertions(+), 43 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 917ddd1e422f..ed618144fed4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -107,6 +107,7 @@ mlxsw_sp_prefix_usage_clear(struct mlxsw_sp_prefix_usage *prefix_usage, } struct mlxsw_sp_fib_key { + struct net_device *dev; unsigned char addr[sizeof(struct in6_addr)]; unsigned char prefix_len; }; @@ -123,6 +124,7 @@ struct mlxsw_sp_fib_entry { struct rhash_head ht_node; struct mlxsw_sp_fib_key key; enum mlxsw_sp_fib_entry_type type; + unsigned int ref_count; u8 added:1; u16 rif; /* used for action local */ struct mlxsw_sp_vr *vr; @@ -171,13 +173,15 @@ static void mlxsw_sp_fib_entry_remove(struct mlxsw_sp_fib *fib, static struct mlxsw_sp_fib_entry * mlxsw_sp_fib_entry_create(struct mlxsw_sp_fib *fib, const void *addr, - size_t addr_len, unsigned char prefix_len) + size_t addr_len, unsigned char prefix_len, + struct net_device *dev) { struct mlxsw_sp_fib_entry *fib_entry; fib_entry = kzalloc(sizeof(*fib_entry), GFP_KERNEL); if (!fib_entry) return NULL; + fib_entry->key.dev = dev; memcpy(fib_entry->key.addr, addr, addr_len); fib_entry->key.prefix_len = prefix_len; return fib_entry; @@ -190,10 +194,13 @@ static void mlxsw_sp_fib_entry_destroy(struct mlxsw_sp_fib_entry *fib_entry) static struct mlxsw_sp_fib_entry * mlxsw_sp_fib_entry_lookup(struct mlxsw_sp_fib *fib, const void *addr, - size_t addr_len, unsigned char prefix_len) + size_t addr_len, unsigned char prefix_len, + struct net_device *dev) { - struct mlxsw_sp_fib_key key = {{ 0 } }; + struct mlxsw_sp_fib_key key; + memset(&key, 0, sizeof(key)); + key.dev = dev; memcpy(key.addr, addr, addr_len); key.prefix_len = prefix_len; return rhashtable_lookup_fast(&fib->ht, &key, mlxsw_sp_fib_ht_params); @@ -1695,6 +1702,79 @@ mlxsw_sp_router_fib4_entry_fini(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_nexthop_group_put(mlxsw_sp, fib_entry); } +static struct mlxsw_sp_fib_entry * +mlxsw_sp_fib_entry_get(struct mlxsw_sp *mlxsw_sp, + const struct switchdev_obj_ipv4_fib *fib4) +{ + struct mlxsw_sp_fib_entry *fib_entry; + struct fib_info *fi = fib4->fi; + struct mlxsw_sp_vr *vr; + int err; + + vr = mlxsw_sp_vr_get(mlxsw_sp, fib4->dst_len, fib4->tb_id, + MLXSW_SP_L3_PROTO_IPV4); + if (IS_ERR(vr)) + return ERR_CAST(vr); + + fib_entry = mlxsw_sp_fib_entry_lookup(vr->fib, &fib4->dst, + sizeof(fib4->dst), + fib4->dst_len, fi->fib_dev); + if (fib_entry) { + /* Already exists, just take a reference */ + fib_entry->ref_count++; + return fib_entry; + } + fib_entry = mlxsw_sp_fib_entry_create(vr->fib, &fib4->dst, + sizeof(fib4->dst), + fib4->dst_len, fi->fib_dev); + if (!fib_entry) { + err = -ENOMEM; + goto err_fib_entry_create; + } + fib_entry->vr = vr; + fib_entry->ref_count = 1; + + err = mlxsw_sp_router_fib4_entry_init(mlxsw_sp, fib4, fib_entry); + if (err) + goto err_fib4_entry_init; + + return fib_entry; + +err_fib4_entry_init: + mlxsw_sp_fib_entry_destroy(fib_entry); +err_fib_entry_create: + mlxsw_sp_vr_put(mlxsw_sp, vr); + + return ERR_PTR(err); +} + +static struct mlxsw_sp_fib_entry * +mlxsw_sp_fib_entry_find(struct mlxsw_sp *mlxsw_sp, + const struct switchdev_obj_ipv4_fib *fib4) +{ + struct mlxsw_sp_vr *vr; + + vr = mlxsw_sp_vr_find(mlxsw_sp, fib4->tb_id, MLXSW_SP_L3_PROTO_IPV4); + if (!vr) + return NULL; + + return mlxsw_sp_fib_entry_lookup(vr->fib, &fib4->dst, + sizeof(fib4->dst), fib4->dst_len, + fib4->fi->fib_dev); +} + +void mlxsw_sp_fib_entry_put(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + struct mlxsw_sp_vr *vr = fib_entry->vr; + + if (--fib_entry->ref_count == 0) { + mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); + mlxsw_sp_fib_entry_destroy(fib_entry); + } + mlxsw_sp_vr_put(mlxsw_sp, vr); +} + static int mlxsw_sp_router_fib4_add_prepare(struct mlxsw_sp_port *mlxsw_sp_port, const struct switchdev_obj_ipv4_fib *fib4, @@ -1703,25 +1783,11 @@ mlxsw_sp_router_fib4_add_prepare(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_router_fib4_add_info *info; struct mlxsw_sp_fib_entry *fib_entry; - struct mlxsw_sp_vr *vr; int err; - vr = mlxsw_sp_vr_get(mlxsw_sp, fib4->dst_len, fib4->tb_id, - MLXSW_SP_L3_PROTO_IPV4); - if (IS_ERR(vr)) - return PTR_ERR(vr); - - fib_entry = mlxsw_sp_fib_entry_create(vr->fib, &fib4->dst, - sizeof(fib4->dst), fib4->dst_len); - if (!fib_entry) { - err = -ENOMEM; - goto err_fib_entry_create; - } - fib_entry->vr = vr; - - err = mlxsw_sp_router_fib4_entry_init(mlxsw_sp, fib4, fib_entry); - if (err) - goto err_fib4_entry_init; + fib_entry = mlxsw_sp_fib_entry_get(mlxsw_sp, fib4); + if (IS_ERR(fib_entry)) + return PTR_ERR(fib_entry); info = kmalloc(sizeof(*info), GFP_KERNEL); if (!info) { @@ -1736,11 +1802,7 @@ mlxsw_sp_router_fib4_add_prepare(struct mlxsw_sp_port *mlxsw_sp_port, return 0; err_alloc_info: - mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); -err_fib4_entry_init: - mlxsw_sp_fib_entry_destroy(fib_entry); -err_fib_entry_create: - mlxsw_sp_vr_put(mlxsw_sp, vr); + mlxsw_sp_fib_entry_put(mlxsw_sp, fib_entry); return err; } @@ -1759,11 +1821,14 @@ mlxsw_sp_router_fib4_add_commit(struct mlxsw_sp_port *mlxsw_sp_port, fib_entry = info->fib_entry; kfree(info); + if (fib_entry->ref_count != 1) + return 0; + vr = fib_entry->vr; - err = mlxsw_sp_fib_entry_insert(fib_entry->vr->fib, fib_entry); + err = mlxsw_sp_fib_entry_insert(vr->fib, fib_entry); if (err) goto err_fib_entry_insert; - err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); + err = mlxsw_sp_fib_entry_update(mlxsw_sp_port->mlxsw_sp, fib_entry); if (err) goto err_fib_entry_add; return 0; @@ -1771,9 +1836,7 @@ mlxsw_sp_router_fib4_add_commit(struct mlxsw_sp_port *mlxsw_sp_port, err_fib_entry_add: mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); err_fib_entry_insert: - mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); - mlxsw_sp_fib_entry_destroy(fib_entry); - mlxsw_sp_vr_put(mlxsw_sp, vr); + mlxsw_sp_fib_entry_put(mlxsw_sp, fib_entry); return err; } @@ -1793,23 +1856,18 @@ int mlxsw_sp_router_fib4_del(struct mlxsw_sp_port *mlxsw_sp_port, { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_fib_entry *fib_entry; - struct mlxsw_sp_vr *vr; - vr = mlxsw_sp_vr_find(mlxsw_sp, fib4->tb_id, MLXSW_SP_L3_PROTO_IPV4); - if (!vr) { - dev_warn(mlxsw_sp->bus_info->dev, "Failed to find virtual router for FIB4 entry being removed.\n"); - return -ENOENT; - } - fib_entry = mlxsw_sp_fib_entry_lookup(vr->fib, &fib4->dst, - sizeof(fib4->dst), fib4->dst_len); + fib_entry = mlxsw_sp_fib_entry_find(mlxsw_sp, fib4); if (!fib_entry) { dev_warn(mlxsw_sp->bus_info->dev, "Failed to find FIB4 entry being removed.\n"); return -ENOENT; } - mlxsw_sp_fib_entry_del(mlxsw_sp_port->mlxsw_sp, fib_entry); - mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); - mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); - mlxsw_sp_fib_entry_destroy(fib_entry); - mlxsw_sp_vr_put(mlxsw_sp, vr); + + if (fib_entry->ref_count == 1) { + mlxsw_sp_fib_entry_del(mlxsw_sp, fib_entry); + mlxsw_sp_fib_entry_remove(fib_entry->vr->fib, fib_entry); + } + + mlxsw_sp_fib_entry_put(mlxsw_sp, fib_entry); return 0; } From 7146da31817aa21b19d7a01e95b94686c3f4bd97 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 1 Sep 2016 10:37:41 +0200 Subject: [PATCH 056/271] mlxsw: spectrum_router: Fix fib entry update path Originally, I expected that there would be needed to call update operation in case RALUE record action is changed. However, that is not needed since write operation takes care of that nicely. Remove prepared construct and always call the write operation. Fixes: 61c503f976b5 ("mlxsw: spectrum_router: Implement fib4 add/del switchdev obj ops") Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index ed618144fed4..49f4e06e5842 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -125,7 +125,6 @@ struct mlxsw_sp_fib_entry { struct mlxsw_sp_fib_key key; enum mlxsw_sp_fib_entry_type type; unsigned int ref_count; - u8 added:1; u16 rif; /* used for action local */ struct mlxsw_sp_vr *vr; struct list_head nexthop_group_node; @@ -1633,11 +1632,8 @@ static int mlxsw_sp_fib_entry_op(struct mlxsw_sp *mlxsw_sp, static int mlxsw_sp_fib_entry_update(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib_entry *fib_entry) { - enum mlxsw_reg_ralue_op op; - - op = !fib_entry->added ? MLXSW_REG_RALUE_OP_WRITE_WRITE : - MLXSW_REG_RALUE_OP_WRITE_UPDATE; - return mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, op); + return mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, + MLXSW_REG_RALUE_OP_WRITE_WRITE); } static int mlxsw_sp_fib_entry_del(struct mlxsw_sp *mlxsw_sp, From de7d62952b01755f9885bbe945b141d9d4338e55 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 1 Sep 2016 10:37:42 +0200 Subject: [PATCH 057/271] mlxsw: spectrum: Fix error path in mlxsw_sp_module_init Add forgotten notifier unregister. Fixes: 99724c18fc66 ("mlxsw: spectrum: Introduce support for router interfaces") Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 7291f2c4b0c7..ec1f7d88948f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -4553,6 +4553,7 @@ static int __init mlxsw_sp_module_init(void) return 0; err_core_driver_register: + unregister_inetaddr_notifier(&mlxsw_sp_inetaddr_nb); unregister_netdevice_notifier(&mlxsw_sp_netdevice_nb); return err; } From e73226384948e9b35823150cadd9a3ea4ca9fc97 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 1 Sep 2016 10:37:43 +0200 Subject: [PATCH 058/271] mlxsw: spectrum_router: Fix netevent notifier registration Currently the notifier is registered for every asic instance, however the same block. Fix this by moving the registration to module init. Fixes: c723c735fa6b ("mlxsw: spectrum_router: Periodically update the kernel's neigh table") Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 9 +++++++++ drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 2 ++ .../ethernet/mellanox/mlxsw/spectrum_router.c | 17 ++--------------- 3 files changed, 13 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index ec1f7d88948f..e16b34718c8c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -56,6 +56,7 @@ #include #include #include +#include #include "spectrum.h" #include "core.h" @@ -4541,18 +4542,25 @@ static struct notifier_block mlxsw_sp_inetaddr_nb __read_mostly = { .priority = 10, /* Must be called before FIB notifier block */ }; +static struct notifier_block mlxsw_sp_router_netevent_nb __read_mostly = { + .notifier_call = mlxsw_sp_router_netevent_event, +}; + static int __init mlxsw_sp_module_init(void) { int err; register_netdevice_notifier(&mlxsw_sp_netdevice_nb); register_inetaddr_notifier(&mlxsw_sp_inetaddr_nb); + register_netevent_notifier(&mlxsw_sp_router_netevent_nb); + err = mlxsw_core_driver_register(&mlxsw_sp_driver); if (err) goto err_core_driver_register; return 0; err_core_driver_register: + unregister_netevent_notifier(&mlxsw_sp_router_netevent_nb); unregister_inetaddr_notifier(&mlxsw_sp_inetaddr_nb); unregister_netdevice_notifier(&mlxsw_sp_netdevice_nb); return err; @@ -4561,6 +4569,7 @@ err_core_driver_register: static void __exit mlxsw_sp_module_exit(void) { mlxsw_core_driver_unregister(&mlxsw_sp_driver); + unregister_netevent_notifier(&mlxsw_sp_router_netevent_nb); unregister_inetaddr_notifier(&mlxsw_sp_inetaddr_nb); unregister_netdevice_notifier(&mlxsw_sp_netdevice_nb); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index ab3feb81bd43..ac48abebe904 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -587,6 +587,8 @@ int mlxsw_sp_router_neigh_construct(struct net_device *dev, struct neighbour *n); void mlxsw_sp_router_neigh_destroy(struct net_device *dev, struct neighbour *n); +int mlxsw_sp_router_netevent_event(struct notifier_block *unused, + unsigned long event, void *ptr); int mlxsw_sp_kvdl_alloc(struct mlxsw_sp *mlxsw_sp, unsigned int entry_count); void mlxsw_sp_kvdl_free(struct mlxsw_sp *mlxsw_sp, int entry_index); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 49f4e06e5842..352259b3c38a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -944,8 +944,8 @@ static void mlxsw_sp_router_neigh_update_hw(struct work_struct *work) mlxsw_sp_port_dev_put(mlxsw_sp_port); } -static int mlxsw_sp_router_netevent_event(struct notifier_block *unused, - unsigned long event, void *ptr) +int mlxsw_sp_router_netevent_event(struct notifier_block *unused, + unsigned long event, void *ptr) { struct mlxsw_sp_neigh_entry *neigh_entry; struct mlxsw_sp_port *mlxsw_sp_port; @@ -1015,10 +1015,6 @@ static int mlxsw_sp_router_netevent_event(struct notifier_block *unused, return NOTIFY_DONE; } -static struct notifier_block mlxsw_sp_router_netevent_nb __read_mostly = { - .notifier_call = mlxsw_sp_router_netevent_event, -}; - static int mlxsw_sp_neigh_init(struct mlxsw_sp *mlxsw_sp) { int err; @@ -1033,10 +1029,6 @@ static int mlxsw_sp_neigh_init(struct mlxsw_sp *mlxsw_sp) */ mlxsw_sp_router_neighs_update_interval_init(mlxsw_sp); - err = register_netevent_notifier(&mlxsw_sp_router_netevent_nb); - if (err) - goto err_register_netevent_notifier; - /* Create the delayed works for the activity_update */ INIT_DELAYED_WORK(&mlxsw_sp->router.neighs_update.dw, mlxsw_sp_router_neighs_update_work); @@ -1045,17 +1037,12 @@ static int mlxsw_sp_neigh_init(struct mlxsw_sp *mlxsw_sp) mlxsw_core_schedule_dw(&mlxsw_sp->router.neighs_update.dw, 0); mlxsw_core_schedule_dw(&mlxsw_sp->router.nexthop_probe_dw, 0); return 0; - -err_register_netevent_notifier: - rhashtable_destroy(&mlxsw_sp->router.neigh_ht); - return err; } static void mlxsw_sp_neigh_fini(struct mlxsw_sp *mlxsw_sp) { cancel_delayed_work_sync(&mlxsw_sp->router.neighs_update.dw); cancel_delayed_work_sync(&mlxsw_sp->router.nexthop_probe_dw); - unregister_netevent_notifier(&mlxsw_sp_router_netevent_nb); rhashtable_destroy(&mlxsw_sp->router.neigh_ht); } From f1de7a28d53cfde8fa290d6c535eb690a4e1b5d5 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Thu, 1 Sep 2016 10:37:44 +0200 Subject: [PATCH 059/271] mlxsw: spectrum: Don't take multiple references on a FID In commit 14d39461b3f4 ("mlxsw: spectrum: Use per-FID struct for the VLAN-aware bridge") I added a per-FID struct, which member ports can take a reference on upon VLAN membership configuration. However, sometimes only the VLAN flags (e.g. egress untagged) are toggled without changing the VLAN membership. In these cases we shouldn't take another reference on the FID. Fixes: 14d39461b3f4 ("mlxsw: spectrum: Use per-FID struct for the VLAN-aware bridge") Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index d1b59cdfacc1..e0d95c419634 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -460,6 +460,9 @@ static int __mlxsw_sp_port_fid_join(struct mlxsw_sp_port *mlxsw_sp_port, { struct mlxsw_sp_fid *f; + if (test_bit(fid, mlxsw_sp_port->active_vlans)) + return 0; + f = mlxsw_sp_fid_find(mlxsw_sp_port->mlxsw_sp, fid); if (!f) { f = mlxsw_sp_fid_create(mlxsw_sp_port->mlxsw_sp, fid); From aad8b6bae7c63fcf860c4c7ce693cb8ac5d4a511 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Thu, 1 Sep 2016 10:37:45 +0200 Subject: [PATCH 060/271] mlxsw: spectrum: Use existing flood setup when adding VLANs When a VLAN is added on a bridge port we should use the existing unicast flood configuration of the port instead of assuming it's enabled. Fixes: 0293038e0c36 ("mlxsw: spectrum: Add support for flood control") Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- .../mellanox/mlxsw/spectrum_switchdev.c | 25 +++++++------------ 1 file changed, 9 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index e0d95c419634..7b654c517b91 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -167,8 +167,8 @@ static int mlxsw_sp_port_attr_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port, } static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, - u16 idx_begin, u16 idx_end, bool set, - bool only_uc) + u16 idx_begin, u16 idx_end, bool uc_set, + bool bm_set) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; u16 local_port = mlxsw_sp_port->local_port; @@ -187,28 +187,22 @@ static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, return -ENOMEM; mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_UC, idx_begin, - table_type, range, local_port, set); + table_type, range, local_port, uc_set); err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl); if (err) goto buffer_out; - /* Flooding control allows one to decide whether a given port will - * flood unicast traffic for which there is no FDB entry. - */ - if (only_uc) - goto buffer_out; - mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_BM, idx_begin, - table_type, range, local_port, set); + table_type, range, local_port, bm_set); err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl); if (err) goto err_flood_bm_set; - else - goto buffer_out; + + goto buffer_out; err_flood_bm_set: mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_UC, idx_begin, - table_type, range, local_port, !set); + table_type, range, local_port, !uc_set); mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl); buffer_out: kfree(sftr_pl); @@ -257,8 +251,7 @@ int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, * the start of the vFIDs range. */ vfid = mlxsw_sp_fid_to_vfid(fid); - return __mlxsw_sp_port_flood_set(mlxsw_sp_vport, vfid, vfid, set, - false); + return __mlxsw_sp_port_flood_set(mlxsw_sp_vport, vfid, vfid, set, set); } static int mlxsw_sp_port_attr_br_flags_set(struct mlxsw_sp_port *mlxsw_sp_port, @@ -520,7 +513,7 @@ static int mlxsw_sp_port_fid_join(struct mlxsw_sp_port *mlxsw_sp_port, } err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, fid_begin, fid_end, - true, false); + mlxsw_sp_port->uc_flood, true); if (err) goto err_port_flood_set; From 561ed23331dfefea4c8eec616463ee4d0c8fa4b7 Mon Sep 17 00:00:00 2001 From: Wu Fengguang Date: Thu, 1 Sep 2016 14:45:12 +0800 Subject: [PATCH 061/271] qed: fix kzalloc-simple.cocci warnings drivers/net/ethernet/qlogic/qed/qed_dcbx.c:1230:13-20: WARNING: kzalloc should be used for dcbx_info, instead of kmalloc/memset drivers/net/ethernet/qlogic/qed/qed_dcbx.c:1192:13-20: WARNING: kzalloc should be used for dcbx_info, instead of kmalloc/memset Use kzalloc rather than kmalloc followed by memset with 0 This considers some simple cases that are common and easy to validate Note in particular that there are no ...s in the rule, so all of the matched code has to be contiguous Generated by: scripts/coccinelle/api/alloc/kzalloc-simple.cocci CC: Sudarsana Reddy Kalluru Signed-off-by: Fengguang Wu Acked-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dcbx.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index ad023fe7d156..3656d2fd673d 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -1189,13 +1189,12 @@ int qed_dcbx_get_config_params(struct qed_hwfn *p_hwfn, return 0; } - dcbx_info = kmalloc(sizeof(*dcbx_info), GFP_KERNEL); + dcbx_info = kzalloc(sizeof(*dcbx_info), GFP_KERNEL); if (!dcbx_info) { DP_ERR(p_hwfn, "Failed to allocate struct qed_dcbx_info\n"); return -ENOMEM; } - memset(dcbx_info, 0, sizeof(*dcbx_info)); rc = qed_dcbx_query_params(p_hwfn, dcbx_info, QED_DCBX_OPERATIONAL_MIB); if (rc) { kfree(dcbx_info); @@ -1227,13 +1226,12 @@ static struct qed_dcbx_get *qed_dcbnl_get_dcbx(struct qed_hwfn *hwfn, { struct qed_dcbx_get *dcbx_info; - dcbx_info = kmalloc(sizeof(*dcbx_info), GFP_KERNEL); + dcbx_info = kzalloc(sizeof(*dcbx_info), GFP_KERNEL); if (!dcbx_info) { DP_ERR(hwfn->cdev, "Failed to allocate memory for dcbx_info\n"); return NULL; } - memset(dcbx_info, 0, sizeof(*dcbx_info)); if (qed_dcbx_query_params(hwfn, dcbx_info, type)) { kfree(dcbx_info); return NULL; From aabdd09d535073c35f746e46c3a5d3286088be3a Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 1 Sep 2016 11:28:59 +0200 Subject: [PATCH 062/271] tg3: Fix for disallow tx coalescing time to be 0 The recent commit 087d7a8c9174 "tg3: Fix for diasllow rx coalescing time to be 0" disallow to set Rx coalescing time to be 0 as this stops generating interrupts for the incoming packets. I found the zero Tx coalescing time stops generating interrupts for outgoing packets as well and fires Tx watchdog later. To avoid this, don't allow to set Tx coalescing time to 0 and also remove subsequent checks that become senseless. Cc: satish.baddipadige@broadcom.com Cc: siva.kallam@broadcom.com Cc: michael.chan@broadcom.com Signed-off-by: Ivan Vecera Acked-by: Siva Reddy Kallam Acked-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/tg3.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 659261218d9f..a2551bcd1027 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -14012,6 +14012,7 @@ static int tg3_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec) if ((ec->rx_coalesce_usecs > MAX_RXCOL_TICKS) || (!ec->rx_coalesce_usecs) || (ec->tx_coalesce_usecs > MAX_TXCOL_TICKS) || + (!ec->tx_coalesce_usecs) || (ec->rx_max_coalesced_frames > MAX_RXMAX_FRAMES) || (ec->tx_max_coalesced_frames > MAX_TXMAX_FRAMES) || (ec->rx_coalesce_usecs_irq > max_rxcoal_tick_int) || @@ -14022,16 +14023,6 @@ static int tg3_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec) (ec->stats_block_coalesce_usecs < min_stat_coal_ticks)) return -EINVAL; - /* No rx interrupts will be generated if both are zero */ - if ((ec->rx_coalesce_usecs == 0) && - (ec->rx_max_coalesced_frames == 0)) - return -EINVAL; - - /* No tx interrupts will be generated if both are zero */ - if ((ec->tx_coalesce_usecs == 0) && - (ec->tx_max_coalesced_frames == 0)) - return -EINVAL; - /* Only copy relevant parameters, ignore all others. */ tp->coal.rx_coalesce_usecs = ec->rx_coalesce_usecs; tp->coal.tx_coalesce_usecs = ec->tx_coalesce_usecs; From d2f394dc4816b7bd1b44981d83509f18f19c53f0 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Thu, 1 Sep 2016 16:22:16 +0200 Subject: [PATCH 063/271] tipc: fix random link resets while adding a second bearer In a dual bearer configuration, if the second tipc link becomes active while the first link still has pending nametable "bulk" updates, it randomly leads to reset of the second link. When a link is established, the function named_distribute(), fills the skb based on node mtu (allows room for TUNNEL_PROTOCOL) with NAME_DISTRIBUTOR message for each PUBLICATION. However, the function named_distribute() allocates the buffer by increasing the node mtu by INT_H_SIZE (to insert NAME_DISTRIBUTOR). This consumes the space allocated for TUNNEL_PROTOCOL. When establishing the second link, the link shall tunnel all the messages in the first link queue including the "bulk" update. As size of the NAME_DISTRIBUTOR messages while tunnelling, exceeds the link mtu the transmission fails (-EMSGSIZE). Thus, the synch point based on the message count of the tunnel packets is never reached leading to link timeout. In this commit, we adjust the size of name distributor message so that they can be tunnelled. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/name_distr.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index 6b626a64b517..a04fe9be1c60 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -62,6 +62,8 @@ static void publ_to_item(struct distr_item *i, struct publication *p) /** * named_prepare_buf - allocate & initialize a publication message + * + * The buffer returned is of size INT_H_SIZE + payload size */ static struct sk_buff *named_prepare_buf(struct net *net, u32 type, u32 size, u32 dest) @@ -141,9 +143,9 @@ static void named_distribute(struct net *net, struct sk_buff_head *list, struct publication *publ; struct sk_buff *skb = NULL; struct distr_item *item = NULL; - uint msg_dsz = (tipc_node_get_mtu(net, dnode, 0) / ITEM_SIZE) * - ITEM_SIZE; - uint msg_rem = msg_dsz; + u32 msg_dsz = ((tipc_node_get_mtu(net, dnode, 0) - INT_H_SIZE) / + ITEM_SIZE) * ITEM_SIZE; + u32 msg_rem = msg_dsz; list_for_each_entry(publ, pls, local_list) { /* Prepare next buffer: */ From da29838dcbde86b3bdf3b5702abb5937b4c31d18 Mon Sep 17 00:00:00 2001 From: Andy Gospodarek Date: Mon, 29 Aug 2016 16:51:30 -0400 Subject: [PATCH 064/271] MAINTAINERS: update to working email address Signed-off-by: Andy Gospodarek Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 71aa5daeae8f..0537211788e5 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2484,7 +2484,7 @@ F: include/net/bluetooth/ BONDING DRIVER M: Jay Vosburgh M: Veaceslav Falico -M: Andy Gospodarek +M: Andy Gospodarek L: netdev@vger.kernel.org W: http://sourceforge.net/projects/bonding/ S: Supported From a036244c068612a43fa8c0f33a0eb4daa4d8dba0 Mon Sep 17 00:00:00 2001 From: Dave Ertman Date: Mon, 29 Aug 2016 17:38:26 -0700 Subject: [PATCH 065/271] i40e: Fix kernel panic on enable/disable LLDP If DCB is configured on the link partner switch with an unsupported traffic class configuration (e.g. non-contiguous TCs), the driver is flagging DCB as disabled. But, for future DCB LLDPDUs, the driver was checking if the interface was DCB capable instead of enabled. This was causing a kernel panic when LLDP was enabled/disabled on the link partner switch. This patch corrects the situation by having the LLDP event handler check the correct flag in the pf structure. It also cleans up the setting and clearing of the enabled flag for other checks. Signed-off-by: Dave Ertman Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher Signed-off-by: David S. Miller --- drivers/net/ethernet/intel/i40e/i40e_main.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 828ed28c3c14..d0b3a1bb82ca 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -5113,9 +5113,13 @@ static int i40e_init_pf_dcb(struct i40e_pf *pf) DCB_CAP_DCBX_VER_IEEE; pf->flags |= I40E_FLAG_DCB_CAPABLE; - /* Enable DCB tagging only when more than one TC */ + /* Enable DCB tagging only when more than one TC + * or explicitly disable if only one TC + */ if (i40e_dcb_get_num_tc(&hw->local_dcbx_config) > 1) pf->flags |= I40E_FLAG_DCB_ENABLED; + else + pf->flags &= ~I40E_FLAG_DCB_ENABLED; dev_dbg(&pf->pdev->dev, "DCBX offload is supported for this PF.\n"); } @@ -5716,7 +5720,7 @@ static int i40e_handle_lldp_event(struct i40e_pf *pf, u8 type; /* Not DCB capable or capability disabled */ - if (!(pf->flags & I40E_FLAG_DCB_CAPABLE)) + if (!(pf->flags & I40E_FLAG_DCB_ENABLED)) return ret; /* Ignore if event is not for Nearest Bridge */ @@ -7896,6 +7900,7 @@ static int i40e_init_interrupt_scheme(struct i40e_pf *pf) #endif I40E_FLAG_RSS_ENABLED | I40E_FLAG_DCB_CAPABLE | + I40E_FLAG_DCB_ENABLED | I40E_FLAG_SRIOV_ENABLED | I40E_FLAG_FD_SB_ENABLED | I40E_FLAG_FD_ATR_ENABLED | @@ -10502,6 +10507,7 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf) I40E_FLAG_FD_SB_ENABLED | I40E_FLAG_FD_ATR_ENABLED | I40E_FLAG_DCB_CAPABLE | + I40E_FLAG_DCB_ENABLED | I40E_FLAG_SRIOV_ENABLED | I40E_FLAG_VMDQ_ENABLED); } else if (!(pf->flags & (I40E_FLAG_RSS_ENABLED | @@ -10525,7 +10531,8 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf) /* Not enough queues for all TCs */ if ((pf->flags & I40E_FLAG_DCB_CAPABLE) && (queues_left < I40E_MAX_TRAFFIC_CLASS)) { - pf->flags &= ~I40E_FLAG_DCB_CAPABLE; + pf->flags &= ~(I40E_FLAG_DCB_CAPABLE | + I40E_FLAG_DCB_ENABLED); dev_info(&pf->pdev->dev, "not enough queues for DCB. DCB is disabled.\n"); } pf->num_lan_qps = max_t(int, pf->rss_size_max, @@ -10922,7 +10929,7 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) err = i40e_init_pf_dcb(pf); if (err) { dev_info(&pdev->dev, "DCB init failed %d, disabled\n", err); - pf->flags &= ~I40E_FLAG_DCB_CAPABLE; + pf->flags &= ~(I40E_FLAG_DCB_CAPABLE & I40E_FLAG_DCB_ENABLED); /* Continue without DCB enabled */ } #endif /* CONFIG_I40E_DCB */ From 57e81d44b0e1aa4dcb479ff8de8fc34cf635d0e8 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Tue, 30 Aug 2016 11:36:26 +0530 Subject: [PATCH 066/271] net: thunderx: Fix for HW issue while padding TSO packet There is a issue in HW where-in while sending GSO sized pkts as part of TSO, if pkt len falls below configured min packet size i.e 60, NIC will zero PAD packet and also updates IP total length. Hence set this value to lessthan min pkt size of MAC + IP + TCP headers, BGX will anyway do the padding to transmit 64 byte pkt including FCS. Signed-off-by: Sunil Goutham Signed-off-by: David S. Miller --- drivers/net/ethernet/cavium/thunder/nic_main.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/cavium/thunder/nic_main.c b/drivers/net/ethernet/cavium/thunder/nic_main.c index 16ed20357c5c..85cc782b9060 100644 --- a/drivers/net/ethernet/cavium/thunder/nic_main.c +++ b/drivers/net/ethernet/cavium/thunder/nic_main.c @@ -251,9 +251,14 @@ static void nic_set_tx_pkt_pad(struct nicpf *nic, int size) int lmac; u64 lmac_cfg; - /* Max value that can be set is 60 */ - if (size > 60) - size = 60; + /* There is a issue in HW where-in while sending GSO sized + * pkts as part of TSO, if pkt len falls below this size + * NIC will zero PAD packet and also updates IP total length. + * Hence set this value to lessthan min pkt size of MAC+IP+TCP + * headers, BGX will do the padding to transmit 64 byte pkt. + */ + if (size > 52) + size = 52; for (lmac = 0; lmac < (MAX_BGX_PER_CN88XX * MAX_LMAC_PER_BGX); lmac++) { lmac_cfg = nic_reg_read(nic, NIC_PF_LMAC_0_7_CFG | (lmac << 3)); From 7ceb8a1319ec64954459d474dd4a8c3c60ff0999 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Tue, 30 Aug 2016 11:36:27 +0530 Subject: [PATCH 067/271] net: thunderx: Fix for issues with multiple CQEs posted for a TSO packet On ThunderX 88xx pass 2.x chips when TSO is offloaded to HW, HW posts a CQE for every TSO segment transmitted. Current code does handles this, but is prone to issues when segment sizes are small resulting in SW processing too many CQEs and also at times frees a SKB which is not yet transmitted. This patch handles the errata in a different way and eliminates issues with earlier approach, TSO packet is submitted to HW with post_cqe=0, so that no CQE is posted upon completion of transmission of TSO packet but a additional HDR + IMMEDIATE descriptors are added to SQ due to which a CQE is posted and will have required info to be used while cleanup in napi. This way only one CQE is posted for a TSO packet. Signed-off-by: Sunil Goutham Signed-off-by: David S. Miller --- drivers/net/ethernet/cavium/thunder/nic.h | 1 + .../net/ethernet/cavium/thunder/nicvf_main.c | 20 ++++-- .../ethernet/cavium/thunder/nicvf_queues.c | 64 +++++++++++++++++-- 3 files changed, 73 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h index 83025bb4737c..e29815d9e6f4 100644 --- a/drivers/net/ethernet/cavium/thunder/nic.h +++ b/drivers/net/ethernet/cavium/thunder/nic.h @@ -279,6 +279,7 @@ struct nicvf { u8 sqs_id; bool sqs_mode; bool hw_tso; + bool t88; /* Receive buffer alloc */ u32 rb_page_offset; diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index a19e73f11d73..3240349615bd 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -513,6 +513,7 @@ static void nicvf_snd_pkt_handler(struct net_device *netdev, struct nicvf *nic = netdev_priv(netdev); struct snd_queue *sq; struct sq_hdr_subdesc *hdr; + struct sq_hdr_subdesc *tso_sqe; sq = &nic->qs->sq[cqe_tx->sq_idx]; @@ -527,17 +528,21 @@ static void nicvf_snd_pkt_handler(struct net_device *netdev, nicvf_check_cqe_tx_errs(nic, cq, cqe_tx); skb = (struct sk_buff *)sq->skbuff[cqe_tx->sqe_ptr]; - /* For TSO offloaded packets only one SQE will have a valid SKB */ if (skb) { + /* Check for dummy descriptor used for HW TSO offload on 88xx */ + if (hdr->dont_send) { + /* Get actual TSO descriptors and free them */ + tso_sqe = + (struct sq_hdr_subdesc *)GET_SQ_DESC(sq, hdr->rsvd2); + nicvf_put_sq_desc(sq, tso_sqe->subdesc_cnt + 1); + } nicvf_put_sq_desc(sq, hdr->subdesc_cnt + 1); prefetch(skb); dev_consume_skb_any(skb); sq->skbuff[cqe_tx->sqe_ptr] = (u64)NULL; } else { - /* In case of HW TSO, HW sends a CQE for each segment of a TSO - * packet instead of a single CQE for the whole TSO packet - * transmitted. Each of this CQE points to the same SQE, so - * avoid freeing same SQE multiple times. + /* In case of SW TSO on 88xx, only last segment will have + * a SKB attached, so just free SQEs here. */ if (!nic->hw_tso) nicvf_put_sq_desc(sq, hdr->subdesc_cnt + 1); @@ -1502,6 +1507,7 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) struct net_device *netdev; struct nicvf *nic; int err, qcount; + u16 sdevid; err = pci_enable_device(pdev); if (err) { @@ -1575,6 +1581,10 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (!pass1_silicon(nic->pdev)) nic->hw_tso = true; + pci_read_config_word(nic->pdev, PCI_SUBSYSTEM_ID, &sdevid); + if (sdevid == 0xA134) + nic->t88 = true; + /* Check if this VF is in QS only mode */ if (nic->sqs_mode) return 0; diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index 0ff8e60deccb..dda3ea3f3bb6 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -938,6 +938,8 @@ static int nicvf_tso_count_subdescs(struct sk_buff *skb) return num_edescs + sh->gso_segs; } +#define POST_CQE_DESC_COUNT 2 + /* Get the number of SQ descriptors needed to xmit this skb */ static int nicvf_sq_subdesc_required(struct nicvf *nic, struct sk_buff *skb) { @@ -948,6 +950,10 @@ static int nicvf_sq_subdesc_required(struct nicvf *nic, struct sk_buff *skb) return subdesc_cnt; } + /* Dummy descriptors to get TSO pkt completion notification */ + if (nic->t88 && nic->hw_tso && skb_shinfo(skb)->gso_size) + subdesc_cnt += POST_CQE_DESC_COUNT; + if (skb_shinfo(skb)->nr_frags) subdesc_cnt += skb_shinfo(skb)->nr_frags; @@ -965,14 +971,21 @@ nicvf_sq_add_hdr_subdesc(struct nicvf *nic, struct snd_queue *sq, int qentry, struct sq_hdr_subdesc *hdr; hdr = (struct sq_hdr_subdesc *)GET_SQ_DESC(sq, qentry); - sq->skbuff[qentry] = (u64)skb; - memset(hdr, 0, SND_QUEUE_DESC_SIZE); hdr->subdesc_type = SQ_DESC_TYPE_HEADER; - /* Enable notification via CQE after processing SQE */ - hdr->post_cqe = 1; - /* No of subdescriptors following this */ - hdr->subdesc_cnt = subdesc_cnt; + + if (nic->t88 && nic->hw_tso && skb_shinfo(skb)->gso_size) { + /* post_cqe = 0, to avoid HW posting a CQE for every TSO + * segment transmitted on 88xx. + */ + hdr->subdesc_cnt = subdesc_cnt - POST_CQE_DESC_COUNT; + } else { + sq->skbuff[qentry] = (u64)skb; + /* Enable notification via CQE after processing SQE */ + hdr->post_cqe = 1; + /* No of subdescriptors following this */ + hdr->subdesc_cnt = subdesc_cnt; + } hdr->tot_len = len; /* Offload checksum calculation to HW */ @@ -1023,6 +1036,37 @@ static inline void nicvf_sq_add_gather_subdesc(struct snd_queue *sq, int qentry, gather->addr = data; } +/* Add HDR + IMMEDIATE subdescriptors right after descriptors of a TSO + * packet so that a CQE is posted as a notifation for transmission of + * TSO packet. + */ +static inline void nicvf_sq_add_cqe_subdesc(struct snd_queue *sq, int qentry, + int tso_sqe, struct sk_buff *skb) +{ + struct sq_imm_subdesc *imm; + struct sq_hdr_subdesc *hdr; + + sq->skbuff[qentry] = (u64)skb; + + hdr = (struct sq_hdr_subdesc *)GET_SQ_DESC(sq, qentry); + memset(hdr, 0, SND_QUEUE_DESC_SIZE); + hdr->subdesc_type = SQ_DESC_TYPE_HEADER; + /* Enable notification via CQE after processing SQE */ + hdr->post_cqe = 1; + /* There is no packet to transmit here */ + hdr->dont_send = 1; + hdr->subdesc_cnt = POST_CQE_DESC_COUNT - 1; + hdr->tot_len = 1; + /* Actual TSO header SQE index, needed for cleanup */ + hdr->rsvd2 = tso_sqe; + + qentry = nicvf_get_nxt_sqentry(sq, qentry); + imm = (struct sq_imm_subdesc *)GET_SQ_DESC(sq, qentry); + memset(imm, 0, SND_QUEUE_DESC_SIZE); + imm->subdesc_type = SQ_DESC_TYPE_IMMEDIATE; + imm->len = 1; +} + /* Segment a TSO packet into 'gso_size' segments and append * them to SQ for transfer */ @@ -1096,7 +1140,7 @@ static int nicvf_sq_append_tso(struct nicvf *nic, struct snd_queue *sq, int nicvf_sq_append_skb(struct nicvf *nic, struct sk_buff *skb) { int i, size; - int subdesc_cnt; + int subdesc_cnt, tso_sqe = 0; int sq_num, qentry; struct queue_set *qs; struct snd_queue *sq; @@ -1131,6 +1175,7 @@ int nicvf_sq_append_skb(struct nicvf *nic, struct sk_buff *skb) /* Add SQ header subdesc */ nicvf_sq_add_hdr_subdesc(nic, sq, qentry, subdesc_cnt - 1, skb, skb->len); + tso_sqe = qentry; /* Add SQ gather subdescs */ qentry = nicvf_get_nxt_sqentry(sq, qentry); @@ -1154,6 +1199,11 @@ int nicvf_sq_append_skb(struct nicvf *nic, struct sk_buff *skb) } doorbell: + if (nic->t88 && skb_shinfo(skb)->gso_size) { + qentry = nicvf_get_nxt_sqentry(sq, qentry); + nicvf_sq_add_cqe_subdesc(sq, qentry, tso_sqe, skb); + } + /* make sure all memory stores are done before ringing doorbell */ smp_wmb(); From d26c638c16cb54f6fb1507e27df93ede692db572 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 30 Aug 2016 10:09:21 +0200 Subject: [PATCH 068/271] ipv6: add missing netconf notif when 'all' is updated The 'default' value was not advertised. Fixes: f3a1bfb11ccb ("rtnl/ipv6: use netconf msg to advertise forwarding status") Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index f418d2eaeddd..2a688171a188 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -778,7 +778,14 @@ static int addrconf_fixup_forwarding(struct ctl_table *table, int *p, int newf) } if (p == &net->ipv6.devconf_all->forwarding) { + int old_dflt = net->ipv6.devconf_dflt->forwarding; + net->ipv6.devconf_dflt->forwarding = newf; + if ((!newf) ^ (!old_dflt)) + inet6_netconf_notify_devconf(net, NETCONFA_FORWARDING, + NETCONFA_IFINDEX_DEFAULT, + net->ipv6.devconf_dflt); + addrconf_forward_change(net, newf); if ((!newf) ^ (!old)) inet6_netconf_notify_devconf(net, NETCONFA_FORWARDING, From 29c994e361009142ec0bca6493cc8f7b0d3c561a Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 30 Aug 2016 10:09:22 +0200 Subject: [PATCH 069/271] netconf: add a notif when settings are created All changes are notified, but the initial state was missing. Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- net/ipv4/devinet.c | 11 +++++++---- net/ipv6/addrconf.c | 9 ++++++++- 2 files changed, 15 insertions(+), 5 deletions(-) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 415e117967c7..062a67ca9a21 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -2232,7 +2232,7 @@ static struct devinet_sysctl_table { }; static int __devinet_sysctl_register(struct net *net, char *dev_name, - struct ipv4_devconf *p) + int ifindex, struct ipv4_devconf *p) { int i; struct devinet_sysctl_table *t; @@ -2255,6 +2255,8 @@ static int __devinet_sysctl_register(struct net *net, char *dev_name, goto free; p->sysctl = t; + + inet_netconf_notify_devconf(net, NETCONFA_ALL, ifindex, p); return 0; free: @@ -2286,7 +2288,7 @@ static int devinet_sysctl_register(struct in_device *idev) if (err) return err; err = __devinet_sysctl_register(dev_net(idev->dev), idev->dev->name, - &idev->cnf); + idev->dev->ifindex, &idev->cnf); if (err) neigh_sysctl_unregister(idev->arp_parms); return err; @@ -2347,11 +2349,12 @@ static __net_init int devinet_init_net(struct net *net) } #ifdef CONFIG_SYSCTL - err = __devinet_sysctl_register(net, "all", all); + err = __devinet_sysctl_register(net, "all", NETCONFA_IFINDEX_ALL, all); if (err < 0) goto err_reg_all; - err = __devinet_sysctl_register(net, "default", dflt); + err = __devinet_sysctl_register(net, "default", + NETCONFA_IFINDEX_DEFAULT, dflt); if (err < 0) goto err_reg_dflt; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 2a688171a188..bdf368eff5ab 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -6032,7 +6032,7 @@ static const struct ctl_table addrconf_sysctl[] = { static int __addrconf_sysctl_register(struct net *net, char *dev_name, struct inet6_dev *idev, struct ipv6_devconf *p) { - int i; + int i, ifindex; struct ctl_table *table; char path[sizeof("net/ipv6/conf/") + IFNAMSIZ]; @@ -6052,6 +6052,13 @@ static int __addrconf_sysctl_register(struct net *net, char *dev_name, if (!p->sysctl_header) goto free; + if (!strcmp(dev_name, "all")) + ifindex = NETCONFA_IFINDEX_ALL; + else if (!strcmp(dev_name, "default")) + ifindex = NETCONFA_IFINDEX_DEFAULT; + else + ifindex = idev->dev->ifindex; + inet6_netconf_notify_devconf(net, NETCONFA_ALL, ifindex, p); return 0; free: From 85a3d4a9356b595d5440c3f1bf07ee7cecca1567 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 30 Aug 2016 17:44:29 +0200 Subject: [PATCH 070/271] net: bridge: don't increment tx_dropped in br_do_proxy_arp pskb_may_pull may fail due to various reasons (e.g. alloc failure), but the skb isn't changed/dropped and processing continues so we shouldn't increment tx_dropped. CC: Kyeyoon Park CC: Roopa Prabhu CC: Stephen Hemminger CC: bridge@lists.linux-foundation.org Fixes: 958501163ddd ("bridge: Add support for IEEE 802.11 Proxy ARP") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_input.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 8e486203d133..abe11f085479 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -80,13 +80,10 @@ static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br, BR_INPUT_SKB_CB(skb)->proxyarp_replied = false; - if (dev->flags & IFF_NOARP) + if ((dev->flags & IFF_NOARP) || + !pskb_may_pull(skb, arp_hdr_len(dev))) return; - if (!pskb_may_pull(skb, arp_hdr_len(dev))) { - dev->stats.tx_dropped++; - return; - } parp = arp_hdr(skb); if (parp->ar_pro != htons(ETH_P_IP) || From 28b346cbc0715ae45b2814d857f1d8a7e6817ed8 Mon Sep 17 00:00:00 2001 From: Neal Cardwell Date: Tue, 30 Aug 2016 11:55:23 -0400 Subject: [PATCH 071/271] tcp: fastopen: fix rcv_wup initialization for TFO server on SYN/data Yuchung noticed that on the first TFO server data packet sent after the (TFO) handshake, the server echoed the TCP timestamp value in the SYN/data instead of the timestamp value in the final ACK of the handshake. This problem did not happen on regular opens. The tcp_replace_ts_recent() logic that decides whether to remember an incoming TS value needs tp->rcv_wup to hold the latest receive sequence number that we have ACKed (latest tp->rcv_nxt we have ACKed). This commit fixes this issue by ensuring that a TFO server properly updates tp->rcv_wup to match tp->rcv_nxt at the time it sends a SYN/ACK for the SYN/data. Reported-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: Yuchung Cheng Signed-off-by: Eric Dumazet Signed-off-by: Soheil Hassas Yeganeh Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path") Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 54d9f9b0120f..62a5751d4fe1 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -226,6 +226,7 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk, tcp_fastopen_add_skb(child, skb); tcp_rsk(req)->rcv_nxt = tp->rcv_nxt; + tp->rcv_wup = tp->rcv_nxt; /* tcp_conn_request() is sending the SYNACK, * and queues the child into listener accept queue. */ From 635c223cfa05af9523146b2f37e119d945f449ae Mon Sep 17 00:00:00 2001 From: Gao Feng Date: Wed, 31 Aug 2016 14:15:05 +0800 Subject: [PATCH 072/271] rps: flow_dissector: Fix uninitialized flow_keys used in __skb_get_hash possibly The original codes depend on that the function parameters are evaluated from left to right. But the parameter's evaluation order is not defined in C standard actually. When flow_keys_have_l4(&keys) is invoked before ___skb_get_hash(skb, &keys, hashrnd) with some compilers or environment, the keys passed to flow_keys_have_l4 is not initialized. Fixes: 6db61d79c1e1 ("flow_dissector: Ignore flow dissector return value from ___skb_get_hash") Acked-by: Eric Dumazet Signed-off-by: Gao Feng Signed-off-by: David S. Miller --- net/core/flow_dissector.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 61ad43f61c5e..52742a02814f 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -680,11 +680,13 @@ EXPORT_SYMBOL_GPL(__skb_get_hash_symmetric); void __skb_get_hash(struct sk_buff *skb) { struct flow_keys keys; + u32 hash; __flow_hash_secret_init(); - __skb_set_sw_hash(skb, ___skb_get_hash(skb, &keys, hashrnd), - flow_keys_have_l4(&keys)); + hash = ___skb_get_hash(skb, &keys, hashrnd); + + __skb_set_sw_hash(skb, hash, flow_keys_have_l4(&keys)); } EXPORT_SYMBOL(__skb_get_hash); From b44e108b6f322eb5f20aa6eba39b468a1ffc10ff Mon Sep 17 00:00:00 2001 From: "Guilherme G. Piccoli" Date: Wed, 31 Aug 2016 12:11:57 -0300 Subject: [PATCH 073/271] bnx2x: don't reset chip on cleanup if PCI function is offline When PCI error is detected, in some architectures (like PowerPC) a slot reset is performed - the driver's error handlers are in charge of "disable" device before the reset, and re-enable it after a successful slot reset. There are two cases though that another path is taken on the code: if the slot reset is not successful or if too many errors already happened in the specific adapter (meaning that possibly the device is experiencing a HW failure that slot reset is not able to solve), the core PCI error mechanism (called EEH in PowerPC) will remove the adapter from the system, since it will consider this as a permanent failure on device. In this case, a path is taken that leads to bnx2x_chip_cleanup() calling bnx2x_reset_hw(), which then tries to perform a HW reset on chip. This reset won't succeed since the HW is in a fault state, which can be seen by multiple messages on kernel log like below: bnx2x: [bnx2x_issue_dmae_with_comp:552(eth1)]DMAE timeout! bnx2x: [bnx2x_write_dmae:600(eth1)]DMAE returned failure -1 After some time, the PCI error mechanism gives up on waiting the driver's correct removal procedure and forcibly remove the adapter from the system. We can see soft lockup while core PCI error mechanism is waiting for driver to accomplish the right removal process. This patch adds a verification to avoid a chip reset whenever the function is in PCI error state - since this case is only reached when we have a device being removed because of a permanent failure, the HW chip reset is not expected to work fine neither is necessary. Also, as a minor improvement in error path, we avoid the MCP information dump in case of non-recoverable PCI error (when adapter is about to be removed), since it will certainly fail. Reported-by: Harsha Thyagaraja Signed-off-by: Guilherme G. Piccoli Acked-By: Yuval Mintz Signed-off-by: David S. Miller --- .../net/ethernet/broadcom/bnx2x/bnx2x_main.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 97e892511666..fa3386bb14f7 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -772,6 +772,11 @@ void bnx2x_fw_dump_lvl(struct bnx2x *bp, const char *lvl) (bp->common.bc_ver & 0xff00) >> 8, (bp->common.bc_ver & 0xff)); + if (pci_channel_offline(bp->pdev)) { + BNX2X_ERR("Cannot dump MCP info while in PCI error\n"); + return; + } + val = REG_RD(bp, MCP_REG_MCPR_CPU_PROGRAM_COUNTER); if (val == REG_RD(bp, MCP_REG_MCPR_CPU_PROGRAM_COUNTER)) BNX2X_ERR("%s" "MCP PC at 0x%x\n", lvl, val); @@ -9415,10 +9420,16 @@ unload_error: /* Release IRQs */ bnx2x_free_irq(bp); - /* Reset the chip */ - rc = bnx2x_reset_hw(bp, reset_code); - if (rc) - BNX2X_ERR("HW_RESET failed\n"); + /* Reset the chip, unless PCI function is offline. If we reach this + * point following a PCI error handling, it means device is really + * in a bad state and we're about to remove it, so reset the chip + * is not a good idea. + */ + if (!pci_channel_offline(bp->pdev)) { + rc = bnx2x_reset_hw(bp, reset_code); + if (rc) + BNX2X_ERR("HW_RESET failed\n"); + } /* Report UNLOAD_DONE to MCP */ bnx2x_send_unload_done(bp, keep_link); From ab34380162cbc9b5172afdadf5136643c687bb73 Mon Sep 17 00:00:00 2001 From: Eli Cooper Date: Fri, 26 Aug 2016 23:52:29 +0800 Subject: [PATCH 074/271] ipv6: Don't unset flowi6_proto in ipxip6_tnl_xmit() Commit 8eb30be0352d0916 ("ipv6: Create ip6_tnl_xmit") unsets flowi6_proto in ip4ip6_tnl_xmit() and ip6ip6_tnl_xmit(). Since xfrm_selector_match() relies on this info, IPv6 packets sent by an ip6tunnel cannot be properly selected by their protocols after removing it. This patch puts flowi6_proto back. Cc: stable@vger.kernel.org Fixes: 8eb30be0352d ("ipv6: Create ip6_tnl_xmit") Signed-off-by: Eli Cooper Signed-off-by: David S. Miller --- net/ipv6/ip6_tunnel.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 7b0481e3738f..888543debe4e 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1174,6 +1174,7 @@ ip4ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) encap_limit = t->parms.encap_limit; memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6)); + fl6.flowi6_proto = IPPROTO_IPIP; dsfield = ipv4_get_dsfield(iph); @@ -1233,6 +1234,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) encap_limit = t->parms.encap_limit; memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6)); + fl6.flowi6_proto = IPPROTO_IPV6; dsfield = ipv6_get_dsfield(ipv6h); if (t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS) From c2f321126e31cd69365e65ecd4a7c774e4fc71d2 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Tue, 30 Aug 2016 21:50:22 +0200 Subject: [PATCH 075/271] ARM: shmobile: fix regulator quirk for Gen2 The current implementation only works if the da9xxx devices are added before their drivers are registered. Only then it can apply the fixes to both devices. Otherwise, the driver for the first device gets probed before the fix for the second device can be applied. This is what fails when using the IP core switcher or when having the i2c master driver as a module. So, we need to disable both da9xxx once we detected one of them. We now use i2c_transfer with hardcoded i2c_messages and device addresses, so we don't need the da9xxx client devices to be instantiated. Because the fixup is used on specific boards only, the addresses are not going to change. Fixes: 663fbb52159cca ("ARM: shmobile: R-Car Gen2: Add da9063/da9210 regulator quirk") Signed-off-by: Wolfram Sang Reviewed-by: Geert Uytterhoeven Tested-by: Geert Uytterhoeven (r8a7791/koelsch) Tested-by: Kuninori Morimoto Signed-off-by: Simon Horman --- .../mach-shmobile/regulator-quirk-rcar-gen2.c | 62 ++++++++----------- 1 file changed, 26 insertions(+), 36 deletions(-) diff --git a/arch/arm/mach-shmobile/regulator-quirk-rcar-gen2.c b/arch/arm/mach-shmobile/regulator-quirk-rcar-gen2.c index 62437b57813e..73e3adbc1330 100644 --- a/arch/arm/mach-shmobile/regulator-quirk-rcar-gen2.c +++ b/arch/arm/mach-shmobile/regulator-quirk-rcar-gen2.c @@ -41,40 +41,27 @@ #define REGULATOR_IRQ_MASK BIT(2) /* IRQ2, active low */ +/* start of DA9210 System Control and Event Registers */ +#define DA9210_REG_MASK_A 0x54 + static void __iomem *irqc; -static const u8 da9063_mask_regs[] = { - DA9063_REG_IRQ_MASK_A, - DA9063_REG_IRQ_MASK_B, - DA9063_REG_IRQ_MASK_C, - DA9063_REG_IRQ_MASK_D, +/* first byte sets the memory pointer, following are consecutive reg values */ +static u8 da9063_irq_clr[] = { DA9063_REG_IRQ_MASK_A, 0xff, 0xff, 0xff, 0xff }; +static u8 da9210_irq_clr[] = { DA9210_REG_MASK_A, 0xff, 0xff }; + +static struct i2c_msg da9xxx_msgs[2] = { + { + .addr = 0x58, + .len = ARRAY_SIZE(da9063_irq_clr), + .buf = da9063_irq_clr, + }, { + .addr = 0x68, + .len = ARRAY_SIZE(da9210_irq_clr), + .buf = da9210_irq_clr, + }, }; -/* DA9210 System Control and Event Registers */ -#define DA9210_REG_MASK_A 0x54 -#define DA9210_REG_MASK_B 0x55 - -static const u8 da9210_mask_regs[] = { - DA9210_REG_MASK_A, - DA9210_REG_MASK_B, -}; - -static void da9xxx_mask_irqs(struct i2c_client *client, const u8 regs[], - unsigned int nregs) -{ - unsigned int i; - - dev_info(&client->dev, "Masking %s interrupt sources\n", client->name); - - for (i = 0; i < nregs; i++) { - int error = i2c_smbus_write_byte_data(client, regs[i], ~0); - if (error) { - dev_err(&client->dev, "i2c error %d\n", error); - return; - } - } -} - static int regulator_quirk_notify(struct notifier_block *nb, unsigned long action, void *data) { @@ -93,12 +80,15 @@ static int regulator_quirk_notify(struct notifier_block *nb, client = to_i2c_client(dev); dev_dbg(dev, "Detected %s\n", client->name); - if ((client->addr == 0x58 && !strcmp(client->name, "da9063"))) - da9xxx_mask_irqs(client, da9063_mask_regs, - ARRAY_SIZE(da9063_mask_regs)); - else if (client->addr == 0x68 && !strcmp(client->name, "da9210")) - da9xxx_mask_irqs(client, da9210_mask_regs, - ARRAY_SIZE(da9210_mask_regs)); + if ((client->addr == 0x58 && !strcmp(client->name, "da9063")) || + (client->addr == 0x68 && !strcmp(client->name, "da9210"))) { + int ret; + + dev_info(&client->dev, "clearing da9063/da9210 interrupts\n"); + ret = i2c_transfer(client->adapter, da9xxx_msgs, ARRAY_SIZE(da9xxx_msgs)); + if (ret != ARRAY_SIZE(da9xxx_msgs)) + dev_err(&client->dev, "i2c error %d\n", ret); + } mon = ioread32(irqc + IRQC_MONITOR); if (mon & REGULATOR_IRQ_MASK) From db7b542e4a789156a4fb61278f2074795fe2eb9c Mon Sep 17 00:00:00 2001 From: Miaoqing Pan Date: Thu, 4 Aug 2016 15:48:34 +0800 Subject: [PATCH 076/271] ath9k: fix AR5416 access GPIO warning The warning was seen on AR5416 chip, which invoke ath9k_hw_gio_get() before the GPIO initialized correctly. WARNING: CPU: 1 PID: 1159 at ~/drivers/net/wireless/ath/ath9k/hw.c:2776 ath9k_hw_gpio_get+0x148/0x1a0 [ath9k_hw] ... CPU: 1 PID: 1159 Comm: systemd-udevd Not tainted 4.7.0-rc7-aptosid-amd64 #1 aptosid 4.7~rc7-1~git92.slh.3 Hardware name: /DH67CL, BIOS BLH6710H.86A.0160.2012.1204.1156 12/04/2012 0000000000000286 00000000f912d633 ffffffff81290fd3 0000000000000000 0000000000000000 ffffffff81063fd4 ffff88040c6dc018 0000000000000000 0000000000000002 0000000000000000 0000000000000100 ffff88040c6dc018 Call Trace: [] ? dump_stack+0x5c/0x79 [] ? __warn+0xb4/0xd0 [] ? ath9k_hw_gpio_get+0x148/0x1a0 [ath9k_hw] Signed-off-by: Miaoqing Pan Reported-by: Stefan Lippers-Hollmann Tested-by: Stefan Lippers-Hollmann Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath9k/hw.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index d1d0c06d627c..14b13f07cd1f 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -2482,6 +2482,8 @@ int ath9k_hw_fill_cap_info(struct ath_hw *ah) return -EINVAL; } + ath9k_gpio_cap_init(ah); + if (AR_SREV_9485(ah) || AR_SREV_9285(ah) || AR_SREV_9330(ah) || @@ -2531,8 +2533,6 @@ int ath9k_hw_fill_cap_info(struct ath_hw *ah) else pCap->hw_caps &= ~ATH9K_HW_CAP_HT; - ath9k_gpio_cap_init(ah); - if (AR_SREV_9160_10_OR_LATER(ah) || AR_SREV_9100(ah)) pCap->rts_aggr_limit = ATH_AMPDU_LIMIT_MAX; else From 2f86953e7436c9b9a4690909c5e2db24799e173b Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Fri, 2 Sep 2016 10:22:54 +0200 Subject: [PATCH 077/271] l2tp: fix use-after-free during module unload Tunnel deletion is delayed by both a workqueue (l2tp_tunnel_delete -> wq -> l2tp_tunnel_del_work) and RCU (sk_destruct -> RCU -> l2tp_tunnel_destruct). By the time l2tp_tunnel_destruct() runs to destroy the tunnel and finish destroying the socket, the private data reserved via the net_generic mechanism has already been freed, but l2tp_tunnel_destruct() actually uses this data. Make sure tunnel deletion for the netns has completed before returning from l2tp_exit_net() by first flushing the tunnel removal workqueue, and then waiting for RCU callbacks to complete. Fixes: 167eb17e0b17 ("l2tp: create tunnel sockets in the right namespace") Signed-off-by: Sabrina Dubroca Signed-off-by: David S. Miller --- net/l2tp/l2tp_core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 1e40dacaa137..a2ed3bda4ddc 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1855,6 +1855,9 @@ static __net_exit void l2tp_exit_net(struct net *net) (void)l2tp_tunnel_delete(tunnel); } rcu_read_unlock_bh(); + + flush_workqueue(l2tp_wq); + rcu_barrier(); } static struct pernet_operations l2tp_net_ops = { From 1358bd5a7477b346dfb6b502051d61f29b11a200 Mon Sep 17 00:00:00 2001 From: Jeremy Linton Date: Thu, 1 Sep 2016 15:15:06 -0500 Subject: [PATCH 078/271] net: smsc911x: Remove multiple exit points from smsc911x_open Rework the error handling in smsc911x open in preparation for the mdio startup being moved here. Signed-off-by: Jeremy Linton Signed-off-by: David S. Miller --- drivers/net/ethernet/smsc/smsc911x.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index ca3134540d2d..c9b0e055c62f 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1520,17 +1520,20 @@ static int smsc911x_open(struct net_device *dev) unsigned int timeout; unsigned int temp; unsigned int intcfg; + int retval; /* if the phy is not yet registered, retry later*/ if (!dev->phydev) { SMSC_WARN(pdata, hw, "phy_dev is NULL"); - return -EAGAIN; + retval = -EAGAIN; + goto out; } /* Reset the LAN911x */ - if (smsc911x_soft_reset(pdata)) { + retval = smsc911x_soft_reset(pdata); + if (retval) { SMSC_WARN(pdata, hw, "soft reset failed"); - return -EIO; + goto out; } smsc911x_reg_write(pdata, HW_CFG, 0x00050000); @@ -1600,7 +1603,8 @@ static int smsc911x_open(struct net_device *dev) if (!pdata->software_irq_signal) { netdev_warn(dev, "ISR failed signaling test (IRQ %d)\n", dev->irq); - return -ENODEV; + retval = -ENODEV; + goto out; } SMSC_TRACE(pdata, ifup, "IRQ handler passed test using IRQ %d", dev->irq); @@ -1646,6 +1650,8 @@ static int smsc911x_open(struct net_device *dev) netif_start_queue(dev); return 0; +out: + return retval; } /* Entry point for stopping the interface */ From aea95dd52db436f406f3f45a455a710774a3a210 Mon Sep 17 00:00:00 2001 From: Jeremy Linton Date: Thu, 1 Sep 2016 15:15:07 -0500 Subject: [PATCH 079/271] net: smsc911x: Fix register_netdev, phy startup, driver unload ordering Move phy startup/shutdown into the smsc911x_open/stop routines. This allows the module to be unloaded because phy_connect_direct is no longer always holding the module use count. This one change also resolves a number of other problems. The link status of a downed interface no longer reflects a stale state. Errors caused by the net device being opened before the mdio/phy was configured. There is also a potential power savings as the phy's don't remain powered when the interface isn't running. Signed-off-by: Jeremy Linton Signed-off-by: David S. Miller --- drivers/net/ethernet/smsc/smsc911x.c | 48 ++++++++++++++-------------- 1 file changed, 24 insertions(+), 24 deletions(-) diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index c9b0e055c62f..823ad3f3716d 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1099,15 +1099,8 @@ static int smsc911x_mii_init(struct platform_device *pdev, goto err_out_free_bus_2; } - if (smsc911x_mii_probe(dev) < 0) { - SMSC_WARN(pdata, probe, "Error registering mii bus"); - goto err_out_unregister_bus_3; - } - return 0; -err_out_unregister_bus_3: - mdiobus_unregister(pdata->mii_bus); err_out_free_bus_2: mdiobus_free(pdata->mii_bus); err_out_1: @@ -1522,18 +1515,20 @@ static int smsc911x_open(struct net_device *dev) unsigned int intcfg; int retval; - /* if the phy is not yet registered, retry later*/ + /* find and start the given phy */ if (!dev->phydev) { - SMSC_WARN(pdata, hw, "phy_dev is NULL"); - retval = -EAGAIN; - goto out; + retval = smsc911x_mii_probe(dev); + if (retval < 0) { + SMSC_WARN(pdata, probe, "Error starting phy"); + goto out; + } } /* Reset the LAN911x */ retval = smsc911x_soft_reset(pdata); if (retval) { SMSC_WARN(pdata, hw, "soft reset failed"); - goto out; + goto mii_free_out; } smsc911x_reg_write(pdata, HW_CFG, 0x00050000); @@ -1604,7 +1599,7 @@ static int smsc911x_open(struct net_device *dev) netdev_warn(dev, "ISR failed signaling test (IRQ %d)\n", dev->irq); retval = -ENODEV; - goto out; + goto mii_free_out; } SMSC_TRACE(pdata, ifup, "IRQ handler passed test using IRQ %d", dev->irq); @@ -1650,6 +1645,10 @@ static int smsc911x_open(struct net_device *dev) netif_start_queue(dev); return 0; + +mii_free_out: + phy_disconnect(dev->phydev); + dev->phydev = NULL; out: return retval; } @@ -1674,8 +1673,12 @@ static int smsc911x_stop(struct net_device *dev) smsc911x_tx_update_txcounters(dev); /* Bring the PHY down */ - if (dev->phydev) + if (dev->phydev) { phy_stop(dev->phydev); + phy_disconnect(dev->phydev); + dev->phydev = NULL; + } + netif_carrier_off(dev); SMSC_TRACE(pdata, ifdown, "Interface stopped"); return 0; @@ -2297,11 +2300,10 @@ static int smsc911x_drv_remove(struct platform_device *pdev) pdata = netdev_priv(dev); BUG_ON(!pdata); BUG_ON(!pdata->ioaddr); - BUG_ON(!dev->phydev); + WARN_ON(dev->phydev); SMSC_TRACE(pdata, ifdown, "Stopping driver"); - phy_disconnect(dev->phydev); mdiobus_unregister(pdata->mii_bus); mdiobus_free(pdata->mii_bus); @@ -2500,6 +2502,12 @@ static int smsc911x_drv_probe(struct platform_device *pdev) netif_carrier_off(dev); + retval = smsc911x_mii_init(pdev, dev); + if (retval) { + SMSC_WARN(pdata, probe, "Error %i initialising mii", retval); + goto out_free_irq; + } + retval = register_netdev(dev); if (retval) { SMSC_WARN(pdata, probe, "Error %i registering device", retval); @@ -2509,12 +2517,6 @@ static int smsc911x_drv_probe(struct platform_device *pdev) "Network interface: \"%s\"", dev->name); } - retval = smsc911x_mii_init(pdev, dev); - if (retval) { - SMSC_WARN(pdata, probe, "Error %i initialising mii", retval); - goto out_unregister_netdev_5; - } - spin_lock_irq(&pdata->mac_lock); /* Check if mac address has been specified when bringing interface up */ @@ -2550,8 +2552,6 @@ static int smsc911x_drv_probe(struct platform_device *pdev) return 0; -out_unregister_netdev_5: - unregister_netdev(dev); out_free_irq: free_irq(dev->irq, dev); out_disable_resources: From a85f00c36ef53ea3cb5ebf3dee4ce9cc6726671f Mon Sep 17 00:00:00 2001 From: Jeremy Linton Date: Thu, 1 Sep 2016 15:15:08 -0500 Subject: [PATCH 080/271] net: smsc911x: Move interrupt handler before open In preparation for the allocating/enabling interrupts in the ndo_open routine move the irq handler before it. Signed-off-by: Jeremy Linton Signed-off-by: David S. Miller --- drivers/net/ethernet/smsc/smsc911x.c | 122 +++++++++++++-------------- 1 file changed, 61 insertions(+), 61 deletions(-) diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 823ad3f3716d..c2e56f06713a 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1507,6 +1507,67 @@ static void smsc911x_disable_irq_chip(struct net_device *dev) smsc911x_reg_write(pdata, INT_STS, 0xFFFFFFFF); } +static irqreturn_t smsc911x_irqhandler(int irq, void *dev_id) +{ + struct net_device *dev = dev_id; + struct smsc911x_data *pdata = netdev_priv(dev); + u32 intsts = smsc911x_reg_read(pdata, INT_STS); + u32 inten = smsc911x_reg_read(pdata, INT_EN); + int serviced = IRQ_NONE; + u32 temp; + + if (unlikely(intsts & inten & INT_STS_SW_INT_)) { + temp = smsc911x_reg_read(pdata, INT_EN); + temp &= (~INT_EN_SW_INT_EN_); + smsc911x_reg_write(pdata, INT_EN, temp); + smsc911x_reg_write(pdata, INT_STS, INT_STS_SW_INT_); + pdata->software_irq_signal = 1; + smp_wmb(); + serviced = IRQ_HANDLED; + } + + if (unlikely(intsts & inten & INT_STS_RXSTOP_INT_)) { + /* Called when there is a multicast update scheduled and + * it is now safe to complete the update */ + SMSC_TRACE(pdata, intr, "RX Stop interrupt"); + smsc911x_reg_write(pdata, INT_STS, INT_STS_RXSTOP_INT_); + if (pdata->multicast_update_pending) + smsc911x_rx_multicast_update_workaround(pdata); + serviced = IRQ_HANDLED; + } + + if (intsts & inten & INT_STS_TDFA_) { + temp = smsc911x_reg_read(pdata, FIFO_INT); + temp |= FIFO_INT_TX_AVAIL_LEVEL_; + smsc911x_reg_write(pdata, FIFO_INT, temp); + smsc911x_reg_write(pdata, INT_STS, INT_STS_TDFA_); + netif_wake_queue(dev); + serviced = IRQ_HANDLED; + } + + if (unlikely(intsts & inten & INT_STS_RXE_)) { + SMSC_TRACE(pdata, intr, "RX Error interrupt"); + smsc911x_reg_write(pdata, INT_STS, INT_STS_RXE_); + serviced = IRQ_HANDLED; + } + + if (likely(intsts & inten & INT_STS_RSFL_)) { + if (likely(napi_schedule_prep(&pdata->napi))) { + /* Disable Rx interrupts */ + temp = smsc911x_reg_read(pdata, INT_EN); + temp &= (~INT_EN_RSFL_EN_); + smsc911x_reg_write(pdata, INT_EN, temp); + /* Schedule a NAPI poll */ + __napi_schedule(&pdata->napi); + } else { + SMSC_WARN(pdata, rx_err, "napi_schedule_prep failed"); + } + serviced = IRQ_HANDLED; + } + + return serviced; +} + static int smsc911x_open(struct net_device *dev) { struct smsc911x_data *pdata = netdev_priv(dev); @@ -1820,67 +1881,6 @@ static void smsc911x_set_multicast_list(struct net_device *dev) spin_unlock_irqrestore(&pdata->mac_lock, flags); } -static irqreturn_t smsc911x_irqhandler(int irq, void *dev_id) -{ - struct net_device *dev = dev_id; - struct smsc911x_data *pdata = netdev_priv(dev); - u32 intsts = smsc911x_reg_read(pdata, INT_STS); - u32 inten = smsc911x_reg_read(pdata, INT_EN); - int serviced = IRQ_NONE; - u32 temp; - - if (unlikely(intsts & inten & INT_STS_SW_INT_)) { - temp = smsc911x_reg_read(pdata, INT_EN); - temp &= (~INT_EN_SW_INT_EN_); - smsc911x_reg_write(pdata, INT_EN, temp); - smsc911x_reg_write(pdata, INT_STS, INT_STS_SW_INT_); - pdata->software_irq_signal = 1; - smp_wmb(); - serviced = IRQ_HANDLED; - } - - if (unlikely(intsts & inten & INT_STS_RXSTOP_INT_)) { - /* Called when there is a multicast update scheduled and - * it is now safe to complete the update */ - SMSC_TRACE(pdata, intr, "RX Stop interrupt"); - smsc911x_reg_write(pdata, INT_STS, INT_STS_RXSTOP_INT_); - if (pdata->multicast_update_pending) - smsc911x_rx_multicast_update_workaround(pdata); - serviced = IRQ_HANDLED; - } - - if (intsts & inten & INT_STS_TDFA_) { - temp = smsc911x_reg_read(pdata, FIFO_INT); - temp |= FIFO_INT_TX_AVAIL_LEVEL_; - smsc911x_reg_write(pdata, FIFO_INT, temp); - smsc911x_reg_write(pdata, INT_STS, INT_STS_TDFA_); - netif_wake_queue(dev); - serviced = IRQ_HANDLED; - } - - if (unlikely(intsts & inten & INT_STS_RXE_)) { - SMSC_TRACE(pdata, intr, "RX Error interrupt"); - smsc911x_reg_write(pdata, INT_STS, INT_STS_RXE_); - serviced = IRQ_HANDLED; - } - - if (likely(intsts & inten & INT_STS_RSFL_)) { - if (likely(napi_schedule_prep(&pdata->napi))) { - /* Disable Rx interrupts */ - temp = smsc911x_reg_read(pdata, INT_EN); - temp &= (~INT_EN_RSFL_EN_); - smsc911x_reg_write(pdata, INT_EN, temp); - /* Schedule a NAPI poll */ - __napi_schedule(&pdata->napi); - } else { - SMSC_WARN(pdata, rx_err, "napi_schedule_prep failed"); - } - serviced = IRQ_HANDLED; - } - - return serviced; -} - #ifdef CONFIG_NET_POLL_CONTROLLER static void smsc911x_poll_controller(struct net_device *dev) { From f252974eaa64f64b940894f24bfa162a8e7f6b0d Mon Sep 17 00:00:00 2001 From: Jeremy Linton Date: Thu, 1 Sep 2016 15:15:09 -0500 Subject: [PATCH 081/271] net: smsc911x: Move interrupt allocation to open/stop The /proc/irq/xx information is incorrect for smsc911x because the request_irq is happening before the register_netdev has the proper device name. Moving it to the open also fixes the case of when the device is renamed. Reported-by: Will Deacon Signed-off-by: Jeremy Linton Tested-by: Will Deacon Signed-off-by: David S. Miller --- drivers/net/ethernet/smsc/smsc911x.c | 47 +++++++++++----------------- 1 file changed, 18 insertions(+), 29 deletions(-) diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index c2e56f06713a..4f8910b7db2e 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1575,6 +1575,7 @@ static int smsc911x_open(struct net_device *dev) unsigned int temp; unsigned int intcfg; int retval; + int irq_flags; /* find and start the given phy */ if (!dev->phydev) { @@ -1645,6 +1646,15 @@ static int smsc911x_open(struct net_device *dev) pdata->software_irq_signal = 0; smp_wmb(); + irq_flags = irq_get_trigger_type(dev->irq); + retval = request_irq(dev->irq, smsc911x_irqhandler, + irq_flags | IRQF_SHARED, dev->name, dev); + if (retval) { + SMSC_WARN(pdata, probe, + "Unable to claim requested irq: %d", dev->irq); + goto mii_free_out; + } + temp = smsc911x_reg_read(pdata, INT_EN); temp |= INT_EN_SW_INT_EN_; smsc911x_reg_write(pdata, INT_EN, temp); @@ -1660,7 +1670,7 @@ static int smsc911x_open(struct net_device *dev) netdev_warn(dev, "ISR failed signaling test (IRQ %d)\n", dev->irq); retval = -ENODEV; - goto mii_free_out; + goto irq_stop_out; } SMSC_TRACE(pdata, ifup, "IRQ handler passed test using IRQ %d", dev->irq); @@ -1707,6 +1717,8 @@ static int smsc911x_open(struct net_device *dev) netif_start_queue(dev); return 0; +irq_stop_out: + free_irq(dev->irq, dev); mii_free_out: phy_disconnect(dev->phydev); dev->phydev = NULL; @@ -1733,6 +1745,8 @@ static int smsc911x_stop(struct net_device *dev) dev->stats.rx_dropped += smsc911x_reg_read(pdata, RX_DROP); smsc911x_tx_update_txcounters(dev); + free_irq(dev->irq, dev); + /* Bring the PHY down */ if (dev->phydev) { phy_stop(dev->phydev); @@ -2308,7 +2322,6 @@ static int smsc911x_drv_remove(struct platform_device *pdev) mdiobus_free(pdata->mii_bus); unregister_netdev(dev); - free_irq(dev->irq, dev); res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smsc911x-memory"); if (!res) @@ -2393,8 +2406,7 @@ static int smsc911x_drv_probe(struct platform_device *pdev) struct smsc911x_data *pdata; struct smsc911x_platform_config *config = dev_get_platdata(&pdev->dev); struct resource *res; - unsigned int intcfg = 0; - int res_size, irq, irq_flags; + int res_size, irq; int retval; res = platform_get_resource_byname(pdev, IORESOURCE_MEM, @@ -2433,7 +2445,6 @@ static int smsc911x_drv_probe(struct platform_device *pdev) pdata = netdev_priv(dev); dev->irq = irq; - irq_flags = irq_get_trigger_type(irq); pdata->ioaddr = ioremap_nocache(res->start, res_size); pdata->dev = dev; @@ -2480,38 +2491,18 @@ static int smsc911x_drv_probe(struct platform_device *pdev) if (retval < 0) goto out_disable_resources; - /* configure irq polarity and type before connecting isr */ - if (pdata->config.irq_polarity == SMSC911X_IRQ_POLARITY_ACTIVE_HIGH) - intcfg |= INT_CFG_IRQ_POL_; - - if (pdata->config.irq_type == SMSC911X_IRQ_TYPE_PUSH_PULL) - intcfg |= INT_CFG_IRQ_TYPE_; - - smsc911x_reg_write(pdata, INT_CFG, intcfg); - - /* Ensure interrupts are globally disabled before connecting ISR */ - smsc911x_disable_irq_chip(dev); - - retval = request_irq(dev->irq, smsc911x_irqhandler, - irq_flags | IRQF_SHARED, dev->name, dev); - if (retval) { - SMSC_WARN(pdata, probe, - "Unable to claim requested irq: %d", dev->irq); - goto out_disable_resources; - } - netif_carrier_off(dev); retval = smsc911x_mii_init(pdev, dev); if (retval) { SMSC_WARN(pdata, probe, "Error %i initialising mii", retval); - goto out_free_irq; + goto out_disable_resources; } retval = register_netdev(dev); if (retval) { SMSC_WARN(pdata, probe, "Error %i registering device", retval); - goto out_free_irq; + goto out_disable_resources; } else { SMSC_TRACE(pdata, probe, "Network interface: \"%s\"", dev->name); @@ -2552,8 +2543,6 @@ static int smsc911x_drv_probe(struct platform_device *pdev) return 0; -out_free_irq: - free_irq(dev->irq, dev); out_disable_resources: pm_runtime_put(&pdev->dev); pm_runtime_disable(&pdev->dev); From a41bd25ae67d3e4052c7f00ee9f2b4ba9219309e Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 25 Aug 2016 18:42:35 +0200 Subject: [PATCH 082/271] sunrpc: fix UDP memory accounting The commit f9b2ee714c5c ("SUNRPC: Move UDP receive data path into a workqueue context"), as a side effect, moved the skb_free_datagram() call outside the scope of the related socket lock, but UDP sockets require such lock to be held for proper memory accounting. Fix it by replacing skb_free_datagram() with skb_free_datagram_locked(). Fixes: f9b2ee714c5c ("SUNRPC: Move UDP receive data path into a workqueue context") Reported-and-tested-by: Jan Stancek Signed-off-by: Paolo Abeni Cc: stable@vger.kernel.org # 4.4+ Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 8ede3bc52481..bf168838a029 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1074,7 +1074,7 @@ static void xs_udp_data_receive(struct sock_xprt *transport) skb = skb_recv_datagram(sk, 0, 1, &err); if (skb != NULL) { xs_udp_data_read_skb(&transport->xprt, sk, skb); - skb_free_datagram(sk, skb); + skb_free_datagram_locked(sk, skb); continue; } if (!test_and_clear_bit(XPRT_SOCK_DATA_READY, &transport->sock_state)) From c49edecd513693ea7530ab18efbd7d6d5b7cbf90 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 3 Sep 2016 12:05:31 -0400 Subject: [PATCH 083/271] NFS: Fix error reporting in nfs_file_write() When doing O_DSYNC writes, the actual write errors are reported through generic_write_sync(), so we must test the result. Reported-by: J. R. Okajima Fixes: 18290650b1c8 ("NFS: Move buffered I/O locking into nfs_file_write()") Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 7d620970f2e1..ca699ddc11c1 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -657,7 +657,10 @@ ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) if (result <= 0) goto out; - written = generic_write_sync(iocb, result); + result = generic_write_sync(iocb, result); + if (result < 0) + goto out; + written = result; iocb->ki_pos += written; /* Return error values */ From bf0291dd2267a2b9a4cd74d65249553d11bb45d6 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 3 Sep 2016 10:39:51 -0400 Subject: [PATCH 084/271] pNFS: Ensure LAYOUTGET and LAYOUTRETURN are properly serialised According to RFC5661, the client is responsible for serialising LAYOUTGET and LAYOUTRETURN to avoid ambiguity. Consider the case where we send both in parallel. Client Server ====== ====== LAYOUTGET(seqid=X) LAYOUTRETURN(seqid=X) LAYOUTGET return seqid=X+1 LAYOUTRETURN return seqid=X+2 Process LAYOUTRETURN Forget layout stateid Process LAYOUTGET Set seqid=X+1 The client processes the layoutget/layoutreturn in the wrong order, and since the result of the layoutreturn was to clear the only existing layout segment, the client forgets the layout stateid. When the LAYOUTGET comes in, it is treated as having a completely new stateid, and so the client sets the wrong sequence id... Fix is to check if there are outstanding LAYOUTGET requests before we send the LAYOUTRETURN (note that LAYOUGET will already wait if it sees an outstanding LAYOUTRETURN). Signed-off-by: Trond Myklebust Cc: stable@vger.kernel.org # v4.5+ Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 6daf034645c8..519ad320f5cd 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -899,6 +899,9 @@ pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo, nfs4_stateid *stateid, enum pnfs_iomode *iomode) { + /* Serialise LAYOUTGET/LAYOUTRETURN */ + if (atomic_read(&lo->plh_outstanding) != 0) + return false; if (test_and_set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) return false; pnfs_get_layout_hdr(lo); From 2a59a0411671ef9daf17ba21da57809c696f4119 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 3 Sep 2016 11:20:04 -0400 Subject: [PATCH 085/271] pNFS: Fix pnfs_set_layout_stateid() to clear NFS_LAYOUT_INVALID_STID If the layout was marked as invalid, we want to ensure to initialise the layout header fields correctly. Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 36 +++++++++++++++++++----------------- 1 file changed, 19 insertions(+), 17 deletions(-) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 519ad320f5cd..cd8b5fca33f6 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -768,17 +768,32 @@ pnfs_destroy_all_layouts(struct nfs_client *clp) pnfs_destroy_layouts_byclid(clp, false); } +static void +pnfs_clear_layoutreturn_info(struct pnfs_layout_hdr *lo) +{ + lo->plh_return_iomode = 0; + lo->plh_return_seq = 0; + clear_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags); +} + /* update lo->plh_stateid with new if is more recent */ void pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo, const nfs4_stateid *new, bool update_barrier) { u32 oldseq, newseq, new_barrier = 0; - bool invalid = !pnfs_layout_is_valid(lo); oldseq = be32_to_cpu(lo->plh_stateid.seqid); newseq = be32_to_cpu(new->seqid); - if (invalid || pnfs_seqid_is_newer(newseq, oldseq)) { + + if (!pnfs_layout_is_valid(lo)) { + nfs4_stateid_copy(&lo->plh_stateid, new); + lo->plh_barrier = newseq; + pnfs_clear_layoutreturn_info(lo); + clear_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags); + return; + } + if (pnfs_seqid_is_newer(newseq, oldseq)) { nfs4_stateid_copy(&lo->plh_stateid, new); /* * Because of wraparound, we want to keep the barrier @@ -790,7 +805,7 @@ pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo, const nfs4_stateid *new, new_barrier = be32_to_cpu(new->seqid); else if (new_barrier == 0) return; - if (invalid || pnfs_seqid_is_newer(new_barrier, lo->plh_barrier)) + if (pnfs_seqid_is_newer(new_barrier, lo->plh_barrier)) lo->plh_barrier = new_barrier; } @@ -886,14 +901,6 @@ void pnfs_clear_layoutreturn_waitbit(struct pnfs_layout_hdr *lo) rpc_wake_up(&NFS_SERVER(lo->plh_inode)->roc_rpcwaitq); } -static void -pnfs_clear_layoutreturn_info(struct pnfs_layout_hdr *lo) -{ - lo->plh_return_iomode = 0; - lo->plh_return_seq = 0; - clear_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags); -} - static bool pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo, nfs4_stateid *stateid, @@ -1801,16 +1808,11 @@ pnfs_layout_process(struct nfs4_layoutget *lgp) */ pnfs_mark_layout_stateid_invalid(lo, &free_me); - nfs4_stateid_copy(&lo->plh_stateid, &res->stateid); - lo->plh_barrier = be32_to_cpu(res->stateid.seqid); + pnfs_set_layout_stateid(lo, &res->stateid, true); } pnfs_get_lseg(lseg); pnfs_layout_insert_lseg(lo, lseg, &free_me); - if (!pnfs_layout_is_valid(lo)) { - pnfs_clear_layoutreturn_info(lo); - clear_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags); - } if (res->return_on_close) From 52ec7be2e27392201adf77892ba883f68df88c99 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 3 Sep 2016 11:05:28 -0400 Subject: [PATCH 086/271] pNFS: Clear out all layout segments if the server unsets lrp->res.lrs_present If the server fails to set lrp->res.lrs_present in the LAYOUTRETURN reply, then that means it believes the client holds no more layout state for that file, and that the layout stateid is now invalid. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index f5aecaabcb7c..c380d2ee4137 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -8190,10 +8190,13 @@ static void nfs4_layoutreturn_release(void *calldata) dprintk("--> %s\n", __func__); spin_lock(&lo->plh_inode->i_lock); - pnfs_mark_matching_lsegs_invalid(lo, &freeme, &lrp->args.range, - be32_to_cpu(lrp->args.stateid.seqid)); - if (lrp->res.lrs_present && pnfs_layout_is_valid(lo)) + if (lrp->res.lrs_present) { + pnfs_mark_matching_lsegs_invalid(lo, &freeme, + &lrp->args.range, + be32_to_cpu(lrp->args.stateid.seqid)); pnfs_set_layout_stateid(lo, &lrp->res.stateid, true); + } else + pnfs_mark_layout_stateid_invalid(lo, &freeme); pnfs_clear_layoutreturn_waitbit(lo); spin_unlock(&lo->plh_inode->i_lock); nfs4_sequence_free_slot(&lrp->res.seq_res); From 609e941a6bcd7ceb1cbb561941c997f6465e8698 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Thu, 1 Sep 2016 06:43:46 -0700 Subject: [PATCH 087/271] iw_cxgb4: call dev_put() on l2t allocation failure Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Signed-off-by: Steve Wise Signed-off-by: Sagi Grimberg --- drivers/infiniband/hw/cxgb4/cm.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index a3a67216bce6..57b5bb56cdbf 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -2077,8 +2077,10 @@ static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip, } ep->l2t = cxgb4_l2t_get(cdev->rdev.lldi.l2t, n, pdev, rt_tos2priority(tos)); - if (!ep->l2t) + if (!ep->l2t) { + dev_put(pdev); goto out; + } ep->mtu = pdev->mtu; ep->tx_chan = cxgb4_port_chan(pdev); ep->smac_idx = cxgb4_tp_smt_idx(adapter_type, From 37eb816c0867b1b0db273d22b530780a0a083980 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Thu, 1 Sep 2016 06:44:52 -0700 Subject: [PATCH 088/271] iw_cxgb4: block module unload until all ep resources are released Otherwise an endpoint can be still closing down causing a touch after free crash. Also WARN_ON if ulps have failed to destroy various resources during device removal. Fixes: ad61a4c7a9b7 ("iw_cxgb4: don't block in destroy_qp awaiting the last deref") Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Signed-off-by: Steve Wise Signed-off-by: Sagi Grimberg --- drivers/infiniband/hw/cxgb4/cm.c | 2 ++ drivers/infiniband/hw/cxgb4/device.c | 5 +++++ drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 1 + 3 files changed, 8 insertions(+) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 57b5bb56cdbf..5621270664f4 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -314,6 +314,8 @@ static void remove_ep_tid(struct c4iw_ep *ep) spin_lock_irqsave(&ep->com.dev->lock, flags); _remove_handle(ep->com.dev, &ep->com.dev->hwtid_idr, ep->hwtid, 0); + if (idr_is_empty(&ep->com.dev->hwtid_idr)) + wake_up(&ep->com.dev->wait); spin_unlock_irqrestore(&ep->com.dev->lock, flags); } diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index ae2e8b23d2dd..63561aa9c69e 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -872,9 +872,13 @@ static void c4iw_rdev_close(struct c4iw_rdev *rdev) static void c4iw_dealloc(struct uld_ctx *ctx) { c4iw_rdev_close(&ctx->dev->rdev); + WARN_ON_ONCE(!idr_is_empty(&ctx->dev->cqidr)); idr_destroy(&ctx->dev->cqidr); + WARN_ON_ONCE(!idr_is_empty(&ctx->dev->qpidr)); idr_destroy(&ctx->dev->qpidr); + WARN_ON_ONCE(!idr_is_empty(&ctx->dev->mmidr)); idr_destroy(&ctx->dev->mmidr); + wait_event(ctx->dev->wait, idr_is_empty(&ctx->dev->hwtid_idr)); idr_destroy(&ctx->dev->hwtid_idr); idr_destroy(&ctx->dev->stid_idr); idr_destroy(&ctx->dev->atid_idr); @@ -992,6 +996,7 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop) mutex_init(&devp->rdev.stats.lock); mutex_init(&devp->db_mutex); INIT_LIST_HEAD(&devp->db_fc_list); + init_waitqueue_head(&devp->wait); devp->avail_ird = devp->rdev.lldi.max_ird_adapter; if (c4iw_debugfs_root) { diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index f6f34a75af27..0e9cd4479bb3 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -263,6 +263,7 @@ struct c4iw_dev { struct idr stid_idr; struct list_head db_fc_list; u32 avail_ird; + wait_queue_head_t wait; }; static inline struct c4iw_dev *to_c4iw_dev(struct ib_device *ibdev) From cdbecc8d24b642b67ae79a0acc2ff18d3d0e677e Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Thu, 1 Sep 2016 09:12:25 -0700 Subject: [PATCH 089/271] nvme_rdma: keep a ref on the ctrl during delete/flush Reviewed-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Signed-off-by: Steve Wise Signed-off-by: Sagi Grimberg --- drivers/nvme/host/rdma.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index ab545fb347a0..15b0c1d025f5 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -1351,9 +1351,15 @@ static int nvme_rdma_device_unplug(struct nvme_rdma_queue *queue) ret = 1; } - /* Queue controller deletion */ + /* + * Queue controller deletion. Keep a reference until all + * work is flushed since delete_work will free the ctrl mem + */ + kref_get(&ctrl->ctrl.kref); queue_work(nvme_rdma_wq, &ctrl->delete_work); flush_work(&ctrl->delete_work); + nvme_put_ctrl(&ctrl->ctrl); + return ret; } @@ -1700,15 +1706,19 @@ static int __nvme_rdma_del_ctrl(struct nvme_rdma_ctrl *ctrl) static int nvme_rdma_del_ctrl(struct nvme_ctrl *nctrl) { struct nvme_rdma_ctrl *ctrl = to_rdma_ctrl(nctrl); - int ret; + int ret = 0; + /* + * Keep a reference until all work is flushed since + * __nvme_rdma_del_ctrl can free the ctrl mem + */ + if (!kref_get_unless_zero(&ctrl->ctrl.kref)) + return -EBUSY; ret = __nvme_rdma_del_ctrl(ctrl); - if (ret) - return ret; - - flush_work(&ctrl->delete_work); - - return 0; + if (!ret) + flush_work(&ctrl->delete_work); + nvme_put_ctrl(&ctrl->ctrl); + return ret; } static void nvme_rdma_remove_ctrl_work(struct work_struct *work) From f361e5a01ed35c0f9a00816d76a910d8a5cb4547 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Fri, 2 Sep 2016 09:01:27 -0700 Subject: [PATCH 090/271] nvme-rdma: destroy nvme queue rdma resources on connect failure After address resolution, the nvme_rdma_queue rdma resources are allocated. If rdma route resolution or the connect fails, or the controller reconnect times out and gives up, then the rdma resources need to be freed. Otherwise, rdma resources are leaked. Reviewed-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Signed-off-by: Steve Wise Signed-off-by: Sagi Grimberg --- drivers/nvme/host/rdma.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 15b0c1d025f5..a9d43f09963f 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -82,6 +82,7 @@ struct nvme_rdma_request { enum nvme_rdma_queue_flags { NVME_RDMA_Q_CONNECTED = (1 << 0), + NVME_RDMA_IB_QUEUE_ALLOCATED = (1 << 1), }; struct nvme_rdma_queue { @@ -480,9 +481,14 @@ out_err: static void nvme_rdma_destroy_queue_ib(struct nvme_rdma_queue *queue) { - struct nvme_rdma_device *dev = queue->device; - struct ib_device *ibdev = dev->dev; + struct nvme_rdma_device *dev; + struct ib_device *ibdev; + if (!test_and_clear_bit(NVME_RDMA_IB_QUEUE_ALLOCATED, &queue->flags)) + return; + + dev = queue->device; + ibdev = dev->dev; rdma_destroy_qp(queue->cm_id); ib_free_cq(queue->ib_cq); @@ -533,6 +539,7 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue, ret = -ENOMEM; goto out_destroy_qp; } + set_bit(NVME_RDMA_IB_QUEUE_ALLOCATED, &queue->flags); return 0; @@ -590,6 +597,7 @@ static int nvme_rdma_init_queue(struct nvme_rdma_ctrl *ctrl, return 0; out_destroy_cm_id: + nvme_rdma_destroy_queue_ib(queue); rdma_destroy_id(queue->cm_id); return ret; } @@ -652,7 +660,7 @@ static int nvme_rdma_init_io_queues(struct nvme_rdma_ctrl *ctrl) return 0; out_free_queues: - for (; i >= 1; i--) + for (i--; i >= 1; i--) nvme_rdma_stop_and_free_queue(&ctrl->queues[i]); return ret; From 334a8f37115bf35e38617315a360a91ac4f2b2c6 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 4 Sep 2016 12:46:35 -0400 Subject: [PATCH 091/271] pNFS: Don't forget the layout stateid if there are outstanding LAYOUTGETs If there are outstanding LAYOUTGET rpc calls, then we want to ensure that we keep the layout stateid around so we that don't inadvertently pick up an old/misordered sequence id. The race is as follows: Client Server ====== ====== LAYOUTGET(seqid) LAYOUTGET(seqid) return LAYOUTGET(seqid+1) return LAYOUTGET(seqid+2) process LAYOUTGET(seqid+2) forget layout process LAYOUTGET(seqid+1) If it forgets the layout stateid before processing seqid+1, then the client will not check the layout->plh_barrier, and so will set the stateid with seqid+1. Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index cd8b5fca33f6..2c93a85eda51 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -365,7 +365,8 @@ pnfs_layout_remove_lseg(struct pnfs_layout_hdr *lo, /* Matched by pnfs_get_layout_hdr in pnfs_layout_insert_lseg */ atomic_dec(&lo->plh_refcount); if (list_empty(&lo->plh_segs)) { - set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags); + if (atomic_read(&lo->plh_outstanding) == 0) + set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags); clear_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags); } rpc_wake_up(&NFS_SERVER(inode)->roc_rpcwaitq); From 24b27fc4cdf9e10c5e79e5923b6b7c2c5c95096c Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Thu, 1 Sep 2016 22:18:34 -0700 Subject: [PATCH 092/271] bonding: Fix bonding crash Following few steps will crash kernel - (a) Create bonding master > modprobe bonding miimon=50 (b) Create macvlan bridge on eth2 > ip link add link eth2 dev mvl0 address aa:0:0:0:0:01 \ type macvlan (c) Now try adding eth2 into the bond > echo +eth2 > /sys/class/net/bond0/bonding/slaves Bonding does lots of things before checking if the device enslaved is busy or not. In this case when the notifier call-chain sends notifications, the bond_netdev_event() assumes that the rx_handler /rx_handler_data is registered while the bond_enslave() hasn't progressed far enough to register rx_handler for the new slave. This patch adds a rx_handler check that can be performed right at the beginning of the enslave code to avoid getting into this situation. Signed-off-by: Mahesh Bandewar Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 7 ++++--- include/linux/netdevice.h | 1 + net/core/dev.c | 16 ++++++++++++++++ 3 files changed, 21 insertions(+), 3 deletions(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 217e8da0628c..9599ed6f1213 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1341,9 +1341,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) slave_dev->name); } - /* already enslaved */ - if (slave_dev->flags & IFF_SLAVE) { - netdev_dbg(bond_dev, "Error: Device was already enslaved\n"); + /* already in-use? */ + if (netdev_is_rx_handler_busy(slave_dev)) { + netdev_err(bond_dev, + "Error: Device is in use and cannot be enslaved\n"); return -EBUSY; } diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 3a788bf0affd..e8d79d4ebcfe 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3267,6 +3267,7 @@ static inline void napi_free_frags(struct napi_struct *napi) napi->skb = NULL; } +bool netdev_is_rx_handler_busy(struct net_device *dev); int netdev_rx_handler_register(struct net_device *dev, rx_handler_func_t *rx_handler, void *rx_handler_data); diff --git a/net/core/dev.c b/net/core/dev.c index dd6ce598de89..ea6312057a71 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3974,6 +3974,22 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, return skb; } +/** + * netdev_is_rx_handler_busy - check if receive handler is registered + * @dev: device to check + * + * Check if a receive handler is already registered for a given device. + * Return true if there one. + * + * The caller must hold the rtnl_mutex. + */ +bool netdev_is_rx_handler_busy(struct net_device *dev) +{ + ASSERT_RTNL(); + return dev && rtnl_dereference(dev->rx_handler); +} +EXPORT_SYMBOL_GPL(netdev_is_rx_handler_busy); + /** * netdev_rx_handler_register - register receive handler * @dev: device to register a handler for From 9b4cdd516dadc1b68c55ba24520194a06adff10c Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Fri, 2 Sep 2016 13:37:11 +0200 Subject: [PATCH 093/271] vxlan: reject multicast destination without an interface Currently, kernel accepts configurations such as: ip l a type vxlan dstport 4789 id 1 group 239.192.0.1 ip l a type vxlan dstport 4789 id 1 group ff0e::110 However, neither of those really works. In the IPv4 case, the interface cannot be brought up ("RTNETLINK answers: No such device"). This is because multicast join will be rejected without the interface being specified. In the IPv6 case, multicast wil be joined on the first interface found. This is not what the user wants as it depends on random factors (order of interfaces). Note that it's possible to add a local address but it doesn't solve anything. For IPv4, it's not considered in the multicast join (thus the same error as above is returned on ifup). This could be added but it wouldn't help for IPv6 anyway. For IPv6, we do need the interface. Just reject a configuration that sets multicast address and does not provide an interface. Nobody can depend on the previous behavior as it never worked. Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index c0dda6fc0921..6358e35d74b7 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2842,6 +2842,9 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, dev->mtu = lowerdev->mtu - (use_ipv6 ? VXLAN6_HEADROOM : VXLAN_HEADROOM); needed_headroom = lowerdev->hard_header_len; + } else if (vxlan_addr_multicast(&dst->remote_ip)) { + pr_info("multicast destination requires interface to be specified\n"); + return -EINVAL; } if (conf->mtu) { From 3555621de7fcceb79c4850f3d07d1ae4f652acf6 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Fri, 2 Sep 2016 13:37:12 +0200 Subject: [PATCH 094/271] vxlan: fix duplicated and wrong error messages vxlan_dev_configure outputs error messages before returning, no need to print again the same mesages in vxlan_newlink. Also, vxlan_dev_configure may return a particular error code for a different reason than vxlan_newlink thinks. Move the remaining error messages into vxlan_dev_configure and let vxlan_newlink just pass on the error code. Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 35 +++++++++-------------------------- 1 file changed, 9 insertions(+), 26 deletions(-) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 6358e35d74b7..6e65832051d6 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2782,14 +2782,15 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, struct net_device *lowerdev = NULL; if (conf->flags & VXLAN_F_GPE) { - if (conf->flags & ~VXLAN_F_ALLOWED_GPE) - return -EINVAL; /* For now, allow GPE only together with COLLECT_METADATA. * This can be relaxed later; in such case, the other side * of the PtP link will have to be provided. */ - if (!(conf->flags & VXLAN_F_COLLECT_METADATA)) + if ((conf->flags & ~VXLAN_F_ALLOWED_GPE) || + !(conf->flags & VXLAN_F_COLLECT_METADATA)) { + pr_info("unsupported combination of extensions\n"); return -EINVAL; + } vxlan_raw_setup(dev); } else { @@ -2877,8 +2878,10 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, tmp->cfg.saddr.sa.sa_family == AF_INET6) == use_ipv6 && tmp->cfg.dst_port == vxlan->cfg.dst_port && (tmp->flags & VXLAN_F_RCV_FLAGS) == - (vxlan->flags & VXLAN_F_RCV_FLAGS)) - return -EEXIST; + (vxlan->flags & VXLAN_F_RCV_FLAGS)) { + pr_info("duplicate VNI %u\n", be32_to_cpu(conf->vni)); + return -EEXIST; + } } dev->ethtool_ops = &vxlan_ethtool_ops; @@ -2912,7 +2915,6 @@ static int vxlan_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[], struct nlattr *data[]) { struct vxlan_config conf; - int err; memset(&conf, 0, sizeof(conf)); @@ -3021,26 +3023,7 @@ static int vxlan_newlink(struct net *src_net, struct net_device *dev, if (tb[IFLA_MTU]) conf.mtu = nla_get_u32(tb[IFLA_MTU]); - err = vxlan_dev_configure(src_net, dev, &conf); - switch (err) { - case -ENODEV: - pr_info("ifindex %d does not exist\n", conf.remote_ifindex); - break; - - case -EPERM: - pr_info("IPv6 is disabled via sysctl\n"); - break; - - case -EEXIST: - pr_info("duplicate VNI %u\n", be32_to_cpu(conf.vni)); - break; - - case -EINVAL: - pr_info("unsupported combination of extensions\n"); - break; - } - - return err; + return vxlan_dev_configure(src_net, dev, &conf); } static void vxlan_dellink(struct net_device *dev, struct list_head *head) From 38f7bd94a97b542de86a2be9229289717e33a7a4 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 1 Sep 2016 14:56:49 -0700 Subject: [PATCH 095/271] Revert "af_unix: Fix splice-bind deadlock" This reverts commit c845acb324aa85a39650a14e7696982ceea75dc1. It turns out that it just replaces one deadlock with another one: we can still get the wrong lock ordering with the readlock due to overlayfs calling back into the filesystem layer and still taking the vfs locks after the readlock. The proper solution ends up being to just split the readlock into two pieces: the bind lock (taken *outside* the vfs locks) and the IO lock (taken *inside* the filesystem locks). The two locks are independent anyway. Signed-off-by: Linus Torvalds Reviewed-by: Shmulik Ladkani Signed-off-by: David S. Miller --- net/unix/af_unix.c | 66 ++++++++++++++++++---------------------------- 1 file changed, 26 insertions(+), 40 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index f1dffe84f0d5..433ae1bbef97 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -954,20 +954,32 @@ fail: return NULL; } -static int unix_mknod(struct dentry *dentry, const struct path *path, umode_t mode, - struct path *res) +static int unix_mknod(const char *sun_path, umode_t mode, struct path *res) { - int err; + struct dentry *dentry; + struct path path; + int err = 0; + /* + * Get the parent directory, calculate the hash for last + * component. + */ + dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0); + err = PTR_ERR(dentry); + if (IS_ERR(dentry)) + return err; - err = security_path_mknod(path, dentry, mode, 0); + /* + * All right, let's create it. + */ + err = security_path_mknod(&path, dentry, mode, 0); if (!err) { - err = vfs_mknod(d_inode(path->dentry), dentry, mode, 0); + err = vfs_mknod(d_inode(path.dentry), dentry, mode, 0); if (!err) { - res->mnt = mntget(path->mnt); + res->mnt = mntget(path.mnt); res->dentry = dget(dentry); } } - + done_path_create(&path, dentry); return err; } @@ -978,12 +990,10 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) struct unix_sock *u = unix_sk(sk); struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr; char *sun_path = sunaddr->sun_path; - int err, name_err; + int err; unsigned int hash; struct unix_address *addr; struct hlist_head *list; - struct path path; - struct dentry *dentry; err = -EINVAL; if (sunaddr->sun_family != AF_UNIX) @@ -999,34 +1009,14 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out; addr_len = err; - name_err = 0; - dentry = NULL; - if (sun_path[0]) { - /* Get the parent directory, calculate the hash for last - * component. - */ - dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0); - - if (IS_ERR(dentry)) { - /* delay report until after 'already bound' check */ - name_err = PTR_ERR(dentry); - dentry = NULL; - } - } - err = mutex_lock_interruptible(&u->readlock); if (err) - goto out_path; + goto out; err = -EINVAL; if (u->addr) goto out_up; - if (name_err) { - err = name_err == -EEXIST ? -EADDRINUSE : name_err; - goto out_up; - } - err = -ENOMEM; addr = kmalloc(sizeof(*addr)+addr_len, GFP_KERNEL); if (!addr) @@ -1037,11 +1027,11 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) addr->hash = hash ^ sk->sk_type; atomic_set(&addr->refcnt, 1); - if (dentry) { - struct path u_path; + if (sun_path[0]) { + struct path path; umode_t mode = S_IFSOCK | (SOCK_INODE(sock)->i_mode & ~current_umask()); - err = unix_mknod(dentry, &path, mode, &u_path); + err = unix_mknod(sun_path, mode, &path); if (err) { if (err == -EEXIST) err = -EADDRINUSE; @@ -1049,9 +1039,9 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out_up; } addr->hash = UNIX_HASH_SIZE; - hash = d_real_inode(dentry)->i_ino & (UNIX_HASH_SIZE - 1); + hash = d_real_inode(path.dentry)->i_ino & (UNIX_HASH_SIZE - 1); spin_lock(&unix_table_lock); - u->path = u_path; + u->path = path; list = &unix_socket_table[hash]; } else { spin_lock(&unix_table_lock); @@ -1074,10 +1064,6 @@ out_unlock: spin_unlock(&unix_table_lock); out_up: mutex_unlock(&u->readlock); -out_path: - if (dentry) - done_path_create(&path, dentry); - out: return err; } From 6e1ce3c3451291142a57c4f3f6f999a29fb5b3bc Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 1 Sep 2016 14:43:53 -0700 Subject: [PATCH 096/271] af_unix: split 'u->readlock' into two: 'iolock' and 'bindlock' Right now we use the 'readlock' both for protecting some of the af_unix IO path and for making the bind be single-threaded. The two are independent, but using the same lock makes for a nasty deadlock due to ordering with regards to filesystem locking. The bind locking would want to nest outside the VSF pathname locking, but the IO locking wants to nest inside some of those same locks. We tried to fix this earlier with commit c845acb324aa ("af_unix: Fix splice-bind deadlock") which moved the readlock inside the vfs locks, but that caused problems with overlayfs that will then call back into filesystem routines that take the lock in the wrong order anyway. Splitting the locks means that we can go back to having the bind lock be the outermost lock, and we don't have any deadlocks with lock ordering. Acked-by: Rainer Weikusat Acked-by: Al Viro Signed-off-by: Linus Torvalds Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- include/net/af_unix.h | 2 +- net/unix/af_unix.c | 45 ++++++++++++++++++++++--------------------- 2 files changed, 24 insertions(+), 23 deletions(-) diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 9b4c418bebd8..fd60eccb59a6 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -52,7 +52,7 @@ struct unix_sock { struct sock sk; struct unix_address *addr; struct path path; - struct mutex readlock; + struct mutex iolock, bindlock; struct sock *peer; struct list_head link; atomic_long_t inflight; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 433ae1bbef97..8309687a56b0 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -661,11 +661,11 @@ static int unix_set_peek_off(struct sock *sk, int val) { struct unix_sock *u = unix_sk(sk); - if (mutex_lock_interruptible(&u->readlock)) + if (mutex_lock_interruptible(&u->iolock)) return -EINTR; sk->sk_peek_off = val; - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); return 0; } @@ -779,7 +779,8 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern) spin_lock_init(&u->lock); atomic_long_set(&u->inflight, 0); INIT_LIST_HEAD(&u->link); - mutex_init(&u->readlock); /* single task reading lock */ + mutex_init(&u->iolock); /* single task reading lock */ + mutex_init(&u->bindlock); /* single task binding lock */ init_waitqueue_head(&u->peer_wait); init_waitqueue_func_entry(&u->peer_wake, unix_dgram_peer_wake_relay); unix_insert_socket(unix_sockets_unbound(sk), sk); @@ -848,7 +849,7 @@ static int unix_autobind(struct socket *sock) int err; unsigned int retries = 0; - err = mutex_lock_interruptible(&u->readlock); + err = mutex_lock_interruptible(&u->bindlock); if (err) return err; @@ -895,7 +896,7 @@ retry: spin_unlock(&unix_table_lock); err = 0; -out: mutex_unlock(&u->readlock); +out: mutex_unlock(&u->bindlock); return err; } @@ -1009,7 +1010,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out; addr_len = err; - err = mutex_lock_interruptible(&u->readlock); + err = mutex_lock_interruptible(&u->bindlock); if (err) goto out; @@ -1063,7 +1064,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) out_unlock: spin_unlock(&unix_table_lock); out_up: - mutex_unlock(&u->readlock); + mutex_unlock(&u->bindlock); out: return err; } @@ -1955,17 +1956,17 @@ static ssize_t unix_stream_sendpage(struct socket *socket, struct page *page, if (false) { alloc_skb: unix_state_unlock(other); - mutex_unlock(&unix_sk(other)->readlock); + mutex_unlock(&unix_sk(other)->iolock); newskb = sock_alloc_send_pskb(sk, 0, 0, flags & MSG_DONTWAIT, &err, 0); if (!newskb) goto err; } - /* we must acquire readlock as we modify already present + /* we must acquire iolock as we modify already present * skbs in the sk_receive_queue and mess with skb->len */ - err = mutex_lock_interruptible(&unix_sk(other)->readlock); + err = mutex_lock_interruptible(&unix_sk(other)->iolock); if (err) { err = flags & MSG_DONTWAIT ? -EAGAIN : -ERESTARTSYS; goto err; @@ -2032,7 +2033,7 @@ alloc_skb: } unix_state_unlock(other); - mutex_unlock(&unix_sk(other)->readlock); + mutex_unlock(&unix_sk(other)->iolock); other->sk_data_ready(other); scm_destroy(&scm); @@ -2041,7 +2042,7 @@ alloc_skb: err_state_unlock: unix_state_unlock(other); err_unlock: - mutex_unlock(&unix_sk(other)->readlock); + mutex_unlock(&unix_sk(other)->iolock); err: kfree_skb(newskb); if (send_sigpipe && !(flags & MSG_NOSIGNAL)) @@ -2109,7 +2110,7 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); do { - mutex_lock(&u->readlock); + mutex_lock(&u->iolock); skip = sk_peek_offset(sk, flags); skb = __skb_try_recv_datagram(sk, flags, &peeked, &skip, &err, @@ -2117,14 +2118,14 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, if (skb) break; - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); if (err != -EAGAIN) break; } while (timeo && !__skb_wait_for_more_packets(sk, &err, &timeo, last)); - if (!skb) { /* implies readlock unlocked */ + if (!skb) { /* implies iolock unlocked */ unix_state_lock(sk); /* Signal EOF on disconnected non-blocking SEQPACKET socket. */ if (sk->sk_type == SOCK_SEQPACKET && err == -EAGAIN && @@ -2189,7 +2190,7 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, out_free: skb_free_datagram(sk, skb); - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); out: return err; } @@ -2284,7 +2285,7 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state) /* Lock the socket to prevent queue disordering * while sleeps in memcpy_tomsg */ - mutex_lock(&u->readlock); + mutex_lock(&u->iolock); if (flags & MSG_PEEK) skip = sk_peek_offset(sk, flags); @@ -2326,7 +2327,7 @@ again: break; } - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); timeo = unix_stream_data_wait(sk, timeo, last, last_len); @@ -2337,7 +2338,7 @@ again: goto out; } - mutex_lock(&u->readlock); + mutex_lock(&u->iolock); goto redo; unlock: unix_state_unlock(sk); @@ -2440,7 +2441,7 @@ unlock: } } while (size); - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); if (state->msg) scm_recv(sock, state->msg, &scm, flags); else @@ -2481,9 +2482,9 @@ static ssize_t skb_unix_socket_splice(struct sock *sk, int ret; struct unix_sock *u = unix_sk(sk); - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); ret = splice_to_pipe(pipe, spd); - mutex_lock(&u->readlock); + mutex_lock(&u->iolock); return ret; } From cc2187a6e037bc64404f63c6d650ff263c2200c0 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sun, 4 Sep 2016 11:37:36 +0200 Subject: [PATCH 097/271] x86/microcode/AMD: Fix load of builtin microcode with randomized memory We do not need to add the randomization offset when the microcode is built in. Reported-and-tested-by: Emanuel Czirai Signed-off-by: Borislav Petkov Link: http://lkml.kernel.org/r/20160904093736.GA11939@pd.tnic Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/microcode/amd.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c index b816971f5da4..620ab06bcf45 100644 --- a/arch/x86/kernel/cpu/microcode/amd.c +++ b/arch/x86/kernel/cpu/microcode/amd.c @@ -54,6 +54,7 @@ static LIST_HEAD(pcache); */ static u8 *container; static size_t container_size; +static bool ucode_builtin; static u32 ucode_new_rev; static u8 amd_ucode_patch[PATCH_MAX_SIZE]; @@ -281,18 +282,22 @@ static bool __init load_builtin_amd_microcode(struct cpio_data *cp, void __init load_ucode_amd_bsp(unsigned int family) { struct cpio_data cp; + bool *builtin; void **data; size_t *size; #ifdef CONFIG_X86_32 data = (void **)__pa_nodebug(&ucode_cpio.data); size = (size_t *)__pa_nodebug(&ucode_cpio.size); + builtin = (bool *)__pa_nodebug(&ucode_builtin); #else data = &ucode_cpio.data; size = &ucode_cpio.size; + builtin = &ucode_builtin; #endif - if (!load_builtin_amd_microcode(&cp, family)) + *builtin = load_builtin_amd_microcode(&cp, family); + if (!*builtin) cp = find_ucode_in_initrd(); if (!(cp.data && cp.size)) @@ -373,7 +378,8 @@ void load_ucode_amd_ap(void) return; /* Add CONFIG_RANDOMIZE_MEMORY offset. */ - cont += PAGE_OFFSET - __PAGE_OFFSET_BASE; + if (!ucode_builtin) + cont += PAGE_OFFSET - __PAGE_OFFSET_BASE; eax = cpuid_eax(0x00000001); eq = (struct equiv_cpu_entry *)(cont + CONTAINER_HDR_SZ); @@ -439,7 +445,8 @@ int __init save_microcode_in_initrd_amd(void) container = cont_va; /* Add CONFIG_RANDOMIZE_MEMORY offset. */ - container += PAGE_OFFSET - __PAGE_OFFSET_BASE; + if (!ucode_builtin) + container += PAGE_OFFSET - __PAGE_OFFSET_BASE; eax = cpuid_eax(0x00000001); eax = ((eax >> 8) & 0xf) + ((eax >> 20) & 0xff); From d7127b5e5fa0551be21b86640f1648b224e36d43 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 26 Aug 2016 08:16:00 +0200 Subject: [PATCH 098/271] locking/barriers: Don't use sizeof(void) in lockless_dereference() My previous commit: 112dc0c8069e ("locking/barriers: Suppress sparse warnings in lockless_dereference()") caused sparse to complain that (in radix-tree.h) we use sizeof(void) since that rcu_dereference()s a void *. Really, all we need is to have the expression *p in here somewhere to make sure p is a pointer type, and sizeof(*p) was the thing that came to my mind first to make sure that's done without really doing anything at runtime. Another thing I had considered was using typeof(*p), but obviously we can't just declare a typeof(*p) variable either, since that may end up being void. Declaring a variable as typeof(*p)* gets around that, and still checks that typeof(*p) is valid, so do that. This type construction can't be done for _________p1 because that will actually be used and causes sparse address space warnings, so keep a separate unused variable for it. Reported-by: Fengguang Wu Signed-off-by: Johannes Berg Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Paul E . McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: kbuild-all@01.org Fixes: 112dc0c8069e ("locking/barriers: Suppress sparse warnings in lockless_dereference()") Link: http://lkml.kernel.org/r/1472192160-4049-1-git-send-email-johannes@sipsolutions.net Signed-off-by: Ingo Molnar --- include/linux/compiler.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 436aa4e42221..668569844d37 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -527,13 +527,14 @@ static __always_inline void __write_once_size(volatile void *p, void *res, int s * object's lifetime is managed by something other than RCU. That * "something other" might be reference counting or simple immortality. * - * The seemingly unused size_t variable is to validate @p is indeed a pointer - * type by making sure it can be dereferenced. + * The seemingly unused variable ___typecheck_p validates that @p is + * indeed a pointer type by using a pointer to typeof(*p) as the type. + * Taking a pointer to typeof(*p) again is needed in case p is void *. */ #define lockless_dereference(p) \ ({ \ typeof(p) _________p1 = READ_ONCE(p); \ - size_t __maybe_unused __size_of_ptr = sizeof(*(p)); \ + typeof(*(p)) *___typecheck_p __maybe_unused; \ smp_read_barrier_depends(); /* Dependency order vs. p above. */ \ (_________p1); \ }) From 58763148758057ffc447bf990321d3ea86d199a0 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 30 Aug 2016 10:15:03 +0200 Subject: [PATCH 099/271] perf/core: Remove WARN from perf_event_read() This effectively reverts commit: 71e7bc2bab77 ("perf/core: Check return value of the perf_event_read() IPI") ... and puts in a comment explaining why we ignore the return value. Reported-by: Vegard Nossum Signed-off-by: Peter Zijlstra (Intel) Cc: David Carrillo-Cisneros Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: 71e7bc2bab77 ("perf/core: Check return value of the perf_event_read() IPI") Signed-off-by: Ingo Molnar --- kernel/events/core.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 3cfabdf7b942..07ac8596a728 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -3549,10 +3549,18 @@ static int perf_event_read(struct perf_event *event, bool group) .group = group, .ret = 0, }; - ret = smp_call_function_single(event->oncpu, __perf_event_read, &data, 1); - /* The event must have been read from an online CPU: */ - WARN_ON_ONCE(ret); - ret = ret ? : data.ret; + /* + * Purposely ignore the smp_call_function_single() return + * value. + * + * If event->oncpu isn't a valid CPU it means the event got + * scheduled out and that will have updated the event count. + * + * Therefore, either way, we'll have an up-to-date event count + * after this. + */ + (void)smp_call_function_single(event->oncpu, __perf_event_read, &data, 1); + ret = data.ret; } else if (event->state == PERF_EVENT_STATE_INACTIVE) { struct perf_event_context *ctx = event->ctx; unsigned long flags; From 135e8c9250dd5c8c9aae5984fde6f230d0cbfeaf Mon Sep 17 00:00:00 2001 From: Balbir Singh Date: Mon, 5 Sep 2016 13:16:40 +1000 Subject: [PATCH 100/271] sched/core: Fix a race between try_to_wake_up() and a woken up task The origin of the issue I've seen is related to a missing memory barrier between check for task->state and the check for task->on_rq. The task being woken up is already awake from a schedule() and is doing the following: do { schedule() set_current_state(TASK_(UN)INTERRUPTIBLE); } while (!cond); The waker, actually gets stuck doing the following in try_to_wake_up(): while (p->on_cpu) cpu_relax(); Analysis: The instance I've seen involves the following race: CPU1 CPU2 while () { if (cond) break; do { schedule(); set_current_state(TASK_UN..) } while (!cond); wakeup_routine() spin_lock_irqsave(wait_lock) raw_spin_lock_irqsave(wait_lock) wake_up_process() } try_to_wake_up() set_current_state(TASK_RUNNING); .. list_del(&waiter.list); CPU2 wakes up CPU1, but before it can get the wait_lock and set current state to TASK_RUNNING the following occurs: CPU3 wakeup_routine() raw_spin_lock_irqsave(wait_lock) if (!list_empty) wake_up_process() try_to_wake_up() raw_spin_lock_irqsave(p->pi_lock) .. if (p->on_rq && ttwu_wakeup()) .. while (p->on_cpu) cpu_relax() .. CPU3 tries to wake up the task on CPU1 again since it finds it on the wait_queue, CPU1 is spinning on wait_lock, but immediately after CPU2, CPU3 got it. CPU3 checks the state of p on CPU1, it is TASK_UNINTERRUPTIBLE and the task is spinning on the wait_lock. Interestingly since p->on_rq is checked under pi_lock, I've noticed that try_to_wake_up() finds p->on_rq to be 0. This was the most confusing bit of the analysis, but p->on_rq is changed under runqueue lock, rq_lock, the p->on_rq check is not reliable without this fix IMHO. The race is visible (based on the analysis) only when ttwu_queue() does a remote wakeup via ttwu_queue_remote. In which case the p->on_rq change is not done uder the pi_lock. The result is that after a while the entire system locks up on the raw_spin_irqlock_save(wait_lock) and the holder spins infintely Reproduction of the issue: The issue can be reproduced after a long run on my system with 80 threads and having to tweak available memory to very low and running memory stress-ng mmapfork test. It usually takes a long time to reproduce. I am trying to work on a test case that can reproduce the issue faster, but thats work in progress. I am still testing the changes on my still in a loop and the tests seem OK thus far. Big thanks to Benjamin and Nick for helping debug this as well. Ben helped catch the missing barrier, Nick caught every missing bit in my theory. Signed-off-by: Balbir Singh [ Updated comment to clarify matching barriers. Many architectures do not have a full barrier in switch_to() so that cannot be relied upon. ] Signed-off-by: Peter Zijlstra (Intel) Acked-by: Benjamin Herrenschmidt Cc: Alexey Kardashevskiy Cc: Linus Torvalds Cc: Nicholas Piggin Cc: Nicholas Piggin Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Link: http://lkml.kernel.org/r/e02cce7b-d9ca-1ad0-7a61-ea97c7582b37@gmail.com Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 2a906f20fba7..44817c640e99 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2016,6 +2016,28 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) success = 1; /* we're going to change ->state */ cpu = task_cpu(p); + /* + * Ensure we load p->on_rq _after_ p->state, otherwise it would + * be possible to, falsely, observe p->on_rq == 0 and get stuck + * in smp_cond_load_acquire() below. + * + * sched_ttwu_pending() try_to_wake_up() + * [S] p->on_rq = 1; [L] P->state + * UNLOCK rq->lock -----. + * \ + * +--- RMB + * schedule() / + * LOCK rq->lock -----' + * UNLOCK rq->lock + * + * [task p] + * [S] p->state = UNINTERRUPTIBLE [L] p->on_rq + * + * Pairs with the UNLOCK+LOCK on rq->lock from the + * last wakeup of our task and the schedule that got our task + * current. + */ + smp_rmb(); if (p->on_rq && ttwu_remote(p, wake_flags)) goto stat; From d4c4fed08f31f3746000c46cb1b20bed2959547a Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 15 Aug 2016 09:05:45 -0600 Subject: [PATCH 101/271] efi: Make for_each_efi_memory_desc_in_map() cope with running on Xen While commit 55f1ea15216 ("efi: Fix for_each_efi_memory_desc_in_map() for empty memmaps") made an attempt to deal with empty memory maps, it didn't address the case where the map field never gets set, as is apparently the case when running under Xen. Reported-by: Tested-by: Cc: Vitaly Kuznetsov Cc: Jiri Slaby Cc: Mark Rutland Cc: # v4.7+ Signed-off-by: Jan Beulich [ Guard the loop with a NULL check instead of pointer underflow ] Signed-off-by: Matt Fleming --- include/linux/efi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/efi.h b/include/linux/efi.h index 7f5a58225385..23cd3ced8c1a 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -946,7 +946,7 @@ extern int efi_memattr_apply_permissions(struct mm_struct *mm, /* Iterate through an efi_memory_map */ #define for_each_efi_memory_desc_in_map(m, md) \ for ((md) = (m)->map; \ - ((void *)(md) + (m)->desc_size) <= (m)->map_end; \ + (md) && ((void *)(md) + (m)->desc_size) <= (m)->map_end; \ (md) = (void *)(md) + (m)->desc_size) /** From 4af9ed578a50cd331a725322cfd9d555251ce788 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Tue, 30 Aug 2016 12:41:37 +0200 Subject: [PATCH 102/271] efi: Fix handling error value in fdt_find_uefi_params of_get_flat_dt_subnode_by_name can return negative value in case of error. Assigning the result to unsigned variable and checking if the variable is lesser than zero is incorrect and always false. The patch fixes it by using signed variable to check the result. The problem has been detected using semantic patch scripts/coccinelle/tests/unsigned_lesser_than_zero.cocci Signed-off-by: Andrzej Hajda Cc: Bartlomiej Zolnierkiewicz Cc: Marek Szyprowski Cc: Ingo Molnar Cc: Thomas Gleixner Cc: "H. Peter Anvin" Cc: Shawn Lin Cc: Mark Rutland Cc: Signed-off-by: Matt Fleming --- drivers/firmware/efi/efi.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c index 5a2631af7410..7dd2e2d37231 100644 --- a/drivers/firmware/efi/efi.c +++ b/drivers/firmware/efi/efi.c @@ -657,9 +657,12 @@ static int __init fdt_find_uefi_params(unsigned long node, const char *uname, } if (subnode) { - node = of_get_flat_dt_subnode_by_name(node, subnode); - if (node < 0) + int err = of_get_flat_dt_subnode_by_name(node, subnode); + + if (err < 0) return 0; + + node = err; } return __find_uefi_params(node, info, dt_params[i].params); From dadb57abc37499f565b23933dbf49b435c3ba8af Mon Sep 17 00:00:00 2001 From: Jeffrey Hugo Date: Mon, 29 Aug 2016 14:38:51 -0600 Subject: [PATCH 103/271] efi/libstub: Allocate headspace in efi_get_memory_map() efi_get_memory_map() allocates a buffer to store the memory map that it retrieves. This buffer may need to be reused by the client after ExitBootServices() is called, at which point allocations are not longer permitted. To support this usecase, provide the allocated buffer size back to the client, and allocate some additional headroom to account for any reasonable growth in the map that is likely to happen between the call to efi_get_memory_map() and the client reusing the buffer. Signed-off-by: Jeffrey Hugo Cc: Ard Biesheuvel Cc: Mark Rutland Cc: Leif Lindholm Cc: Ingo Molnar Cc: Signed-off-by: Matt Fleming --- arch/x86/boot/compressed/eboot.c | 18 ++-- .../firmware/efi/libstub/efi-stub-helper.c | 96 +++++++++++++------ drivers/firmware/efi/libstub/fdt.c | 17 +++- drivers/firmware/efi/libstub/random.c | 12 ++- include/linux/efi.h | 15 ++- 5 files changed, 110 insertions(+), 48 deletions(-) diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c index ff574dad95cc..c5b7c7b4f0d7 100644 --- a/arch/x86/boot/compressed/eboot.c +++ b/arch/x86/boot/compressed/eboot.c @@ -1008,7 +1008,7 @@ static efi_status_t exit_boot(struct boot_params *boot_params, void *handle, bool is64) { struct efi_info *efi = &boot_params->efi_info; - unsigned long map_sz, key, desc_size; + unsigned long map_sz, key, desc_size, buff_size; efi_memory_desc_t *mem_map; struct setup_data *e820ext; const char *signature; @@ -1019,14 +1019,20 @@ static efi_status_t exit_boot(struct boot_params *boot_params, bool called_exit = false; u8 nr_entries; int i; + struct efi_boot_memmap map; - nr_desc = 0; - e820ext = NULL; - e820ext_size = 0; + nr_desc = 0; + e820ext = NULL; + e820ext_size = 0; + map.map = &mem_map; + map.map_size = &map_sz; + map.desc_size = &desc_size; + map.desc_ver = &desc_version; + map.key_ptr = &key; + map.buff_size = &buff_size; get_map: - status = efi_get_memory_map(sys_table, &mem_map, &map_sz, &desc_size, - &desc_version, &key); + status = efi_get_memory_map(sys_table, &map); if (status != EFI_SUCCESS) return status; diff --git a/drivers/firmware/efi/libstub/efi-stub-helper.c b/drivers/firmware/efi/libstub/efi-stub-helper.c index 3bd127f95315..29368ac69221 100644 --- a/drivers/firmware/efi/libstub/efi-stub-helper.c +++ b/drivers/firmware/efi/libstub/efi-stub-helper.c @@ -41,6 +41,8 @@ static unsigned long __chunk_size = EFI_READ_CHUNK_SIZE; #define EFI_ALLOC_ALIGN EFI_PAGE_SIZE #endif +#define EFI_MMAP_NR_SLACK_SLOTS 8 + struct file_info { efi_file_handle_t *handle; u64 size; @@ -63,49 +65,62 @@ void efi_printk(efi_system_table_t *sys_table_arg, char *str) } } +static inline bool mmap_has_headroom(unsigned long buff_size, + unsigned long map_size, + unsigned long desc_size) +{ + unsigned long slack = buff_size - map_size; + + return slack / desc_size >= EFI_MMAP_NR_SLACK_SLOTS; +} + efi_status_t efi_get_memory_map(efi_system_table_t *sys_table_arg, - efi_memory_desc_t **map, - unsigned long *map_size, - unsigned long *desc_size, - u32 *desc_ver, - unsigned long *key_ptr) + struct efi_boot_memmap *map) { efi_memory_desc_t *m = NULL; efi_status_t status; unsigned long key; u32 desc_version; - *map_size = sizeof(*m) * 32; + *map->desc_size = sizeof(*m); + *map->map_size = *map->desc_size * 32; + *map->buff_size = *map->map_size; again: - /* - * Add an additional efi_memory_desc_t because we're doing an - * allocation which may be in a new descriptor region. - */ - *map_size += sizeof(*m); status = efi_call_early(allocate_pool, EFI_LOADER_DATA, - *map_size, (void **)&m); + *map->map_size, (void **)&m); if (status != EFI_SUCCESS) goto fail; - *desc_size = 0; + *map->desc_size = 0; key = 0; - status = efi_call_early(get_memory_map, map_size, m, - &key, desc_size, &desc_version); - if (status == EFI_BUFFER_TOO_SMALL) { + status = efi_call_early(get_memory_map, map->map_size, m, + &key, map->desc_size, &desc_version); + if (status == EFI_BUFFER_TOO_SMALL || + !mmap_has_headroom(*map->buff_size, *map->map_size, + *map->desc_size)) { efi_call_early(free_pool, m); + /* + * Make sure there is some entries of headroom so that the + * buffer can be reused for a new map after allocations are + * no longer permitted. Its unlikely that the map will grow to + * exceed this headroom once we are ready to trigger + * ExitBootServices() + */ + *map->map_size += *map->desc_size * EFI_MMAP_NR_SLACK_SLOTS; + *map->buff_size = *map->map_size; goto again; } if (status != EFI_SUCCESS) efi_call_early(free_pool, m); - if (key_ptr && status == EFI_SUCCESS) - *key_ptr = key; - if (desc_ver && status == EFI_SUCCESS) - *desc_ver = desc_version; + if (map->key_ptr && status == EFI_SUCCESS) + *map->key_ptr = key; + if (map->desc_ver && status == EFI_SUCCESS) + *map->desc_ver = desc_version; fail: - *map = m; + *map->map = m; return status; } @@ -113,13 +128,20 @@ fail: unsigned long get_dram_base(efi_system_table_t *sys_table_arg) { efi_status_t status; - unsigned long map_size; + unsigned long map_size, buff_size; unsigned long membase = EFI_ERROR; struct efi_memory_map map; efi_memory_desc_t *md; + struct efi_boot_memmap boot_map; - status = efi_get_memory_map(sys_table_arg, (efi_memory_desc_t **)&map.map, - &map_size, &map.desc_size, NULL, NULL); + boot_map.map = (efi_memory_desc_t **)&map.map; + boot_map.map_size = &map_size; + boot_map.desc_size = &map.desc_size; + boot_map.desc_ver = NULL; + boot_map.key_ptr = NULL; + boot_map.buff_size = &buff_size; + + status = efi_get_memory_map(sys_table_arg, &boot_map); if (status != EFI_SUCCESS) return membase; @@ -144,15 +166,22 @@ efi_status_t efi_high_alloc(efi_system_table_t *sys_table_arg, unsigned long size, unsigned long align, unsigned long *addr, unsigned long max) { - unsigned long map_size, desc_size; + unsigned long map_size, desc_size, buff_size; efi_memory_desc_t *map; efi_status_t status; unsigned long nr_pages; u64 max_addr = 0; int i; + struct efi_boot_memmap boot_map; - status = efi_get_memory_map(sys_table_arg, &map, &map_size, &desc_size, - NULL, NULL); + boot_map.map = ↦ + boot_map.map_size = &map_size; + boot_map.desc_size = &desc_size; + boot_map.desc_ver = NULL; + boot_map.key_ptr = NULL; + boot_map.buff_size = &buff_size; + + status = efi_get_memory_map(sys_table_arg, &boot_map); if (status != EFI_SUCCESS) goto fail; @@ -230,14 +259,21 @@ efi_status_t efi_low_alloc(efi_system_table_t *sys_table_arg, unsigned long size, unsigned long align, unsigned long *addr) { - unsigned long map_size, desc_size; + unsigned long map_size, desc_size, buff_size; efi_memory_desc_t *map; efi_status_t status; unsigned long nr_pages; int i; + struct efi_boot_memmap boot_map; - status = efi_get_memory_map(sys_table_arg, &map, &map_size, &desc_size, - NULL, NULL); + boot_map.map = ↦ + boot_map.map_size = &map_size; + boot_map.desc_size = &desc_size; + boot_map.desc_ver = NULL; + boot_map.key_ptr = NULL; + boot_map.buff_size = &buff_size; + + status = efi_get_memory_map(sys_table_arg, &boot_map); if (status != EFI_SUCCESS) goto fail; diff --git a/drivers/firmware/efi/libstub/fdt.c b/drivers/firmware/efi/libstub/fdt.c index e58abfa953cc..bec0fa8d8746 100644 --- a/drivers/firmware/efi/libstub/fdt.c +++ b/drivers/firmware/efi/libstub/fdt.c @@ -175,13 +175,21 @@ efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table, unsigned long fdt_addr, unsigned long fdt_size) { - unsigned long map_size, desc_size; + unsigned long map_size, desc_size, buff_size; u32 desc_ver; unsigned long mmap_key; efi_memory_desc_t *memory_map, *runtime_map; unsigned long new_fdt_size; efi_status_t status; int runtime_entry_count = 0; + struct efi_boot_memmap map; + + map.map = &runtime_map; + map.map_size = &map_size; + map.desc_size = &desc_size; + map.desc_ver = &desc_ver; + map.key_ptr = &mmap_key; + map.buff_size = &buff_size; /* * Get a copy of the current memory map that we will use to prepare @@ -189,8 +197,7 @@ efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table, * subsequent allocations adding entries, since they could not affect * the number of EFI_MEMORY_RUNTIME regions. */ - status = efi_get_memory_map(sys_table, &runtime_map, &map_size, - &desc_size, &desc_ver, &mmap_key); + status = efi_get_memory_map(sys_table, &map); if (status != EFI_SUCCESS) { pr_efi_err(sys_table, "Unable to retrieve UEFI memory map.\n"); return status; @@ -199,6 +206,7 @@ efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table, pr_efi(sys_table, "Exiting boot services and installing virtual address map...\n"); + map.map = &memory_map; /* * Estimate size of new FDT, and allocate memory for it. We * will allocate a bigger buffer if this ends up being too @@ -218,8 +226,7 @@ efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table, * we can get the memory map key needed for * exit_boot_services(). */ - status = efi_get_memory_map(sys_table, &memory_map, &map_size, - &desc_size, &desc_ver, &mmap_key); + status = efi_get_memory_map(sys_table, &map); if (status != EFI_SUCCESS) goto fail_free_new_fdt; diff --git a/drivers/firmware/efi/libstub/random.c b/drivers/firmware/efi/libstub/random.c index 53f6d3fe6d86..0c9f58c5ba50 100644 --- a/drivers/firmware/efi/libstub/random.c +++ b/drivers/firmware/efi/libstub/random.c @@ -73,12 +73,20 @@ efi_status_t efi_random_alloc(efi_system_table_t *sys_table_arg, unsigned long random_seed) { unsigned long map_size, desc_size, total_slots = 0, target_slot; + unsigned long buff_size; efi_status_t status; efi_memory_desc_t *memory_map; int map_offset; + struct efi_boot_memmap map; - status = efi_get_memory_map(sys_table_arg, &memory_map, &map_size, - &desc_size, NULL, NULL); + map.map = &memory_map; + map.map_size = &map_size; + map.desc_size = &desc_size; + map.desc_ver = NULL; + map.key_ptr = NULL; + map.buff_size = &buff_size; + + status = efi_get_memory_map(sys_table_arg, &map); if (status != EFI_SUCCESS) return status; diff --git a/include/linux/efi.h b/include/linux/efi.h index 23cd3ced8c1a..943fee524176 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -118,6 +118,15 @@ typedef struct { u32 imagesize; } efi_capsule_header_t; +struct efi_boot_memmap { + efi_memory_desc_t **map; + unsigned long *map_size; + unsigned long *desc_size; + u32 *desc_ver; + unsigned long *key_ptr; + unsigned long *buff_size; +}; + /* * EFI capsule flags */ @@ -1371,11 +1380,7 @@ char *efi_convert_cmdline(efi_system_table_t *sys_table_arg, efi_loaded_image_t *image, int *cmd_line_len); efi_status_t efi_get_memory_map(efi_system_table_t *sys_table_arg, - efi_memory_desc_t **map, - unsigned long *map_size, - unsigned long *desc_size, - u32 *desc_ver, - unsigned long *key_ptr); + struct efi_boot_memmap *map); efi_status_t efi_low_alloc(efi_system_table_t *sys_table_arg, unsigned long size, unsigned long align, From fc07716ba803483be91bc4b2344f9c84985e6f07 Mon Sep 17 00:00:00 2001 From: Jeffrey Hugo Date: Mon, 29 Aug 2016 14:38:52 -0600 Subject: [PATCH 104/271] efi/libstub: Introduce ExitBootServices helper The spec allows ExitBootServices to fail with EFI_INVALID_PARAMETER if a race condition has occurred where the EFI has updated the memory map after the stub grabbed a reference to the map. The spec defines a retry proceedure with specific requirements to handle this scenario. This scenario was previously observed on x86 - commit d3768d885c6c ("x86, efi: retry ExitBootServices() on failure") but the current fix is not spec compliant and the scenario is now observed on the Qualcomm Technologies QDF2432 via the FDT stub which does not handle the error and thus causes boot failures. The user will notice the boot failure as the kernel is not executed and the system may drop back to a UEFI shell, but will be unresponsive to input and the system will require a power cycle to recover. Add a helper to the stub library that correctly adheres to the spec in the case of EFI_INVALID_PARAMETER from ExitBootServices and can be universally used across all stub implementations. Signed-off-by: Jeffrey Hugo Cc: Ard Biesheuvel Cc: Mark Rutland Cc: Leif Lindholm Cc: Ingo Molnar Cc: Signed-off-by: Matt Fleming --- .../firmware/efi/libstub/efi-stub-helper.c | 73 +++++++++++++++++++ include/linux/efi.h | 10 +++ 2 files changed, 83 insertions(+) diff --git a/drivers/firmware/efi/libstub/efi-stub-helper.c b/drivers/firmware/efi/libstub/efi-stub-helper.c index 29368ac69221..aded10662020 100644 --- a/drivers/firmware/efi/libstub/efi-stub-helper.c +++ b/drivers/firmware/efi/libstub/efi-stub-helper.c @@ -740,3 +740,76 @@ char *efi_convert_cmdline(efi_system_table_t *sys_table_arg, *cmd_line_len = options_bytes; return (char *)cmdline_addr; } + +/* + * Handle calling ExitBootServices according to the requirements set out by the + * spec. Obtains the current memory map, and returns that info after calling + * ExitBootServices. The client must specify a function to perform any + * processing of the memory map data prior to ExitBootServices. A client + * specific structure may be passed to the function via priv. The client + * function may be called multiple times. + */ +efi_status_t efi_exit_boot_services(efi_system_table_t *sys_table_arg, + void *handle, + struct efi_boot_memmap *map, + void *priv, + efi_exit_boot_map_processing priv_func) +{ + efi_status_t status; + + status = efi_get_memory_map(sys_table_arg, map); + + if (status != EFI_SUCCESS) + goto fail; + + status = priv_func(sys_table_arg, map, priv); + if (status != EFI_SUCCESS) + goto free_map; + + status = efi_call_early(exit_boot_services, handle, *map->key_ptr); + + if (status == EFI_INVALID_PARAMETER) { + /* + * The memory map changed between efi_get_memory_map() and + * exit_boot_services(). Per the UEFI Spec v2.6, Section 6.4: + * EFI_BOOT_SERVICES.ExitBootServices we need to get the + * updated map, and try again. The spec implies one retry + * should be sufficent, which is confirmed against the EDK2 + * implementation. Per the spec, we can only invoke + * get_memory_map() and exit_boot_services() - we cannot alloc + * so efi_get_memory_map() cannot be used, and we must reuse + * the buffer. For all practical purposes, the headroom in the + * buffer should account for any changes in the map so the call + * to get_memory_map() is expected to succeed here. + */ + *map->map_size = *map->buff_size; + status = efi_call_early(get_memory_map, + map->map_size, + *map->map, + map->key_ptr, + map->desc_size, + map->desc_ver); + + /* exit_boot_services() was called, thus cannot free */ + if (status != EFI_SUCCESS) + goto fail; + + status = priv_func(sys_table_arg, map, priv); + /* exit_boot_services() was called, thus cannot free */ + if (status != EFI_SUCCESS) + goto fail; + + status = efi_call_early(exit_boot_services, handle, *map->key_ptr); + } + + /* exit_boot_services() was called, thus cannot free */ + if (status != EFI_SUCCESS) + goto fail; + + return EFI_SUCCESS; + +free_map: + efi_call_early(free_pool, *map->map); +fail: + return status; +} diff --git a/include/linux/efi.h b/include/linux/efi.h index 943fee524176..0148a3046b48 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -1462,4 +1462,14 @@ extern void efi_call_virt_check_flags(unsigned long flags, const char *call); arch_efi_call_virt_teardown(); \ }) +typedef efi_status_t (*efi_exit_boot_map_processing)( + efi_system_table_t *sys_table_arg, + struct efi_boot_memmap *map, + void *priv); + +efi_status_t efi_exit_boot_services(efi_system_table_t *sys_table, + void *handle, + struct efi_boot_memmap *map, + void *priv, + efi_exit_boot_map_processing priv_func); #endif /* _LINUX_EFI_H */ From ed9cc156c42ff0c0bf9b1d09df48a12bf0873473 Mon Sep 17 00:00:00 2001 From: Jeffrey Hugo Date: Mon, 29 Aug 2016 14:38:53 -0600 Subject: [PATCH 105/271] efi/libstub: Use efi_exit_boot_services() in FDT The FDT code directly calls ExitBootServices. This is inadvisable as the UEFI spec details a complex set of errors, race conditions, and API interactions that the caller of ExitBootServices must get correct. The FDT code does not handle EFI_INVALID_PARAMETER as required by the spec, which causes intermittent boot failures on the Qualcomm Technologies QDF2432. Call the efi_exit_boot_services() helper intead, which handles the EFI_INVALID_PARAMETER scenario properly. Signed-off-by: Jeffrey Hugo Cc: Ard Biesheuvel Cc: Mark Rutland Cc: Leif Lindholm Cc: Ingo Molnar Cc: Signed-off-by: Matt Fleming --- drivers/firmware/efi/libstub/fdt.c | 37 ++++++++++++++++++++++-------- 1 file changed, 27 insertions(+), 10 deletions(-) diff --git a/drivers/firmware/efi/libstub/fdt.c b/drivers/firmware/efi/libstub/fdt.c index bec0fa8d8746..a6a93116a8f0 100644 --- a/drivers/firmware/efi/libstub/fdt.c +++ b/drivers/firmware/efi/libstub/fdt.c @@ -152,6 +152,27 @@ fdt_set_fail: #define EFI_FDT_ALIGN EFI_PAGE_SIZE #endif +struct exit_boot_struct { + efi_memory_desc_t *runtime_map; + int *runtime_entry_count; +}; + +static efi_status_t exit_boot_func(efi_system_table_t *sys_table_arg, + struct efi_boot_memmap *map, + void *priv) +{ + struct exit_boot_struct *p = priv; + /* + * Update the memory map with virtual addresses. The function will also + * populate @runtime_map with copies of just the EFI_MEMORY_RUNTIME + * entries so that we can pass it straight to SetVirtualAddressMap() + */ + efi_get_virtmap(*map->map, *map->map_size, *map->desc_size, + p->runtime_map, p->runtime_entry_count); + + return EFI_SUCCESS; +} + /* * Allocate memory for a new FDT, then add EFI, commandline, and * initrd related fields to the FDT. This routine increases the @@ -183,6 +204,7 @@ efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table, efi_status_t status; int runtime_entry_count = 0; struct efi_boot_memmap map; + struct exit_boot_struct priv; map.map = &runtime_map; map.map_size = &map_size; @@ -257,16 +279,11 @@ efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table, } } - /* - * Update the memory map with virtual addresses. The function will also - * populate @runtime_map with copies of just the EFI_MEMORY_RUNTIME - * entries so that we can pass it straight into SetVirtualAddressMap() - */ - efi_get_virtmap(memory_map, map_size, desc_size, runtime_map, - &runtime_entry_count); - - /* Now we are ready to exit_boot_services.*/ - status = sys_table->boottime->exit_boot_services(handle, mmap_key); + sys_table->boottime->free_pool(memory_map); + priv.runtime_map = runtime_map; + priv.runtime_entry_count = &runtime_entry_count; + status = efi_exit_boot_services(sys_table, handle, &map, &priv, + exit_boot_func); if (status == EFI_SUCCESS) { efi_set_virtual_address_map_t *svam; From d64934019f6cc39202e2f78063709f61ca5cb364 Mon Sep 17 00:00:00 2001 From: Jeffrey Hugo Date: Mon, 29 Aug 2016 14:38:54 -0600 Subject: [PATCH 106/271] x86/efi: Use efi_exit_boot_services() The eboot code directly calls ExitBootServices. This is inadvisable as the UEFI spec details a complex set of errors, race conditions, and API interactions that the caller of ExitBootServices must get correct. The eboot code attempts allocations after calling ExitBootSerives which is not permitted per the spec. Call the efi_exit_boot_services() helper intead, which handles the allocation scenario properly. Signed-off-by: Jeffrey Hugo Cc: Ard Biesheuvel Cc: Mark Rutland Cc: Leif Lindholm Cc: Ingo Molnar Cc: Signed-off-by: Matt Fleming --- arch/x86/boot/compressed/eboot.c | 134 +++++++++++++++---------------- 1 file changed, 66 insertions(+), 68 deletions(-) diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c index c5b7c7b4f0d7..94dd4a31f5b3 100644 --- a/arch/x86/boot/compressed/eboot.c +++ b/arch/x86/boot/compressed/eboot.c @@ -1004,85 +1004,87 @@ static efi_status_t alloc_e820ext(u32 nr_desc, struct setup_data **e820ext, return status; } +struct exit_boot_struct { + struct boot_params *boot_params; + struct efi_info *efi; + struct setup_data *e820ext; + __u32 e820ext_size; + bool is64; +}; + +static efi_status_t exit_boot_func(efi_system_table_t *sys_table_arg, + struct efi_boot_memmap *map, + void *priv) +{ + static bool first = true; + const char *signature; + __u32 nr_desc; + efi_status_t status; + struct exit_boot_struct *p = priv; + + if (first) { + nr_desc = *map->buff_size / *map->desc_size; + if (nr_desc > ARRAY_SIZE(p->boot_params->e820_map)) { + u32 nr_e820ext = nr_desc - + ARRAY_SIZE(p->boot_params->e820_map); + + status = alloc_e820ext(nr_e820ext, &p->e820ext, + &p->e820ext_size); + if (status != EFI_SUCCESS) + return status; + } + first = false; + } + + signature = p->is64 ? EFI64_LOADER_SIGNATURE : EFI32_LOADER_SIGNATURE; + memcpy(&p->efi->efi_loader_signature, signature, sizeof(__u32)); + + p->efi->efi_systab = (unsigned long)sys_table_arg; + p->efi->efi_memdesc_size = *map->desc_size; + p->efi->efi_memdesc_version = *map->desc_ver; + p->efi->efi_memmap = (unsigned long)*map->map; + p->efi->efi_memmap_size = *map->map_size; + +#ifdef CONFIG_X86_64 + p->efi->efi_systab_hi = (unsigned long)sys_table_arg >> 32; + p->efi->efi_memmap_hi = (unsigned long)*map->map >> 32; +#endif + + return EFI_SUCCESS; +} + static efi_status_t exit_boot(struct boot_params *boot_params, void *handle, bool is64) { - struct efi_info *efi = &boot_params->efi_info; unsigned long map_sz, key, desc_size, buff_size; efi_memory_desc_t *mem_map; struct setup_data *e820ext; - const char *signature; __u32 e820ext_size; - __u32 nr_desc, prev_nr_desc; efi_status_t status; __u32 desc_version; - bool called_exit = false; - u8 nr_entries; - int i; struct efi_boot_memmap map; + struct exit_boot_struct priv; - nr_desc = 0; - e820ext = NULL; - e820ext_size = 0; - map.map = &mem_map; - map.map_size = &map_sz; - map.desc_size = &desc_size; - map.desc_ver = &desc_version; - map.key_ptr = &key; - map.buff_size = &buff_size; - -get_map: - status = efi_get_memory_map(sys_table, &map); + map.map = &mem_map; + map.map_size = &map_sz; + map.desc_size = &desc_size; + map.desc_ver = &desc_version; + map.key_ptr = &key; + map.buff_size = &buff_size; + priv.boot_params = boot_params; + priv.efi = &boot_params->efi_info; + priv.e820ext = NULL; + priv.e820ext_size = 0; + priv.is64 = is64; + /* Might as well exit boot services now */ + status = efi_exit_boot_services(sys_table, handle, &map, &priv, + exit_boot_func); if (status != EFI_SUCCESS) return status; - prev_nr_desc = nr_desc; - nr_desc = map_sz / desc_size; - if (nr_desc > prev_nr_desc && - nr_desc > ARRAY_SIZE(boot_params->e820_map)) { - u32 nr_e820ext = nr_desc - ARRAY_SIZE(boot_params->e820_map); - - status = alloc_e820ext(nr_e820ext, &e820ext, &e820ext_size); - if (status != EFI_SUCCESS) - goto free_mem_map; - - efi_call_early(free_pool, mem_map); - goto get_map; /* Allocated memory, get map again */ - } - - signature = is64 ? EFI64_LOADER_SIGNATURE : EFI32_LOADER_SIGNATURE; - memcpy(&efi->efi_loader_signature, signature, sizeof(__u32)); - - efi->efi_systab = (unsigned long)sys_table; - efi->efi_memdesc_size = desc_size; - efi->efi_memdesc_version = desc_version; - efi->efi_memmap = (unsigned long)mem_map; - efi->efi_memmap_size = map_sz; - -#ifdef CONFIG_X86_64 - efi->efi_systab_hi = (unsigned long)sys_table >> 32; - efi->efi_memmap_hi = (unsigned long)mem_map >> 32; -#endif - - /* Might as well exit boot services now */ - status = efi_call_early(exit_boot_services, handle, key); - if (status != EFI_SUCCESS) { - /* - * ExitBootServices() will fail if any of the event - * handlers change the memory map. In which case, we - * must be prepared to retry, but only once so that - * we're guaranteed to exit on repeated failures instead - * of spinning forever. - */ - if (called_exit) - goto free_mem_map; - - called_exit = true; - efi_call_early(free_pool, mem_map); - goto get_map; - } - + e820ext = priv.e820ext; + e820ext_size = priv.e820ext_size; /* Historic? */ boot_params->alt_mem_k = 32 * 1024; @@ -1091,10 +1093,6 @@ get_map: return status; return EFI_SUCCESS; - -free_mem_map: - efi_call_early(free_pool, mem_map); - return status; } /* From 4d21cef3ea00ba3ac508eb61fb8db70e3e31df67 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Fri, 2 Sep 2016 12:33:49 +0200 Subject: [PATCH 107/271] KVM: s390: vsie: fix riccbd We store the address of riccbd at the wrong location, overwriting gvrd. This means that our nested guest will not be able to use runtime instrumentation. Also, a memory leak, if our KVM guest actually sets gvrd. Not noticed until now, as KVM guests never make use of gvrd and runtime instrumentation wasn't completely tested yet. Reported-by: Fan Zhang Reviewed-by: Cornelia Huck Signed-off-by: David Hildenbrand Signed-off-by: Cornelia Huck --- arch/s390/kvm/vsie.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index c106488b4137..d8673e243f13 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -584,7 +584,7 @@ static int pin_blocks(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) /* Validity 0x0044 will be checked by SIE */ if (rc) goto unpin; - scb_s->gvrd = hpa; + scb_s->riccbd = hpa; } return 0; unpin: From e12c8f36f3f7a60d55938c5aed5999278fa92bcb Mon Sep 17 00:00:00 2001 From: Wanpeng Li Date: Tue, 30 Aug 2016 16:14:00 +0800 Subject: [PATCH 108/271] KVM: lapic: adjust preemption timer correctly when goes TSC backward MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit TSC_OFFSET will be adjusted if discovers TSC backward during vCPU load. The preemption timer, which relies on the guest tsc to reprogram its preemption timer value, is also reprogrammed if vCPU is scheded in to a different pCPU. However, the current implementation reprogram preemption timer before TSC_OFFSET is adjusted to the right value, resulting in the preemption timer firing prematurely. This patch fix it by adjusting TSC_OFFSET before reprogramming preemption timer if TSC backward. Cc: Paolo Bonzini Cc: Radim Krċmář Cc: Yunhong Jiang Signed-off-by: Wanpeng Li Signed-off-by: Paolo Bonzini --- arch/x86/kvm/x86.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 19f9f9e05c2a..699f8726539a 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2743,16 +2743,16 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) if (tsc_delta < 0) mark_tsc_unstable("KVM discovered backwards TSC"); - if (kvm_lapic_hv_timer_in_use(vcpu) && - kvm_x86_ops->set_hv_timer(vcpu, - kvm_get_lapic_tscdeadline_msr(vcpu))) - kvm_lapic_switch_to_sw_timer(vcpu); if (check_tsc_unstable()) { u64 offset = kvm_compute_tsc_offset(vcpu, vcpu->arch.last_guest_tsc); kvm_x86_ops->write_tsc_offset(vcpu, offset); vcpu->arch.tsc_catchup = 1; } + if (kvm_lapic_hv_timer_in_use(vcpu) && + kvm_x86_ops->set_hv_timer(vcpu, + kvm_get_lapic_tscdeadline_msr(vcpu))) + kvm_lapic_switch_to_sw_timer(vcpu); /* * On a host with synchronized TSC, there is no need to update * kvmclock on vcpu->cpu migration From 79d102cbfd2e9d94257fcc7c82807ef1cdf80322 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Mon, 5 Sep 2016 17:30:07 +0200 Subject: [PATCH 109/271] perf/x86/intel/cqm: Check cqm/mbm enabled state in event init Yanqiu Zhang reported kernel panic when using mbm event on system where CQM is detected but without mbm event support, like with perf: # perf stat -e 'intel_cqm/event=3/' -a BUG: unable to handle kernel NULL pointer dereference at 0000000000000020 IP: [] update_sample+0xbc/0xe0 ... [] __intel_mbm_event_init+0x18/0x20 [] flush_smp_call_function_queue+0x7b/0x160 [] generic_smp_call_function_single_interrupt+0x13/0x60 [] smp_call_function_interrupt+0x27/0x40 [] call_function_interrupt+0x8c/0xa0 ... The reason is that we currently allow to init mbm event even if mbm support is not detected. Adding checks for both cqm and mbm events and support into cqm's event_init. Fixes: 33c3cc7acfd9 ("perf/x86/mbm: Add Intel Memory B/W Monitoring enumeration and init") Reported-by: Yanqiu Zhang Signed-off-by: Jiri Olsa Acked-by: Peter Zijlstra Cc: Vikas Shivappa Cc: Tony Luck Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1473089407-21857-1-git-send-email-jolsa@kernel.org Signed-off-by: Thomas Gleixner --- arch/x86/events/intel/cqm.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/arch/x86/events/intel/cqm.c b/arch/x86/events/intel/cqm.c index 783c49ddef29..8f82b02934fa 100644 --- a/arch/x86/events/intel/cqm.c +++ b/arch/x86/events/intel/cqm.c @@ -458,6 +458,11 @@ static void __intel_cqm_event_count(void *info); static void init_mbm_sample(u32 rmid, u32 evt_type); static void __intel_mbm_event_count(void *info); +static bool is_cqm_event(int e) +{ + return (e == QOS_L3_OCCUP_EVENT_ID); +} + static bool is_mbm_event(int e) { return (e >= QOS_MBM_TOTAL_EVENT_ID && e <= QOS_MBM_LOCAL_EVENT_ID); @@ -1366,6 +1371,10 @@ static int intel_cqm_event_init(struct perf_event *event) (event->attr.config > QOS_MBM_LOCAL_EVENT_ID)) return -EINVAL; + if ((is_cqm_event(event->attr.config) && !cqm_enabled) || + (is_mbm_event(event->attr.config) && !mbm_enabled)) + return -EINVAL; + /* unsupported modes and filters */ if (event->attr.exclude_user || event->attr.exclude_kernel || From d2896d4b55b2e32b423072a4124d7da4dc1e6cb1 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 22 Aug 2016 09:01:17 +0100 Subject: [PATCH 110/271] arm: KVM: Fix idmap overlap detection when the kernel is idmap'ed We're trying hard to detect when the HYP idmap overlaps with the HYP va, as it makes the teardown of a cpu dangerous. But there is one case where an overlap is completely safe, which is when the whole of the kernel is idmap'ed, which is likely to happen on 32bit when RAM is at 0x8000000 and we're using a 2G/2G VA split. In that case, we can proceed safely. Reported-by: Christoffer Dall Signed-off-by: Marc Zyngier Signed-off-by: Christoffer Dall --- arch/arm/kvm/mmu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c index 29d0b23af2a9..a3faafe70081 100644 --- a/arch/arm/kvm/mmu.c +++ b/arch/arm/kvm/mmu.c @@ -1714,7 +1714,8 @@ int kvm_mmu_init(void) kern_hyp_va(PAGE_OFFSET), kern_hyp_va(~0UL)); if (hyp_idmap_start >= kern_hyp_va(PAGE_OFFSET) && - hyp_idmap_start < kern_hyp_va(~0UL)) { + hyp_idmap_start < kern_hyp_va(~0UL) && + hyp_idmap_start != (unsigned long)__hyp_idmap_text_start) { /* * The idmap page is intersecting with the VA space, * it is not safe to continue further. From d31ed3f05763644840c654a384eaefa94c097ba2 Mon Sep 17 00:00:00 2001 From: Jan Leupold Date: Wed, 6 Jul 2016 13:22:35 +0200 Subject: [PATCH 111/271] drm: atmel-hlcdc: Fix vertical scaling The code is applying the same scaling for the X and Y components, thus making the scaling feature only functional when both components have the same scaling factor. Do the s/_w/_h/ replacement where appropriate to fix vertical scaling. Signed-off-by: Jan Leupold Fixes: 1a396789f65a2 ("drm: add Atmel HLCDC Display Controller support") Cc: Signed-off-by: Boris Brezillon --- drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c index 016c191221f3..52c527f6642a 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c @@ -320,19 +320,19 @@ atmel_hlcdc_plane_update_pos_and_size(struct atmel_hlcdc_plane *plane, u32 *coeff_tab = heo_upscaling_ycoef; u32 max_memsize; - if (state->crtc_w < state->src_w) + if (state->crtc_h < state->src_h) coeff_tab = heo_downscaling_ycoef; for (i = 0; i < ARRAY_SIZE(heo_upscaling_ycoef); i++) atmel_hlcdc_layer_update_cfg(&plane->layer, 33 + i, 0xffffffff, coeff_tab[i]); - factor = ((8 * 256 * state->src_w) - (256 * 4)) / - state->crtc_w; + factor = ((8 * 256 * state->src_h) - (256 * 4)) / + state->crtc_h; factor++; - max_memsize = ((factor * state->crtc_w) + (256 * 4)) / + max_memsize = ((factor * state->crtc_h) + (256 * 4)) / 2048; - if (max_memsize > state->src_w) + if (max_memsize > state->src_h) factor--; factor_reg |= (factor << 16) | 0x80000000; } From 1ba7db07ccc2825669d6e376632316813a072887 Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Mon, 11 Jul 2016 12:19:40 +0200 Subject: [PATCH 112/271] drm/atmel-hlcdc: Make ->reset() implementation static The atmel_hlcdc_crtc_reset() function is never used outside the file and can be static. This avoids a warning from sparse. Signed-off-by: Thierry Reding --- drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c index a978381ef95b..9b17a66cf0e1 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c @@ -387,7 +387,7 @@ void atmel_hlcdc_crtc_irq(struct drm_crtc *c) atmel_hlcdc_crtc_finish_page_flip(drm_crtc_to_atmel_hlcdc_crtc(c)); } -void atmel_hlcdc_crtc_reset(struct drm_crtc *crtc) +static void atmel_hlcdc_crtc_reset(struct drm_crtc *crtc) { struct atmel_hlcdc_crtc_state *state; From cd28e716c6869d2f06e64bcd679d0a45dd8a6295 Mon Sep 17 00:00:00 2001 From: Ping Gao Date: Tue, 6 Sep 2016 12:04:11 +0800 Subject: [PATCH 113/271] drm/i915: enable vGPU detection for all vGPU capability is handled by GVT-g host driver, not needed to put extra HW check for vGPU detection. And we'll actually support vGPU from BDW. Signed-off-by: Ping Gao Signed-off-by: Zhenyu Wang Reviewed-by: Joonas Lahtinen Acked-by: Chris Wilson Cc: drm-intel-fixes@lists.freedesktop.org Signed-off-by: Jani Nikula Link: http://patchwork.freedesktop.org/patch/msgid/20160906040412.1274-2-zhenyuw@linux.intel.com (cherry picked from commit 8ef89995c735f978d5dfcb3ca6bce70d41728c91) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_vgpu.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_vgpu.c b/drivers/gpu/drm/i915/i915_vgpu.c index f6acb5a0e701..b81cfb3b22ec 100644 --- a/drivers/gpu/drm/i915/i915_vgpu.c +++ b/drivers/gpu/drm/i915/i915_vgpu.c @@ -65,9 +65,6 @@ void i915_check_vgpu(struct drm_i915_private *dev_priv) BUILD_BUG_ON(sizeof(struct vgt_if) != VGT_PVINFO_SIZE); - if (!IS_HASWELL(dev_priv)) - return; - magic = __raw_i915_read64(dev_priv, vgtif_reg(magic)); if (magic != VGT_MAGIC) return; From 557b1a8cae25e36ac2f125d93f003e60a7d0d014 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Tue, 6 Sep 2016 12:04:12 +0800 Subject: [PATCH 114/271] drm/i915: disable 48bit full PPGTT when vGPU is active Disable 48bit full PPGTT on vGPU too for now. Signed-off-by: Zhi Wang Signed-off-by: Zhenyu Wang Reviewed-by: Joonas Lahtinen Acked-by: Chris Wilson Cc: drm-intel-fixes@lists.freedesktop.org Signed-off-by: Jani Nikula Link: http://patchwork.freedesktop.org/patch/msgid/20160906040412.1274-3-zhenyuw@linux.intel.com (cherry picked from commit e320d40022128845dfff900422ea9fd69f576c98) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_gem_gtt.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 7a30af79d799..f38ceffd82c3 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -122,8 +122,11 @@ int intel_sanitize_enable_ppgtt(struct drm_i915_private *dev_priv, has_full_48bit_ppgtt = IS_BROADWELL(dev_priv) || INTEL_GEN(dev_priv) >= 9; - if (intel_vgpu_active(dev_priv)) - has_full_ppgtt = false; /* emulation is too hard */ + if (intel_vgpu_active(dev_priv)) { + /* emulation is too hard */ + has_full_ppgtt = false; + has_full_48bit_ppgtt = false; + } if (!has_aliasing_ppgtt) return 0; @@ -158,7 +161,7 @@ int intel_sanitize_enable_ppgtt(struct drm_i915_private *dev_priv, return 0; } - if (INTEL_GEN(dev_priv) >= 8 && i915.enable_execlists) + if (INTEL_GEN(dev_priv) >= 8 && i915.enable_execlists && has_full_ppgtt) return has_full_48bit_ppgtt ? 3 : 2; else return has_aliasing_ppgtt ? 1 : 0; From 03c2778a938aaba0893f6d6cdc29511d91a79848 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Fri, 2 Sep 2016 14:39:50 -0400 Subject: [PATCH 115/271] ipv6: release dst in ping_v6_sendmsg Neither the failure or success paths of ping_v6_sendmsg release the dst it acquires. This leads to a flood of warnings from "net/core/dst.c:288 dst_release" on older kernels that don't have 8bf4ada2e21378816b28205427ee6b0e1ca4c5f1 backported. That patch optimistically hoped this had been fixed post 3.10, but it seems at least one case wasn't, where I've seen this triggered a lot from machines doing unprivileged icmp sockets. Cc: Martin Lau Signed-off-by: Dave Jones Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- net/ipv6/ping.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c index 0900352c924c..0e983b694ee8 100644 --- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -126,8 +126,10 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) rt = (struct rt6_info *) dst; np = inet6_sk(sk); - if (!np) - return -EBADF; + if (!np) { + err = -EBADF; + goto dst_err_out; + } if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr)) fl6.flowi6_oif = np->mcast_oif; @@ -163,6 +165,9 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) } release_sock(sk); +dst_err_out: + dst_release(dst); + if (err) return err; From 78d506e1b7071b24850fd5ac22b896c459b0a04c Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 6 Sep 2016 11:22:49 -0400 Subject: [PATCH 116/271] xprtrdma: Revert 3d4cf35bd4fa ("xprtrdma: Reply buffer exhaustion...") Receive buffer exhaustion, if it were to actually occur, would be catastrophic. However, when there are no reply buffers to post, that means all of them have already been posted and are waiting for incoming replies. By design, there can never be more RPCs in flight than there are available receive buffers. A receive buffer can be left posted after an RPC exits without a received reply; say, due to a credential problem or a soft timeout. This does not result in fewer posted receive buffers than there are pending RPCs, and there is already logic in xprtrdma to deal appropriately with this case. It also looks like the "+ 2" that was removed was accidentally accommodating the number of extra receive buffers needed for receiving backchannel requests. That will need to be addressed by another patch. Fixes: 3d4cf35bd4fa ("xprtrdma: Reply buffer exhaustion can be...") Signed-off-by: Chuck Lever Reviewed-by: Anna Schumaker Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/verbs.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 536d0be3f61b..fefcba982415 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -923,7 +923,7 @@ rpcrdma_buffer_create(struct rpcrdma_xprt *r_xprt) } INIT_LIST_HEAD(&buf->rb_recv_bufs); - for (i = 0; i < buf->rb_max_requests; i++) { + for (i = 0; i < buf->rb_max_requests + 2; i++) { struct rpcrdma_rep *rep; rep = rpcrdma_create_rep(r_xprt); @@ -1076,6 +1076,8 @@ rpcrdma_put_mw(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mw *mw) /* * Get a set of request/reply buffers. + * + * Reply buffer (if available) is attached to send buffer upon return. */ struct rpcrdma_req * rpcrdma_buffer_get(struct rpcrdma_buffer *buffers) @@ -1094,13 +1096,13 @@ rpcrdma_buffer_get(struct rpcrdma_buffer *buffers) out_reqbuf: spin_unlock(&buffers->rb_lock); - pr_warn("rpcrdma: out of request buffers (%p)\n", buffers); + pr_warn("RPC: %s: out of request buffers\n", __func__); return NULL; out_repbuf: - list_add(&req->rl_free, &buffers->rb_send_bufs); spin_unlock(&buffers->rb_lock); - pr_warn("rpcrdma: out of reply buffers (%p)\n", buffers); - return NULL; + pr_warn("RPC: %s: out of reply buffers\n", __func__); + req->rl_reply = NULL; + return req; } /* From 05c974669ecec510a85d8534099bb75404e82c41 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 6 Sep 2016 11:22:58 -0400 Subject: [PATCH 117/271] xprtrdma: Fix receive buffer accounting An RPC can terminate before its reply arrives, if a credential problem or a soft timeout occurs. After this happens, xprtrdma reports it is out of Receive buffers. A Receive buffer is posted before each RPC is sent, and returned to the buffer pool when a reply is received. If no reply is received for an RPC, that Receive buffer remains posted. But xprtrdma tries to post another when the next RPC is sent. If this happens a few dozen times, there are no receive buffers left to be posted at send time. I don't see a way for a transport connection to recover at that point, and it will spit warnings and unnecessarily delay RPCs on occasion for its remaining lifetime. Commit 1e465fd4ff47 ("xprtrdma: Replace send and receive arrays") removed a little bit of logic to detect this case and not provide a Receive buffer so no more buffers are posted, and then transport operation continues correctly. We didn't understand what that logic did, and it wasn't commented, so it was removed as part of the overhaul to support backchannel requests. Restore it, but be wary of the need to keep extra Receives posted to deal with backchannel requests. Fixes: 1e465fd4ff47 ("xprtrdma: Replace send and receive arrays") Signed-off-by: Chuck Lever Reviewed-by: Anna Schumaker Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/verbs.c | 41 +++++++++++++++++++++++---------- net/sunrpc/xprtrdma/xprt_rdma.h | 1 + 2 files changed, 30 insertions(+), 12 deletions(-) diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index fefcba982415..799cce6cbe45 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -51,6 +51,7 @@ #include #include #include +#include #include #include /* try_module_get()/module_put() */ @@ -923,7 +924,7 @@ rpcrdma_buffer_create(struct rpcrdma_xprt *r_xprt) } INIT_LIST_HEAD(&buf->rb_recv_bufs); - for (i = 0; i < buf->rb_max_requests + 2; i++) { + for (i = 0; i < buf->rb_max_requests + RPCRDMA_MAX_BC_REQUESTS; i++) { struct rpcrdma_rep *rep; rep = rpcrdma_create_rep(r_xprt); @@ -1018,6 +1019,7 @@ rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf) rep = rpcrdma_buffer_get_rep_locked(buf); rpcrdma_destroy_rep(ia, rep); } + buf->rb_send_count = 0; spin_lock(&buf->rb_reqslock); while (!list_empty(&buf->rb_allreqs)) { @@ -1032,6 +1034,7 @@ rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf) spin_lock(&buf->rb_reqslock); } spin_unlock(&buf->rb_reqslock); + buf->rb_recv_count = 0; rpcrdma_destroy_mrs(buf); } @@ -1074,6 +1077,23 @@ rpcrdma_put_mw(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mw *mw) spin_unlock(&buf->rb_mwlock); } +static struct rpcrdma_rep * +rpcrdma_buffer_get_rep(struct rpcrdma_buffer *buffers) +{ + /* If an RPC previously completed without a reply (say, a + * credential problem or a soft timeout occurs) then hold off + * on supplying more Receive buffers until the number of new + * pending RPCs catches up to the number of posted Receives. + */ + if (unlikely(buffers->rb_send_count < buffers->rb_recv_count)) + return NULL; + + if (unlikely(list_empty(&buffers->rb_recv_bufs))) + return NULL; + buffers->rb_recv_count++; + return rpcrdma_buffer_get_rep_locked(buffers); +} + /* * Get a set of request/reply buffers. * @@ -1087,10 +1107,9 @@ rpcrdma_buffer_get(struct rpcrdma_buffer *buffers) spin_lock(&buffers->rb_lock); if (list_empty(&buffers->rb_send_bufs)) goto out_reqbuf; + buffers->rb_send_count++; req = rpcrdma_buffer_get_req_locked(buffers); - if (list_empty(&buffers->rb_recv_bufs)) - goto out_repbuf; - req->rl_reply = rpcrdma_buffer_get_rep_locked(buffers); + req->rl_reply = rpcrdma_buffer_get_rep(buffers); spin_unlock(&buffers->rb_lock); return req; @@ -1098,11 +1117,6 @@ out_reqbuf: spin_unlock(&buffers->rb_lock); pr_warn("RPC: %s: out of request buffers\n", __func__); return NULL; -out_repbuf: - spin_unlock(&buffers->rb_lock); - pr_warn("RPC: %s: out of reply buffers\n", __func__); - req->rl_reply = NULL; - return req; } /* @@ -1119,9 +1133,12 @@ rpcrdma_buffer_put(struct rpcrdma_req *req) req->rl_reply = NULL; spin_lock(&buffers->rb_lock); + buffers->rb_send_count--; list_add_tail(&req->rl_free, &buffers->rb_send_bufs); - if (rep) + if (rep) { + buffers->rb_recv_count--; list_add_tail(&rep->rr_list, &buffers->rb_recv_bufs); + } spin_unlock(&buffers->rb_lock); } @@ -1135,8 +1152,7 @@ rpcrdma_recv_buffer_get(struct rpcrdma_req *req) struct rpcrdma_buffer *buffers = req->rl_buffer; spin_lock(&buffers->rb_lock); - if (!list_empty(&buffers->rb_recv_bufs)) - req->rl_reply = rpcrdma_buffer_get_rep_locked(buffers); + req->rl_reply = rpcrdma_buffer_get_rep(buffers); spin_unlock(&buffers->rb_lock); } @@ -1150,6 +1166,7 @@ rpcrdma_recv_buffer_put(struct rpcrdma_rep *rep) struct rpcrdma_buffer *buffers = &rep->rr_rxprt->rx_buf; spin_lock(&buffers->rb_lock); + buffers->rb_recv_count--; list_add_tail(&rep->rr_list, &buffers->rb_recv_bufs); spin_unlock(&buffers->rb_lock); } diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index 670fad57153a..a71b0f5897d8 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -321,6 +321,7 @@ struct rpcrdma_buffer { char *rb_pool; spinlock_t rb_lock; /* protect buf lists */ + int rb_send_count, rb_recv_count; struct list_head rb_send_bufs; struct list_head rb_recv_bufs; u32 rb_max_requests; From 007e4ba3ee137f4700f39aa6dbaf01a71047c5f6 Mon Sep 17 00:00:00 2001 From: Helmut Buchsbaum Date: Sun, 4 Sep 2016 18:09:47 +0200 Subject: [PATCH 118/271] net: macb: initialize checksum when using checksum offloading I'm still struggling to get this fix right.. Changes since v2: - do not blindly modify SKB contents according to Dave's legitimate objection Changes since v1: - dropped disabling HW checksum offload for Zynq - initialize checksum similar to net/ethernet/freescale/fec_main.c -- >8 -- MACB/GEM needs the checksum field initialized to 0 to get correct results on transmit in all cases, e.g. on Zynq, UDP packets with payload <= 2 otherwise contain a wrong checksums. Signed-off-by: Helmut Buchsbaum Signed-off-by: David S. Miller --- drivers/net/ethernet/cadence/macb.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 89c0cfa9719f..d954a97b0b0b 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -1323,6 +1323,24 @@ dma_error: return 0; } +static inline int macb_clear_csum(struct sk_buff *skb) +{ + /* no change for packets without checksum offloading */ + if (skb->ip_summed != CHECKSUM_PARTIAL) + return 0; + + /* make sure we can modify the header */ + if (unlikely(skb_cow_head(skb, 0))) + return -1; + + /* initialize checksum field + * This is required - at least for Zynq, which otherwise calculates + * wrong UDP header checksums for UDP packets with UDP data len <=2 + */ + *(__sum16 *)(skb_checksum_start(skb) + skb->csum_offset) = 0; + return 0; +} + static int macb_start_xmit(struct sk_buff *skb, struct net_device *dev) { u16 queue_index = skb_get_queue_mapping(skb); @@ -1362,6 +1380,11 @@ static int macb_start_xmit(struct sk_buff *skb, struct net_device *dev) return NETDEV_TX_BUSY; } + if (macb_clear_csum(skb)) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + /* Map socket buffer for DMA transfer */ if (!macb_tx_map(bp, queue, skb)) { dev_kfree_skb_any(skb); From daa7ee8dfa64233789221685e73a6ef8159f80de Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Sun, 4 Sep 2016 23:02:21 +0530 Subject: [PATCH 119/271] net: smsc: remove build warning of duplicate definition The build of m32r was giving warning: In file included from drivers/net/ethernet/smsc/smc91x.c:92:0: drivers/net/ethernet/smsc/smc91x.h:448:0: warning: "SMC_inb" redefined #define SMC_inb(ioaddr, reg) ({ BUG(); 0; }) drivers/net/ethernet/smsc/smc91x.h:106:0: note: this is the location of the previous definition #define SMC_inb(a, r) inb(((u32)a) + (r)) drivers/net/ethernet/smsc/smc91x.h:449:0: warning: "SMC_outb" redefined #define SMC_outb(x, ioaddr, reg) BUG() drivers/net/ethernet/smsc/smc91x.h:108:0: note: this is the location of the previous definition #define SMC_outb(v, a, r) outb(v, ((u32)a) + (r)) Signed-off-by: Sudip Mukherjee Signed-off-by: David S. Miller --- drivers/net/ethernet/smsc/smc91x.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/smsc/smc91x.h b/drivers/net/ethernet/smsc/smc91x.h index e17671c9d1b0..ea8465467469 100644 --- a/drivers/net/ethernet/smsc/smc91x.h +++ b/drivers/net/ethernet/smsc/smc91x.h @@ -470,7 +470,9 @@ smc_pxa_dma_insw(void __iomem *ioaddr, struct smc_local *lp, int reg, int dma, #endif #if ! SMC_CAN_USE_8BIT +#undef SMC_inb #define SMC_inb(ioaddr, reg) ({ BUG(); 0; }) +#undef SMC_outb #define SMC_outb(x, ioaddr, reg) BUG() #define SMC_insb(a, r, p, l) BUG() #define SMC_outsb(a, r, p, l) BUG() From 5a56a0b3a45dd0cc5b2f7bec6afd053a474ed9f5 Mon Sep 17 00:00:00 2001 From: Mark Tomlinson Date: Mon, 5 Sep 2016 10:20:20 +1200 Subject: [PATCH 120/271] net: Don't delete routes in different VRFs When deleting an IP address from an interface, there is a clean-up of routes which refer to this local address. However, there was no check to see that the VRF matched. This meant that deletion wasn't confined to the VRF it should have been. To solve this, a new field has been added to fib_info to hold a table id. When removing fib entries corresponding to a local ip address, this table id is also used in the comparison. The table id is populated when the fib_info is created. This was already done in some places, but not in ip_rt_ioctl(). This has now been fixed. Fixes: 021dd3b8a142 ("net: Add routes to the table associated with the device") Acked-by: David Ahern Tested-by: David Ahern Signed-off-by: Mark Tomlinson Signed-off-by: David S. Miller --- include/net/ip_fib.h | 3 ++- net/ipv4/fib_frontend.c | 3 ++- net/ipv4/fib_semantics.c | 8 ++++++-- 3 files changed, 10 insertions(+), 4 deletions(-) diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 4079fc18ffe4..7d4a72e75f33 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -111,6 +111,7 @@ struct fib_info { unsigned char fib_scope; unsigned char fib_type; __be32 fib_prefsrc; + u32 fib_tb_id; u32 fib_priority; u32 *fib_metrics; #define fib_mtu fib_metrics[RTAX_MTU-1] @@ -319,7 +320,7 @@ void fib_flush_external(struct net *net); /* Exported by fib_semantics.c */ int ip_fib_check_default(__be32 gw, struct net_device *dev); int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force); -int fib_sync_down_addr(struct net *net, __be32 local); +int fib_sync_down_addr(struct net_device *dev, __be32 local); int fib_sync_up(struct net_device *dev, unsigned int nh_flags); extern u32 fib_multipath_secret __read_mostly; diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index ef2ebeb89d0f..1b25daf8c7f1 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -509,6 +509,7 @@ static int rtentry_to_fib_config(struct net *net, int cmd, struct rtentry *rt, if (!dev) return -ENODEV; cfg->fc_oif = dev->ifindex; + cfg->fc_table = l3mdev_fib_table(dev); if (colon) { struct in_ifaddr *ifa; struct in_device *in_dev = __in_dev_get_rtnl(dev); @@ -1027,7 +1028,7 @@ no_promotions: * First of all, we scan fib_info list searching * for stray nexthop entries, then ignite fib_flush. */ - if (fib_sync_down_addr(dev_net(dev), ifa->ifa_local)) + if (fib_sync_down_addr(dev, ifa->ifa_local)) fib_flush(dev_net(dev)); } } diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 539fa264e67d..e9f56225e53f 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -1057,6 +1057,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg) fi->fib_priority = cfg->fc_priority; fi->fib_prefsrc = cfg->fc_prefsrc; fi->fib_type = cfg->fc_type; + fi->fib_tb_id = cfg->fc_table; fi->fib_nhs = nhs; change_nexthops(fi) { @@ -1337,18 +1338,21 @@ nla_put_failure: * referring to it. * - device went down -> we must shutdown all nexthops going via it. */ -int fib_sync_down_addr(struct net *net, __be32 local) +int fib_sync_down_addr(struct net_device *dev, __be32 local) { int ret = 0; unsigned int hash = fib_laddr_hashfn(local); struct hlist_head *head = &fib_info_laddrhash[hash]; + struct net *net = dev_net(dev); + int tb_id = l3mdev_fib_table(dev); struct fib_info *fi; if (!fib_info_laddrhash || local == 0) return 0; hlist_for_each_entry(fi, head, fib_lhash) { - if (!net_eq(fi->fib_net, net)) + if (!net_eq(fi->fib_net, net) || + fi->fib_tb_id != tb_id) continue; if (fi->fib_prefsrc == local) { fi->fib_flags |= RTNH_F_DEAD; From 9d13744bb75078175ab49408f2abb980e4dbccc9 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 5 Sep 2016 01:57:35 -0400 Subject: [PATCH 121/271] bnxt_en: Fix TX push operation on ARM64. There is a code path where we are calling __iowrite64_copy() on an address that is not 64-bit aligned. This causes an exception on some architectures such as arm64. Fix that code path by using __iowrite32_copy(). Reported-by: JD Zheng Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 2cf79100c9cb..228c964e709a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -353,8 +353,8 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev) push_len = (length + sizeof(*tx_push) + 7) / 8; if (push_len > 16) { __iowrite64_copy(txr->tx_doorbell, tx_push_buf, 16); - __iowrite64_copy(txr->tx_doorbell + 4, tx_push_buf + 1, - push_len - 16); + __iowrite32_copy(txr->tx_doorbell + 4, tx_push_buf + 1, + (push_len - 16) << 1); } else { __iowrite64_copy(txr->tx_doorbell, tx_push_buf, push_len); From 751eb6b6042a596b0080967c1a529a9fe98dac1d Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Mon, 5 Sep 2016 16:06:31 +0800 Subject: [PATCH 122/271] ipv6: addrconf: fix dev refcont leak when DAD failed In general, when DAD detected IPv6 duplicate address, ifp->state will be set to INET6_IFADDR_STATE_ERRDAD and DAD is stopped by a delayed work, the call tree should be like this: ndisc_recv_ns -> addrconf_dad_failure <- missing ifp put -> addrconf_mod_dad_work -> schedule addrconf_dad_work() -> addrconf_dad_stop() <- missing ifp hold before call it addrconf_dad_failure() called with ifp refcont holding but not put. addrconf_dad_work() call addrconf_dad_stop() without extra holding refcount. This will not cause any issue normally. But the race between addrconf_dad_failure() and addrconf_dad_work() may cause ifp refcount leak and netdevice can not be unregister, dmesg show the following messages: IPv6: eth0: IPv6 duplicate address fe80::XX:XXXX:XXXX:XX detected! ... unregister_netdevice: waiting for eth0 to become free. Usage count = 1 Cc: stable@vger.kernel.org Fixes: c15b1ccadb32 ("ipv6: move DAD and addrconf_verify processing to workqueue") Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index bdf368eff5ab..2f1f5d439788 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1948,6 +1948,7 @@ errdad: spin_unlock_bh(&ifp->lock); addrconf_mod_dad_work(ifp, 0); + in6_ifa_put(ifp); } /* Join to solicited addr multicast group. @@ -3857,6 +3858,7 @@ static void addrconf_dad_work(struct work_struct *w) addrconf_dad_begin(ifp); goto out; } else if (action == DAD_ABORT) { + in6_ifa_hold(ifp); addrconf_dad_stop(ifp, 1); if (disable_ipv6) addrconf_ifdown(idev->dev, 0); From 0bd2223594a4dcddc1e34b15774a3a4776f7749e Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Thu, 1 Sep 2016 14:25:43 +0100 Subject: [PATCH 123/271] crypto: cryptd - initialize child shash_desc on import When calling .import() on a cryptd ahash_request, the structure members that describe the child transform in the shash_desc need to be initialized like they are when calling .init() Cc: stable@vger.kernel.org Signed-off-by: Ard Biesheuvel Signed-off-by: Herbert Xu --- crypto/cryptd.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/crypto/cryptd.c b/crypto/cryptd.c index 77207b41940c..0c654e59f215 100644 --- a/crypto/cryptd.c +++ b/crypto/cryptd.c @@ -631,9 +631,14 @@ static int cryptd_hash_export(struct ahash_request *req, void *out) static int cryptd_hash_import(struct ahash_request *req, const void *in) { - struct cryptd_hash_request_ctx *rctx = ahash_request_ctx(req); + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct cryptd_hash_ctx *ctx = crypto_ahash_ctx(tfm); + struct shash_desc *desc = cryptd_shash_desc(req); - return crypto_shash_import(&rctx->desc, in); + desc->tfm = ctx->child; + desc->flags = req->base.flags; + + return crypto_shash_import(desc, in); } static int cryptd_create_hash(struct crypto_template *tmpl, struct rtattr **tb, From e34f2ff40e0339f6a379e1ecf49e8f2759056453 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Giedrius=20Statkevi=C4=8Dius?= Date: Thu, 1 Sep 2016 20:47:02 +0300 Subject: [PATCH 124/271] ath9k: bring back direction setting in ath9k_{start_stop} MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A regression was introduced in commit id 79d4db1214a ("ath9k: cleanup led_pin initial") that broken the WLAN status led on my laptop with AR9287 after suspending and resuming. Steps to reproduce: * Suspend (laptop) * Resume (laptop) * Observe that the WLAN led no longer turns ON/OFF depending on the status and is always red Even though for my case it only needs to be set to OUT in ath9k_start but for consistency bring back the IN direction setting as well. Fixes: 79d4db1214a0 ("ath9k: cleanup led_pin initial") Cc: Miaoqing Pan Cc: Kalle Valo Cc: # 4.7+ Link: https://bugzilla.kernel.org/show_bug.cgi?id=151711 Signed-off-by: Giedrius Statkevičius [kvalo@qca.qualcomm.com: improve commit log] Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath9k/main.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index 6ca4337a98c4..7cb65c303f8d 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -718,9 +718,12 @@ static int ath9k_start(struct ieee80211_hw *hw) if (!ath_complete_reset(sc, false)) ah->reset_power_on = false; - if (ah->led_pin >= 0) + if (ah->led_pin >= 0) { ath9k_hw_set_gpio(ah, ah->led_pin, (ah->config.led_active_high) ? 1 : 0); + ath9k_hw_gpio_request_out(ah, ah->led_pin, NULL, + AR_GPIO_OUTPUT_MUX_AS_OUTPUT); + } /* * Reset key cache to sane defaults (all entries cleared) instead of @@ -864,9 +867,11 @@ static void ath9k_stop(struct ieee80211_hw *hw) spin_lock_bh(&sc->sc_pcu_lock); - if (ah->led_pin >= 0) + if (ah->led_pin >= 0) { ath9k_hw_set_gpio(ah, ah->led_pin, (ah->config.led_active_high) ? 0 : 1); + ath9k_hw_gpio_request_in(ah, ah->led_pin, NULL); + } ath_prepare_reset(sc); From ded89912156b1a47d940a0c954c43afbabd0c42c Mon Sep 17 00:00:00 2001 From: Arend Van Spriel Date: Mon, 5 Sep 2016 10:45:47 +0100 Subject: [PATCH 125/271] brcmfmac: avoid potential stack overflow in brcmf_cfg80211_start_ap() User-space can choose to omit NL80211_ATTR_SSID and only provide raw IE TLV data. When doing so it can provide SSID IE with length exceeding the allowed size. The driver further processes this IE copying it into a local variable without checking the length. Hence stack can be corrupted and used as exploit. Cc: stable@vger.kernel.org # v4.7 Reported-by: Daxing Guo Reviewed-by: Hante Meuleman Reviewed-by: Pieter-Paul Giesberts Reviewed-by: Franky Lin Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo --- drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 5db56a74e7df..b8aec5e5ef93 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -4527,7 +4527,7 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev, (u8 *)&settings->beacon.head[ie_offset], settings->beacon.head_len - ie_offset, WLAN_EID_SSID); - if (!ssid_ie) + if (!ssid_ie || ssid_ie->len > IEEE80211_MAX_SSID_LEN) return -EINVAL; memcpy(ssid_le.SSID, ssid_ie->data, ssid_ie->len); From fc2780b66b15092ac68272644a522c1624c48547 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 26 Aug 2016 11:59:26 +0100 Subject: [PATCH 126/271] drm/i915: Add GEN7_PCODE_MIN_FREQ_TABLE_GT_RATIO_OUT_OF_RANGE to SNB MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit According to the CI test machines, SNB also uses the GEN7_PCODE_MIN_FREQ_TABLE_GT_RATIO_OUT_OF_RANGE value to report a bad GEN6_PCODE_MIN_FREQ_TABLE request. [ 157.744641] WARNING: CPU: 5 PID: 9238 at drivers/gpu/drm/i915/intel_pm.c:7760 sandybridge_pcode_write+0x141/0x200 [i915] [ 157.744642] Missing switch case (16) in gen6_check_mailbox_status [ 157.744642] Modules linked in: snd_hda_intel i915 ax88179_178a usbnet mii x86_pkg_temp_thermal intel_powerclamp coretemp crct10dif_pclmul crc32_pclmul ghash_clmulni_intel snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec snd_hwdep snd_hda_core mei_me lpc_ich snd_pcm mei broadcom bcm_phy_lib tg3 ptp pps_core [last unloaded: vgem] [ 157.744658] CPU: 5 PID: 9238 Comm: drv_hangman Tainted: G U W 4.8.0-rc3-CI-CI_DRM_1589+ #1 [ 157.744658] Hardware name: Dell Inc. XPS 8300 /0Y2MRG, BIOS A06 10/17/2011 [ 157.744659] 0000000000000000 ffff88011f093a98 ffffffff81426415 ffff88011f093ae8 [ 157.744662] 0000000000000000 ffff88011f093ad8 ffffffff8107d2a6 00001e50810d3c9f [ 157.744663] ffff880128680000 0000000000000008 0000000000000000 ffff88012868a650 [ 157.744665] Call Trace: [ 157.744669] [] dump_stack+0x67/0x92 [ 157.744672] [] __warn+0xc6/0xe0 [ 157.744673] [] warn_slowpath_fmt+0x4a/0x50 [ 157.744685] [] sandybridge_pcode_write+0x141/0x200 [i915] [ 157.744697] [] intel_enable_gt_powersave+0x64a/0x1330 [i915] [ 157.744712] [] ? i9xx_emit_request+0x1b/0x80 [i915] [ 157.744725] [] __i915_add_request+0x1e3/0x370 [i915] [ 157.744738] [] i915_gem_do_execbuffer.isra.16+0xced/0x1b80 [i915] [ 157.744740] [] ? __might_fault+0x3e/0x90 [ 157.744752] [] i915_gem_execbuffer2+0xc2/0x2a0 [i915] [ 157.744753] [] drm_ioctl+0x207/0x4c0 [ 157.744765] [] ? i915_gem_execbuffer+0x360/0x360 [i915] [ 157.744767] [] ? debug_lockdep_rcu_enabled+0x1d/0x20 [ 157.744769] [] do_vfs_ioctl+0x8e/0x680 [ 157.744770] [] ? __might_fault+0x87/0x90 [ 157.744771] [] ? __might_fault+0x3e/0x90 [ 157.744773] [] ? trace_hardirqs_on_caller+0x122/0x1b0 [ 157.744774] [] SyS_ioctl+0x3c/0x70 [ 157.744776] [] entry_SYSCALL_64_fastpath+0x1c/0xac Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97491 Fixes: 87660502f1a4 ("drm/i915/gen6+: Interpret mailbox error flags") Signed-off-by: Chris Wilson Cc: Lyude Cc: Matt Roper Cc: Maarten Lankhorst Cc: Daniel Vetter Cc: Ville Syrjälä Cc: stable@vger.kernel.org Link: http://patchwork.freedesktop.org/patch/msgid/20160826105926.3413-1-chris@chris-wilson.co.uk Acked-by: Mika Kuoppala (cherry picked from commit 7850d1c35344c7bd6a357240f2f9f60fc2c097b5) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_pm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 53e13c10e4ea..2d2481392824 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -7859,6 +7859,7 @@ static inline int gen6_check_mailbox_status(struct drm_i915_private *dev_priv) case GEN6_PCODE_ILLEGAL_CMD: return -ENXIO; case GEN6_PCODE_MIN_FREQ_TABLE_GT_RATIO_OUT_OF_RANGE: + case GEN7_PCODE_MIN_FREQ_TABLE_GT_RATIO_OUT_OF_RANGE: return -EOVERFLOW; case GEN6_PCODE_TIMEOUT: return -ETIMEDOUT; From 82469c59d222f839ded5cd282172258e026f9112 Mon Sep 17 00:00:00 2001 From: Gabriel Krisman Bertazi Date: Tue, 6 Sep 2016 17:39:13 -0300 Subject: [PATCH 127/271] nvme: Don't suspend admin queue that wasn't created This fixes a regression in my previous commit c21377f8366c ("nvme: Suspend all queues before deletion"), which provoked an Oops in the removal path when removing a device that became IO incapable very early at probe (i.e. after a failed EEH recovery). Turns out, if the error occurred very early at the probe path, before even configuring the admin queue, we might try to suspend the uninitialized admin queue, accessing bad memory. Fixes: c21377f8366c ("nvme: Suspend all queues before deletion") Signed-off-by: Gabriel Krisman Bertazi Reviewed-by: Jay Freyensee Signed-off-by: Jens Axboe --- drivers/nvme/host/pci.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 8dcf5a960951..be84a84a40f7 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -1693,7 +1693,12 @@ static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown) nvme_suspend_queue(dev->queues[i]); if (csts & NVME_CSTS_CFS || !(csts & NVME_CSTS_RDY)) { - nvme_suspend_queue(dev->queues[0]); + /* A device might become IO incapable very soon during + * probe, before the admin queue is configured. Thus, + * queue_count can be 0 here. + */ + if (dev->queue_count) + nvme_suspend_queue(dev->queues[0]); } else { nvme_disable_io_queues(dev); nvme_disable_admin_queue(dev, shutdown); From 9dd4aaef194e45d96c2d0b6232a9cbc5430ad789 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 6 Sep 2016 20:22:21 -0700 Subject: [PATCH 128/271] MAINTAINERS: Update CPMAC email address Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 0537211788e5..c74792178b99 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3268,7 +3268,7 @@ S: Maintained F: drivers/net/wan/cosa* CPMAC ETHERNET DRIVER -M: Florian Fainelli +M: Florian Fainelli L: netdev@vger.kernel.org S: Maintained F: drivers/net/ethernet/ti/cpmac.c From db91e2370e087967cb6b6425c092188767fb5e00 Mon Sep 17 00:00:00 2001 From: Daniel Baluta Date: Wed, 7 Sep 2016 19:49:34 +0930 Subject: [PATCH 129/271] tools/lguest: Don't bork the terminal in case of wrong args Running lguest without arguments or with a wrong argument name borks the terminal, because the cleanup handler is set up too late in the initialization process. Signed-off-by: Daniel Baluta Signed-off-by: Rusty Russell Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Ingo Molnar --- tools/lguest/lguest.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/lguest/lguest.c b/tools/lguest/lguest.c index d9836c5eb694..11c8d9bc762e 100644 --- a/tools/lguest/lguest.c +++ b/tools/lguest/lguest.c @@ -3266,6 +3266,9 @@ int main(int argc, char *argv[]) } } + /* If we exit via err(), this kills all the threads, restores tty. */ + atexit(cleanup_devices); + /* We always have a console device, and it's always device 1. */ setup_console(); @@ -3369,9 +3372,6 @@ int main(int argc, char *argv[]) /* Ensure that we terminate if a device-servicing child dies. */ signal(SIGCHLD, kill_launcher); - /* If we exit via err(), this kills all the threads, restores tty. */ - atexit(cleanup_devices); - /* If requested, chroot to a directory */ if (chroot_path) { if (chroot(chroot_path) != 0) From c291b015158577be533dd5a959dfc09bab119eed Mon Sep 17 00:00:00 2001 From: Dou Liyang Date: Wed, 7 Sep 2016 10:21:33 +0800 Subject: [PATCH 130/271] x86/apic: Fix num_processors value in case of failure If the topology package map check of the APIC ID and the CPU is a failure, we don't generate the processor info for that APIC ID yet we increase disabled_cpus by one - which is buggy. Only increase num_processors once we are sure we don't fail. Signed-off-by: Dou Liyang Acked-by: David Rientjes Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1473214893-16481-1-git-send-email-douly.fnst@cn.fujitsu.com [ Rewrote the changelog. ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/apic.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 50c95af0f017..f3e9b2df4b16 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -2093,7 +2093,6 @@ int generic_processor_info(int apicid, int version) return -EINVAL; } - num_processors++; if (apicid == boot_cpu_physical_apicid) { /* * x86_bios_cpu_apicid is required to have processors listed @@ -2116,10 +2115,13 @@ int generic_processor_info(int apicid, int version) pr_warning("APIC: Package limit reached. Processor %d/0x%x ignored.\n", thiscpu, apicid); + disabled_cpus++; return -ENOSPC; } + num_processors++; + /* * Validate version */ From 7e9d2850a8db4e0d85a20bb692198bf2cc4be3b7 Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Thu, 8 Sep 2016 11:11:00 +0200 Subject: [PATCH 131/271] ARM: dts: STiH410: Handle interconnect clock required by EHCI/OHCI (USB) The STiH4{07,10} platform contains some interconnect clocks which are used by various IPs. If this clock isn't handled correctly by ST's EHCI/OHCI drivers, their hub won't be found, the following error be shown and the result will be non-working USB: [ 97.221963] hub 2-1:1.0: hub_ext_port_status failed (err = -110) Cc: stable@vger.kernel.org Tested-by: Peter Griffin Signed-off-by: Lee Jones Acked-by: Patrice Chotard --- arch/arm/boot/dts/stih410.dtsi | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/arch/arm/boot/dts/stih410.dtsi b/arch/arm/boot/dts/stih410.dtsi index 18ed1ad10d32..40318869c733 100644 --- a/arch/arm/boot/dts/stih410.dtsi +++ b/arch/arm/boot/dts/stih410.dtsi @@ -41,7 +41,8 @@ compatible = "st,st-ohci-300x"; reg = <0x9a03c00 0x100>; interrupts = ; - clocks = <&clk_s_c0_flexgen CLK_TX_ICN_DISP_0>; + clocks = <&clk_s_c0_flexgen CLK_TX_ICN_DISP_0>, + <&clk_s_c0_flexgen CLK_RX_ICN_DISP_0>; resets = <&powerdown STIH407_USB2_PORT0_POWERDOWN>, <&softreset STIH407_USB2_PORT0_SOFTRESET>; reset-names = "power", "softreset"; @@ -57,7 +58,8 @@ interrupts = ; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usb0>; - clocks = <&clk_s_c0_flexgen CLK_TX_ICN_DISP_0>; + clocks = <&clk_s_c0_flexgen CLK_TX_ICN_DISP_0>, + <&clk_s_c0_flexgen CLK_RX_ICN_DISP_0>; resets = <&powerdown STIH407_USB2_PORT0_POWERDOWN>, <&softreset STIH407_USB2_PORT0_SOFTRESET>; reset-names = "power", "softreset"; @@ -71,7 +73,8 @@ compatible = "st,st-ohci-300x"; reg = <0x9a83c00 0x100>; interrupts = ; - clocks = <&clk_s_c0_flexgen CLK_TX_ICN_DISP_0>; + clocks = <&clk_s_c0_flexgen CLK_TX_ICN_DISP_0>, + <&clk_s_c0_flexgen CLK_RX_ICN_DISP_0>; resets = <&powerdown STIH407_USB2_PORT1_POWERDOWN>, <&softreset STIH407_USB2_PORT1_SOFTRESET>; reset-names = "power", "softreset"; @@ -87,7 +90,8 @@ interrupts = ; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usb1>; - clocks = <&clk_s_c0_flexgen CLK_TX_ICN_DISP_0>; + clocks = <&clk_s_c0_flexgen CLK_TX_ICN_DISP_0>, + <&clk_s_c0_flexgen CLK_RX_ICN_DISP_0>; resets = <&powerdown STIH407_USB2_PORT1_POWERDOWN>, <&softreset STIH407_USB2_PORT1_SOFTRESET>; reset-names = "power", "softreset"; From 78567f135d9bbbaf4538f63656d3e4d957c35fe9 Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Thu, 8 Sep 2016 11:11:00 +0200 Subject: [PATCH 132/271] ARM: dts: STiH407-family: Provide interconnect clock for consumption in ST SDHCI The STiH4{07,10} platform contains some interconnect clocks which are used by various IPs. If these clocks aren't handled correctly by ST's SDHCI driver MMC will break and the following output can be observed: [ 13.916949] mmc0: Timeout waiting for hardware interrupt. [ 13.922349] sdhci: =========== REGISTER DUMP (mmc0)=========== [ 13.928175] sdhci: Sys addr: 0x00000000 | Version: 0x00001002 [ 13.933999] sdhci: Blk size: 0x00007040 | Blk cnt: 0x00000001 [ 13.939825] sdhci: Argument: 0x00fffff0 | Trn mode: 0x00000013 [ 13.945650] sdhci: Present: 0x1fff0206 | Host ctl: 0x00000011 [ 13.951475] sdhci: Power: 0x0000000f | Blk gap: 0x00000080 [ 13.957300] sdhci: Wake-up: 0x00000000 | Clock: 0x00003f07 [ 13.963126] sdhci: Timeout: 0x00000004 | Int stat: 0x00000000 [ 13.968952] sdhci: Int enab: 0x02ff008b | Sig enab: 0x02ff008b [ 13.974777] sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000 [ 13.980602] sdhci: Caps: 0x21ed3281 | Caps_1: 0x00000000 [ 13.986428] sdhci: Cmd: 0x0000063a | Max curr: 0x00000000 [ 13.992252] sdhci: Host ctl2: 0x00000000 [ 13.996166] sdhci: ADMA Err: 0x00000000 | ADMA Ptr: 0x7c048200 [ 14.001990] sdhci: =========================================== [ 14.009802] mmc0: Got data interrupt 0x02000000 even though no data operation was in progress. Cc: stable@vger.kernel.org Tested-by: Peter Griffin Signed-off-by: Lee Jones Acked-by: Patrice Chotard --- arch/arm/boot/dts/stih407-family.dtsi | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/arm/boot/dts/stih407-family.dtsi b/arch/arm/boot/dts/stih407-family.dtsi index d294e82447a2..8b063ab10c19 100644 --- a/arch/arm/boot/dts/stih407-family.dtsi +++ b/arch/arm/boot/dts/stih407-family.dtsi @@ -550,8 +550,9 @@ interrupt-names = "mmcirq"; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_mmc0>; - clock-names = "mmc"; - clocks = <&clk_s_c0_flexgen CLK_MMC_0>; + clock-names = "mmc", "icn"; + clocks = <&clk_s_c0_flexgen CLK_MMC_0>, + <&clk_s_c0_flexgen CLK_RX_ICN_HVA>; bus-width = <8>; non-removable; }; @@ -565,8 +566,9 @@ interrupt-names = "mmcirq"; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_sd1>; - clock-names = "mmc"; - clocks = <&clk_s_c0_flexgen CLK_MMC_1>; + clock-names = "mmc", "icn"; + clocks = <&clk_s_c0_flexgen CLK_MMC_1>, + <&clk_s_c0_flexgen CLK_RX_ICN_HVA>; resets = <&softreset STIH407_MMC1_SOFTRESET>; bus-width = <4>; }; From a4497a86fb9b855c5ac8503fdc959393b00bb643 Mon Sep 17 00:00:00 2001 From: Prarit Bhargava Date: Thu, 8 Sep 2016 10:15:28 -0400 Subject: [PATCH 133/271] x86, clock: Fix kvm guest tsc initialization MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When booting a kvm guest on AMD with the latest kernel the following messages are displayed in the boot log: tsc: Unable to calibrate against PIT tsc: HPET/PMTIMER calibration failed aa297292d708 ("x86/tsc: Enumerate SKL cpu_khz and tsc_khz via CPUID") introduced a change to account for a difference in cpu and tsc frequencies for Intel SKL processors. Before this change the native tsc set x86_platform.calibrate_tsc to native_calibrate_tsc() which is a hardware calibration of the tsc, and in tsc_init() executed tsc_khz = x86_platform.calibrate_tsc(); cpu_khz = tsc_khz; The kvm code changed x86_platform.calibrate_tsc to kvm_get_tsc_khz() and executed the same tsc_init() function. This meant that KVM guests did not execute the native hardware calibration function. After aa297292d708, there are separate native calibrations for cpu_khz and tsc_khz. The code sets x86_platform.calibrate_tsc to native_calibrate_tsc() which is now an Intel specific calibration function, and x86_platform.calibrate_cpu to native_calibrate_cpu() which is the "old" native_calibrate_tsc() function (ie, the native hardware calibration function). tsc_init() now does cpu_khz = x86_platform.calibrate_cpu(); tsc_khz = x86_platform.calibrate_tsc(); if (tsc_khz == 0) tsc_khz = cpu_khz; else if (abs(cpu_khz - tsc_khz) * 10 > tsc_khz) cpu_khz = tsc_khz; The kvm code should not call the hardware initialization in native_calibrate_cpu(), as it isn't applicable for kvm and it didn't do that prior to aa297292d708. This patch resolves this issue by setting x86_platform.calibrate_cpu to kvm_get_tsc_khz(). v2: I had originally set x86_platform.calibrate_cpu to cpu_khz_from_cpuid(), however, pbonzini pointed out that the CPUID leaf in that function is not available in KVM. I have changed the function pointer to kvm_get_tsc_khz(). Fixes: aa297292d708 ("x86/tsc: Enumerate SKL cpu_khz and tsc_khz via CPUID") Signed-off-by: Prarit Bhargava Cc: Paolo Bonzini Cc: Radim Krčmář Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: x86@kernel.org Cc: Len Brown Cc: "Peter Zijlstra (Intel)" Cc: Borislav Petkov Cc: Adrian Hunter Cc: "Christopher S. Hall" Cc: David Woodhouse Cc: kvm@vger.kernel.org Signed-off-by: Paolo Bonzini --- arch/x86/kernel/kvmclock.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index 1d39bfbd26bb..3692249a70f1 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -289,6 +289,7 @@ void __init kvmclock_init(void) put_cpu(); x86_platform.calibrate_tsc = kvm_get_tsc_khz; + x86_platform.calibrate_cpu = kvm_get_tsc_khz; x86_platform.get_wallclock = kvm_get_wallclock; x86_platform.set_wallclock = kvm_set_wallclock; #ifdef CONFIG_X86_LOCAL_APIC From 015282c9eb6da05bfad6ff009078f91e06c0c98f Mon Sep 17 00:00:00 2001 From: Wenbo Wang Date: Thu, 8 Sep 2016 12:12:11 -0400 Subject: [PATCH 134/271] nvme/quirk: Add a delay before checking device ready for memblaze device Signed-off-by: Wenbo Wang Signed-off-by: Jens Axboe --- drivers/nvme/host/pci.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index be84a84a40f7..60f7eab11865 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -2117,6 +2117,8 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_IDENTIFY_CNS, }, { PCI_DEVICE(0x1c58, 0x0003), /* HGST adapter */ .driver_data = NVME_QUIRK_DELAY_BEFORE_CHK_RDY, }, + { PCI_DEVICE(0x1c5f, 0x0540), /* Memblaze Pblaze4 adapter */ + .driver_data = NVME_QUIRK_DELAY_BEFORE_CHK_RDY, }, { PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_EXPRESS, 0xffffff) }, { PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2001) }, { 0, } From 47a7b0d8888c04c9746812820b6e60553cc77bbc Mon Sep 17 00:00:00 2001 From: Guoqing Jiang Date: Sun, 4 Sep 2016 22:17:28 -0400 Subject: [PATCH 135/271] md-cluster: make md-cluster also can work when compiled into kernel The md-cluster is compiled as module by default, if it is compiled by built-in way, then we can't make md-cluster works. [64782.630008] md/raid1:md127: active with 2 out of 2 mirrors [64782.630528] md-cluster module not found. [64782.630530] md127: Could not setup cluster service (-2) Fixes: edb39c9 ("Introduce md_cluster_operations to handle cluster functions") Cc: stable@vger.kernel.org (v4.1+) Reported-by: Marc Smith Reviewed-by: NeilBrown Signed-off-by: Guoqing Jiang Signed-off-by: Shaohua Li --- drivers/md/md.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 67642bacd597..915e84d631a2 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -7610,16 +7610,12 @@ EXPORT_SYMBOL(unregister_md_cluster_operations); int md_setup_cluster(struct mddev *mddev, int nodes) { - int err; - - err = request_module("md-cluster"); - if (err) { - pr_err("md-cluster module not found.\n"); - return -ENOENT; - } - + if (!md_cluster_ops) + request_module("md-cluster"); spin_lock(&pers_lock); + /* ensure module won't be unloaded */ if (!md_cluster_ops || !try_module_get(md_cluster_mod)) { + pr_err("can't find md-cluster module or get it's reference.\n"); spin_unlock(&pers_lock); return -ENOENT; } From 76061f631c2ea4ab9c4d66f3a96ecc5737f5aaf7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 7 Sep 2016 08:34:11 -0700 Subject: [PATCH 136/271] tcp: fastopen: avoid negative sk_forward_alloc When DATA and/or FIN are carried in a SYN/ACK message or SYN message, we append an skb in socket receive queue, but we forget to call sk_forced_mem_schedule(). Effect is that the socket has a negative sk->sk_forward_alloc as long as the message is not read by the application. Josh Hunt fixed a similar issue in commit d22e15371811 ("tcp: fix tcp fin memory accounting") Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path") Signed-off-by: Eric Dumazet Reviewed-by: Josh Hunt Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 62a5751d4fe1..4e777a3243f9 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -150,6 +150,7 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb) tp->segs_in = 0; tcp_segs_in(tp, skb); __skb_pull(skb, tcp_hdrlen(skb)); + sk_forced_mem_schedule(sk, skb->truesize); skb_set_owner_r(skb, sk); TCP_SKB_CB(skb)->seq++; From 0dbf657c3993a269fa0cd7dabbb9c02741d4aefa Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Wed, 7 Sep 2016 19:07:57 +0300 Subject: [PATCH 137/271] net/mlx5e: Fix xmit_more counter race issue Update the xmit_more counter before notifying the HW, to prevent a possible use-after-free of the skb. Fixes: c8cf78fe100b ("net/mlx5e: Add ethtool counter for TX xmit_more") Signed-off-by: Tariq Toukan Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 988eca99ee0f..eb0e72537f10 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -356,6 +356,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) sq->stats.stopped++; } + sq->stats.xmit_more += skb->xmit_more; if (!skb->xmit_more || netif_xmit_stopped(sq->txq)) { int bf_sz = 0; @@ -375,7 +376,6 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) sq->stats.packets++; sq->stats.bytes += num_bytes; - sq->stats.xmit_more += skb->xmit_more; return NETDEV_TX_OK; dma_unmap_wqe_err: From e7e31ca43d6bedf1c551b1f9c7e78d51c9a45790 Mon Sep 17 00:00:00 2001 From: Bodong Wang Date: Wed, 7 Sep 2016 19:07:58 +0300 Subject: [PATCH 138/271] net/mlx5e: Move an_disable_cap bit to a new position Previous an_disable_cap position bit31 is deprecated to be use in driver with newer firmware. New firmware will advertise the same capability in bit29. Old capability didn't allow setting more than one protocol for a specific speed when autoneg is off, while newer firmware will allow this and it is indicated in the new capability location. Signed-off-by: Bodong Wang Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- include/linux/mlx5/mlx5_ifc.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 21bc4557b67a..d1f9a581aca8 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -6710,9 +6710,10 @@ struct mlx5_ifc_pude_reg_bits { }; struct mlx5_ifc_ptys_reg_bits { - u8 an_disable_cap[0x1]; + u8 reserved_at_0[0x1]; u8 an_disable_admin[0x1]; - u8 reserved_at_2[0x6]; + u8 an_disable_cap[0x1]; + u8 reserved_at_3[0x5]; u8 local_port[0x8]; u8 reserved_at_10[0xd]; u8 proto_mask[0x3]; From 7abc2110779b133e54ce681a4d0be3d89304eda1 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Wed, 7 Sep 2016 19:07:59 +0300 Subject: [PATCH 139/271] net/mlx5e: Prevent casting overflow On 64 bits architectures unsigned long is longer than u32, casting to unsigned long will result in overflow. We need to first allocate an unsigned long variable, then assign the wanted value. Fixes: 665bc53969d7 ('net/mlx5e: Use new ethtool get/set link ksettings API') Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index d0cf8fa22659..98e1a4a44268 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -659,9 +659,10 @@ out: static void ptys2ethtool_supported_link(unsigned long *supported_modes, u32 eth_proto_cap) { + unsigned long proto_cap = eth_proto_cap; int proto; - for_each_set_bit(proto, (unsigned long *)ð_proto_cap, MLX5E_LINK_MODES_NUMBER) + for_each_set_bit(proto, &proto_cap, MLX5E_LINK_MODES_NUMBER) bitmap_or(supported_modes, supported_modes, ptys2ethtool_table[proto].supported, __ETHTOOL_LINK_MODE_MASK_NBITS); @@ -670,9 +671,10 @@ static void ptys2ethtool_supported_link(unsigned long *supported_modes, static void ptys2ethtool_adver_link(unsigned long *advertising_modes, u32 eth_proto_cap) { + unsigned long proto_cap = eth_proto_cap; int proto; - for_each_set_bit(proto, (unsigned long *)ð_proto_cap, MLX5E_LINK_MODES_NUMBER) + for_each_set_bit(proto, &proto_cap, MLX5E_LINK_MODES_NUMBER) bitmap_or(advertising_modes, advertising_modes, ptys2ethtool_table[proto].advertised, __ETHTOOL_LINK_MODE_MASK_NBITS); From 4e39883d9c7426f8246ef86a32dc6a6f06f5aace Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Wed, 7 Sep 2016 19:08:00 +0300 Subject: [PATCH 140/271] net/mlx5e: Fix global PFC counters replication Currently when reading global PFC statistics we left the counter iterator out of the equation and we ended up reading the same counter over and over again. Instead of reading the counter at index 0 on every iteration we now read the counter at index (i). Fixes: e989d5a532ce ('net/mlx5e: Expose flow control counters to ethtool') Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 98e1a4a44268..7a346bb2ed00 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -331,7 +331,7 @@ static void mlx5e_get_ethtool_stats(struct net_device *dev, if (mlx5e_query_global_pause_combined(priv)) { for (i = 0; i < NUM_PPORT_PER_PRIO_PFC_COUNTERS; i++) { data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pport.per_prio_counters[0], - pport_per_prio_pfc_stats_desc, 0); + pport_per_prio_pfc_stats_desc, i); } } From cd17d230dd060a12f7451c0caeedb3fd5158eaf9 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Wed, 7 Sep 2016 19:08:01 +0300 Subject: [PATCH 141/271] net/mlx5e: Fix parsing of vlan packets when updating lro header Currently vlan tagged packets were not parsed correctly and assumed to be regular IPv4/IPv6 packets. We should check for 802.1Q/802.1ad tags and update the lro header accordingly. This fixes the use case where LRO is on and rxvlan is off (vlan stripping is off). Fixes: e586b3b0baee ('net/mlx5: Ethernet Datapath files') Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- .../net/ethernet/mellanox/mlx5/core/en_rx.c | 22 +++++++++++++------ 1 file changed, 15 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index b6f8ebbdb487..e7c969df3dad 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -637,24 +637,32 @@ bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq) static void mlx5e_lro_update_hdr(struct sk_buff *skb, struct mlx5_cqe64 *cqe, u32 cqe_bcnt) { - struct ethhdr *eth = (struct ethhdr *)(skb->data); - struct iphdr *ipv4 = (struct iphdr *)(skb->data + ETH_HLEN); - struct ipv6hdr *ipv6 = (struct ipv6hdr *)(skb->data + ETH_HLEN); + struct ethhdr *eth = (struct ethhdr *)(skb->data); + struct iphdr *ipv4; + struct ipv6hdr *ipv6; struct tcphdr *tcp; + int network_depth = 0; + __be16 proto; + u16 tot_len; u8 l4_hdr_type = get_cqe_l4_hdr_type(cqe); int tcp_ack = ((CQE_L4_HDR_TYPE_TCP_ACK_NO_DATA == l4_hdr_type) || (CQE_L4_HDR_TYPE_TCP_ACK_AND_DATA == l4_hdr_type)); - u16 tot_len = cqe_bcnt - ETH_HLEN; + skb->mac_len = ETH_HLEN; + proto = __vlan_get_protocol(skb, eth->h_proto, &network_depth); - if (eth->h_proto == htons(ETH_P_IP)) { - tcp = (struct tcphdr *)(skb->data + ETH_HLEN + + ipv4 = (struct iphdr *)(skb->data + network_depth); + ipv6 = (struct ipv6hdr *)(skb->data + network_depth); + tot_len = cqe_bcnt - network_depth; + + if (proto == htons(ETH_P_IP)) { + tcp = (struct tcphdr *)(skb->data + network_depth + sizeof(struct iphdr)); ipv6 = NULL; skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; } else { - tcp = (struct tcphdr *)(skb->data + ETH_HLEN + + tcp = (struct tcphdr *)(skb->data + network_depth + sizeof(struct ipv6hdr)); ipv4 = NULL; skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6; From db7196a0d0984b933ccf2cd6a60e26abf466e8a3 Mon Sep 17 00:00:00 2001 From: Artem Germanov Date: Wed, 7 Sep 2016 10:49:36 -0700 Subject: [PATCH 142/271] tcp: cwnd does not increase in TCP YeAH Commit 76174004a0f19785a328f40388e87e982bbf69b9 (tcp: do not slow start when cwnd equals ssthresh ) introduced regression in TCP YeAH. Using 100ms delay 1% loss virtual ethernet link kernel 4.2 shows bandwidth ~500KB/s for single TCP connection and kernel 4.3 and above (including 4.8-rc4) shows bandwidth ~100KB/s. That is caused by stalled cwnd when cwnd equals ssthresh. This patch fixes it by proper increasing cwnd in this case. Signed-off-by: Artem Germanov Acked-by: Dmitry Adamushko Signed-off-by: David S. Miller --- net/ipv4/tcp_yeah.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/tcp_yeah.c b/net/ipv4/tcp_yeah.c index 028eb046ea40..9c5fc973267f 100644 --- a/net/ipv4/tcp_yeah.c +++ b/net/ipv4/tcp_yeah.c @@ -76,7 +76,7 @@ static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) if (!tcp_is_cwnd_limited(sk)) return; - if (tp->snd_cwnd <= tp->snd_ssthresh) + if (tcp_in_slow_start(tp)) tcp_slow_start(tp, acked); else if (!yeah->doing_reno_now) { From 312fada1f9f87fb55ace4b5a55a70a9eea5100fd Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 7 Sep 2016 19:45:35 +0100 Subject: [PATCH 143/271] nfp: remove linux/version.h includes Remove unnecessary version.h includes. Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Reviewed-by: Dinan Gunawardena Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 1 - drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c | 1 - drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c | 1 - 3 files changed, 3 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 88678c172b19..e4fe0f080f34 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -41,7 +41,6 @@ * Chris Telfer */ -#include #include #include #include diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c index 7d7933d00b8f..4c9897220969 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c @@ -40,7 +40,6 @@ * Brad Petrus */ -#include #include #include #include diff --git a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c index 37abef016a0a..6f22b0e12ac7 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c @@ -38,7 +38,6 @@ * Rolf Neugebauer */ -#include #include #include #include From 313b345cbff566340022c82267a377e1e493ef90 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 7 Sep 2016 19:45:36 +0100 Subject: [PATCH 144/271] nfp: drop support for old firmware ABIs Be more strict about FW versions. Drop support for old transitional revisions which were never used in production. Dropping support for FW ABI version 0.0.0.0 is particularly useful because 0 could just be uninitialized memory. Signed-off-by: Jakub Kicinski Reviewed-by: Dinan Gunawardena Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c index 6f22b0e12ac7..f7062cb648e1 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c @@ -133,7 +133,7 @@ static int nfp_netvf_pci_probe(struct pci_dev *pdev, } nfp_net_get_fw_version(&fw_ver, ctrl_bar); - if (fw_ver.class != NFP_NET_CFG_VERSION_CLASS_GENERIC) { + if (fw_ver.resv || fw_ver.class != NFP_NET_CFG_VERSION_CLASS_GENERIC) { dev_err(&pdev->dev, "Unknown Firmware ABI %d.%d.%d.%d\n", fw_ver.resv, fw_ver.class, fw_ver.major, fw_ver.minor); err = -EINVAL; @@ -141,9 +141,7 @@ static int nfp_netvf_pci_probe(struct pci_dev *pdev, } /* Determine stride */ - if (nfp_net_fw_ver_eq(&fw_ver, 0, 0, 0, 0) || - nfp_net_fw_ver_eq(&fw_ver, 0, 0, 0, 1) || - nfp_net_fw_ver_eq(&fw_ver, 0, 0, 0x12, 0x48)) { + if (nfp_net_fw_ver_eq(&fw_ver, 0, 0, 0, 1)) { stride = 2; tx_bar_no = NFP_NET_Q0_BAR; rx_bar_no = NFP_NET_Q1_BAR; From ebecefc820ec791a216382a68b937577bb69d3f7 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 7 Sep 2016 19:45:37 +0100 Subject: [PATCH 145/271] nfp: don't pad frames on receive There is no need to pad frames to ETH_ZLEN on RX. Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Reviewed-by: Dinan Gunawardena Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index e4fe0f080f34..252e4924de0f 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -1440,10 +1440,6 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget) nfp_net_set_hash(nn->netdev, skb, rxd); - /* Pad small frames to minimum */ - if (skb_put_padto(skb, 60)) - break; - /* Stats update */ u64_stats_update_begin(&r_vec->rx_sync); r_vec->rx_pkts++; From 293f293637b55db4f9f522a5a72514e98a541076 Mon Sep 17 00:00:00 2001 From: Suzuki K Poulose Date: Thu, 8 Sep 2016 16:25:49 +0100 Subject: [PATCH 146/271] kvm-arm: Unmap shadow pagetables properly On arm/arm64, we depend on the kvm_unmap_hva* callbacks (via mmu_notifiers::invalidate_*) to unmap the stage2 pagetables when the userspace buffer gets unmapped. However, when the Hypervisor process exits without explicit unmap of the guest buffers, the only notifier we get is kvm_arch_flush_shadow_all() (via mmu_notifier::release ) which does nothing on arm. Later this causes us to access pages that were already released [via exit_mmap() -> unmap_vmas()] when we actually get to unmap the stage2 pagetable [via kvm_arch_destroy_vm() -> kvm_free_stage2_pgd()]. This triggers crashes with CONFIG_DEBUG_PAGEALLOC, which unmaps any free'd pages from the linear map. [ 757.644120] Unable to handle kernel paging request at virtual address ffff800661e00000 [ 757.652046] pgd = ffff20000b1a2000 [ 757.655471] [ffff800661e00000] *pgd=00000047fffe3003, *pud=00000047fcd8c003, *pmd=00000047fcc7c003, *pte=00e8004661e00712 [ 757.666492] Internal error: Oops: 96000147 [#3] PREEMPT SMP [ 757.672041] Modules linked in: [ 757.675100] CPU: 7 PID: 3630 Comm: qemu-system-aar Tainted: G D 4.8.0-rc1 #3 [ 757.683240] Hardware name: AppliedMicro X-Gene Mustang Board/X-Gene Mustang Board, BIOS 3.06.15 Aug 19 2016 [ 757.692938] task: ffff80069cdd3580 task.stack: ffff8006adb7c000 [ 757.698840] PC is at __flush_dcache_area+0x1c/0x40 [ 757.703613] LR is at kvm_flush_dcache_pmd+0x60/0x70 [ 757.708469] pc : [] lr : [] pstate: 20000145 ... [ 758.357249] [] __flush_dcache_area+0x1c/0x40 [ 758.363059] [] unmap_stage2_range+0x458/0x5f0 [ 758.368954] [] kvm_free_stage2_pgd+0x34/0x60 [ 758.374761] [] kvm_arch_destroy_vm+0x20/0x68 [ 758.380570] [] kvm_put_kvm+0x210/0x358 [ 758.385860] [] kvm_vm_release+0x2c/0x40 [ 758.391239] [] __fput+0x114/0x2e8 [ 758.396096] [] ____fput+0xc/0x18 [ 758.400869] [] task_work_run+0x108/0x138 [ 758.406332] [] do_exit+0x48c/0x10e8 [ 758.411363] [] do_group_exit+0x6c/0x130 [ 758.416739] [] get_signal+0x284/0xa18 [ 758.421943] [] do_signal+0x158/0x860 [ 758.427060] [] do_notify_resume+0x6c/0x88 [ 758.432608] [] work_pending+0x10/0x14 [ 758.437812] Code: 9ac32042 8b010001 d1000443 8a230000 (d50b7e20) This patch fixes the issue by moving the kvm_free_stage2_pgd() to kvm_arch_flush_shadow_all(). Cc: # 3.9+ Tested-by: Itaru Kitayama Reported-by: Itaru Kitayama Reported-by: James Morse Cc: Marc Zyngier Cc: Catalin Marinas Cc: Christoffer Dall Signed-off-by: Suzuki K Poulose Signed-off-by: Christoffer Dall --- arch/arm/kvm/arm.c | 2 -- arch/arm/kvm/mmu.c | 1 + 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index 75f130ef6504..c94b90d43772 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -158,8 +158,6 @@ void kvm_arch_destroy_vm(struct kvm *kvm) { int i; - kvm_free_stage2_pgd(kvm); - for (i = 0; i < KVM_MAX_VCPUS; ++i) { if (kvm->vcpus[i]) { kvm_arch_vcpu_free(kvm->vcpus[i]); diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c index a3faafe70081..e9a5c0e0c115 100644 --- a/arch/arm/kvm/mmu.c +++ b/arch/arm/kvm/mmu.c @@ -1894,6 +1894,7 @@ void kvm_arch_memslots_updated(struct kvm *kvm, struct kvm_memslots *slots) void kvm_arch_flush_shadow_all(struct kvm *kvm) { + kvm_free_stage2_pgd(kvm); } void kvm_arch_flush_shadow_memslot(struct kvm *kvm, From 6b7b554d34fdbc5dc9fae9d4ca9dd37f6346be3d Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Wed, 3 Aug 2016 15:12:44 +0100 Subject: [PATCH 147/271] ARM: dts: Remove use of skeleton.dtsi from bcm283x.dtsi This file is included from DTS files under arch/arm64 too (via broadcom/bcm2837-rpi-3-b.dts and broadcom/bcm2837.dtsi). There is a desire not to have skeleton.dtsi for ARM64. See commit 3ebee5a2e141 ("arm64: dts: kill skeleton.dtsi") for rationale for its removal. As well as the addition of #*-cells also requires adding the device_type to the rpi memory node explicitly. Note that this change results in the removal of an empty /aliases node from bcm2835-rpi-a.dtb and bcm2835-rpi-a-plus.dtb. I have no hardware to check if this is a problem or not. It also results in some reordering of the nodes in the DTBs (the /aliases and /memory nodes come later). This isn't supposed to matter but, again, I've no hardware to check if it is true in this particular case. Signed-off-by: Ian Campbell Acked-by: Mark Rutland Tested-by: Stefan Wahren Cc: Catalin Marinas Cc: Will Deacon Cc: Mark Rutland Cc: Rob Herring Cc: Frank Rowand Cc: Eric Anholt Cc: Stephen Warren Cc: Lee Jones Cc: Gerd Hoffmann Cc: devicetree@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-rpi-kernel@lists.infradead.org Cc: arm@kernel.org Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/bcm2835-rpi.dtsi | 1 + arch/arm/boot/dts/bcm283x.dtsi | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/bcm2835-rpi.dtsi b/arch/arm/boot/dts/bcm2835-rpi.dtsi index caf2707680c1..e9b47b2bbc33 100644 --- a/arch/arm/boot/dts/bcm2835-rpi.dtsi +++ b/arch/arm/boot/dts/bcm2835-rpi.dtsi @@ -2,6 +2,7 @@ / { memory { + device_type = "memory"; reg = <0 0x10000000>; }; diff --git a/arch/arm/boot/dts/bcm283x.dtsi b/arch/arm/boot/dts/bcm283x.dtsi index b98252232d20..445624a1a1de 100644 --- a/arch/arm/boot/dts/bcm283x.dtsi +++ b/arch/arm/boot/dts/bcm283x.dtsi @@ -2,7 +2,6 @@ #include #include #include -#include "skeleton.dtsi" /* This include file covers the common peripherals and configuration between * bcm2835 and bcm2836 implementations, leaving the CPU configuration to @@ -13,6 +12,8 @@ compatible = "brcm,bcm2835"; model = "BCM2835"; interrupt-parent = <&intc>; + #address-cells = <1>; + #size-cells = <1>; chosen { bootargs = "earlyprintk console=ttyAMA0"; From 76aa7591688001839ec9ca838041f2d55d49ab92 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Wed, 3 Aug 2016 15:12:45 +0100 Subject: [PATCH 148/271] ARM64: dts: bcm: Use a symlink to R-Pi dtsi files from arch=arm The ../../../arm... style cross-references added by commit 9d56c22a7861 ("ARM: bcm2835: Add devicetree for the Raspberry Pi 3.") do not work in the context of the split device-tree repository[0] (where the directory structure differs). As with commit 8ee57b8182c4 ("ARM64: dts: vexpress: Use a symlink to vexpress-v2m-rs1.dtsi from arch=arm") use symlinks instead. [0] https://git.kernel.org/cgit/linux/kernel/git/devicetree/devicetree-rebasing.git/ Signed-off-by: Ian Campbell Acked-by: Mark Rutland Cc: Catalin Marinas Cc: Will Deacon Cc: Mark Rutland Cc: Rob Herring Cc: Frank Rowand Cc: Eric Anholt Cc: Stephen Warren Cc: Lee Jones Cc: Gerd Hoffmann Cc: devicetree@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-rpi-kernel@lists.infradead.org Cc: arm@kernel.org Signed-off-by: Arnd Bergmann --- arch/arm64/boot/dts/broadcom/bcm2835-rpi.dtsi | 1 + arch/arm64/boot/dts/broadcom/bcm2837-rpi-3-b.dts | 4 ++-- arch/arm64/boot/dts/broadcom/bcm2837.dtsi | 2 +- arch/arm64/boot/dts/broadcom/bcm283x-rpi-smsc9514.dtsi | 1 + arch/arm64/boot/dts/broadcom/bcm283x.dtsi | 1 + 5 files changed, 6 insertions(+), 3 deletions(-) create mode 120000 arch/arm64/boot/dts/broadcom/bcm2835-rpi.dtsi create mode 120000 arch/arm64/boot/dts/broadcom/bcm283x-rpi-smsc9514.dtsi create mode 120000 arch/arm64/boot/dts/broadcom/bcm283x.dtsi diff --git a/arch/arm64/boot/dts/broadcom/bcm2835-rpi.dtsi b/arch/arm64/boot/dts/broadcom/bcm2835-rpi.dtsi new file mode 120000 index 000000000000..3937b77cb310 --- /dev/null +++ b/arch/arm64/boot/dts/broadcom/bcm2835-rpi.dtsi @@ -0,0 +1 @@ +../../../../arm/boot/dts/bcm2835-rpi.dtsi \ No newline at end of file diff --git a/arch/arm64/boot/dts/broadcom/bcm2837-rpi-3-b.dts b/arch/arm64/boot/dts/broadcom/bcm2837-rpi-3-b.dts index 6f47dd2bb1db..7841b724e340 100644 --- a/arch/arm64/boot/dts/broadcom/bcm2837-rpi-3-b.dts +++ b/arch/arm64/boot/dts/broadcom/bcm2837-rpi-3-b.dts @@ -1,7 +1,7 @@ /dts-v1/; #include "bcm2837.dtsi" -#include "../../../../arm/boot/dts/bcm2835-rpi.dtsi" -#include "../../../../arm/boot/dts/bcm283x-rpi-smsc9514.dtsi" +#include "bcm2835-rpi.dtsi" +#include "bcm283x-rpi-smsc9514.dtsi" / { compatible = "raspberrypi,3-model-b", "brcm,bcm2837"; diff --git a/arch/arm64/boot/dts/broadcom/bcm2837.dtsi b/arch/arm64/boot/dts/broadcom/bcm2837.dtsi index f2a31d06845d..8216bbb29fe0 100644 --- a/arch/arm64/boot/dts/broadcom/bcm2837.dtsi +++ b/arch/arm64/boot/dts/broadcom/bcm2837.dtsi @@ -1,4 +1,4 @@ -#include "../../../../arm/boot/dts/bcm283x.dtsi" +#include "bcm283x.dtsi" / { compatible = "brcm,bcm2836"; diff --git a/arch/arm64/boot/dts/broadcom/bcm283x-rpi-smsc9514.dtsi b/arch/arm64/boot/dts/broadcom/bcm283x-rpi-smsc9514.dtsi new file mode 120000 index 000000000000..dca7c057d5a5 --- /dev/null +++ b/arch/arm64/boot/dts/broadcom/bcm283x-rpi-smsc9514.dtsi @@ -0,0 +1 @@ +../../../../arm/boot/dts/bcm283x-rpi-smsc9514.dtsi \ No newline at end of file diff --git a/arch/arm64/boot/dts/broadcom/bcm283x.dtsi b/arch/arm64/boot/dts/broadcom/bcm283x.dtsi new file mode 120000 index 000000000000..5f54e4cab99b --- /dev/null +++ b/arch/arm64/boot/dts/broadcom/bcm283x.dtsi @@ -0,0 +1 @@ +../../../../arm/boot/dts/bcm283x.dtsi \ No newline at end of file From c94455558337eece474eebb6a16b905f98930418 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 8 Sep 2016 10:43:58 -0700 Subject: [PATCH 149/271] raid5: fix a small race condition commit 5f9d1fde7d54a5(raid5: fix memory leak of bio integrity data) moves bio_reset to bio_endio. But it introduces a small race condition. It does bio_reset after raid5_release_stripe, which could make the stripe reusable and hence reuse the bio just before bio_reset. Moving bio_reset before raid5_release_stripe is called should fix the race. Reported-and-tested-by: Stefan Priebe - Profihost AG Signed-off-by: Shaohua Li --- drivers/md/raid5.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index b95c54c0e596..ee7fc3701700 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -2423,10 +2423,10 @@ static void raid5_end_read_request(struct bio * bi) } } rdev_dec_pending(rdev, conf->mddev); + bio_reset(bi); clear_bit(R5_LOCKED, &sh->dev[i].flags); set_bit(STRIPE_HANDLE, &sh->state); raid5_release_stripe(sh); - bio_reset(bi); } static void raid5_end_write_request(struct bio *bi) @@ -2498,6 +2498,7 @@ static void raid5_end_write_request(struct bio *bi) if (sh->batch_head && bi->bi_error && !replacement) set_bit(STRIPE_BATCH_ERR, &sh->batch_head->state); + bio_reset(bi); if (!test_and_clear_bit(R5_DOUBLE_LOCKED, &sh->dev[i].flags)) clear_bit(R5_LOCKED, &sh->dev[i].flags); set_bit(STRIPE_HANDLE, &sh->state); @@ -2505,7 +2506,6 @@ static void raid5_end_write_request(struct bio *bi) if (sh->batch_head && sh != sh->batch_head) raid5_release_stripe(sh->batch_head); - bio_reset(bi); } static void raid5_build_block(struct stripe_head *sh, int i, int previous) From ea90383837334bcebe842e719ad4d8c966f4ef51 Mon Sep 17 00:00:00 2001 From: Baoyou Xie Date: Thu, 8 Sep 2016 19:03:20 +0800 Subject: [PATCH 150/271] drm/vc4: mark vc4_bo_cache_purge() static We get 1 warning when building kernel with W=1: drivers/gpu/drm/vc4/vc4_bo.c:147:6: warning: no previous prototype for 'vc4_bo_cache_purge' [-Wmissing-prototypes] In fact, this function is only used in the file in which it is declared and don't need a declaration, but can be made static. So this patch marks it 'static'. Signed-off-by: Baoyou Xie Signed-off-by: Eric Anholt --- drivers/gpu/drm/vc4/vc4_bo.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c index 59adcf8532dd..3f6704cf6608 100644 --- a/drivers/gpu/drm/vc4/vc4_bo.c +++ b/drivers/gpu/drm/vc4/vc4_bo.c @@ -144,7 +144,7 @@ static struct list_head *vc4_get_cache_list_for_size(struct drm_device *dev, return &vc4->bo_cache.size_list[page_index]; } -void vc4_bo_cache_purge(struct drm_device *dev) +static void vc4_bo_cache_purge(struct drm_device *dev) { struct vc4_dev *vc4 = to_vc4_dev(dev); From 7d762e49c2117d3829eb3355f2617aea080ed3a7 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Fri, 9 Sep 2016 18:08:23 +0200 Subject: [PATCH 151/271] perf/x86/amd/uncore: Prevent use after free The resent conversion of the cpu hotplug support in the uncore driver introduced a regression due to the way the callbacks are invoked at initialization time. The old code called the prepare/starting/online function on each online cpu as a block. The new code registers the hotplug callbacks in the core for each state. The core invokes the callbacks at each registration on all online cpus. The code implicitely relied on the prepare/starting/online callbacks being called as combo on a particular cpu, which was not obvious and completely undocumented. The resulting subtle wreckage happens due to the way how the uncore code manages shared data structures for cpus which share an uncore resource in hardware. The sharing is determined in the cpu starting callback, but the prepare callback allocates per cpu data for the upcoming cpu because potential sharing is unknown at this point. If the starting callback finds a online cpu which shares the hardware resource it takes a refcount on the percpu data of that cpu and puts the own data structure into a 'free_at_online' pointer of that shared data structure. The online callback frees that. With the old model this worked because in a starting callback only one non unused structure (the one of the starting cpu) was available. The new code allocates the data structures for all cpus when the prepare callback is registered. Now the starting function iterates through all online cpus and looks for a data structure (skipping its own) which has a matching hardware id. The id member of the data structure is initialized to 0, but the hardware id can be 0 as well. The resulting wreckage is: CPU0 finds a matching id on CPU1, takes a refcount on CPU1 data and puts its own data structure into CPU1s data structure to be freed. CPU1 skips CPU0 because the data structure is its allegedly unsued own. It finds a matching id on CPU2, takes a refcount on CPU1 data and puts its own data structure into CPU2s data structure to be freed. .... Now the online callbacks are invoked. CPU0 has a pointer to CPU1s data and frees the original CPU0 data. So far so good. CPU1 has a pointer to CPU2s data and frees the original CPU1 data, which is still referenced by CPU0 ---> Booom So there are two issues to be solved here: 1) The id field must be initialized at allocation time to a value which cannot be a valid hardware id, i.e. -1 This prevents the above scenario, but now CPU1 and CPU2 both stick their own data structure into the free_at_online pointer of CPU0. So we leak CPU1s data structure. 2) Fix the memory leak described in #1 Instead of having a single pointer, use a hlist to enqueue the superflous data structures which are then freed by the first cpu invoking the online callback. Ideally we should know the sharing _before_ invoking the prepare callback, but that's way beyond the scope of this bug fix. [ tglx: Rewrote changelog ] Fixes: 96b2bd3866a0 ("perf/x86/amd/uncore: Convert to hotplug state machine") Reported-and-tested-by: Eric Sandeen Signed-off-by: Sebastian Andrzej Siewior Cc: Borislav Petkov Link: http://lkml.kernel.org/r/20160909160822.lowgmkdwms2dheyv@linutronix.de Signed-off-by: Thomas Gleixner --- arch/x86/events/amd/uncore.c | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) diff --git a/arch/x86/events/amd/uncore.c b/arch/x86/events/amd/uncore.c index e6131d4454e6..65577f081d07 100644 --- a/arch/x86/events/amd/uncore.c +++ b/arch/x86/events/amd/uncore.c @@ -29,6 +29,8 @@ #define COUNTER_SHIFT 16 +static HLIST_HEAD(uncore_unused_list); + struct amd_uncore { int id; int refcnt; @@ -39,7 +41,7 @@ struct amd_uncore { cpumask_t *active_mask; struct pmu *pmu; struct perf_event *events[MAX_COUNTERS]; - struct amd_uncore *free_when_cpu_online; + struct hlist_node node; }; static struct amd_uncore * __percpu *amd_uncore_nb; @@ -306,6 +308,7 @@ static int amd_uncore_cpu_up_prepare(unsigned int cpu) uncore_nb->msr_base = MSR_F15H_NB_PERF_CTL; uncore_nb->active_mask = &amd_nb_active_mask; uncore_nb->pmu = &amd_nb_pmu; + uncore_nb->id = -1; *per_cpu_ptr(amd_uncore_nb, cpu) = uncore_nb; } @@ -319,6 +322,7 @@ static int amd_uncore_cpu_up_prepare(unsigned int cpu) uncore_l2->msr_base = MSR_F16H_L2I_PERF_CTL; uncore_l2->active_mask = &amd_l2_active_mask; uncore_l2->pmu = &amd_l2_pmu; + uncore_l2->id = -1; *per_cpu_ptr(amd_uncore_l2, cpu) = uncore_l2; } @@ -348,7 +352,7 @@ amd_uncore_find_online_sibling(struct amd_uncore *this, continue; if (this->id == that->id) { - that->free_when_cpu_online = this; + hlist_add_head(&this->node, &uncore_unused_list); this = that; break; } @@ -388,13 +392,23 @@ static int amd_uncore_cpu_starting(unsigned int cpu) return 0; } +static void uncore_clean_online(void) +{ + struct amd_uncore *uncore; + struct hlist_node *n; + + hlist_for_each_entry_safe(uncore, n, &uncore_unused_list, node) { + hlist_del(&uncore->node); + kfree(uncore); + } +} + static void uncore_online(unsigned int cpu, struct amd_uncore * __percpu *uncores) { struct amd_uncore *uncore = *per_cpu_ptr(uncores, cpu); - kfree(uncore->free_when_cpu_online); - uncore->free_when_cpu_online = NULL; + uncore_clean_online(); if (cpu == uncore->cpu) cpumask_set_cpu(cpu, uncore->active_mask); From 2561d309dfd1555e781484af757ed0115035ddb3 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 17 Aug 2016 16:02:32 -0400 Subject: [PATCH 152/271] alpha: fix copy_from_user() it should clear the destination even when access_ok() fails. Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/alpha/include/asm/uaccess.h | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) diff --git a/arch/alpha/include/asm/uaccess.h b/arch/alpha/include/asm/uaccess.h index c419b43c461d..466e42e96bfa 100644 --- a/arch/alpha/include/asm/uaccess.h +++ b/arch/alpha/include/asm/uaccess.h @@ -371,14 +371,6 @@ __copy_tofrom_user_nocheck(void *to, const void *from, long len) return __cu_len; } -extern inline long -__copy_tofrom_user(void *to, const void *from, long len, const void __user *validate) -{ - if (__access_ok((unsigned long)validate, len, get_fs())) - len = __copy_tofrom_user_nocheck(to, from, len); - return len; -} - #define __copy_to_user(to, from, n) \ ({ \ __chk_user_ptr(to); \ @@ -393,17 +385,22 @@ __copy_tofrom_user(void *to, const void *from, long len, const void __user *vali #define __copy_to_user_inatomic __copy_to_user #define __copy_from_user_inatomic __copy_from_user - extern inline long copy_to_user(void __user *to, const void *from, long n) { - return __copy_tofrom_user((__force void *)to, from, n, to); + if (likely(__access_ok((unsigned long)to, n, get_fs()))) + n = __copy_tofrom_user_nocheck((__force void *)to, from, n); + return n; } extern inline long copy_from_user(void *to, const void __user *from, long n) { - return __copy_tofrom_user(to, (__force void *)from, n, from); + if (likely(__access_ok((unsigned long)from, n, get_fs()))) + n = __copy_tofrom_user_nocheck(to, (__force void *)from, n); + else + memset(to, 0, n); + return n; } extern void __do_clear_user(void); From 2545e5da080b4839dd859e3b09343a884f6ab0e3 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 17 Aug 2016 16:36:37 -0400 Subject: [PATCH 153/271] asm-generic: make copy_from_user() zero the destination properly ... in all cases, including the failing access_ok() Note that some architectures using asm-generic/uaccess.h have __copy_from_user() not zeroing the tail on failure halfway through. This variant works either way. Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- include/asm-generic/uaccess.h | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h index 1bfa602958f2..04e21a41796a 100644 --- a/include/asm-generic/uaccess.h +++ b/include/asm-generic/uaccess.h @@ -257,11 +257,13 @@ extern int __get_user_bad(void) __attribute__((noreturn)); static inline long copy_from_user(void *to, const void __user * from, unsigned long n) { + unsigned long res = n; might_fault(); - if (access_ok(VERIFY_READ, from, n)) - return __copy_from_user(to, from, n); - else - return n; + if (likely(access_ok(VERIFY_READ, from, n))) + res = __copy_from_user(to, from, n); + if (unlikely(res)) + memset(to + (n - res), 0, res); + return res; } static inline long copy_to_user(void __user *to, From 40d259045500c6431633f215516c1f94bac8c62f Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 8 Sep 2016 08:16:01 +0200 Subject: [PATCH 154/271] mlxsw: spectrum_router: Fix error path in mlxsw_sp_router_init When neigh_init fails, we have to do proper cleanup including router_fini call. Fixes: 6cf3c971dc84cb ("mlxsw: spectrum_router: Add private neigh table") Signed-off-by: Jiri Pirko Acked-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 352259b3c38a..3f5c51da6d3e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -1517,7 +1517,14 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) return err; mlxsw_sp_lpm_init(mlxsw_sp); mlxsw_sp_vrs_init(mlxsw_sp); - return mlxsw_sp_neigh_init(mlxsw_sp); + err = mlxsw_sp_neigh_init(mlxsw_sp); + if (err) + goto err_neigh_init; + return 0; + +err_neigh_init: + __mlxsw_sp_router_fini(mlxsw_sp); + return err; } void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) From 3247ff2b318f3eeaaf32163ce39ddaac23e9a6f9 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Thu, 8 Sep 2016 08:16:02 +0200 Subject: [PATCH 155/271] mlxsw: spectrum: Set port type before setting its address During port init, we currently set the port's type to Ethernet after setting its MAC address. However, the hardware documentation states this should be the other way around. Align the driver with the hardware documentation and set the port's MAC address after setting its type. Fixes: 56ade8fe3fe1 ("mlxsw: spectrum: Add initial support for Spectrum ASIC") Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index e16b34718c8c..d48873bcbddf 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2106,6 +2106,13 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, dev->netdev_ops = &mlxsw_sp_port_netdev_ops; dev->ethtool_ops = &mlxsw_sp_port_ethtool_ops; + err = mlxsw_sp_port_swid_set(mlxsw_sp_port, 0); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set SWID\n", + mlxsw_sp_port->local_port); + goto err_port_swid_set; + } + err = mlxsw_sp_port_dev_addr_init(mlxsw_sp_port); if (err) { dev_err(mlxsw_sp->bus_info->dev, "Port %d: Unable to init port mac address\n", @@ -2131,13 +2138,6 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, goto err_port_system_port_mapping_set; } - err = mlxsw_sp_port_swid_set(mlxsw_sp_port, 0); - if (err) { - dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set SWID\n", - mlxsw_sp_port->local_port); - goto err_port_swid_set; - } - err = mlxsw_sp_port_speed_by_width_set(mlxsw_sp_port, width); if (err) { dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to enable speeds\n", @@ -2219,10 +2219,10 @@ err_port_buffers_init: err_port_admin_status_set: err_port_mtu_set: err_port_speed_by_width_set: - mlxsw_sp_port_swid_set(mlxsw_sp_port, MLXSW_PORT_SWID_DISABLED_PORT); -err_port_swid_set: err_port_system_port_mapping_set: err_dev_addr_init: + mlxsw_sp_port_swid_set(mlxsw_sp_port, MLXSW_PORT_SWID_DISABLED_PORT); +err_port_swid_set: free_percpu(mlxsw_sp_port->pcpu_stats); err_alloc_stats: kfree(mlxsw_sp_port->untagged_vlans); From 7303a1475008bee5c3e82a06a282568415690d72 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 8 Sep 2016 17:54:11 +0800 Subject: [PATCH 156/271] sctp: identify chunks that need to be fragmented at IP level Previously, without GSO, it was easy to identify it: if the chunk didn't fit and there was no data chunk in the packet yet, we could fragment at IP level. So if there was an auth chunk and we were bundling a big data chunk, it would fragment regardless of the size of the auth chunk. This also works for the context of PMTU reductions. But with GSO, we cannot distinguish such PMTU events anymore, as the packet is allowed to exceed PMTU. So we need another check: to ensure that the chunk that we are adding, actually fits the current PMTU. If it doesn't, trigger a flush and let it be fragmented at IP level in the next round. Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- net/sctp/output.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/net/sctp/output.c b/net/sctp/output.c index 1f1682b9a6a8..31b7bc35895d 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -878,7 +878,7 @@ static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, struct sctp_chunk *chunk, u16 chunk_len) { - size_t psize, pmtu; + size_t psize, pmtu, maxsize; sctp_xmit_t retval = SCTP_XMIT_OK; psize = packet->size; @@ -906,6 +906,17 @@ static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, goto out; } + /* Similarly, if this chunk was built before a PMTU + * reduction, we have to fragment it at IP level now. So + * if the packet already contains something, we need to + * flush. + */ + maxsize = pmtu - packet->overhead; + if (packet->auth) + maxsize -= WORD_ROUND(packet->auth->skb->len); + if (chunk_len > maxsize) + retval = SCTP_XMIT_PMTU_FULL; + /* It is also okay to fragment if the chunk we are * adding is a control chunk, but only if current packet * is not a GSO one otherwise it causes fragmentation of From 47b02f7294a483387772a46931da942b2ca9d845 Mon Sep 17 00:00:00 2001 From: Lars Persson Date: Thu, 8 Sep 2016 13:24:21 +0200 Subject: [PATCH 157/271] dwc_eth_qos: do not register semi-initialized device We move register_netdev() to the end of dwceqos_probe() to close any races where the netdev callbacks are called before the initialization has finished. Reported-by: Pavel Andrianov Signed-off-by: Lars Persson Signed-off-by: David S. Miller --- drivers/net/ethernet/synopsys/dwc_eth_qos.c | 38 ++++++++++----------- 1 file changed, 18 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index 5a3941bf250f..4490ebaed127 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -1246,7 +1246,7 @@ static int dwceqos_mii_init(struct net_local *lp) lp->mii_bus->read = &dwceqos_mdio_read; lp->mii_bus->write = &dwceqos_mdio_write; lp->mii_bus->priv = lp; - lp->mii_bus->parent = &lp->ndev->dev; + lp->mii_bus->parent = &lp->pdev->dev; of_address_to_resource(lp->pdev->dev.of_node, 0, &res); snprintf(lp->mii_bus->id, MII_BUS_ID_SIZE, "%.8llx", @@ -2853,25 +2853,17 @@ static int dwceqos_probe(struct platform_device *pdev) ndev->features = ndev->hw_features; - netif_napi_add(ndev, &lp->napi, dwceqos_rx_poll, NAPI_POLL_WEIGHT); - - ret = register_netdev(ndev); - if (ret) { - dev_err(&pdev->dev, "Cannot register net device, aborting.\n"); - goto err_out_clk_dis_aper; - } - lp->phy_ref_clk = devm_clk_get(&pdev->dev, "phy_ref_clk"); if (IS_ERR(lp->phy_ref_clk)) { dev_err(&pdev->dev, "phy_ref_clk clock not found.\n"); ret = PTR_ERR(lp->phy_ref_clk); - goto err_out_unregister_netdev; + goto err_out_clk_dis_aper; } ret = clk_prepare_enable(lp->phy_ref_clk); if (ret) { dev_err(&pdev->dev, "Unable to enable device clock.\n"); - goto err_out_unregister_netdev; + goto err_out_clk_dis_aper; } lp->phy_node = of_parse_phandle(lp->pdev->dev.of_node, @@ -2880,7 +2872,7 @@ static int dwceqos_probe(struct platform_device *pdev) ret = of_phy_register_fixed_link(lp->pdev->dev.of_node); if (ret < 0) { dev_err(&pdev->dev, "invalid fixed-link"); - goto err_out_unregister_clk_notifier; + goto err_out_clk_dis_phy; } lp->phy_node = of_node_get(lp->pdev->dev.of_node); @@ -2889,7 +2881,7 @@ static int dwceqos_probe(struct platform_device *pdev) ret = of_get_phy_mode(lp->pdev->dev.of_node); if (ret < 0) { dev_err(&lp->pdev->dev, "error in getting phy i/f\n"); - goto err_out_unregister_clk_notifier; + goto err_out_clk_dis_phy; } lp->phy_interface = ret; @@ -2897,14 +2889,14 @@ static int dwceqos_probe(struct platform_device *pdev) ret = dwceqos_mii_init(lp); if (ret) { dev_err(&lp->pdev->dev, "error in dwceqos_mii_init\n"); - goto err_out_unregister_clk_notifier; + goto err_out_clk_dis_phy; } ret = dwceqos_mii_probe(ndev); if (ret != 0) { netdev_err(ndev, "mii_probe fail.\n"); ret = -ENXIO; - goto err_out_unregister_clk_notifier; + goto err_out_clk_dis_phy; } dwceqos_set_umac_addr(lp, lp->ndev->dev_addr, 0); @@ -2922,7 +2914,7 @@ static int dwceqos_probe(struct platform_device *pdev) if (ret) { dev_err(&lp->pdev->dev, "Unable to retrieve DT, error %d\n", ret); - goto err_out_unregister_clk_notifier; + goto err_out_clk_dis_phy; } dev_info(&lp->pdev->dev, "pdev->id %d, baseaddr 0x%08lx, irq %d\n", pdev->id, ndev->base_addr, ndev->irq); @@ -2932,18 +2924,24 @@ static int dwceqos_probe(struct platform_device *pdev) if (ret) { dev_err(&lp->pdev->dev, "Unable to request IRQ %d, error %d\n", ndev->irq, ret); - goto err_out_unregister_clk_notifier; + goto err_out_clk_dis_phy; } if (netif_msg_probe(lp)) netdev_dbg(ndev, "net_local@%p\n", lp); + netif_napi_add(ndev, &lp->napi, dwceqos_rx_poll, NAPI_POLL_WEIGHT); + + ret = register_netdev(ndev); + if (ret) { + dev_err(&pdev->dev, "Cannot register net device, aborting.\n"); + goto err_out_clk_dis_phy; + } + return 0; -err_out_unregister_clk_notifier: +err_out_clk_dis_phy: clk_disable_unprepare(lp->phy_ref_clk); -err_out_unregister_netdev: - unregister_netdev(ndev); err_out_clk_dis_aper: clk_disable_unprepare(lp->apb_pclk); err_out_free_netdev: From c2f57fb97da5730509a50e316f353d3da17f6c25 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 8 Sep 2016 16:25:15 +0200 Subject: [PATCH 158/271] drivers: net: phy: mdio-xgene: Add hardware dependency The mdio-xgene driver is only useful on X-Gene SoC. Signed-off-by: Jean Delvare Cc: Iyappan Subramanian Cc: David S. Miller Acked-by: Iyappan Subramanian Signed-off-by: David S. Miller --- drivers/net/phy/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig index 47a64342cc16..b4863e4e522b 100644 --- a/drivers/net/phy/Kconfig +++ b/drivers/net/phy/Kconfig @@ -303,6 +303,7 @@ config MDIO_HISI_FEMAC config MDIO_XGENE tristate "APM X-Gene SoC MDIO bus controller" + depends on ARCH_XGENE || COMPILE_TEST help This module provides a driver for the MDIO busses found in the APM X-Gene SoC's. From 4214ebf4654798309364d0c678b799e402f38288 Mon Sep 17 00:00:00 2001 From: Sachin Prabhu Date: Fri, 29 Jul 2016 22:38:19 +0100 Subject: [PATCH 159/271] Fix memory leaks in cifs_do_mount() Fix memory leaks introduced by the patch fs/cifs: make share unaccessible at root level mountable Also move allocation of cifs_sb->prepath to cifs_setup_cifs_sb(). Signed-off-by: Sachin Prabhu Tested-by: Aurelien Aptel Signed-off-by: Steve French --- fs/cifs/cifsfs.c | 20 ++++++++------------ fs/cifs/cifsproto.h | 2 +- fs/cifs/connect.c | 10 +++++++++- 3 files changed, 18 insertions(+), 14 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 6bbec5e784cd..cc9cdab52dab 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -686,26 +686,22 @@ cifs_do_mount(struct file_system_type *fs_type, cifs_sb->mountdata = kstrndup(data, PAGE_SIZE, GFP_KERNEL); if (cifs_sb->mountdata == NULL) { root = ERR_PTR(-ENOMEM); - goto out_cifs_sb; + goto out_free; } - if (volume_info->prepath) { - cifs_sb->prepath = kstrdup(volume_info->prepath, GFP_KERNEL); - if (cifs_sb->prepath == NULL) { - root = ERR_PTR(-ENOMEM); - goto out_cifs_sb; - } + rc = cifs_setup_cifs_sb(volume_info, cifs_sb); + if (rc) { + root = ERR_PTR(rc); + goto out_free; } - cifs_setup_cifs_sb(volume_info, cifs_sb); - rc = cifs_mount(cifs_sb, volume_info); if (rc) { if (!(flags & MS_SILENT)) cifs_dbg(VFS, "cifs_mount failed w/return code = %d\n", rc); root = ERR_PTR(rc); - goto out_mountdata; + goto out_free; } mnt_data.vol = volume_info; @@ -752,9 +748,9 @@ out: cifs_cleanup_volume_info(volume_info); return root; -out_mountdata: +out_free: + kfree(cifs_sb->prepath); kfree(cifs_sb->mountdata); -out_cifs_sb: kfree(cifs_sb); out_nls: unload_nls(volume_info->local_nls); diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 1243bd326591..95dab43646f0 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -184,7 +184,7 @@ extern int cifs_read_from_socket(struct TCP_Server_Info *server, char *buf, unsigned int to_read); extern int cifs_read_page_from_socket(struct TCP_Server_Info *server, struct page *page, unsigned int to_read); -extern void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, +extern int cifs_setup_cifs_sb(struct smb_vol *pvolume_info, struct cifs_sb_info *cifs_sb); extern int cifs_match_super(struct super_block *, void *); extern void cifs_cleanup_volume_info(struct smb_vol *pvolume_info); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 7ae03283bd61..4546926680d4 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -3222,7 +3222,7 @@ void reset_cifs_unix_caps(unsigned int xid, struct cifs_tcon *tcon, } } -void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, +int cifs_setup_cifs_sb(struct smb_vol *pvolume_info, struct cifs_sb_info *cifs_sb) { INIT_DELAYED_WORK(&cifs_sb->prune_tlinks, cifs_prune_tlinks); @@ -3316,6 +3316,14 @@ void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, if ((pvolume_info->cifs_acl) && (pvolume_info->dynperm)) cifs_dbg(VFS, "mount option dynperm ignored if cifsacl mount option supported\n"); + + if (pvolume_info->prepath) { + cifs_sb->prepath = kstrdup(pvolume_info->prepath, GFP_KERNEL); + if (cifs_sb->prepath == NULL) + return -ENOMEM; + } + + return 0; } static void From c1d8b24d18192764fe82067ec6aa8d4c3bf094e0 Mon Sep 17 00:00:00 2001 From: Sachin Prabhu Date: Fri, 29 Jul 2016 22:38:20 +0100 Subject: [PATCH 160/271] Compare prepaths when comparing superblocks The patch fs/cifs: make share unaccessible at root level mountable makes use of prepaths when any component of the underlying path is inaccessible. When mounting 2 separate shares having different prepaths but are other wise similar in other respects, we end up sharing superblocks when we shouldn't be doing so. Signed-off-by: Sachin Prabhu Tested-by: Aurelien Aptel Signed-off-by: Steve French --- fs/cifs/connect.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 4546926680d4..2e4f4bad8b1e 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2781,6 +2781,24 @@ compare_mount_options(struct super_block *sb, struct cifs_mnt_data *mnt_data) return 1; } +static int +match_prepath(struct super_block *sb, struct cifs_mnt_data *mnt_data) +{ + struct cifs_sb_info *old = CIFS_SB(sb); + struct cifs_sb_info *new = mnt_data->cifs_sb; + + if (old->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) { + if (!(new->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH)) + return 0; + /* The prepath should be null terminated strings */ + if (strcmp(new->prepath, old->prepath)) + return 0; + + return 1; + } + return 0; +} + int cifs_match_super(struct super_block *sb, void *data) { @@ -2808,7 +2826,8 @@ cifs_match_super(struct super_block *sb, void *data) if (!match_server(tcp_srv, volume_info) || !match_session(ses, volume_info) || - !match_tcon(tcon, volume_info->UNC)) { + !match_tcon(tcon, volume_info->UNC) || + !match_prepath(sb, mnt_data)) { rc = 0; goto out; } From 348c1bfa84dfc47da1f1234b7f2bf09fa798edea Mon Sep 17 00:00:00 2001 From: Sachin Prabhu Date: Fri, 29 Jul 2016 22:38:21 +0100 Subject: [PATCH 161/271] Move check for prefix path to within cifs_get_root() Signed-off-by: Sachin Prabhu Tested-by: Aurelien Aptel Signed-off-by: Steve French --- fs/cifs/cifsfs.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index cc9cdab52dab..14ae4b8e1a3c 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -609,6 +609,9 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb) char *s, *p; char sep; + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) + return dget(sb->s_root); + full_path = cifs_build_path_to_root(vol, cifs_sb, cifs_sb_master_tcon(cifs_sb)); if (full_path == NULL) @@ -731,11 +734,7 @@ cifs_do_mount(struct file_system_type *fs_type, sb->s_flags |= MS_ACTIVE; } - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) - root = dget(sb->s_root); - else - root = cifs_get_root(volume_info, sb); - + root = cifs_get_root(volume_info, sb); if (IS_ERR(root)) goto out_super; From 767ae08678c2c796bcd7f582ee457aee20a28a1e Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Tue, 6 Sep 2016 16:23:49 +0300 Subject: [PATCH 162/271] perf/core: Fix a race between mmap_close() and set_output() of AUX events In the mmap_close() path we need to stop all the AUX events that are writing data to the AUX area that we are unmapping, before we can safely free the pages. To determine if an event needs to be stopped, we're comparing its ->rb against the one that's getting unmapped. However, a SET_OUTPUT ioctl may turn up inside an AUX transaction and swizzle event::rb to some other ring buffer, but the transaction will keep writing data to the old ring buffer until the event gets scheduled out. At this point, mmap_close() will skip over such an event and will proceed to free the AUX area, while it's still being used by this event, which will set off a warning in the mmap_close() path and cause a memory corruption. To avoid this, always stop an AUX event before its ->rb is updated; this will release the (potentially) last reference on the AUX area of the buffer. If the event gets restarted, its new ring buffer will be used. If another SET_OUTPUT comes and switches it back to the old ring buffer that's getting unmapped, it's also fine: this ring buffer's aux_mmap_count will be zero and AUX transactions won't start any more. Reported-by: Vince Weaver Signed-off-by: Alexander Shishkin Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160906132353.19887-2-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- kernel/events/core.c | 31 +++++++++++++++++++++++++------ 1 file changed, 25 insertions(+), 6 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 07ac8596a728..a54f2c2cdb20 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -2496,11 +2496,11 @@ static int __perf_event_stop(void *info) return 0; } -static int perf_event_restart(struct perf_event *event) +static int perf_event_stop(struct perf_event *event, int restart) { struct stop_event_data sd = { .event = event, - .restart = 1, + .restart = restart, }; int ret = 0; @@ -4845,6 +4845,19 @@ static void ring_buffer_attach(struct perf_event *event, spin_unlock_irqrestore(&rb->event_lock, flags); } + /* + * Avoid racing with perf_mmap_close(AUX): stop the event + * before swizzling the event::rb pointer; if it's getting + * unmapped, its aux_mmap_count will be 0 and it won't + * restart. See the comment in __perf_pmu_output_stop(). + * + * Data will inevitably be lost when set_output is done in + * mid-air, but then again, whoever does it like this is + * not in for the data anyway. + */ + if (has_aux(event)) + perf_event_stop(event, 0); + rcu_assign_pointer(event->rb, rb); if (old_rb) { @@ -6120,7 +6133,7 @@ static void perf_event_addr_filters_exec(struct perf_event *event, void *data) raw_spin_unlock_irqrestore(&ifh->lock, flags); if (restart) - perf_event_restart(event); + perf_event_stop(event, 1); } void perf_event_exec(void) @@ -6164,7 +6177,13 @@ static void __perf_event_output_stop(struct perf_event *event, void *data) /* * In case of inheritance, it will be the parent that links to the - * ring-buffer, but it will be the child that's actually using it: + * ring-buffer, but it will be the child that's actually using it. + * + * We are using event::rb to determine if the event should be stopped, + * however this may race with ring_buffer_attach() (through set_output), + * which will make us skip the event that actually needs to be stopped. + * So ring_buffer_attach() has to stop an aux event before re-assigning + * its rb pointer. */ if (rcu_dereference(parent->rb) == rb) ro->err = __perf_event_stop(&sd); @@ -6678,7 +6697,7 @@ static void __perf_addr_filters_adjust(struct perf_event *event, void *data) raw_spin_unlock_irqrestore(&ifh->lock, flags); if (restart) - perf_event_restart(event); + perf_event_stop(event, 1); } /* @@ -7867,7 +7886,7 @@ static void perf_event_addr_filters_apply(struct perf_event *event) mmput(mm); restart: - perf_event_restart(event); + perf_event_stop(event, 1); } /* From b79ccadd6bb10e72cf784a298ca6dc1398eb9a24 Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Tue, 6 Sep 2016 16:23:50 +0300 Subject: [PATCH 163/271] perf/core: Fix aux_mmap_count vs aux_refcount order The order of accesses to ring buffer's aux_mmap_count and aux_refcount has to be preserved across the users, namely perf_mmap_close() and perf_aux_output_begin(), otherwise the inversion can result in the latter holding the last reference to the aux buffer and subsequently free'ing it in atomic context, triggering a warning. > ------------[ cut here ]------------ > WARNING: CPU: 0 PID: 257 at kernel/events/ring_buffer.c:541 __rb_free_aux+0x11a/0x130 > CPU: 0 PID: 257 Comm: stopbug Not tainted 4.8.0-rc1+ #2596 > Call Trace: > [] __warn+0xcb/0xf0 > [] warn_slowpath_null+0x1d/0x20 > [] __rb_free_aux+0x11a/0x130 > [] rb_free_aux+0x18/0x20 > [] perf_aux_output_begin+0x163/0x1e0 > [] bts_event_start+0x3a/0xd0 > [] bts_event_add+0x5d/0x80 > [] event_sched_in.isra.104+0xf6/0x2f0 > [] group_sched_in+0x6e/0x190 > [] ctx_sched_in+0x2fe/0x5f0 > [] perf_event_sched_in+0x60/0x80 > [] ctx_resched+0x5b/0x90 > [] __perf_event_enable+0x1e1/0x240 > [] event_function+0xa9/0x180 > [] ? perf_cgroup_attach+0x70/0x70 > [] remote_function+0x3f/0x50 > [] flush_smp_call_function_queue+0x83/0x150 > [] generic_smp_call_function_single_interrupt+0x13/0x60 > [] smp_call_function_single_interrupt+0x27/0x40 > [] call_function_single_interrupt+0x89/0x90 > [] finish_task_switch+0xa6/0x210 > [] ? finish_task_switch+0x67/0x210 > [] __schedule+0x3dd/0xb50 > [] schedule+0x35/0x80 > [] sys_sched_yield+0x61/0x70 > [] entry_SYSCALL_64_fastpath+0x18/0xa8 > ---[ end trace 6235f556f5ea83a9 ]--- This patch puts the checks in perf_aux_output_begin() in the same order as that of perf_mmap_close(). Reported-by: Vince Weaver Signed-off-by: Alexander Shishkin Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160906132353.19887-3-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- kernel/events/ring_buffer.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index ae9b90dc9a5a..257fa460b846 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -330,15 +330,22 @@ void *perf_aux_output_begin(struct perf_output_handle *handle, if (!rb) return NULL; - if (!rb_has_aux(rb) || !atomic_inc_not_zero(&rb->aux_refcount)) + if (!rb_has_aux(rb)) goto err; /* - * If rb::aux_mmap_count is zero (and rb_has_aux() above went through), - * the aux buffer is in perf_mmap_close(), about to get freed. + * If aux_mmap_count is zero, the aux buffer is in perf_mmap_close(), + * about to get freed, so we leave immediately. + * + * Checking rb::aux_mmap_count and rb::refcount has to be done in + * the same order, see perf_mmap_close. Otherwise we end up freeing + * aux pages in this path, which is a bug, because in_atomic(). */ if (!atomic_read(&rb->aux_mmap_count)) - goto err_put; + goto err; + + if (!atomic_inc_not_zero(&rb->aux_refcount)) + goto err; /* * Nesting is not supported for AUX area, make sure nested From a9a94401c2b5805c71e39427b1af1bf1b9f67cd0 Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Tue, 6 Sep 2016 16:23:51 +0300 Subject: [PATCH 164/271] perf/x86/intel/bts: Fix confused ordering of PMU callbacks The intel_bts driver is using a CPU-local 'started' variable to order callbacks and PMIs and make sure that AUX transactions don't get messed up. However, the ordering rules in regard to this variable is a complete mess, which recently resulted in perf_fuzzer-triggered warnings and panics. The general ordering rule that is patch is enforcing is that this cpu-local variable be set only when the cpu-local AUX transaction is active; consequently, this variable is to be checked before the AUX related bits can be touched. Reported-by: Vince Weaver Signed-off-by: Alexander Shishkin Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160906132353.19887-4-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/bts.c | 104 +++++++++++++++++++++++++++--------- 1 file changed, 80 insertions(+), 24 deletions(-) diff --git a/arch/x86/events/intel/bts.c b/arch/x86/events/intel/bts.c index 0a6e393a2e62..61e1d713b114 100644 --- a/arch/x86/events/intel/bts.c +++ b/arch/x86/events/intel/bts.c @@ -31,7 +31,17 @@ struct bts_ctx { struct perf_output_handle handle; struct debug_store ds_back; - int started; + int state; +}; + +/* BTS context states: */ +enum { + /* no ongoing AUX transactions */ + BTS_STATE_STOPPED = 0, + /* AUX transaction is on, BTS tracing is disabled */ + BTS_STATE_INACTIVE, + /* AUX transaction is on, BTS tracing is running */ + BTS_STATE_ACTIVE, }; static DEFINE_PER_CPU(struct bts_ctx, bts_ctx); @@ -204,6 +214,15 @@ static void bts_update(struct bts_ctx *bts) static int bts_buffer_reset(struct bts_buffer *buf, struct perf_output_handle *handle); +/* + * Ordering PMU callbacks wrt themselves and the PMI is done by means + * of bts::state, which: + * - is set when bts::handle::event is valid, that is, between + * perf_aux_output_begin() and perf_aux_output_end(); + * - is zero otherwise; + * - is ordered against bts::handle::event with a compiler barrier. + */ + static void __bts_event_start(struct perf_event *event) { struct bts_ctx *bts = this_cpu_ptr(&bts_ctx); @@ -221,10 +240,13 @@ static void __bts_event_start(struct perf_event *event) /* * local barrier to make sure that ds configuration made it - * before we enable BTS + * before we enable BTS and bts::state goes ACTIVE */ wmb(); + /* INACTIVE/STOPPED -> ACTIVE */ + WRITE_ONCE(bts->state, BTS_STATE_ACTIVE); + intel_pmu_enable_bts(config); } @@ -251,9 +273,6 @@ static void bts_event_start(struct perf_event *event, int flags) __bts_event_start(event); - /* PMI handler: this counter is running and likely generating PMIs */ - ACCESS_ONCE(bts->started) = 1; - return; fail_end_stop: @@ -263,30 +282,34 @@ fail_stop: event->hw.state = PERF_HES_STOPPED; } -static void __bts_event_stop(struct perf_event *event) +static void __bts_event_stop(struct perf_event *event, int state) { + struct bts_ctx *bts = this_cpu_ptr(&bts_ctx); + + /* ACTIVE -> INACTIVE(PMI)/STOPPED(->stop()) */ + WRITE_ONCE(bts->state, state); + /* * No extra synchronization is mandated by the documentation to have * BTS data stores globally visible. */ intel_pmu_disable_bts(); - - if (event->hw.state & PERF_HES_STOPPED) - return; - - ACCESS_ONCE(event->hw.state) |= PERF_HES_STOPPED; } static void bts_event_stop(struct perf_event *event, int flags) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct bts_ctx *bts = this_cpu_ptr(&bts_ctx); - struct bts_buffer *buf = perf_get_aux(&bts->handle); + struct bts_buffer *buf = NULL; + int state = READ_ONCE(bts->state); - /* PMI handler: don't restart this counter */ - ACCESS_ONCE(bts->started) = 0; + if (state == BTS_STATE_ACTIVE) + __bts_event_stop(event, BTS_STATE_STOPPED); - __bts_event_stop(event); + if (state != BTS_STATE_STOPPED) + buf = perf_get_aux(&bts->handle); + + event->hw.state |= PERF_HES_STOPPED; if (flags & PERF_EF_UPDATE) { bts_update(bts); @@ -296,6 +319,7 @@ static void bts_event_stop(struct perf_event *event, int flags) bts->handle.head = local_xchg(&buf->data_size, buf->nr_pages << PAGE_SHIFT); + perf_aux_output_end(&bts->handle, local_xchg(&buf->data_size, 0), !!local_xchg(&buf->lost, 0)); } @@ -310,8 +334,20 @@ static void bts_event_stop(struct perf_event *event, int flags) void intel_bts_enable_local(void) { struct bts_ctx *bts = this_cpu_ptr(&bts_ctx); + int state = READ_ONCE(bts->state); - if (bts->handle.event && bts->started) + /* + * Here we transition from INACTIVE to ACTIVE; + * if we instead are STOPPED from the interrupt handler, + * stay that way. Can't be ACTIVE here though. + */ + if (WARN_ON_ONCE(state == BTS_STATE_ACTIVE)) + return; + + if (state == BTS_STATE_STOPPED) + return; + + if (bts->handle.event) __bts_event_start(bts->handle.event); } @@ -319,8 +355,15 @@ void intel_bts_disable_local(void) { struct bts_ctx *bts = this_cpu_ptr(&bts_ctx); + /* + * Here we transition from ACTIVE to INACTIVE; + * do nothing for STOPPED or INACTIVE. + */ + if (READ_ONCE(bts->state) != BTS_STATE_ACTIVE) + return; + if (bts->handle.event) - __bts_event_stop(bts->handle.event); + __bts_event_stop(bts->handle.event, BTS_STATE_INACTIVE); } static int @@ -407,9 +450,13 @@ int intel_bts_interrupt(void) struct perf_event *event = bts->handle.event; struct bts_buffer *buf; s64 old_head; - int err; + int err = -ENOSPC; - if (!event || !bts->started) + /* + * this is wrapped in intel_bts_enable_local/intel_bts_disable_local, + * so we can only be INACTIVE or STOPPED + */ + if (READ_ONCE(bts->state) == BTS_STATE_STOPPED) return 0; buf = perf_get_aux(&bts->handle); @@ -432,12 +479,21 @@ int intel_bts_interrupt(void) !!local_xchg(&buf->lost, 0)); buf = perf_aux_output_begin(&bts->handle, event); - if (!buf) - return 1; + if (buf) + err = bts_buffer_reset(buf, &bts->handle); - err = bts_buffer_reset(buf, &bts->handle); - if (err) - perf_aux_output_end(&bts->handle, 0, false); + if (err) { + WRITE_ONCE(bts->state, BTS_STATE_STOPPED); + + if (buf) { + /* + * BTS_STATE_STOPPED should be visible before + * cleared handle::event + */ + barrier(); + perf_aux_output_end(&bts->handle, 0, false); + } + } return 1; } From 4d4c474124649198d9b0a065c06f9362cf18e14e Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Tue, 6 Sep 2016 16:23:52 +0300 Subject: [PATCH 165/271] perf/x86/intel/bts: Fix BTS PMI detection Since BTS doesn't have a dedicated PMI status bit, the driver needs to take extra care to check for the condition that triggers it to avoid spurious NMI warnings. Regardless of the local BTS context state, the only way of knowing that the NMI is ours is to compare the write pointer against the interrupt threshold. Reported-by: Vince Weaver Signed-off-by: Alexander Shishkin Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160906132353.19887-5-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/bts.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/arch/x86/events/intel/bts.c b/arch/x86/events/intel/bts.c index 61e1d713b114..9233edf993e1 100644 --- a/arch/x86/events/intel/bts.c +++ b/arch/x86/events/intel/bts.c @@ -446,26 +446,37 @@ bts_buffer_reset(struct bts_buffer *buf, struct perf_output_handle *handle) int intel_bts_interrupt(void) { + struct debug_store *ds = this_cpu_ptr(&cpu_hw_events)->ds; struct bts_ctx *bts = this_cpu_ptr(&bts_ctx); struct perf_event *event = bts->handle.event; struct bts_buffer *buf; s64 old_head; - int err = -ENOSPC; + int err = -ENOSPC, handled = 0; + + /* + * The only surefire way of knowing if this NMI is ours is by checking + * the write ptr against the PMI threshold. + */ + if (ds->bts_index >= ds->bts_interrupt_threshold) + handled = 1; /* * this is wrapped in intel_bts_enable_local/intel_bts_disable_local, * so we can only be INACTIVE or STOPPED */ if (READ_ONCE(bts->state) == BTS_STATE_STOPPED) - return 0; + return handled; buf = perf_get_aux(&bts->handle); + if (!buf) + return handled; + /* * Skip snapshot counters: they don't use the interrupt, but * there's no other way of telling, because the pointer will * keep moving */ - if (!buf || buf->snapshot) + if (buf->snapshot) return 0; old_head = local_read(&buf->head); @@ -473,7 +484,7 @@ int intel_bts_interrupt(void) /* no new data */ if (old_head == local_read(&buf->head)) - return 0; + return handled; perf_aux_output_end(&bts->handle, local_xchg(&buf->data_size, 0), !!local_xchg(&buf->lost, 0)); From ef9ef3befa0d76008e988a9ed9fe439e803351b9 Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Tue, 6 Sep 2016 16:23:53 +0300 Subject: [PATCH 166/271] perf/x86/intel/bts: Kill a silly warning At the moment, intel_bts will WARN() out if there is more than one event writing to the same ring buffer, via SET_OUTPUT, and will only send data from one event to a buffer. There is no reason to have this warning in, so kill it. Signed-off-by: Alexander Shishkin Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160906132353.19887-6-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/bts.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/x86/events/intel/bts.c b/arch/x86/events/intel/bts.c index 9233edf993e1..bdcd6510992c 100644 --- a/arch/x86/events/intel/bts.c +++ b/arch/x86/events/intel/bts.c @@ -378,8 +378,6 @@ bts_buffer_reset(struct bts_buffer *buf, struct perf_output_handle *handle) return 0; head = handle->head & ((buf->nr_pages << PAGE_SHIFT) - 1); - if (WARN_ON_ONCE(head != local_read(&buf->head))) - return -EINVAL; phys = &buf->buf[buf->cur_buf]; space = phys->offset + phys->displacement + phys->size - head; From 8ef9b8455a2a3049efa9e46e8a6402b972a3eb41 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 7 Sep 2016 14:42:55 +0200 Subject: [PATCH 167/271] perf/x86/intel: Fix PEBSv3 record drain Alexander hit the WARN_ON_ONCE(!event) on his Skylake while running the perf fuzzer. This means the PEBSv3 record included a status bit for an inactive event, something that _should_ not happen. Move the code that filters the status bits against our known PEBS events up a spot to guarantee we only deal with events we know about. Further add "continue" statements to the WARN_ON_ONCE()s such that we'll not die nor generate silly events in case we ever do hit them again. Reported-by: Alexander Shishkin Tested-by: Alexander Shishkin Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Kan Liang Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: stable@vger.kernel.org Fixes: a3d86542de88 ("perf/x86/intel/pebs: Add PEBSv3 decoding") Signed-off-by: Ingo Molnar --- arch/x86/events/intel/ds.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 7ce9f3f669e6..9b983a474253 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -1274,18 +1274,18 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs) struct pebs_record_nhm *p = at; u64 pebs_status; - /* PEBS v3 has accurate status bits */ + pebs_status = p->status & cpuc->pebs_enabled; + pebs_status &= (1ULL << x86_pmu.max_pebs_events) - 1; + + /* PEBS v3 has more accurate status bits */ if (x86_pmu.intel_cap.pebs_format >= 3) { - for_each_set_bit(bit, (unsigned long *)&p->status, - MAX_PEBS_EVENTS) + for_each_set_bit(bit, (unsigned long *)&pebs_status, + x86_pmu.max_pebs_events) counts[bit]++; continue; } - pebs_status = p->status & cpuc->pebs_enabled; - pebs_status &= (1ULL << x86_pmu.max_pebs_events) - 1; - /* * On some CPUs the PEBS status can be zero when PEBS is * racing with clearing of GLOBAL_STATUS. @@ -1333,8 +1333,11 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs) continue; event = cpuc->events[bit]; - WARN_ON_ONCE(!event); - WARN_ON_ONCE(!event->attr.precise_ip); + if (WARN_ON_ONCE(!event)) + continue; + + if (WARN_ON_ONCE(!event->attr.precise_ip)) + continue; /* log dropped samples number */ if (error[bit]) From e1487888eccc83e9eb5a3659955f79b039dc7945 Mon Sep 17 00:00:00 2001 From: Chris Brandt Date: Wed, 7 Sep 2016 14:57:09 -0400 Subject: [PATCH 168/271] net: ethernet: renesas: sh_eth: add POST registers for rz Due to a mistake in the hardware manual, the FWSLC and POST1-4 registers were not documented and left out of the driver for RZ/A making the CAM feature non-operational. Additionally, when the offset values for POST1-4 are left blank, the driver attempts to set them using an offset of 0xFFFF which can cause a memory corruption or panic. This patch fixes the panic and properly enables CAM. Reported-by: Daniel Palmer Signed-off-by: Chris Brandt Acked-by: Sergei Shtylyov Signed-off-by: David S. Miller --- drivers/net/ethernet/renesas/sh_eth.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 799d58d86e6d..054e795df90f 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -201,9 +201,14 @@ static const u16 sh_eth_offset_fast_rz[SH_ETH_MAX_REGISTER_OFFSET] = { [ARSTR] = 0x0000, [TSU_CTRST] = 0x0004, + [TSU_FWSLC] = 0x0038, [TSU_VTAG0] = 0x0058, [TSU_ADSBSY] = 0x0060, [TSU_TEN] = 0x0064, + [TSU_POST1] = 0x0070, + [TSU_POST2] = 0x0074, + [TSU_POST3] = 0x0078, + [TSU_POST4] = 0x007c, [TSU_ADRH0] = 0x0100, [TXNLCR0] = 0x0080, @@ -2786,6 +2791,8 @@ static void sh_eth_tsu_init(struct sh_eth_private *mdp) { if (sh_eth_is_rz_fast_ether(mdp)) { sh_eth_tsu_write(mdp, 0, TSU_TEN); /* Disable all CAM entry */ + sh_eth_tsu_write(mdp, TSU_FWSLC_POSTENU | TSU_FWSLC_POSTENL, + TSU_FWSLC); /* Enable POST registers */ return; } From b519d408ea32040b1c7e10b155a3ee9a36660947 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 11 Sep 2016 14:50:01 -0400 Subject: [PATCH 169/271] NFSv4.1: Fix the CREATE_SESSION slot number accounting Ensure that we conform to the algorithm described in RFC5661, section 18.36.4 for when to bump the sequence id. In essence we do it for all cases except when the RPC call timed out, or in case of the server returning NFS4ERR_DELAY or NFS4ERR_STALE_CLIENTID. Signed-off-by: Trond Myklebust Cc: stable@vger.kernel.org --- fs/nfs/nfs4proc.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index c380d2ee4137..a9dec32ba9ba 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7570,12 +7570,20 @@ static int _nfs4_proc_create_session(struct nfs_client *clp, status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); trace_nfs4_create_session(clp, status); + switch (status) { + case -NFS4ERR_STALE_CLIENTID: + case -NFS4ERR_DELAY: + case -ETIMEDOUT: + case -EACCES: + case -EAGAIN: + goto out; + }; + + clp->cl_seqid++; if (!status) { /* Verify the session's negotiated channel_attrs values */ status = nfs4_verify_channel_attrs(&args, &res); /* Increment the clientid slot sequence id */ - if (clp->cl_seqid == res.seqid) - clp->cl_seqid++; if (status) goto out; nfs4_update_session(session, &res); From 74a9e9054456658f047bf71d9108844defde793d Mon Sep 17 00:00:00 2001 From: Kamal Heib Date: Sun, 11 Sep 2016 10:56:17 +0300 Subject: [PATCH 170/271] net/mlx4_en: Fix the return value of mlx4_en_dcbnl_set_all() mlx4_en_dcbnl_set_all() returns u8, so return value can't be negative in case of failure. Fixes: af7d51852631 ("net/mlx4_en: Add DCB PFC support through CEE netlink commands") Signed-off-by: Kamal Heib Signed-off-by: Rana Shahout Reported-by: Dan Carpenter Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- .../net/ethernet/mellanox/mlx4/en_dcb_nl.c | 21 ++++++++++--------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c index 99c6bbdff501..97081e5bafd1 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c @@ -158,10 +158,9 @@ static u8 mlx4_en_dcbnl_set_all(struct net_device *netdev) struct mlx4_en_priv *priv = netdev_priv(netdev); struct mlx4_en_dev *mdev = priv->mdev; struct mlx4_en_cee_config *dcb_cfg = &priv->cee_params.dcb_cfg; - int err = 0; if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) - return -EINVAL; + return 1; if (dcb_cfg->pfc_state) { int tc; @@ -199,15 +198,17 @@ static u8 mlx4_en_dcbnl_set_all(struct net_device *netdev) en_dbg(DRV, priv, "Set pfc off\n"); } - err = mlx4_SET_PORT_general(mdev->dev, priv->port, - priv->rx_skb_size + ETH_FCS_LEN, - priv->prof->tx_pause, - priv->prof->tx_ppp, - priv->prof->rx_pause, - priv->prof->rx_ppp); - if (err) + if (mlx4_SET_PORT_general(mdev->dev, priv->port, + priv->rx_skb_size + ETH_FCS_LEN, + priv->prof->tx_pause, + priv->prof->tx_ppp, + priv->prof->rx_pause, + priv->prof->rx_ppp)) { en_err(priv, "Failed setting pause params\n"); - return err; + return 1; + } + + return 0; } static u8 mlx4_en_dcbnl_get_state(struct net_device *dev) From c677071741343381f4f555867c04af7ec4a90869 Mon Sep 17 00:00:00 2001 From: Kamal Heib Date: Sun, 11 Sep 2016 10:56:18 +0300 Subject: [PATCH 171/271] net/mlx4_en: Fix the return value of mlx4_en_dcbnl_set_state() mlx4_en_dcbnl_set_state() returns u8, the return value from mlx4_en_setup_tc() could be negative in case of failure, so fix that. Fixes: af7d51852631 ("net/mlx4_en: Add DCB PFC support through CEE netlink commands") Signed-off-by: Kamal Heib Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c index 97081e5bafd1..316a70714434 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c @@ -239,7 +239,10 @@ static u8 mlx4_en_dcbnl_set_state(struct net_device *dev, u8 state) priv->flags &= ~MLX4_EN_FLAG_DCB_ENABLED; } - return mlx4_en_setup_tc(dev, num_tcs); + if (mlx4_en_setup_tc(dev, num_tcs)) + return 1; + + return 0; } /* On success returns a non-zero 802.1p user priority bitmap From 564ed9b187082da79726a2f25bb88a58ebeec7e8 Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Sun, 11 Sep 2016 10:56:19 +0300 Subject: [PATCH 172/271] net/mlx4_en: Fixes for DCBX This patch adds a capability check before enabling DCBX. In addition, it re-organizes the relevant data structures, and fixes a typo in a define. Fixes: af7d51852631 ("net/mlx4_en: Add DCB PFC support through CEE netlink commands") Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- .../net/ethernet/mellanox/mlx4/en_dcb_nl.c | 31 +++++++++---------- .../net/ethernet/mellanox/mlx4/en_netdev.c | 21 +++++-------- drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 15 ++------- drivers/net/ethernet/mellanox/mlx4/port.c | 4 +-- 4 files changed, 28 insertions(+), 43 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c index 316a70714434..b04760a5034b 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c @@ -94,7 +94,7 @@ static u8 mlx4_en_dcbnl_getcap(struct net_device *dev, int capid, u8 *cap) *cap = true; break; case DCB_CAP_ATTR_DCBX: - *cap = priv->cee_params.dcbx_cap; + *cap = priv->dcbx_cap; break; case DCB_CAP_ATTR_PFC_TCS: *cap = 1 << mlx4_max_tc(priv->mdev->dev); @@ -111,14 +111,14 @@ static u8 mlx4_en_dcbnl_getpfcstate(struct net_device *netdev) { struct mlx4_en_priv *priv = netdev_priv(netdev); - return priv->cee_params.dcb_cfg.pfc_state; + return priv->cee_config.pfc_state; } static void mlx4_en_dcbnl_setpfcstate(struct net_device *netdev, u8 state) { struct mlx4_en_priv *priv = netdev_priv(netdev); - priv->cee_params.dcb_cfg.pfc_state = state; + priv->cee_config.pfc_state = state; } static void mlx4_en_dcbnl_get_pfc_cfg(struct net_device *netdev, int priority, @@ -126,7 +126,7 @@ static void mlx4_en_dcbnl_get_pfc_cfg(struct net_device *netdev, int priority, { struct mlx4_en_priv *priv = netdev_priv(netdev); - *setting = priv->cee_params.dcb_cfg.tc_config[priority].dcb_pfc; + *setting = priv->cee_config.dcb_pfc[priority]; } static void mlx4_en_dcbnl_set_pfc_cfg(struct net_device *netdev, int priority, @@ -134,8 +134,8 @@ static void mlx4_en_dcbnl_set_pfc_cfg(struct net_device *netdev, int priority, { struct mlx4_en_priv *priv = netdev_priv(netdev); - priv->cee_params.dcb_cfg.tc_config[priority].dcb_pfc = setting; - priv->cee_params.dcb_cfg.pfc_state = true; + priv->cee_config.dcb_pfc[priority] = setting; + priv->cee_config.pfc_state = true; } static int mlx4_en_dcbnl_getnumtcs(struct net_device *netdev, int tcid, u8 *num) @@ -157,12 +157,11 @@ static u8 mlx4_en_dcbnl_set_all(struct net_device *netdev) { struct mlx4_en_priv *priv = netdev_priv(netdev); struct mlx4_en_dev *mdev = priv->mdev; - struct mlx4_en_cee_config *dcb_cfg = &priv->cee_params.dcb_cfg; - if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + if (!(priv->dcbx_cap & DCB_CAP_DCBX_VER_CEE)) return 1; - if (dcb_cfg->pfc_state) { + if (priv->cee_config.pfc_state) { int tc; priv->prof->rx_pause = 0; @@ -170,7 +169,7 @@ static u8 mlx4_en_dcbnl_set_all(struct net_device *netdev) for (tc = 0; tc < CEE_DCBX_MAX_PRIO; tc++) { u8 tc_mask = 1 << tc; - switch (dcb_cfg->tc_config[tc].dcb_pfc) { + switch (priv->cee_config.dcb_pfc[tc]) { case pfc_disabled: priv->prof->tx_ppp &= ~tc_mask; priv->prof->rx_ppp &= ~tc_mask; @@ -226,7 +225,7 @@ static u8 mlx4_en_dcbnl_set_state(struct net_device *dev, u8 state) struct mlx4_en_priv *priv = netdev_priv(dev); int num_tcs = 0; - if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + if (!(priv->dcbx_cap & DCB_CAP_DCBX_VER_CEE)) return 1; if (!!(state) == !!(priv->flags & MLX4_EN_FLAG_DCB_ENABLED)) @@ -256,7 +255,7 @@ static int mlx4_en_dcbnl_getapp(struct net_device *netdev, u8 idtype, u16 id) .selector = idtype, .protocol = id, }; - if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + if (!(priv->dcbx_cap & DCB_CAP_DCBX_VER_CEE)) return 0; return dcb_getapp(netdev, &app); @@ -268,7 +267,7 @@ static int mlx4_en_dcbnl_setapp(struct net_device *netdev, u8 idtype, struct mlx4_en_priv *priv = netdev_priv(netdev); struct dcb_app app; - if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + if (!(priv->dcbx_cap & DCB_CAP_DCBX_VER_CEE)) return -EINVAL; memset(&app, 0, sizeof(struct dcb_app)); @@ -437,7 +436,7 @@ static u8 mlx4_en_dcbnl_getdcbx(struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); - return priv->cee_params.dcbx_cap; + return priv->dcbx_cap; } static u8 mlx4_en_dcbnl_setdcbx(struct net_device *dev, u8 mode) @@ -446,7 +445,7 @@ static u8 mlx4_en_dcbnl_setdcbx(struct net_device *dev, u8 mode) struct ieee_ets ets = {0}; struct ieee_pfc pfc = {0}; - if (mode == priv->cee_params.dcbx_cap) + if (mode == priv->dcbx_cap) return 0; if ((mode & DCB_CAP_DCBX_LLD_MANAGED) || @@ -455,7 +454,7 @@ static u8 mlx4_en_dcbnl_setdcbx(struct net_device *dev, u8 mode) !(mode & DCB_CAP_DCBX_HOST)) goto err; - priv->cee_params.dcbx_cap = mode; + priv->dcbx_cap = mode; ets.ets_cap = IEEE_8021QAZ_MAX_TCS; pfc.pfc_cap = IEEE_8021QAZ_MAX_TCS; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 4198e9bf89d0..fedb829276f4 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -71,10 +71,11 @@ int mlx4_en_setup_tc(struct net_device *dev, u8 up) #ifdef CONFIG_MLX4_EN_DCB if (!mlx4_is_slave(priv->mdev->dev)) { if (up) { - priv->flags |= MLX4_EN_FLAG_DCB_ENABLED; + if (priv->dcbx_cap) + priv->flags |= MLX4_EN_FLAG_DCB_ENABLED; } else { priv->flags &= ~MLX4_EN_FLAG_DCB_ENABLED; - priv->cee_params.dcb_cfg.pfc_state = false; + priv->cee_config.pfc_state = false; } } #endif /* CONFIG_MLX4_EN_DCB */ @@ -3048,9 +3049,6 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, struct mlx4_en_priv *priv; int i; int err; -#ifdef CONFIG_MLX4_EN_DCB - struct tc_configuration *tc; -#endif dev = alloc_etherdev_mqs(sizeof(struct mlx4_en_priv), MAX_TX_RINGS, MAX_RX_RINGS); @@ -3117,16 +3115,13 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, priv->msg_enable = MLX4_EN_MSG_LEVEL; #ifdef CONFIG_MLX4_EN_DCB if (!mlx4_is_slave(priv->mdev->dev)) { - priv->cee_params.dcbx_cap = DCB_CAP_DCBX_VER_CEE | - DCB_CAP_DCBX_HOST | - DCB_CAP_DCBX_VER_IEEE; + priv->dcbx_cap = DCB_CAP_DCBX_VER_CEE | DCB_CAP_DCBX_HOST | + DCB_CAP_DCBX_VER_IEEE; priv->flags |= MLX4_EN_DCB_ENABLED; - priv->cee_params.dcb_cfg.pfc_state = false; + priv->cee_config.pfc_state = false; - for (i = 0; i < MLX4_EN_NUM_UP; i++) { - tc = &priv->cee_params.dcb_cfg.tc_config[i]; - tc->dcb_pfc = pfc_disabled; - } + for (i = 0; i < MLX4_EN_NUM_UP; i++) + priv->cee_config.dcb_pfc[i] = pfc_disabled; if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETS_CFG) { dev->dcbnl_ops = &mlx4_en_dcbnl_ops; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 2c2913dcae98..9099dbd04951 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -482,20 +482,10 @@ enum dcb_pfc_type { pfc_enabled_rx }; -struct tc_configuration { - enum dcb_pfc_type dcb_pfc; -}; - struct mlx4_en_cee_config { bool pfc_state; - struct tc_configuration tc_config[MLX4_EN_NUM_UP]; + enum dcb_pfc_type dcb_pfc[MLX4_EN_NUM_UP]; }; - -struct mlx4_en_cee_params { - u8 dcbx_cap; - struct mlx4_en_cee_config dcb_cfg; -}; - #endif struct ethtool_flow_id { @@ -624,7 +614,8 @@ struct mlx4_en_priv { struct ieee_ets ets; u16 maxrate[IEEE_8021QAZ_MAX_TCS]; enum dcbnl_cndd_states cndd_state[IEEE_8021QAZ_MAX_TCS]; - struct mlx4_en_cee_params cee_params; + struct mlx4_en_cee_config cee_config; + u8 dcbx_cap; #endif #ifdef CONFIG_RFS_ACCEL spinlock_t filters_lock; diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index 3d2095e5c61c..c5b2064297a1 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -52,7 +52,7 @@ #define MLX4_FLAG_V_IGNORE_FCS_MASK 0x2 #define MLX4_IGNORE_FCS_MASK 0x1 -#define MLNX4_TX_MAX_NUMBER 8 +#define MLX4_TC_MAX_NUMBER 8 void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table) { @@ -2022,7 +2022,7 @@ int mlx4_max_tc(struct mlx4_dev *dev) u8 num_tc = dev->caps.max_tc_eth; if (!num_tc) - num_tc = MLNX4_TX_MAX_NUMBER; + num_tc = MLX4_TC_MAX_NUMBER; return num_tc; } From 7a61fc86af4a0c957c5b8c5777ab21d5cc286748 Mon Sep 17 00:00:00 2001 From: Moshe Shemesh Date: Sun, 11 Sep 2016 10:56:20 +0300 Subject: [PATCH 173/271] net/mlx4_en: Fix panic on xmit while port is down When port is down, tx drop counter update is not needed. Updating the counter in this case can cause a kernel panic as when the port is down, ring can be NULL. Fixes: 63a664b7e92b ("net/mlx4_en: fix tx_dropped bug") Signed-off-by: Moshe Shemesh Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_tx.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index 9df87ca0515a..e2509bba3e7c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -818,7 +818,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) real_size = get_real_size(skb, shinfo, dev, &lso_header_size, &inline_ok, &fragptr); if (unlikely(!real_size)) - goto tx_drop; + goto tx_drop_count; /* Align descriptor to TXBB size */ desc_size = ALIGN(real_size, TXBB_SIZE); @@ -826,7 +826,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) if (unlikely(nr_txbb > MAX_DESC_TXBBS)) { if (netif_msg_tx_err(priv)) en_warn(priv, "Oversized header or SG list\n"); - goto tx_drop; + goto tx_drop_count; } bf_ok = ring->bf_enabled; @@ -1071,9 +1071,10 @@ tx_drop_unmap: PCI_DMA_TODEVICE); } +tx_drop_count: + ring->tx_dropped++; tx_drop: dev_kfree_skb_any(skb); - ring->tx_dropped++; return NETDEV_TX_OK; } @@ -1106,7 +1107,7 @@ netdev_tx_t mlx4_en_xmit_frame(struct mlx4_en_rx_alloc *frame, goto tx_drop; if (mlx4_en_is_tx_ring_full(ring)) - goto tx_drop; + goto tx_drop_count; /* fetch ring->cons far ahead before needing it to avoid stall */ ring_cons = READ_ONCE(ring->cons); @@ -1176,7 +1177,8 @@ netdev_tx_t mlx4_en_xmit_frame(struct mlx4_en_rx_alloc *frame, return NETDEV_TX_OK; -tx_drop: +tx_drop_count: ring->tx_dropped++; +tx_drop: return NETDEV_TX_BUSY; } From bd00a240dc52e28706fbbe3aceda63e6c291b433 Mon Sep 17 00:00:00 2001 From: "Gautham R. Shenoy" Date: Wed, 7 Sep 2016 10:46:30 +0530 Subject: [PATCH 174/271] powerpc/powernv: Fix restore of SPRs upon wake up from hypervisor state loss pnv_wakeup_tb_loss() currently expects cr4 to be "eq" if the CPU is waking up from a complete hypervisor state loss. Hence, it currently restores the SPR contents only if cr4 is "eq". However, after commit bcef83a00dc4 ("powerpc/powernv: Add platform support for stop instruction"), on ISA v3.0 CPUs, the function pnv_restore_hyp_resource() sets cr4 to contain the result of the comparison between the state the CPU has woken up from and the first deep stop state before calling pnv_wakeup_tb_loss(). Thus if the CPU woke up from a state that is deeper than the first deep stop state, cr4 will have "gt" set and hence, pnv_wakeup_tb_loss() will fail to restore the SPRs on waking up from such a state. Fix the code in pnv_wakeup_tb_loss() to restore the SPR states when cr4 is "eq" or "gt". Fixes: bcef83a00dc4 ("powerpc/powernv: Add platform support for stop instruction") Signed-off-by: Gautham R. Shenoy Reviewed-by: Shreyas B. Prabhu Signed-off-by: Michael Ellerman --- arch/powerpc/kernel/idle_book3s.S | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/powerpc/kernel/idle_book3s.S b/arch/powerpc/kernel/idle_book3s.S index 2265c6398a17..bd739fed26e3 100644 --- a/arch/powerpc/kernel/idle_book3s.S +++ b/arch/powerpc/kernel/idle_book3s.S @@ -411,7 +411,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300) * * r13 - PACA * cr3 - gt if waking up with partial/complete hypervisor state loss - * cr4 - eq if waking up from complete hypervisor state loss. + * cr4 - gt or eq if waking up from complete hypervisor state loss. */ _GLOBAL(pnv_wakeup_tb_loss) ld r1,PACAR1(r13) @@ -453,7 +453,7 @@ lwarx_loop2: * At this stage * cr2 - eq if first thread to wakeup in core * cr3- gt if waking up with partial/complete hypervisor state loss - * cr4 - eq if waking up from complete hypervisor state loss. + * cr4 - gt or eq if waking up from complete hypervisor state loss. */ ori r15,r15,PNV_CORE_IDLE_LOCK_BIT @@ -481,7 +481,7 @@ first_thread_in_subcore: * If waking up from sleep, subcore state is not lost. Hence * skip subcore state restore */ - bne cr4,subcore_state_restored + blt cr4,subcore_state_restored /* Restore per-subcore state */ ld r4,_SDR1(r1) @@ -526,7 +526,7 @@ timebase_resync: * If waking up from sleep, per core state is not lost, skip to * clear_lock. */ - bne cr4,clear_lock + blt cr4,clear_lock /* * First thread in the core to wake up and its waking up with @@ -557,7 +557,7 @@ common_exit: * If waking up from sleep, hypervisor state is not lost. Hence * skip hypervisor state restore. */ - bne cr4,hypervisor_state_restored + blt cr4,hypervisor_state_restored /* Waking up from winkle */ From ffed15d3ce3f710b94e6f402e1ca2318f7d7c0e2 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Mon, 12 Sep 2016 12:48:28 +1000 Subject: [PATCH 175/271] powerpc/kernel: Fix size of NUM_CPU_FTR_KEYS on 32-bit The number of CPU feature keys is meant to map 1:1 to the number of CPU feature flags defined in cputable.h, and the latter must fit in an unsigned long. In commit 4db7327194db ("powerpc: Add option to use jump label for cpu_has_feature()"), I incorrectly defined NUM_CPU_FTR_KEYS to 64. There should be no real adverse consequences of this bug, other than us allocating too many keys. Fix it by using BITS_PER_LONG. Fixes: 4db7327194db ("powerpc: Add option to use jump label for cpu_has_feature()") Tested-by: Meelis Roos Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/cpu_has_feature.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/include/asm/cpu_has_feature.h b/arch/powerpc/include/asm/cpu_has_feature.h index 2ef55f8968a2..b312b152461b 100644 --- a/arch/powerpc/include/asm/cpu_has_feature.h +++ b/arch/powerpc/include/asm/cpu_has_feature.h @@ -15,7 +15,7 @@ static inline bool early_cpu_has_feature(unsigned long feature) #ifdef CONFIG_JUMP_LABEL_FEATURE_CHECKS #include -#define NUM_CPU_FTR_KEYS 64 +#define NUM_CPU_FTR_KEYS BITS_PER_LONG extern struct static_key_true cpu_feature_keys[NUM_CPU_FTR_KEYS]; From f190fd92458da3e869b4e2c6289e2c617490ae53 Mon Sep 17 00:00:00 2001 From: Daniele Palmas Date: Fri, 2 Sep 2016 10:37:56 +0200 Subject: [PATCH 176/271] USB: serial: simple: add support for another Infineon flashloader This patch adds support for Infineon flashloader 0x8087/0x0801. The flashloader is used in Telit LE940B modem family with Telit flashing application. Signed-off-by: Daniele Palmas Cc: stable Signed-off-by: Johan Hovold --- drivers/usb/serial/usb-serial-simple.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/serial/usb-serial-simple.c b/drivers/usb/serial/usb-serial-simple.c index a204782ae530..e98b6e57b703 100644 --- a/drivers/usb/serial/usb-serial-simple.c +++ b/drivers/usb/serial/usb-serial-simple.c @@ -54,7 +54,8 @@ DEVICE(funsoft, FUNSOFT_IDS); /* Infineon Flashloader driver */ #define FLASHLOADER_IDS() \ { USB_DEVICE_INTERFACE_CLASS(0x058b, 0x0041, USB_CLASS_CDC_DATA) }, \ - { USB_DEVICE(0x8087, 0x0716) } + { USB_DEVICE(0x8087, 0x0716) }, \ + { USB_DEVICE(0x8087, 0x0801) } DEVICE(flashloader, FLASHLOADER_IDS); /* Google Serial USB SubClass */ From 981b178964d03f6f8e6cca01568c17d1dbafdf0e Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Thu, 8 Sep 2016 11:11:35 +0200 Subject: [PATCH 177/271] dt-bindings: mmc: sdhci-st: Mention the discretionary "icn" clock The interconnect (ICN) clock is required for functional working of MMC on some ST platforms. When not supplied it can result in broken MMC and the following output: [ 13.916949] mmc0: Timeout waiting for hardware interrupt. [ 13.922349] sdhci: =========== REGISTER DUMP (mmc0)=========== [ 13.928175] sdhci: Sys addr: 0x00000000 | Version: 0x00001002 [ 13.933999] sdhci: Blk size: 0x00007040 | Blk cnt: 0x00000001 [ 13.939825] sdhci: Argument: 0x00fffff0 | Trn mode: 0x00000013 [ 13.945650] sdhci: Present: 0x1fff0206 | Host ctl: 0x00000011 [ 13.951475] sdhci: Power: 0x0000000f | Blk gap: 0x00000080 [ 13.957300] sdhci: Wake-up: 0x00000000 | Clock: 0x00003f07 [ 13.963126] sdhci: Timeout: 0x00000004 | Int stat: 0x00000000 [ 13.968952] sdhci: Int enab: 0x02ff008b | Sig enab: 0x02ff008b [ 13.974777] sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000 [ 13.980602] sdhci: Caps: 0x21ed3281 | Caps_1: 0x00000000 [ 13.986428] sdhci: Cmd: 0x0000063a | Max curr: 0x00000000 [ 13.992252] sdhci: Host ctl2: 0x00000000 [ 13.996166] sdhci: ADMA Err: 0x00000000 | ADMA Ptr: 0x7c048200 [ 14.001990] sdhci: =========================================== [ 14.009802] mmc0: Got data interrupt 0x02000000 even though no data operation was in progress. Signed-off-by: Lee Jones Signed-off-by: Ulf Hansson --- Documentation/devicetree/bindings/mmc/sdhci-st.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/mmc/sdhci-st.txt b/Documentation/devicetree/bindings/mmc/sdhci-st.txt index 88faa91125bf..3cd4c43a3260 100644 --- a/Documentation/devicetree/bindings/mmc/sdhci-st.txt +++ b/Documentation/devicetree/bindings/mmc/sdhci-st.txt @@ -10,7 +10,7 @@ Required properties: subsystem (mmcss) inside the FlashSS (available in STiH407 SoC family). -- clock-names: Should be "mmc". +- clock-names: Should be "mmc" and "icn". (NB: The latter is not compulsory) See: Documentation/devicetree/bindings/resource-names.txt - clocks: Phandle to the clock. See: Documentation/devicetree/bindings/clock/clock-bindings.txt From 3ae50f4512ce831e8b63eb54ad969417ff30ada7 Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Thu, 8 Sep 2016 11:11:36 +0200 Subject: [PATCH 178/271] mmc: sdhci-st: Handle interconnect clock Some ST platforms contain interconnect (ICN) clocks which must be handed correctly in order to obtain full functionality of a given IP. In this case, if the ICN clocks are not handled properly by the ST SDHCI driver MMC will break and the following output can be observed: [ 13.916949] mmc0: Timeout waiting for hardware interrupt. [ 13.922349] sdhci: =========== REGISTER DUMP (mmc0)=========== [ 13.928175] sdhci: Sys addr: 0x00000000 | Version: 0x00001002 [ 13.933999] sdhci: Blk size: 0x00007040 | Blk cnt: 0x00000001 [ 13.939825] sdhci: Argument: 0x00fffff0 | Trn mode: 0x00000013 [ 13.945650] sdhci: Present: 0x1fff0206 | Host ctl: 0x00000011 [ 13.951475] sdhci: Power: 0x0000000f | Blk gap: 0x00000080 [ 13.957300] sdhci: Wake-up: 0x00000000 | Clock: 0x00003f07 [ 13.963126] sdhci: Timeout: 0x00000004 | Int stat: 0x00000000 [ 13.968952] sdhci: Int enab: 0x02ff008b | Sig enab: 0x02ff008b [ 13.974777] sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000 [ 13.980602] sdhci: Caps: 0x21ed3281 | Caps_1: 0x00000000 [ 13.986428] sdhci: Cmd: 0x0000063a | Max curr: 0x00000000 [ 13.992252] sdhci: Host ctl2: 0x00000000 [ 13.996166] sdhci: ADMA Err: 0x00000000 | ADMA Ptr: 0x7c048200 [ 14.001990] sdhci: =========================================== [ 14.009802] mmc0: Got data interrupt 0x02000000 even though no data operation was in progress. A decent point was raised about minimising the use of a local variable that we 'could' do without. I've chosen consistency over the possibility of reducing the local variable count by 1. Thinking that it's more important for the code to be grouped and authoured in a similar manner/style for greater maintainability/readability. Cc: stable@vger.kernel.org Tested-by: Peter Griffin Signed-off-by: Lee Jones Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-st.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/drivers/mmc/host/sdhci-st.c b/drivers/mmc/host/sdhci-st.c index c95ba83366a0..ed92ce729dde 100644 --- a/drivers/mmc/host/sdhci-st.c +++ b/drivers/mmc/host/sdhci-st.c @@ -28,6 +28,7 @@ struct st_mmc_platform_data { struct reset_control *rstc; + struct clk *icnclk; void __iomem *top_ioaddr; }; @@ -353,7 +354,7 @@ static int sdhci_st_probe(struct platform_device *pdev) struct sdhci_host *host; struct st_mmc_platform_data *pdata; struct sdhci_pltfm_host *pltfm_host; - struct clk *clk; + struct clk *clk, *icnclk; int ret = 0; u16 host_version; struct resource *res; @@ -365,6 +366,11 @@ static int sdhci_st_probe(struct platform_device *pdev) return PTR_ERR(clk); } + /* ICN clock isn't compulsory, but use it if it's provided. */ + icnclk = devm_clk_get(&pdev->dev, "icn"); + if (IS_ERR(icnclk)) + icnclk = NULL; + rstc = devm_reset_control_get(&pdev->dev, NULL); if (IS_ERR(rstc)) rstc = NULL; @@ -389,6 +395,7 @@ static int sdhci_st_probe(struct platform_device *pdev) } clk_prepare_enable(clk); + clk_prepare_enable(icnclk); /* Configure the FlashSS Top registers for setting eMMC TX/RX delay */ res = platform_get_resource_byname(pdev, IORESOURCE_MEM, @@ -400,6 +407,7 @@ static int sdhci_st_probe(struct platform_device *pdev) } pltfm_host->clk = clk; + pdata->icnclk = icnclk; /* Configure the Arasan HC inside the flashSS */ st_mmcss_cconfig(np, host); @@ -422,6 +430,7 @@ static int sdhci_st_probe(struct platform_device *pdev) return 0; err_out: + clk_disable_unprepare(icnclk); clk_disable_unprepare(clk); err_of: sdhci_pltfm_free(pdev); @@ -442,6 +451,8 @@ static int sdhci_st_remove(struct platform_device *pdev) ret = sdhci_pltfm_unregister(pdev); + clk_disable_unprepare(pdata->icnclk); + if (rstc) reset_control_assert(rstc); @@ -462,6 +473,7 @@ static int sdhci_st_suspend(struct device *dev) if (pdata->rstc) reset_control_assert(pdata->rstc); + clk_disable_unprepare(pdata->icnclk); clk_disable_unprepare(pltfm_host->clk); out: return ret; @@ -475,6 +487,7 @@ static int sdhci_st_resume(struct device *dev) struct device_node *np = dev->of_node; clk_prepare_enable(pltfm_host->clk); + clk_prepare_enable(pdata->icnclk); if (pdata->rstc) reset_control_deassert(pdata->rstc); From ad8d52b897a14711e026889053befbbee7fd51ba Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 31 Aug 2016 08:49:43 +0100 Subject: [PATCH 179/271] pcmcia: ds: fix suspend/resume PCMCIA suspend/resume no longer works since the commit mentioned below, as the callbacks are no longer made. Convert the driver to the new dev_pm_ops, which restores the suspend/resume functionality. Tested on the arm arch Assabet platform. Fixes: aa8e54b559479 ("PM / sleep: Go direct_complete if driver has no callbacks") Signed-off-by: Russell King --- drivers/pcmcia/ds.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c index 489ea1098c96..69b5e811ea2b 100644 --- a/drivers/pcmcia/ds.c +++ b/drivers/pcmcia/ds.c @@ -977,7 +977,7 @@ static int pcmcia_bus_uevent(struct device *dev, struct kobj_uevent_env *env) /************************ runtime PM support ***************************/ -static int pcmcia_dev_suspend(struct device *dev, pm_message_t state); +static int pcmcia_dev_suspend(struct device *dev); static int pcmcia_dev_resume(struct device *dev); static int runtime_suspend(struct device *dev) @@ -985,7 +985,7 @@ static int runtime_suspend(struct device *dev) int rc; device_lock(dev); - rc = pcmcia_dev_suspend(dev, PMSG_SUSPEND); + rc = pcmcia_dev_suspend(dev); device_unlock(dev); return rc; } @@ -1135,7 +1135,7 @@ ATTRIBUTE_GROUPS(pcmcia_dev); /* PM support, also needed for reset */ -static int pcmcia_dev_suspend(struct device *dev, pm_message_t state) +static int pcmcia_dev_suspend(struct device *dev) { struct pcmcia_device *p_dev = to_pcmcia_dev(dev); struct pcmcia_driver *p_drv = NULL; @@ -1410,6 +1410,9 @@ static struct class_interface pcmcia_bus_interface __refdata = { .remove_dev = &pcmcia_bus_remove_socket, }; +static const struct dev_pm_ops pcmcia_bus_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(pcmcia_dev_suspend, pcmcia_dev_resume) +}; struct bus_type pcmcia_bus_type = { .name = "pcmcia", @@ -1418,8 +1421,7 @@ struct bus_type pcmcia_bus_type = { .dev_groups = pcmcia_dev_groups, .probe = pcmcia_device_probe, .remove = pcmcia_device_remove, - .suspend = pcmcia_dev_suspend, - .resume = pcmcia_dev_resume, + .pm = &pcmcia_bus_pm_ops, }; From 6dec04e8f30a8cf1e782500244d8601c1f8505ad Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 31 Aug 2016 08:49:43 +0100 Subject: [PATCH 180/271] pcmcia: sa11xx_base: fix reporting of timing information Fix the reporting of the currently programmed timing information. These entries have been showing zero due to the clock rate being a factor of 1000 too big. With this change, we go from: I/O : 165 (0) attribute: 300 (0) common : 300 (0) to: I/O : 165 (172) attribute: 300 (316) common : 300 (316) Signed-off-by: Russell King --- drivers/pcmcia/sa11xx_base.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pcmcia/sa11xx_base.c b/drivers/pcmcia/sa11xx_base.c index 9f6ec87b9f9e..af37b7ec1c73 100644 --- a/drivers/pcmcia/sa11xx_base.c +++ b/drivers/pcmcia/sa11xx_base.c @@ -144,7 +144,7 @@ static int sa1100_pcmcia_show_timing(struct soc_pcmcia_socket *skt, char *buf) { struct soc_pcmcia_timing timing; - unsigned int clock = clk_get_rate(skt->clk); + unsigned int clock = clk_get_rate(skt->clk) / 1000; unsigned long mecr = MECR; char *p = buf; From cbd5a16820e576d26bf985ad62b8c4cdf792fb45 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 31 Aug 2016 08:49:43 +0100 Subject: [PATCH 181/271] pcmcia: sa11xx_base: add units to the timing information Add units to the timing information, so we know that the numbers are nanoseconds. The output changes from: I/O : 165 (172) attribute: 300 (316) common : 300 (316) to: I/O : 165ns (172ns) attribute: 300ns (316ns) common : 300ns (316ns) Signed-off-by: Russell King --- drivers/pcmcia/sa11xx_base.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/pcmcia/sa11xx_base.c b/drivers/pcmcia/sa11xx_base.c index af37b7ec1c73..48140ac73ed6 100644 --- a/drivers/pcmcia/sa11xx_base.c +++ b/drivers/pcmcia/sa11xx_base.c @@ -150,13 +150,13 @@ sa1100_pcmcia_show_timing(struct soc_pcmcia_socket *skt, char *buf) soc_common_pcmcia_get_timing(skt, &timing); - p+=sprintf(p, "I/O : %u (%u)\n", timing.io, + p+=sprintf(p, "I/O : %uns (%uns)\n", timing.io, sa1100_pcmcia_cmd_time(clock, MECR_BSIO_GET(mecr, skt->nr))); - p+=sprintf(p, "attribute: %u (%u)\n", timing.attr, + p+=sprintf(p, "attribute: %uns (%uns)\n", timing.attr, sa1100_pcmcia_cmd_time(clock, MECR_BSA_GET(mecr, skt->nr))); - p+=sprintf(p, "common : %u (%u)\n", timing.mem, + p+=sprintf(p, "common : %uns (%uns)\n", timing.mem, sa1100_pcmcia_cmd_time(clock, MECR_BSM_GET(mecr, skt->nr))); return p - buf; From a466ebd2fc6a793e55f028a008b9f094d7d30fe3 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 31 Aug 2016 08:49:43 +0100 Subject: [PATCH 182/271] pcmcia: soc_common: fix SS_STSCHG polarity SS_STSCHG should be set for an IO card when the BVD1 signal is asserted low, not high. Signed-off-by: Russell King --- drivers/pcmcia/soc_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pcmcia/soc_common.c b/drivers/pcmcia/soc_common.c index eed5e9c05353..d5ca760c4eb2 100644 --- a/drivers/pcmcia/soc_common.c +++ b/drivers/pcmcia/soc_common.c @@ -235,7 +235,7 @@ static unsigned int soc_common_pcmcia_skt_state(struct soc_pcmcia_socket *skt) stat |= skt->cs_state.Vcc ? SS_POWERON : 0; if (skt->cs_state.flags & SS_IOCARD) - stat |= state.bvd1 ? SS_STSCHG : 0; + stat |= state.bvd1 ? 0 : SS_STSCHG; else { if (state.bvd1 == 0) stat |= SS_BATDEAD; From 3f8df892b2312011f2ba73aedc0a192d70b8844e Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 2 Sep 2016 10:14:20 +0100 Subject: [PATCH 183/271] pcmcia: sa1111: fix propagation of lowlevel board init return code When testing Lubbock, it was noticed that the sa1111 pcmcia driver bound but was not functional due to no sockets being registered. This is because the return code from the lowlevel board initialisation was not being propagated out of the probe function. Fix this. Tested-by: Robert Jarzmik Signed-off-by: Russell King --- drivers/pcmcia/sa1111_badge4.c | 20 +++++++------------ drivers/pcmcia/sa1111_generic.c | 22 +++++++++++++++----- drivers/pcmcia/sa1111_jornada720.c | 25 ++++++++++------------- drivers/pcmcia/sa1111_lubbock.c | 32 ++++++++++++------------------ drivers/pcmcia/sa1111_neponset.c | 26 ++++++++++-------------- 5 files changed, 57 insertions(+), 68 deletions(-) diff --git a/drivers/pcmcia/sa1111_badge4.c b/drivers/pcmcia/sa1111_badge4.c index 12f0dd091477..2f490930430d 100644 --- a/drivers/pcmcia/sa1111_badge4.c +++ b/drivers/pcmcia/sa1111_badge4.c @@ -134,20 +134,14 @@ static struct pcmcia_low_level badge4_pcmcia_ops = { int pcmcia_badge4_init(struct sa1111_dev *dev) { - int ret = -ENODEV; + printk(KERN_INFO + "%s: badge4_pcmvcc=%d, badge4_pcmvpp=%d, badge4_cfvcc=%d\n", + __func__, + badge4_pcmvcc, badge4_pcmvpp, badge4_cfvcc); - if (machine_is_badge4()) { - printk(KERN_INFO - "%s: badge4_pcmvcc=%d, badge4_pcmvpp=%d, badge4_cfvcc=%d\n", - __func__, - badge4_pcmvcc, badge4_pcmvpp, badge4_cfvcc); - - sa11xx_drv_pcmcia_ops(&badge4_pcmcia_ops); - ret = sa1111_pcmcia_add(dev, &badge4_pcmcia_ops, - sa11xx_drv_pcmcia_add_one); - } - - return ret; + sa11xx_drv_pcmcia_ops(&badge4_pcmcia_ops); + return sa1111_pcmcia_add(dev, &badge4_pcmcia_ops, + sa11xx_drv_pcmcia_add_one); } static int __init pcmv_setup(char *s) diff --git a/drivers/pcmcia/sa1111_generic.c b/drivers/pcmcia/sa1111_generic.c index a1531feb8460..3d95dffcff7a 100644 --- a/drivers/pcmcia/sa1111_generic.c +++ b/drivers/pcmcia/sa1111_generic.c @@ -18,6 +18,7 @@ #include #include +#include #include #include "sa1111_generic.h" @@ -203,19 +204,30 @@ static int pcmcia_probe(struct sa1111_dev *dev) sa1111_writel(PCSSR_S0_SLEEP | PCSSR_S1_SLEEP, base + PCSSR); sa1111_writel(PCCR_S0_FLT | PCCR_S1_FLT, base + PCCR); + ret = -ENODEV; #ifdef CONFIG_SA1100_BADGE4 - pcmcia_badge4_init(dev); + if (machine_is_badge4()) + ret = pcmcia_badge4_init(dev); #endif #ifdef CONFIG_SA1100_JORNADA720 - pcmcia_jornada720_init(dev); + if (machine_is_jornada720()) + ret = pcmcia_jornada720_init(dev); #endif #ifdef CONFIG_ARCH_LUBBOCK - pcmcia_lubbock_init(dev); + if (machine_is_lubbock()) + ret = pcmcia_lubbock_init(dev); #endif #ifdef CONFIG_ASSABET_NEPONSET - pcmcia_neponset_init(dev); + if (machine_is_assabet()) + ret = pcmcia_neponset_init(dev); #endif - return 0; + + if (ret) { + release_mem_region(dev->res.start, 512); + sa1111_disable_device(dev); + } + + return ret; } static int pcmcia_remove(struct sa1111_dev *dev) diff --git a/drivers/pcmcia/sa1111_jornada720.c b/drivers/pcmcia/sa1111_jornada720.c index c2c30580c83f..480a3ede27c8 100644 --- a/drivers/pcmcia/sa1111_jornada720.c +++ b/drivers/pcmcia/sa1111_jornada720.c @@ -94,22 +94,17 @@ static struct pcmcia_low_level jornada720_pcmcia_ops = { int pcmcia_jornada720_init(struct sa1111_dev *sadev) { - int ret = -ENODEV; + unsigned int pin = GPIO_A0 | GPIO_A1 | GPIO_A2 | GPIO_A3; - if (machine_is_jornada720()) { - unsigned int pin = GPIO_A0 | GPIO_A1 | GPIO_A2 | GPIO_A3; + /* Fixme: why messing around with SA11x0's GPIO1? */ + GRER |= 0x00000002; - GRER |= 0x00000002; + /* Set GPIO_A<3:1> to be outputs for PCMCIA/CF power controller: */ + sa1111_set_io_dir(sadev, pin, 0, 0); + sa1111_set_io(sadev, pin, 0); + sa1111_set_sleep_io(sadev, pin, 0); - /* Set GPIO_A<3:1> to be outputs for PCMCIA/CF power controller: */ - sa1111_set_io_dir(sadev, pin, 0, 0); - sa1111_set_io(sadev, pin, 0); - sa1111_set_sleep_io(sadev, pin, 0); - - sa11xx_drv_pcmcia_ops(&jornada720_pcmcia_ops); - ret = sa1111_pcmcia_add(sadev, &jornada720_pcmcia_ops, - sa11xx_drv_pcmcia_add_one); - } - - return ret; + sa11xx_drv_pcmcia_ops(&jornada720_pcmcia_ops); + return sa1111_pcmcia_add(sadev, &jornada720_pcmcia_ops, + sa11xx_drv_pcmcia_add_one); } diff --git a/drivers/pcmcia/sa1111_lubbock.c b/drivers/pcmcia/sa1111_lubbock.c index c5caf5790451..df2b6b28b1d1 100644 --- a/drivers/pcmcia/sa1111_lubbock.c +++ b/drivers/pcmcia/sa1111_lubbock.c @@ -210,27 +210,21 @@ static struct pcmcia_low_level lubbock_pcmcia_ops = { int pcmcia_lubbock_init(struct sa1111_dev *sadev) { - int ret = -ENODEV; + /* + * Set GPIO_A<3:0> to be outputs for the MAX1600, + * and switch to standby mode. + */ + sa1111_set_io_dir(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0, 0); + sa1111_set_io(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0); + sa1111_set_sleep_io(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0); - if (machine_is_lubbock()) { - /* - * Set GPIO_A<3:0> to be outputs for the MAX1600, - * and switch to standby mode. - */ - sa1111_set_io_dir(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0, 0); - sa1111_set_io(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0); - sa1111_set_sleep_io(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0); + /* Set CF Socket 1 power to standby mode. */ + lubbock_set_misc_wr((1 << 15) | (1 << 14), 0); - /* Set CF Socket 1 power to standby mode. */ - lubbock_set_misc_wr((1 << 15) | (1 << 14), 0); - - pxa2xx_drv_pcmcia_ops(&lubbock_pcmcia_ops); - pxa2xx_configure_sockets(&sadev->dev); - ret = sa1111_pcmcia_add(sadev, &lubbock_pcmcia_ops, - pxa2xx_drv_pcmcia_add_one); - } - - return ret; + pxa2xx_drv_pcmcia_ops(&lubbock_pcmcia_ops); + pxa2xx_configure_sockets(&sadev->dev); + return sa1111_pcmcia_add(sadev, &lubbock_pcmcia_ops, + pxa2xx_drv_pcmcia_add_one); } MODULE_LICENSE("GPL"); diff --git a/drivers/pcmcia/sa1111_neponset.c b/drivers/pcmcia/sa1111_neponset.c index 1d78739c4c07..019c395eb4bf 100644 --- a/drivers/pcmcia/sa1111_neponset.c +++ b/drivers/pcmcia/sa1111_neponset.c @@ -110,20 +110,14 @@ static struct pcmcia_low_level neponset_pcmcia_ops = { int pcmcia_neponset_init(struct sa1111_dev *sadev) { - int ret = -ENODEV; - - if (machine_is_assabet()) { - /* - * Set GPIO_A<3:0> to be outputs for the MAX1600, - * and switch to standby mode. - */ - sa1111_set_io_dir(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0, 0); - sa1111_set_io(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0); - sa1111_set_sleep_io(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0); - sa11xx_drv_pcmcia_ops(&neponset_pcmcia_ops); - ret = sa1111_pcmcia_add(sadev, &neponset_pcmcia_ops, - sa11xx_drv_pcmcia_add_one); - } - - return ret; + /* + * Set GPIO_A<3:0> to be outputs for the MAX1600, + * and switch to standby mode. + */ + sa1111_set_io_dir(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0, 0); + sa1111_set_io(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0); + sa1111_set_sleep_io(sadev, GPIO_A0|GPIO_A1|GPIO_A2|GPIO_A3, 0); + sa11xx_drv_pcmcia_ops(&neponset_pcmcia_ops); + return sa1111_pcmcia_add(sadev, &neponset_pcmcia_ops, + sa11xx_drv_pcmcia_add_one); } From 817ed5748e40bbc5b5f2aa0c3094c4a7adfb8881 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Thu, 1 Sep 2016 08:31:08 +0200 Subject: [PATCH 184/271] pcmcia: lubbock: fix sockets configuration On lubbock board, the probe of the driver crashes by dereferencing very early a platform_data structure which is not set, in pxa2xx_configure_sockets(). The stack fixed is : [ 0.244353] SA1111 Microprocessor Companion Chip: silicon revision 1, metal revision 1 [ 0.256321] sa1111 sa1111: Providing IRQ336-390 [ 0.340899] clocksource: Switched to clocksource oscr0 [ 0.472263] Unable to handle kernel NULL pointer dereference at virtual address 00000004 [ 0.480469] pgd = c0004000 [ 0.483432] [00000004] *pgd=00000000 [ 0.487105] Internal error: Oops: f5 [#1] ARM [ 0.491497] Modules linked in: [ 0.494650] CPU: 0 PID: 1 Comm: swapper Not tainted 4.8.0-rc3-00080-g1aaa68426f0c-dirty #2068 [ 0.503229] Hardware name: Intel DBPXA250 Development Platform (aka Lubbock) [ 0.510344] task: c3e42000 task.stack: c3e44000 [ 0.514984] PC is at pxa2xx_configure_sockets+0x4/0x24 (drivers/pcmcia/pxa2xx_base.c:227) [ 0.520193] LR is at pcmcia_lubbock_init+0x1c/0x38 [ 0.525079] pc : [] lr : [] psr: a0000053 [ 0.525079] sp : c3e45e70 ip : 100019ff fp : 00000000 [ 0.536651] r10: c0828900 r9 : c0434838 r8 : 00000000 [ 0.541953] r7 : c0820700 r6 : c0857b30 r5 : c3ec1400 r4 : c0820758 [ 0.548549] r3 : 00000000 r2 : 0000000c r1 : c3c09c40 r0 : c3ec1400 [ 0.555154] Flags: NzCv IRQs on FIQs off Mode SVC_32 ISA ARM Segment none [ 0.562450] Control: 0000397f Table: a0004000 DAC: 00000053 [ 0.568257] Process swapper (pid: 1, stack limit = 0xc3e44190) [ 0.574154] Stack: (0xc3e45e70 to 0xc3e46000) [ 0.578610] 5e60: c4849800 00000000 c3ec1400 c024769c [ 0.586928] 5e80: 00000000 c3ec140c c3c0ee0c c3ec1400 c3ec1434 c020c410 c3ec1400 c3ec1434 [ 0.595244] 5ea0: c0820700 c080b408 c0828900 c020c5f8 00000000 c0820700 c020c578 c020ac5c [ 0.603560] 5ec0: c3e687cc c3e71e10 c0820700 00000000 c3c02de0 c020bae4 c03c62f7 c03c62f7 [ 0.611872] 5ee0: c3e68780 c0820700 c042e034 00000000 c043c440 c020cdec c080b408 00000005 [ 0.620188] 5f00: c042e034 c00096c0 c0034440 c01c730c 20000053 ffffffff 00000000 00000000 [ 0.628502] 5f20: 00000000 c3ffcb87 c3ffcb90 c00346ac c3e66ba0 c03f7914 00000092 00000005 [ 0.636811] 5f40: 00000005 c03f847c 00000091 c03f847c 00000000 00000005 c0434828 00000005 [ 0.645125] 5f60: c043482c 00000092 c043c440 c0828900 c0434838 c0418d2c 00000005 00000005 [ 0.653430] 5f80: 00000000 c041858c 00000000 c032e9f0 00000000 00000000 00000000 00000000 [ 0.661729] 5fa0: 00000000 c032e9f8 00000000 c000f0f0 00000000 00000000 00000000 00000000 [ 0.670020] 5fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 [ 0.678311] 5fe0: 00000000 00000000 00000000 00000000 00000013 00000000 00000000 00000000 [ 0.686673] (pxa2xx_configure_sockets) from pcmcia_lubbock_init (/drivers/pcmcia/sa1111_lubbock.c:161) [ 0.696026] (pcmcia_lubbock_init) from pcmcia_probe (/drivers/pcmcia/sa1111_generic.c:213) [ 0.704358] (pcmcia_probe) from driver_probe_device (/drivers/base/dd.c:378 /drivers/base/dd.c:499) [ 0.712848] (driver_probe_device) from __driver_attach (/./include/linux/device.h:983 /drivers/base/dd.c:733) [ 0.721414] (__driver_attach) from bus_for_each_dev (/drivers/base/bus.c:313) [ 0.729723] (bus_for_each_dev) from bus_add_driver (/drivers/base/bus.c:708) [ 0.738036] (bus_add_driver) from driver_register (/drivers/base/driver.c:169) [ 0.746185] (driver_register) from do_one_initcall (/init/main.c:778) [ 0.754561] (do_one_initcall) from kernel_init_freeable (/init/main.c:843 /init/main.c:851 /init/main.c:869 /init/main.c:1016) [ 0.763409] (kernel_init_freeable) from kernel_init (/init/main.c:944) [ 0.771660] (kernel_init) from ret_from_fork (/arch/arm/kernel/entry-common.S:119) [ 0.779347] Code: c03c6305 c03c631e c03c632e e5903048 (e993000c) All code ======== 0: c03c6305 eorsgt r6, ip, r5, lsl #6 4: c03c631e eorsgt r6, ip, lr, lsl r3 8: c03c632e eorsgt r6, ip, lr, lsr #6 c: e5903048 ldr r3, [r0, #72] ; 0x48 10:* e993000c ldmib r3, {r2, r3} <-- trapping instruction Signed-off-by: Robert Jarzmik Signed-off-by: Russell King --- drivers/pcmcia/pxa2xx_base.c | 9 +++++---- drivers/pcmcia/pxa2xx_base.h | 2 +- drivers/pcmcia/sa1111_lubbock.c | 2 +- 3 files changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/pcmcia/pxa2xx_base.c b/drivers/pcmcia/pxa2xx_base.c index 483f919e0d2e..91b5f5724cba 100644 --- a/drivers/pcmcia/pxa2xx_base.c +++ b/drivers/pcmcia/pxa2xx_base.c @@ -214,9 +214,8 @@ pxa2xx_pcmcia_frequency_change(struct soc_pcmcia_socket *skt, } #endif -void pxa2xx_configure_sockets(struct device *dev) +void pxa2xx_configure_sockets(struct device *dev, struct pcmcia_low_level *ops) { - struct pcmcia_low_level *ops = dev->platform_data; /* * We have at least one socket, so set MECR:CIT * (Card Is There) @@ -322,7 +321,7 @@ static int pxa2xx_drv_pcmcia_probe(struct platform_device *dev) goto err1; } - pxa2xx_configure_sockets(&dev->dev); + pxa2xx_configure_sockets(&dev->dev, ops); dev_set_drvdata(&dev->dev, sinfo); return 0; @@ -348,7 +347,9 @@ static int pxa2xx_drv_pcmcia_remove(struct platform_device *dev) static int pxa2xx_drv_pcmcia_resume(struct device *dev) { - pxa2xx_configure_sockets(dev); + struct pcmcia_low_level *ops = (struct pcmcia_low_level *)dev->platform_data; + + pxa2xx_configure_sockets(dev, ops); return 0; } diff --git a/drivers/pcmcia/pxa2xx_base.h b/drivers/pcmcia/pxa2xx_base.h index b609b45469ed..e58c7a415418 100644 --- a/drivers/pcmcia/pxa2xx_base.h +++ b/drivers/pcmcia/pxa2xx_base.h @@ -1,4 +1,4 @@ int pxa2xx_drv_pcmcia_add_one(struct soc_pcmcia_socket *skt); void pxa2xx_drv_pcmcia_ops(struct pcmcia_low_level *ops); -void pxa2xx_configure_sockets(struct device *dev); +void pxa2xx_configure_sockets(struct device *dev, struct pcmcia_low_level *ops); diff --git a/drivers/pcmcia/sa1111_lubbock.c b/drivers/pcmcia/sa1111_lubbock.c index df2b6b28b1d1..e741f499c875 100644 --- a/drivers/pcmcia/sa1111_lubbock.c +++ b/drivers/pcmcia/sa1111_lubbock.c @@ -222,7 +222,7 @@ int pcmcia_lubbock_init(struct sa1111_dev *sadev) lubbock_set_misc_wr((1 << 15) | (1 << 14), 0); pxa2xx_drv_pcmcia_ops(&lubbock_pcmcia_ops); - pxa2xx_configure_sockets(&sadev->dev); + pxa2xx_configure_sockets(&sadev->dev, &lubbock_pcmcia_ops); return sa1111_pcmcia_add(sadev, &lubbock_pcmcia_ops, pxa2xx_drv_pcmcia_add_one); } From cb034407ec3f816540f359300cda1122faabdbbd Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 3 Sep 2016 10:21:51 +0100 Subject: [PATCH 185/271] ARM: sa1111: fix error code propagation in sa1111_probe() Ensure that we propagate the platform_get_irq() error code out of the probe function. This allows probe deferrals to work correctly should platform_get_irq() not be able to resolve the interrupt in a DT environment at probe time. Signed-off-by: Russell King --- arch/arm/common/sa1111.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/common/sa1111.c b/arch/arm/common/sa1111.c index fb0a0a4dfea4..332b92317fd8 100644 --- a/arch/arm/common/sa1111.c +++ b/arch/arm/common/sa1111.c @@ -1017,7 +1017,7 @@ static int sa1111_probe(struct platform_device *pdev) return -EINVAL; irq = platform_get_irq(pdev, 0); if (irq < 0) - return -ENXIO; + return irq; return __sa1111_probe(&pdev->dev, mem, irq); } From 7c0091eceab231b59e51b80bbcf5a2205a0fa905 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 6 Sep 2016 14:21:46 +0100 Subject: [PATCH 186/271] ARM: sa1111: fix pcmcia interrupt mask polarity The polarity of the high IRQs was being calculated using SA1111_IRQMASK_HI(), but this assumes a Linux interrupt number, not a hardware interrupt number. Hence, the resulting mask was incorrect. Fix this. Signed-off-by: Russell King --- arch/arm/common/sa1111.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/common/sa1111.c b/arch/arm/common/sa1111.c index 332b92317fd8..cfa61b857cad 100644 --- a/arch/arm/common/sa1111.c +++ b/arch/arm/common/sa1111.c @@ -472,8 +472,8 @@ static int sa1111_setup_irq(struct sa1111 *sachip, unsigned irq_base) * specifies that S0ReadyInt and S1ReadyInt should be '1'. */ sa1111_writel(0, irqbase + SA1111_INTPOL0); - sa1111_writel(SA1111_IRQMASK_HI(IRQ_S0_READY_NINT) | - SA1111_IRQMASK_HI(IRQ_S1_READY_NINT), + sa1111_writel(BIT(IRQ_S0_READY_NINT & 31) | + BIT(IRQ_S1_READY_NINT & 31), irqbase + SA1111_INTPOL1); /* clear all IRQs */ From 06dfe5cc0cc684e735cb0232fdb756d30780b05d Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 6 Sep 2016 14:34:05 +0100 Subject: [PATCH 187/271] ARM: sa1111: fix pcmcia suspend/resume SA1111 PCMCIA was broken when PCMCIA switched to using dev_pm_ops for the PCMCIA socket class. PCMCIA used to handle suspend/resume via the socket hosting device, which happened at normal device suspend/resume time. However, the referenced commit changed this: much of the resume now happens much earlier, in the noirq resume handler of dev_pm_ops. However, on SA1111, the PCMCIA device is not accessible as the SA1111 has not been resumed at _noirq time. It's slightly worse than that, because the SA1111 has already been put to sleep at _noirq time, so suspend doesn't work properly. Fix this by converting the core SA1111 code to use dev_pm_ops as well, and performing its own suspend/resume at noirq time. This fixes these errors in the kernel log: pcmcia_socket pcmcia_socket0: time out after reset pcmcia_socket pcmcia_socket1: time out after reset and the resulting lack of PCMCIA cards after a S2RAM cycle. Fixes: d7646f7632549 ("pcmcia: use dev_pm_ops for class pcmcia_socket_class") Signed-off-by: Russell King --- arch/arm/common/sa1111.c | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) diff --git a/arch/arm/common/sa1111.c b/arch/arm/common/sa1111.c index cfa61b857cad..7838659b870a 100644 --- a/arch/arm/common/sa1111.c +++ b/arch/arm/common/sa1111.c @@ -869,9 +869,9 @@ struct sa1111_save_data { #ifdef CONFIG_PM -static int sa1111_suspend(struct platform_device *dev, pm_message_t state) +static int sa1111_suspend_noirq(struct device *dev) { - struct sa1111 *sachip = platform_get_drvdata(dev); + struct sa1111 *sachip = dev_get_drvdata(dev); struct sa1111_save_data *save; unsigned long flags; unsigned int val; @@ -934,9 +934,9 @@ static int sa1111_suspend(struct platform_device *dev, pm_message_t state) * restored by their respective drivers, and must be called * via LDM after this function. */ -static int sa1111_resume(struct platform_device *dev) +static int sa1111_resume_noirq(struct device *dev) { - struct sa1111 *sachip = platform_get_drvdata(dev); + struct sa1111 *sachip = dev_get_drvdata(dev); struct sa1111_save_data *save; unsigned long flags, id; void __iomem *base; @@ -952,7 +952,7 @@ static int sa1111_resume(struct platform_device *dev) id = sa1111_readl(sachip->base + SA1111_SKID); if ((id & SKID_ID_MASK) != SKID_SA1111_ID) { __sa1111_remove(sachip); - platform_set_drvdata(dev, NULL); + dev_set_drvdata(dev, NULL); kfree(save); return 0; } @@ -1003,8 +1003,8 @@ static int sa1111_resume(struct platform_device *dev) } #else -#define sa1111_suspend NULL -#define sa1111_resume NULL +#define sa1111_suspend_noirq NULL +#define sa1111_resume_noirq NULL #endif static int sa1111_probe(struct platform_device *pdev) @@ -1038,6 +1038,11 @@ static int sa1111_remove(struct platform_device *pdev) return 0; } +static struct dev_pm_ops sa1111_pm_ops = { + .suspend_noirq = sa1111_suspend_noirq, + .resume_noirq = sa1111_resume_noirq, +}; + /* * Not sure if this should be on the system bus or not yet. * We really want some way to register a system device at @@ -1050,10 +1055,9 @@ static int sa1111_remove(struct platform_device *pdev) static struct platform_driver sa1111_device_driver = { .probe = sa1111_probe, .remove = sa1111_remove, - .suspend = sa1111_suspend, - .resume = sa1111_resume, .driver = { .name = "sa1111", + .pm = &sa1111_pm_ops, }, }; From 87d5dd62c07f90ed2b0d6718f5c666f69e7d39b0 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 6 Sep 2016 16:09:17 +0100 Subject: [PATCH 188/271] ARM: sa1111: fix missing clk_disable() SA1111 forgets to call clk_disable() in the probe error cleanup path. Add the necessary call. Signed-off-by: Russell King --- arch/arm/common/sa1111.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/arm/common/sa1111.c b/arch/arm/common/sa1111.c index 7838659b870a..2e076c492005 100644 --- a/arch/arm/common/sa1111.c +++ b/arch/arm/common/sa1111.c @@ -754,7 +754,7 @@ static int __sa1111_probe(struct device *me, struct resource *mem, int irq) if (sachip->irq != NO_IRQ) { ret = sa1111_setup_irq(sachip, pd->irq_base); if (ret) - goto err_unmap; + goto err_clk; } #ifdef CONFIG_ARCH_SA1100 @@ -799,6 +799,8 @@ static int __sa1111_probe(struct device *me, struct resource *mem, int irq) return 0; + err_clk: + clk_disable(sachip->clk); err_unmap: iounmap(sachip->base); err_clk_unprep: From 6b3142b2b852cd5e3216d1aa800a0a49377e6e1c Mon Sep 17 00:00:00 2001 From: Stefan Agner Date: Wed, 7 Sep 2016 21:56:09 +0100 Subject: [PATCH 189/271] ARM: 8612/1: LPAE: initialize cache policy correctly The cachepolicy variable gets initialized using a masked pmd value. So far, the pmd has been masked with flags valid for the 2-page table format, but the 3-page table format requires a different mask. On LPAE, this lead to a wrong assumption of what initial cache policy has been used. Later a check forces the cache policy to writealloc and prints the following warning: Forcing write-allocate cache policy for SMP This patch introduces a new definition PMD_SECT_CACHE_MASK for both page table formats which masks in all cache flags in both cases. Signed-off-by: Stefan Agner Signed-off-by: Russell King --- arch/arm/include/asm/pgtable-2level-hwdef.h | 1 + arch/arm/include/asm/pgtable-3level-hwdef.h | 1 + arch/arm/mm/mmu.c | 2 +- 3 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/arm/include/asm/pgtable-2level-hwdef.h b/arch/arm/include/asm/pgtable-2level-hwdef.h index d0131ee6f6af..3f82e9da7cec 100644 --- a/arch/arm/include/asm/pgtable-2level-hwdef.h +++ b/arch/arm/include/asm/pgtable-2level-hwdef.h @@ -47,6 +47,7 @@ #define PMD_SECT_WB (PMD_SECT_CACHEABLE | PMD_SECT_BUFFERABLE) #define PMD_SECT_MINICACHE (PMD_SECT_TEX(1) | PMD_SECT_CACHEABLE) #define PMD_SECT_WBWA (PMD_SECT_TEX(1) | PMD_SECT_CACHEABLE | PMD_SECT_BUFFERABLE) +#define PMD_SECT_CACHE_MASK (PMD_SECT_TEX(1) | PMD_SECT_CACHEABLE | PMD_SECT_BUFFERABLE) #define PMD_SECT_NONSHARED_DEV (PMD_SECT_TEX(2)) /* diff --git a/arch/arm/include/asm/pgtable-3level-hwdef.h b/arch/arm/include/asm/pgtable-3level-hwdef.h index f8f1cff62065..4cd664abfcd3 100644 --- a/arch/arm/include/asm/pgtable-3level-hwdef.h +++ b/arch/arm/include/asm/pgtable-3level-hwdef.h @@ -62,6 +62,7 @@ #define PMD_SECT_WT (_AT(pmdval_t, 2) << 2) /* normal inner write-through */ #define PMD_SECT_WB (_AT(pmdval_t, 3) << 2) /* normal inner write-back */ #define PMD_SECT_WBWA (_AT(pmdval_t, 7) << 2) /* normal inner write-alloc */ +#define PMD_SECT_CACHE_MASK (_AT(pmdval_t, 7) << 2) /* * + Level 3 descriptor (PTE) diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c index 6344913f0804..30fe03f95c85 100644 --- a/arch/arm/mm/mmu.c +++ b/arch/arm/mm/mmu.c @@ -137,7 +137,7 @@ void __init init_default_cache_policy(unsigned long pmd) initial_pmd_value = pmd; - pmd &= PMD_SECT_TEX(1) | PMD_SECT_BUFFERABLE | PMD_SECT_CACHEABLE; + pmd &= PMD_SECT_CACHE_MASK; for (i = 0; i < ARRAY_SIZE(cache_policies); i++) if (cache_policies[i].pmd == pmd) { From 07f56e6646228da27122e81d5a5a232fdf3d3a50 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 2 Sep 2016 22:08:45 +0100 Subject: [PATCH 190/271] ARM: locomo: fix locomo irq handling Accidentally booting Collie on Assabet reveals that the locomo driver incorrectly overwrites gpio-sa1100's chip data for its parent interrupt, leading to oops in sa1100_gpio_unmask() and sa1100_update_edge_regs() when "gpio: sa1100: convert to use IO accessors" is applied. Fix locomo to use the handler data rather than chip data for its parent interrupt. Signed-off-by: Russell King --- arch/arm/common/locomo.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/arch/arm/common/locomo.c b/arch/arm/common/locomo.c index 0e97b4b871f9..6c7b06854fce 100644 --- a/arch/arm/common/locomo.c +++ b/arch/arm/common/locomo.c @@ -140,7 +140,7 @@ static struct locomo_dev_info locomo_devices[] = { static void locomo_handler(struct irq_desc *desc) { - struct locomo *lchip = irq_desc_get_chip_data(desc); + struct locomo *lchip = irq_desc_get_handler_data(desc); int req, i; /* Acknowledge the parent IRQ */ @@ -200,8 +200,7 @@ static void locomo_setup_irq(struct locomo *lchip) * Install handler for IRQ_LOCOMO_HW. */ irq_set_irq_type(lchip->irq, IRQ_TYPE_EDGE_FALLING); - irq_set_chip_data(lchip->irq, lchip); - irq_set_chained_handler(lchip->irq, locomo_handler); + irq_set_chained_handler_and_data(lchip->irq, locomo_handler, lchip); /* Install handlers for IRQ_LOCOMO_* */ for ( ; irq <= lchip->irq_base + 3; irq++) { From 1a57c286d8ced1e4144c6201a19bbb70827edee6 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 4 Sep 2016 21:45:56 +0100 Subject: [PATCH 191/271] ARM: pxa/lubbock: add pcmcia clock Add the required PCMCIA clock for the SA1111 "1800" device. This clock is used to compute timing information for the PCMCIA interface in the SoC device, rather than the SA1111. Hence, the provision of this clock is a convenience for the driver and does not reflect the hardware, so this must not be copied into DT. Acked-by: Robert Jarzmik Tested-by: Robert Jarzmik Signed-off-by: Russell King --- arch/arm/mach-pxa/lubbock.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/arch/arm/mach-pxa/lubbock.c b/arch/arm/mach-pxa/lubbock.c index 7245f3359564..d6159f8ef0c2 100644 --- a/arch/arm/mach-pxa/lubbock.c +++ b/arch/arm/mach-pxa/lubbock.c @@ -137,6 +137,18 @@ static struct pxa2xx_udc_mach_info udc_info __initdata = { // no D+ pullup; lubbock can't connect/disconnect in software }; +static void lubbock_init_pcmcia(void) +{ + struct clk *clk; + + /* Add an alias for the SA1111 PCMCIA clock */ + clk = clk_get_sys("pxa2xx-pcmcia", NULL); + if (!IS_ERR(clk)) { + clkdev_create(clk, NULL, "1800"); + clk_put(clk); + } +} + static struct resource sa1111_resources[] = { [0] = { .start = 0x10000000, @@ -467,6 +479,8 @@ static void __init lubbock_init(void) pxa_set_btuart_info(NULL); pxa_set_stuart_info(NULL); + lubbock_init_pcmcia(); + clk_add_alias("SA1111_CLK", NULL, "GPIO11_CLK", NULL); pxa_set_udc_info(&udc_info); pxa_set_fb_info(NULL, &sharp_lm8v31); From e89ca58f9c901c8c4cfb09f96d879b186bb01492 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Fri, 2 Sep 2016 09:01:54 -0700 Subject: [PATCH 192/271] nvme-rdma: add DELETING queue flag When we get a surprise disconnect from the target we queue a periodic reconnect (which is the sane thing to do...). We only move the queues out of CONNECTED when we retry to reconnect (after 10 seconds in the default case) but we stop the blk queues immediately so we are not bothered with traffic from now on. If delete() is kicking off in this period the queues are still in CONNECTED state. Part of the delete sequence is trying to issue ctrl shutdown if the admin queue is CONNECTED (which it is!). This request is issued but stuck in blk-mq waiting for the queues to start again. This might be the one preventing us from forward progress... The patch separates the queue flags to CONNECTED and DELETING. Now we will move out of CONNECTED as soon as error recovery kicks in (before stopping the queues) and DELETING is on when we start the queue deletion. Signed-off-by: Sagi Grimberg --- drivers/nvme/host/rdma.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index a9d43f09963f..eeb08b658640 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -83,6 +83,7 @@ struct nvme_rdma_request { enum nvme_rdma_queue_flags { NVME_RDMA_Q_CONNECTED = (1 << 0), NVME_RDMA_IB_QUEUE_ALLOCATED = (1 << 1), + NVME_RDMA_Q_DELETING = (1 << 2), }; struct nvme_rdma_queue { @@ -559,6 +560,7 @@ static int nvme_rdma_init_queue(struct nvme_rdma_ctrl *ctrl, queue = &ctrl->queues[idx]; queue->ctrl = ctrl; + queue->flags = 0; init_completion(&queue->cm_done); if (idx > 0) @@ -616,7 +618,7 @@ static void nvme_rdma_free_queue(struct nvme_rdma_queue *queue) static void nvme_rdma_stop_and_free_queue(struct nvme_rdma_queue *queue) { - if (!test_and_clear_bit(NVME_RDMA_Q_CONNECTED, &queue->flags)) + if (test_and_set_bit(NVME_RDMA_Q_DELETING, &queue->flags)) return; nvme_rdma_stop_queue(queue); nvme_rdma_free_queue(queue); @@ -769,8 +771,13 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work) { struct nvme_rdma_ctrl *ctrl = container_of(work, struct nvme_rdma_ctrl, err_work); + int i; nvme_stop_keep_alive(&ctrl->ctrl); + + for (i = 0; i < ctrl->queue_count; i++) + clear_bit(NVME_RDMA_Q_CONNECTED, &ctrl->queues[i].flags); + if (ctrl->queue_count > 1) nvme_stop_queues(&ctrl->ctrl); blk_mq_stop_hw_queues(ctrl->ctrl.admin_q); @@ -1350,7 +1357,7 @@ static int nvme_rdma_device_unplug(struct nvme_rdma_queue *queue) cancel_delayed_work_sync(&ctrl->reconnect_work); /* Disable the queue so ctrl delete won't free it */ - if (test_and_clear_bit(NVME_RDMA_Q_CONNECTED, &queue->flags)) { + if (!test_and_set_bit(NVME_RDMA_Q_DELETING, &queue->flags)) { /* Free this queue ourselves */ nvme_rdma_stop_queue(queue); nvme_rdma_destroy_queue_ib(queue); From e87a911fed07e368c6f97e75152e6297a7dfba48 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Fri, 2 Sep 2016 09:01:54 -0700 Subject: [PATCH 193/271] nvme-rdma: use ib_client API to detect device removal Change nvme-rdma to use the IB Client API to detect device removal. This has the wonderful benefit of being able to blow away all the ib/rdma_cm resources for the device being removed. No craziness about not destroying the cm_id handling the event. No deadlocks due to broken iw_cm/rdma_cm/iwarp dependencies. And no need to have a bound cm_id around during controller recovery/reconnect to catch device removal events. We don't use the device_add aspect of the ib_client service since we only want to create resources for an IB device if we have a target utilizing that device. Reviewed-by: Christoph Hellwig Signed-off-by: Steve Wise Signed-off-by: Sagi Grimberg --- drivers/nvme/host/rdma.c | 108 +++++++++++++++------------------------ 1 file changed, 40 insertions(+), 68 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index eeb08b658640..d6bdf55a969e 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -1320,64 +1320,6 @@ out_destroy_queue_ib: return ret; } -/** - * nvme_rdma_device_unplug() - Handle RDMA device unplug - * @queue: Queue that owns the cm_id that caught the event - * - * DEVICE_REMOVAL event notifies us that the RDMA device is about - * to unplug so we should take care of destroying our RDMA resources. - * This event will be generated for each allocated cm_id. - * - * In our case, the RDMA resources are managed per controller and not - * only per queue. So the way we handle this is we trigger an implicit - * controller deletion upon the first DEVICE_REMOVAL event we see, and - * hold the event inflight until the controller deletion is completed. - * - * One exception that we need to handle is the destruction of the cm_id - * that caught the event. Since we hold the callout until the controller - * deletion is completed, we'll deadlock if the controller deletion will - * call rdma_destroy_id on this queue's cm_id. Thus, we claim ownership - * of destroying this queue before-hand, destroy the queue resources, - * then queue the controller deletion which won't destroy this queue and - * we destroy the cm_id implicitely by returning a non-zero rc to the callout. - */ -static int nvme_rdma_device_unplug(struct nvme_rdma_queue *queue) -{ - struct nvme_rdma_ctrl *ctrl = queue->ctrl; - int ret = 0; - - /* Own the controller deletion */ - if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_DELETING)) - return 0; - - dev_warn(ctrl->ctrl.device, - "Got rdma device removal event, deleting ctrl\n"); - - /* Get rid of reconnect work if its running */ - cancel_delayed_work_sync(&ctrl->reconnect_work); - - /* Disable the queue so ctrl delete won't free it */ - if (!test_and_set_bit(NVME_RDMA_Q_DELETING, &queue->flags)) { - /* Free this queue ourselves */ - nvme_rdma_stop_queue(queue); - nvme_rdma_destroy_queue_ib(queue); - - /* Return non-zero so the cm_id will destroy implicitly */ - ret = 1; - } - - /* - * Queue controller deletion. Keep a reference until all - * work is flushed since delete_work will free the ctrl mem - */ - kref_get(&ctrl->ctrl.kref); - queue_work(nvme_rdma_wq, &ctrl->delete_work); - flush_work(&ctrl->delete_work); - nvme_put_ctrl(&ctrl->ctrl); - - return ret; -} - static int nvme_rdma_cm_handler(struct rdma_cm_id *cm_id, struct rdma_cm_event *ev) { @@ -1419,8 +1361,8 @@ static int nvme_rdma_cm_handler(struct rdma_cm_id *cm_id, nvme_rdma_error_recovery(queue->ctrl); break; case RDMA_CM_EVENT_DEVICE_REMOVAL: - /* return 1 means impliciy CM ID destroy */ - return nvme_rdma_device_unplug(queue); + /* device removal is handled via the ib_client API */ + break; default: dev_err(queue->ctrl->ctrl.device, "Unexpected RDMA CM event (%d)\n", ev->event); @@ -2030,27 +1972,57 @@ static struct nvmf_transport_ops nvme_rdma_transport = { .create_ctrl = nvme_rdma_create_ctrl, }; +static void nvme_rdma_add_one(struct ib_device *ib_device) +{ +} + +static void nvme_rdma_remove_one(struct ib_device *ib_device, void *client_data) +{ + struct nvme_rdma_ctrl *ctrl; + + /* Delete all controllers using this device */ + mutex_lock(&nvme_rdma_ctrl_mutex); + list_for_each_entry(ctrl, &nvme_rdma_ctrl_list, list) { + if (ctrl->device->dev != ib_device) + continue; + dev_info(ctrl->ctrl.device, + "Removing ctrl: NQN \"%s\", addr %pISp\n", + ctrl->ctrl.opts->subsysnqn, &ctrl->addr); + __nvme_rdma_del_ctrl(ctrl); + } + mutex_unlock(&nvme_rdma_ctrl_mutex); + + flush_workqueue(nvme_rdma_wq); +} + +static struct ib_client nvme_rdma_ib_client = { + .name = "nvme_rdma", + .add = nvme_rdma_add_one, + .remove = nvme_rdma_remove_one +}; + static int __init nvme_rdma_init_module(void) { + int ret; + nvme_rdma_wq = create_workqueue("nvme_rdma_wq"); if (!nvme_rdma_wq) return -ENOMEM; + ret = ib_register_client(&nvme_rdma_ib_client); + if (ret) { + destroy_workqueue(nvme_rdma_wq); + return ret; + } + nvmf_register_transport(&nvme_rdma_transport); return 0; } static void __exit nvme_rdma_cleanup_module(void) { - struct nvme_rdma_ctrl *ctrl; - nvmf_unregister_transport(&nvme_rdma_transport); - - mutex_lock(&nvme_rdma_ctrl_mutex); - list_for_each_entry(ctrl, &nvme_rdma_ctrl_list, list) - __nvme_rdma_del_ctrl(ctrl); - mutex_unlock(&nvme_rdma_ctrl_mutex); - + ib_unregister_client(&nvme_rdma_ib_client); destroy_workqueue(nvme_rdma_wq); } From 1bda18de8f15a611a61d1a935b679db2e153338a Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Mon, 5 Sep 2016 16:24:38 +0100 Subject: [PATCH 194/271] nvme-rdma: fix null pointer dereference on req->mr If there is an error on req->mr, req->mr is set to null, however the following statement sets req->mr->need_inval causing a null pointer dereference. Fix this by bailing out to label 'out' to immediately return and hence skip over the offending null pointer dereference. Fixes: f5b7b559e1488 ("nvme-rdma: Get rid of duplicate variable") Signed-off-by: Colin Ian King Reviewed-by: Christoph Hellwig Signed-off-by: Sagi Grimberg --- drivers/nvme/host/rdma.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index d6bdf55a969e..c2c2c28e6eb5 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -293,6 +293,7 @@ static int nvme_rdma_reinit_request(void *data, struct request *rq) if (IS_ERR(req->mr)) { ret = PTR_ERR(req->mr); req->mr = NULL; + goto out; } req->mr->need_inval = false; From 2cfe199ca5a8816ee80fe15bcf202dd1020aaea0 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 6 Sep 2016 14:58:06 +0200 Subject: [PATCH 195/271] nvme-rdma: add back dependency on CONFIG_BLOCK A recent change removed the dependency on BLK_DEV_NVME, which implies the dependency on PCI and BLOCK. We don't need CONFIG_PCI, but without CONFIG_BLOCK we get tons of build errors, e.g. In file included from drivers/nvme/host/core.c:16:0: linux/blk-mq.h:182:33: error: 'struct gendisk' declared inside parameter list will not be visible outside of this definition or declaration [-Werror] drivers/nvme/host/core.c: In function 'nvme_setup_rw': drivers/nvme/host/core.c:295:21: error: implicit declaration of function 'rq_data_dir' [-Werror=implicit-function-declaration] drivers/nvme/host/nvme.h: In function 'nvme_map_len': drivers/nvme/host/nvme.h:217:6: error: implicit declaration of function 'req_op' [-Werror=implicit-function-declaration] drivers/nvme/host/scsi.c: In function 'nvme_trans_bdev_limits_page': drivers/nvme/host/scsi.c:768:85: error: implicit declaration of function 'queue_max_hw_sectors' [-Werror=implicit-function-declaration] This adds back the specific CONFIG_BLOCK dependency to avoid broken configurations. Signed-off-by: Arnd Bergmann Fixes: aa71987472a9 ("nvme: fabrics drivers don't need the nvme-pci driver") Signed-off-by: Sagi Grimberg --- drivers/nvme/host/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/nvme/host/Kconfig b/drivers/nvme/host/Kconfig index 0c644f7bdf80..4b6cfff4b5be 100644 --- a/drivers/nvme/host/Kconfig +++ b/drivers/nvme/host/Kconfig @@ -31,6 +31,7 @@ config NVME_FABRICS config NVME_RDMA tristate "NVM Express over Fabrics RDMA host driver" depends on INFINIBAND + depends on BLOCK select NVME_CORE select NVME_FABRICS select SG_POOL From bf2c4b6f9b74c2ee1dd3c050b181e9b9c86fbcdb Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 1 Sep 2016 10:50:38 -0400 Subject: [PATCH 196/271] svcauth_gss: Revert 64c59a3726f2 ("Remove unnecessary allocation") rsc_lookup steals the passed-in memory to avoid doing an allocation of its own, so we can't just pass in a pointer to memory that someone else is using. If we really want to avoid allocation there then maybe we should preallocate somwhere, or reference count these handles. For now we should revert. On occasion I see this on my server: kernel: kernel BUG at /home/cel/src/linux/linux-2.6/mm/slub.c:3851! kernel: invalid opcode: 0000 [#1] SMP kernel: Modules linked in: cts rpcsec_gss_krb5 sb_edac edac_core x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd btrfs xor iTCO_wdt iTCO_vendor_support raid6_pq pcspkr i2c_i801 i2c_smbus lpc_ich mfd_core mei_me sg mei shpchp wmi ioatdma ipmi_si ipmi_msghandler acpi_pad acpi_power_meter rpcrdma ib_ipoib rdma_ucm ib_ucm ib_uverbs ib_umad rdma_cm ib_cm iw_cm nfsd nfs_acl lockd grace auth_rpcgss sunrpc ip_tables xfs libcrc32c mlx4_ib mlx4_en ib_core sr_mod cdrom sd_mod ast drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm drm crc32c_intel igb mlx4_core ahci libahci libata ptp pps_core dca i2c_algo_bit i2c_core dm_mirror dm_region_hash dm_log dm_mod kernel: CPU: 7 PID: 145 Comm: kworker/7:2 Not tainted 4.8.0-rc4-00006-g9d06b0b #15 kernel: Hardware name: Supermicro Super Server/X10SRL-F, BIOS 1.0c 09/09/2015 kernel: Workqueue: events do_cache_clean [sunrpc] kernel: task: ffff8808541d8000 task.stack: ffff880854344000 kernel: RIP: 0010:[] [] kfree+0x155/0x180 kernel: RSP: 0018:ffff880854347d70 EFLAGS: 00010246 kernel: RAX: ffffea0020fe7660 RBX: ffff88083f9db064 RCX: 146ff0f9d5ec5600 kernel: RDX: 000077ff80000000 RSI: ffff880853f01500 RDI: ffff88083f9db064 kernel: RBP: ffff880854347d88 R08: ffff8808594ee000 R09: ffff88087fdd8780 kernel: R10: 0000000000000000 R11: ffffea0020fe76c0 R12: ffff880853f01500 kernel: R13: ffffffffa013cf76 R14: ffffffffa013cff0 R15: ffffffffa04253a0 kernel: FS: 0000000000000000(0000) GS:ffff88087fdc0000(0000) knlGS:0000000000000000 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 kernel: CR2: 00007fed60b020c3 CR3: 0000000001c06000 CR4: 00000000001406e0 kernel: Stack: kernel: ffff8808589f2f00 ffff880853f01500 0000000000000001 ffff880854347da0 kernel: ffffffffa013cf76 ffff8808589f2f00 ffff880854347db8 ffffffffa013d006 kernel: ffff8808589f2f20 ffff880854347e00 ffffffffa0406f60 0000000057c7044f kernel: Call Trace: kernel: [] rsc_free+0x16/0x90 [auth_rpcgss] kernel: [] rsc_put+0x16/0x30 [auth_rpcgss] kernel: [] cache_clean+0x2e0/0x300 [sunrpc] kernel: [] do_cache_clean+0xe/0x70 [sunrpc] kernel: [] process_one_work+0x1ff/0x3b0 kernel: [] worker_thread+0x2bc/0x4a0 kernel: [] ? rescuer_thread+0x3a0/0x3a0 kernel: [] kthread+0xe4/0xf0 kernel: [] ret_from_fork+0x1f/0x40 kernel: [] ? kthread_stop+0x110/0x110 kernel: Code: f7 ff ff eb 3b 65 8b 05 da 30 e2 7e 89 c0 48 0f a3 05 a0 38 b8 00 0f 92 c0 84 c0 0f 85 d1 fe ff ff 0f 1f 44 00 00 e9 f5 fe ff ff <0f> 0b 49 8b 03 31 f6 f6 c4 40 0f 85 62 ff ff ff e9 61 ff ff ff kernel: RIP [] kfree+0x155/0x180 kernel: RSP kernel: ---[ end trace 3fdec044969def26 ]--- It seems to be most common after a server reboot where a client has been using a Kerberos mount, and reconnects to continue its workload. Signed-off-by: Chuck Lever Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields --- net/sunrpc/auth_gss/svcauth_gss.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c index 1d281816f2bf..d8582028b346 100644 --- a/net/sunrpc/auth_gss/svcauth_gss.c +++ b/net/sunrpc/auth_gss/svcauth_gss.c @@ -569,9 +569,10 @@ gss_svc_searchbyctx(struct cache_detail *cd, struct xdr_netobj *handle) struct rsc *found; memset(&rsci, 0, sizeof(rsci)); - rsci.handle.data = handle->data; - rsci.handle.len = handle->len; + if (dup_to_netobj(&rsci.handle, handle->data, handle->len)) + return NULL; found = rsc_lookup(cd, &rsci); + rsc_free(&rsci); if (!found) return NULL; if (cache_check(cd, &found->h, NULL)) From ebf9ff753c041b296241990aef76163bbb2cc9c8 Mon Sep 17 00:00:00 2001 From: Boris Brezillon Date: Tue, 13 Sep 2016 15:58:28 +0200 Subject: [PATCH 197/271] genirq: Provide irq_gc_{lock_irqsave,unlock_irqrestore}() helpers Some irqchip drivers need to take the generic chip lock outside of the irq context. Provide the irq_gc_{lock_irqsave,unlock_irqrestore}() helpers to allow one to disable irqs while entering a critical section protected by gc->lock. Note that we do not provide optimized version of these helpers for !SMP, because they are not called from the hot-path. [ tglx: Added a comment when these helpers should be [not] used ] Signed-off-by: Boris Brezillon Cc: Jason Cooper Cc: Marc Zyngier Cc: Nicolas Ferre Cc: stable@vger.kernel.org Cc: Alexandre Belloni Link: http://lkml.kernel.org/r/1473775109-4192-1-git-send-email-boris.brezillon@free-electrons.com Signed-off-by: Thomas Gleixner --- include/linux/irq.h | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/include/linux/irq.h b/include/linux/irq.h index b52424eaa0ed..0ac26c892fe2 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -945,6 +945,16 @@ static inline void irq_gc_lock(struct irq_chip_generic *gc) { } static inline void irq_gc_unlock(struct irq_chip_generic *gc) { } #endif +/* + * The irqsave variants are for usage in non interrupt code. Do not use + * them in irq_chip callbacks. Use irq_gc_lock() instead. + */ +#define irq_gc_lock_irqsave(gc, flags) \ + raw_spin_lock_irqsave(&(gc)->lock, flags) + +#define irq_gc_unlock_irqrestore(gc, flags) \ + raw_spin_unlock_irqrestore(&(gc)->lock, flags) + static inline void irq_reg_writel(struct irq_chip_generic *gc, u32 val, int reg_offset) { From 5eb0d6eb3fac3daa60d9190eed9fa41cf809c756 Mon Sep 17 00:00:00 2001 From: Boris Brezillon Date: Tue, 13 Sep 2016 15:58:29 +0200 Subject: [PATCH 198/271] irqchip/atmel-aic: Fix potential deadlock in ->xlate() aic5_irq_domain_xlate() and aic_irq_domain_xlate() take the generic chip lock without disabling interrupts, which can lead to a deadlock if an interrupt occurs while the lock is held in one of these functions. Replace irq_gc_{lock,unlock}() calls by irq_gc_{lock_irqsave,unlock_irqrestore}() ones to prevent this bug from happening. Fixes: b1479ebb7720 ("irqchip: atmel-aic: Add atmel AIC/AIC5 drivers") Signed-off-by: Boris Brezillon Acked-by: Marc Zyngier Cc: Jason Cooper Cc: Nicolas Ferre Cc: stable@vger.kernel.org Cc: Alexandre Belloni Link: http://lkml.kernel.org/r/1473775109-4192-2-git-send-email-boris.brezillon@free-electrons.com Signed-off-by: Thomas Gleixner --- drivers/irqchip/irq-atmel-aic.c | 5 +++-- drivers/irqchip/irq-atmel-aic5.c | 5 +++-- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/irqchip/irq-atmel-aic.c b/drivers/irqchip/irq-atmel-aic.c index 112e17c2768b..37f952dd9fc9 100644 --- a/drivers/irqchip/irq-atmel-aic.c +++ b/drivers/irqchip/irq-atmel-aic.c @@ -176,6 +176,7 @@ static int aic_irq_domain_xlate(struct irq_domain *d, { struct irq_domain_chip_generic *dgc = d->gc; struct irq_chip_generic *gc; + unsigned long flags; unsigned smr; int idx; int ret; @@ -194,11 +195,11 @@ static int aic_irq_domain_xlate(struct irq_domain *d, gc = dgc->gc[idx]; - irq_gc_lock(gc); + irq_gc_lock_irqsave(gc, flags); smr = irq_reg_readl(gc, AT91_AIC_SMR(*out_hwirq)); aic_common_set_priority(intspec[2], &smr); irq_reg_writel(gc, smr, AT91_AIC_SMR(*out_hwirq)); - irq_gc_unlock(gc); + irq_gc_unlock_irqrestore(gc, flags); return ret; } diff --git a/drivers/irqchip/irq-atmel-aic5.c b/drivers/irqchip/irq-atmel-aic5.c index 4f0d068e1abe..2a624d87a035 100644 --- a/drivers/irqchip/irq-atmel-aic5.c +++ b/drivers/irqchip/irq-atmel-aic5.c @@ -258,6 +258,7 @@ static int aic5_irq_domain_xlate(struct irq_domain *d, unsigned int *out_type) { struct irq_chip_generic *bgc = irq_get_domain_generic_chip(d, 0); + unsigned long flags; unsigned smr; int ret; @@ -269,12 +270,12 @@ static int aic5_irq_domain_xlate(struct irq_domain *d, if (ret) return ret; - irq_gc_lock(bgc); + irq_gc_lock_irqsave(bgc, flags); irq_reg_writel(bgc, *out_hwirq, AT91_AIC5_SSR); smr = irq_reg_readl(bgc, AT91_AIC5_SMR); aic_common_set_priority(intspec[2], &smr); irq_reg_writel(bgc, smr, AT91_AIC5_SMR); - irq_gc_unlock(bgc); + irq_gc_unlock_irqrestore(bgc, flags); return ret; } From 035ee288ae7ade4152f1c3cf23a587b04fdc526c Mon Sep 17 00:00:00 2001 From: Lukas Wunner Date: Tue, 6 Sep 2016 06:20:46 +0200 Subject: [PATCH 199/271] PCI: Fix bridge_d3 update on device removal Starting with v4.8, we allow a PCIe port to runtime suspend to D3hot if the port itself and its children satisfy a number of conditions. Once a child is removed, we recheck those conditions in case the removed device was blocking the port from suspending. The rechecking needs to happen *after* the device has been removed from the bus it resides on. Otherwise when walking the port's subordinate bus in pci_bridge_d3_update(), the device being removed would erroneously still be taken into account. However the device is removed from the bus_list in pci_destroy_dev() and we currently recheck *before* that. Fix it. Fixes: 9d26d3a8f1b0 ("PCI: Put PCIe ports into D3 during suspend") Signed-off-by: Lukas Wunner Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg Acked-by: Rafael J. Wysocki --- drivers/pci/remove.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c index d1ef7acf6930..f9357e09e9b3 100644 --- a/drivers/pci/remove.c +++ b/drivers/pci/remove.c @@ -40,6 +40,7 @@ static void pci_destroy_dev(struct pci_dev *dev) list_del(&dev->bus_list); up_write(&pci_bus_sem); + pci_bridge_d3_device_removed(dev); pci_free_resources(dev); put_device(&dev->dev); } @@ -96,8 +97,6 @@ static void pci_remove_bus_device(struct pci_dev *dev) dev->subordinate = NULL; } - pci_bridge_d3_device_removed(dev); - pci_destroy_dev(dev); } From 9ad18b75c2f6e4a78ce204e79f37781f8815c0fa Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 17 Aug 2016 23:19:01 -0400 Subject: [PATCH 200/271] asm-generic: make get_user() clear the destination on errors both for access_ok() failures and for faults halfway through Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- include/asm-generic/uaccess.h | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h index 04e21a41796a..32901d11f8c4 100644 --- a/include/asm-generic/uaccess.h +++ b/include/asm-generic/uaccess.h @@ -230,14 +230,18 @@ extern int __put_user_bad(void) __attribute__((noreturn)); might_fault(); \ access_ok(VERIFY_READ, __p, sizeof(*ptr)) ? \ __get_user((x), (__typeof__(*(ptr)) *)__p) : \ - -EFAULT; \ + ((x) = (__typeof__(*(ptr)))0,-EFAULT); \ }) #ifndef __get_user_fn static inline int __get_user_fn(size_t size, const void __user *ptr, void *x) { - size = __copy_from_user(x, ptr, size); - return size ? -EFAULT : size; + size_t n = __copy_from_user(x, ptr, size); + if (unlikely(n)) { + memset(x + (size - n), 0, n); + return -EFAULT; + } + return 0; } #define __get_user_fn(sz, u, k) __get_user_fn(sz, u, k) From eb47e0293baaa3044022059f1fa9ff474bfe35cb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 18 Aug 2016 19:34:00 -0400 Subject: [PATCH 201/271] cris: buggered copy_from_user/copy_to_user/clear_user * copy_from_user() on access_ok() failure ought to zero the destination * none of those primitives should skip the access_ok() check in case of small constant size. Cc: stable@vger.kernel.org Acked-by: Jesper Nilsson Signed-off-by: Al Viro --- arch/cris/include/asm/uaccess.h | 71 +++++++++++++++------------------ 1 file changed, 32 insertions(+), 39 deletions(-) diff --git a/arch/cris/include/asm/uaccess.h b/arch/cris/include/asm/uaccess.h index e3530d0f13ee..56c7d5750abd 100644 --- a/arch/cris/include/asm/uaccess.h +++ b/arch/cris/include/asm/uaccess.h @@ -194,30 +194,6 @@ extern unsigned long __copy_user(void __user *to, const void *from, unsigned lon extern unsigned long __copy_user_zeroing(void *to, const void __user *from, unsigned long n); extern unsigned long __do_clear_user(void __user *to, unsigned long n); -static inline unsigned long -__generic_copy_to_user(void __user *to, const void *from, unsigned long n) -{ - if (access_ok(VERIFY_WRITE, to, n)) - return __copy_user(to, from, n); - return n; -} - -static inline unsigned long -__generic_copy_from_user(void *to, const void __user *from, unsigned long n) -{ - if (access_ok(VERIFY_READ, from, n)) - return __copy_user_zeroing(to, from, n); - return n; -} - -static inline unsigned long -__generic_clear_user(void __user *to, unsigned long n) -{ - if (access_ok(VERIFY_WRITE, to, n)) - return __do_clear_user(to, n); - return n; -} - static inline long __strncpy_from_user(char *dst, const char __user *src, long count) { @@ -282,7 +258,7 @@ __constant_copy_from_user(void *to, const void __user *from, unsigned long n) else if (n == 24) __asm_copy_from_user_24(to, from, ret); else - ret = __generic_copy_from_user(to, from, n); + ret = __copy_user_zeroing(to, from, n); return ret; } @@ -333,7 +309,7 @@ __constant_copy_to_user(void __user *to, const void *from, unsigned long n) else if (n == 24) __asm_copy_to_user_24(to, from, ret); else - ret = __generic_copy_to_user(to, from, n); + ret = __copy_user(to, from, n); return ret; } @@ -366,26 +342,43 @@ __constant_clear_user(void __user *to, unsigned long n) else if (n == 24) __asm_clear_24(to, ret); else - ret = __generic_clear_user(to, n); + ret = __do_clear_user(to, n); return ret; } -#define clear_user(to, n) \ - (__builtin_constant_p(n) ? \ - __constant_clear_user(to, n) : \ - __generic_clear_user(to, n)) +static inline size_t clear_user(void __user *to, size_t n) +{ + if (unlikely(!access_ok(VERIFY_WRITE, to, n))) + return n; + if (__builtin_constant_p(n)) + return __constant_clear_user(to, n); + else + return __do_clear_user(to, n); +} -#define copy_from_user(to, from, n) \ - (__builtin_constant_p(n) ? \ - __constant_copy_from_user(to, from, n) : \ - __generic_copy_from_user(to, from, n)) +static inline size_t copy_from_user(void *to, const void __user *from, size_t n) +{ + if (unlikely(!access_ok(VERIFY_READ, from, n))) { + memset(to, 0, n); + return n; + } + if (__builtin_constant_p(n)) + return __constant_copy_from_user(to, from, n); + else + return __copy_user_zeroing(to, from, n); +} -#define copy_to_user(to, from, n) \ - (__builtin_constant_p(n) ? \ - __constant_copy_to_user(to, from, n) : \ - __generic_copy_to_user(to, from, n)) +static inline size_t copy_to_user(void __user *to, const void *from, size_t n) +{ + if (unlikely(!access_ok(VERIFY_WRITE, to, n))) + return n; + if (__builtin_constant_p(n)) + return __constant_copy_to_user(to, from, n); + else + return __copy_user(to, from, n); +} /* We let the __ versions of copy_from/to_user inline, because they're often * used in fast paths and have only a small space overhead. From 3b8767a8f00cc6538ba6b1cf0f88502e2fd2eb90 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 18 Aug 2016 20:54:02 -0400 Subject: [PATCH 202/271] frv: fix clear_user() It should check access_ok(). Otherwise a bunch of places turn into trivially exploitable rootholes. Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/frv/include/asm/uaccess.h | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/arch/frv/include/asm/uaccess.h b/arch/frv/include/asm/uaccess.h index 3ac9a59d65d4..87d9e34c5df8 100644 --- a/arch/frv/include/asm/uaccess.h +++ b/arch/frv/include/asm/uaccess.h @@ -263,19 +263,25 @@ do { \ extern long __memset_user(void *dst, unsigned long count); extern long __memcpy_user(void *dst, const void *src, unsigned long count); -#define clear_user(dst,count) __memset_user(____force(dst), (count)) +#define __clear_user(dst,count) __memset_user(____force(dst), (count)) #define __copy_from_user_inatomic(to, from, n) __memcpy_user((to), ____force(from), (n)) #define __copy_to_user_inatomic(to, from, n) __memcpy_user(____force(to), (from), (n)) #else -#define clear_user(dst,count) (memset(____force(dst), 0, (count)), 0) +#define __clear_user(dst,count) (memset(____force(dst), 0, (count)), 0) #define __copy_from_user_inatomic(to, from, n) (memcpy((to), ____force(from), (n)), 0) #define __copy_to_user_inatomic(to, from, n) (memcpy(____force(to), (from), (n)), 0) #endif -#define __clear_user clear_user +static inline unsigned long __must_check +clear_user(void __user *to, unsigned long n) +{ + if (likely(__access_ok(to, n))) + n = __clear_user(to, n); + return n; +} static inline unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n) From f35c1e0671728d1c9abc405d05ef548b5fcb2fc4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 18 Aug 2016 21:16:49 -0400 Subject: [PATCH 203/271] hexagon: fix strncpy_from_user() error return It's -EFAULT, not -1 (and contrary to the comment in there, __strnlen_user() can return 0 - on faults). Cc: stable@vger.kernel.org Acked-by: Richard Kuo Signed-off-by: Al Viro --- arch/hexagon/include/asm/uaccess.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/hexagon/include/asm/uaccess.h b/arch/hexagon/include/asm/uaccess.h index f000a382bc7f..f61cfb28e9f2 100644 --- a/arch/hexagon/include/asm/uaccess.h +++ b/arch/hexagon/include/asm/uaccess.h @@ -103,7 +103,8 @@ static inline long hexagon_strncpy_from_user(char *dst, const char __user *src, { long res = __strnlen_user(src, n); - /* return from strnlen can't be zero -- that would be rubbish. */ + if (unlikely(!res)) + return -EFAULT; if (res > n) { copy_from_user(dst, src, n); From a5e541f796f17228793694d64b507f5f57db4cd7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 18 Aug 2016 21:31:41 -0400 Subject: [PATCH 204/271] ia64: copy_from_user() should zero the destination on access_ok() failure Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/ia64/include/asm/uaccess.h | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) diff --git a/arch/ia64/include/asm/uaccess.h b/arch/ia64/include/asm/uaccess.h index 465c70982f40..6c2d2c8c24e8 100644 --- a/arch/ia64/include/asm/uaccess.h +++ b/arch/ia64/include/asm/uaccess.h @@ -272,20 +272,17 @@ __copy_from_user (void *to, const void __user *from, unsigned long count) __cu_len; \ }) -#define copy_from_user(to, from, n) \ -({ \ - void *__cu_to = (to); \ - const void __user *__cu_from = (from); \ - long __cu_len = (n); \ - \ - __chk_user_ptr(__cu_from); \ - if (__access_ok(__cu_from, __cu_len, get_fs())) { \ - if (!__builtin_constant_p(n)) \ - check_object_size(__cu_to, __cu_len, false); \ - __cu_len = __copy_user((__force void __user *) __cu_to, __cu_from, __cu_len); \ - } \ - __cu_len; \ -}) +static inline unsigned long +copy_from_user(void *to, const void __user *from, unsigned long n) +{ + if (!__builtin_constant_p(n)) + check_object_size(to, n, false); + if (likely(__access_ok(from, n, get_fs()))) + n = __copy_user((__force void __user *) to, from, n); + else + memset(to, 0, n); + return n; +} #define __copy_in_user(to, from, size) __copy_user((to), (from), (size)) From 8ae95ed4ae5fc7c3391ed668b2014c9e2079533b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 18 Aug 2016 22:08:20 -0400 Subject: [PATCH 205/271] metag: copy_from_user() should zero the destination on access_ok() failure Cc: stable@vger.kernel.org Acked-by: James Hogan Signed-off-by: Al Viro --- arch/metag/include/asm/uaccess.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/metag/include/asm/uaccess.h b/arch/metag/include/asm/uaccess.h index 8282cbce7e39..273e61225c27 100644 --- a/arch/metag/include/asm/uaccess.h +++ b/arch/metag/include/asm/uaccess.h @@ -204,8 +204,9 @@ extern unsigned long __must_check __copy_user_zeroing(void *to, static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) { - if (access_ok(VERIFY_READ, from, n)) + if (likely(access_ok(VERIFY_READ, from, n))) return __copy_user_zeroing(to, from, n); + memset(to, 0, n); return n; } From 05d9d0b96e53c52a113fd783c0c97c830c8dc7af Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Fri, 19 Aug 2016 12:10:02 -0700 Subject: [PATCH 206/271] ARC: uaccess: get_user to zero out dest in cause of fault Al reported potential issue with ARC get_user() as it wasn't clearing out destination pointer in case of fault due to bad address etc. Verified using following | { | u32 bogus1 = 0xdeadbeef; | u64 bogus2 = 0xdead; | int rc1, rc2; | | pr_info("Orig values %x %llx\n", bogus1, bogus2); | rc1 = get_user(bogus1, (u32 __user *)0x40000000); | rc2 = get_user(bogus2, (u64 __user *)0x50000000); | pr_info("access %d %d, new values %x %llx\n", | rc1, rc2, bogus1, bogus2); | } | [ARCLinux]# insmod /mnt/kernel-module/qtn.ko | Orig values deadbeef dead | access -14 -14, new values 0 0 Reported-by: Al Viro Cc: Linus Torvalds Cc: linux-snps-arc@lists.infradead.org Cc: linux-kernel@vger.kernel.org Cc: stable@vger.kernel.org Signed-off-by: Vineet Gupta Signed-off-by: Al Viro --- arch/arc/include/asm/uaccess.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/arch/arc/include/asm/uaccess.h b/arch/arc/include/asm/uaccess.h index a78d5670884f..41faf17cd28d 100644 --- a/arch/arc/include/asm/uaccess.h +++ b/arch/arc/include/asm/uaccess.h @@ -83,7 +83,10 @@ "2: ;nop\n" \ " .section .fixup, \"ax\"\n" \ " .align 4\n" \ - "3: mov %0, %3\n" \ + "3: # return -EFAULT\n" \ + " mov %0, %3\n" \ + " # zero out dst ptr\n" \ + " mov %1, 0\n" \ " j 2b\n" \ " .previous\n" \ " .section __ex_table, \"a\"\n" \ @@ -101,7 +104,11 @@ "2: ;nop\n" \ " .section .fixup, \"ax\"\n" \ " .align 4\n" \ - "3: mov %0, %3\n" \ + "3: # return -EFAULT\n" \ + " mov %0, %3\n" \ + " # zero out dst ptr\n" \ + " mov %1, 0\n" \ + " mov %R1, 0\n" \ " j 2b\n" \ " .previous\n" \ " .section __ex_table, \"a\"\n" \ From e69d700535ac43a18032b3c399c69bf4639e89a2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Aug 2016 16:18:53 -0400 Subject: [PATCH 207/271] mips: copy_from_user() must zero the destination on access_ok() failure Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/mips/include/asm/uaccess.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h index 11b965f98d95..21a2aaba20d5 100644 --- a/arch/mips/include/asm/uaccess.h +++ b/arch/mips/include/asm/uaccess.h @@ -14,6 +14,7 @@ #include #include #include +#include #include /* @@ -1170,6 +1171,8 @@ extern size_t __copy_in_user_eva(void *__to, const void *__from, size_t __n); __cu_len = __invoke_copy_from_user(__cu_to, \ __cu_from, \ __cu_len); \ + } else { \ + memset(__cu_to, 0, __cu_len); \ } \ } \ __cu_len; \ From 43403eabf558d2800b429cd886e996fd555aa542 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Aug 2016 16:32:02 -0400 Subject: [PATCH 208/271] mn10300: failing __get_user() and get_user() should zero Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/mn10300/include/asm/uaccess.h | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mn10300/include/asm/uaccess.h b/arch/mn10300/include/asm/uaccess.h index 20f7bf6de384..d012e877a95a 100644 --- a/arch/mn10300/include/asm/uaccess.h +++ b/arch/mn10300/include/asm/uaccess.h @@ -166,6 +166,7 @@ struct __large_struct { unsigned long buf[100]; }; "2:\n" \ " .section .fixup,\"ax\"\n" \ "3:\n\t" \ + " mov 0,%1\n" \ " mov %3,%0\n" \ " jmp 2b\n" \ " .previous\n" \ From ae7cc577ec2a4a6151c9e928fd1f595d953ecef1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Aug 2016 16:33:10 -0400 Subject: [PATCH 209/271] mn10300: copy_from_user() should zero on access_ok() failure... Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/mn10300/lib/usercopy.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/mn10300/lib/usercopy.c b/arch/mn10300/lib/usercopy.c index 7826e6c364e7..ce8899e5e171 100644 --- a/arch/mn10300/lib/usercopy.c +++ b/arch/mn10300/lib/usercopy.c @@ -9,7 +9,7 @@ * as published by the Free Software Foundation; either version * 2 of the Licence, or (at your option) any later version. */ -#include +#include unsigned long __generic_copy_to_user(void *to, const void *from, unsigned long n) @@ -24,6 +24,8 @@ __generic_copy_from_user(void *to, const void *from, unsigned long n) { if (access_ok(VERIFY_READ, from, n)) __copy_user_zeroing(to, from, n); + else + memset(to, 0, n); return n; } From e33d1f6f72cc82fcfc3d1fb20c9e3ad83b1928fa Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Aug 2016 16:36:36 -0400 Subject: [PATCH 210/271] nios2: copy_from_user() should zero the tail of destination Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/nios2/include/asm/uaccess.h | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h index caa51ff85a3c..2b4b9e919675 100644 --- a/arch/nios2/include/asm/uaccess.h +++ b/arch/nios2/include/asm/uaccess.h @@ -102,9 +102,12 @@ extern long __copy_to_user(void __user *to, const void *from, unsigned long n); static inline long copy_from_user(void *to, const void __user *from, unsigned long n) { - if (!access_ok(VERIFY_READ, from, n)) - return n; - return __copy_from_user(to, from, n); + unsigned long res = n; + if (access_ok(VERIFY_READ, from, n)) + res = __copy_from_user(to, from, n); + if (unlikely(res)) + memset(to + (n - res), 0, res); + return res; } static inline long copy_to_user(void __user *to, const void *from, From 2e29f50ad5e23db37dde9be71410d95d50241ecd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Aug 2016 16:39:01 -0400 Subject: [PATCH 211/271] nios2: fix __get_user() a) should not leave crap on fault b) should _not_ require access_ok() in any cases. Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/nios2/include/asm/uaccess.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h index 2b4b9e919675..0ab82324c817 100644 --- a/arch/nios2/include/asm/uaccess.h +++ b/arch/nios2/include/asm/uaccess.h @@ -142,7 +142,7 @@ extern long strnlen_user(const char __user *s, long n); #define __get_user_unknown(val, size, ptr, err) do { \ err = 0; \ - if (copy_from_user(&(val), ptr, size)) { \ + if (__copy_from_user(&(val), ptr, size)) { \ err = -EFAULT; \ } \ } while (0) @@ -169,7 +169,7 @@ do { \ ({ \ long __gu_err = -EFAULT; \ const __typeof__(*(ptr)) __user *__gu_ptr = (ptr); \ - unsigned long __gu_val; \ + unsigned long __gu_val = 0; \ __get_user_common(__gu_val, sizeof(*(ptr)), __gu_ptr, __gu_err);\ (x) = (__force __typeof__(x))__gu_val; \ __gu_err; \ From acb2505d0119033a80c85ac8d02dccae41271667 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Aug 2016 17:05:21 -0400 Subject: [PATCH 212/271] openrisc: fix copy_from_user() ... that should zero on faults. Also remove the helpful logics wrt range truncation copied from ppc32. Where it had ever been needed only in case of copy_from_user() *and* had not been merged into the mainline until a month after the need had disappeared. A decade before openrisc went into mainline, I might add... Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/openrisc/include/asm/uaccess.h | 33 +++++++++-------------------- 1 file changed, 10 insertions(+), 23 deletions(-) diff --git a/arch/openrisc/include/asm/uaccess.h b/arch/openrisc/include/asm/uaccess.h index a6bd07ca3d6c..cbad29b5a131 100644 --- a/arch/openrisc/include/asm/uaccess.h +++ b/arch/openrisc/include/asm/uaccess.h @@ -273,28 +273,20 @@ __copy_tofrom_user(void *to, const void *from, unsigned long size); static inline unsigned long copy_from_user(void *to, const void *from, unsigned long n) { - unsigned long over; + unsigned long res = n; - if (access_ok(VERIFY_READ, from, n)) - return __copy_tofrom_user(to, from, n); - if ((unsigned long)from < TASK_SIZE) { - over = (unsigned long)from + n - TASK_SIZE; - return __copy_tofrom_user(to, from, n - over) + over; - } - return n; + if (likely(access_ok(VERIFY_READ, from, n))) + n = __copy_tofrom_user(to, from, n); + if (unlikely(res)) + memset(to + (n - res), 0, res); + return res; } static inline unsigned long copy_to_user(void *to, const void *from, unsigned long n) { - unsigned long over; - - if (access_ok(VERIFY_WRITE, to, n)) - return __copy_tofrom_user(to, from, n); - if ((unsigned long)to < TASK_SIZE) { - over = (unsigned long)to + n - TASK_SIZE; - return __copy_tofrom_user(to, from, n - over) + over; - } + if (likely(access_ok(VERIFY_WRITE, to, n))) + n = __copy_tofrom_user(to, from, n); return n; } @@ -303,13 +295,8 @@ extern unsigned long __clear_user(void *addr, unsigned long size); static inline __must_check unsigned long clear_user(void *addr, unsigned long size) { - - if (access_ok(VERIFY_WRITE, addr, size)) - return __clear_user(addr, size); - if ((unsigned long)addr < TASK_SIZE) { - unsigned long over = (unsigned long)addr + size - TASK_SIZE; - return __clear_user(addr, size - over) + over; - } + if (likely(access_ok(VERIFY_WRITE, addr, size))) + size = __clear_user(addr, size); return size; } From aace880feea38875fbc919761b77e5732a3659ef Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Aug 2016 19:03:37 -0400 Subject: [PATCH 213/271] parisc: fix copy_from_user() Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/parisc/include/asm/uaccess.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/parisc/include/asm/uaccess.h b/arch/parisc/include/asm/uaccess.h index 0f59fd9ca205..37a1bee96444 100644 --- a/arch/parisc/include/asm/uaccess.h +++ b/arch/parisc/include/asm/uaccess.h @@ -10,6 +10,7 @@ #include #include +#include #define VERIFY_READ 0 #define VERIFY_WRITE 1 @@ -221,13 +222,14 @@ static inline unsigned long __must_check copy_from_user(void *to, unsigned long n) { int sz = __compiletime_object_size(to); - int ret = -EFAULT; + unsigned long ret = n; if (likely(sz == -1 || !__builtin_constant_p(n) || sz >= n)) ret = __copy_from_user(to, from, n); else copy_from_user_overflow(); - + if (unlikely(ret)) + memset(to + (n - ret), 0, ret); return ret; } From 224264657b8b228f949b42346e09ed8c90136a8e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 21 Aug 2016 19:16:26 -0400 Subject: [PATCH 214/271] ppc32: fix copy_from_user() should clear on access_ok() failures. Also remove the useless range truncation logics. Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/powerpc/include/asm/uaccess.h | 25 ++----------------------- 1 file changed, 2 insertions(+), 23 deletions(-) diff --git a/arch/powerpc/include/asm/uaccess.h b/arch/powerpc/include/asm/uaccess.h index c1dc6c14deb8..c2ce5dd33673 100644 --- a/arch/powerpc/include/asm/uaccess.h +++ b/arch/powerpc/include/asm/uaccess.h @@ -308,40 +308,23 @@ extern unsigned long __copy_tofrom_user(void __user *to, static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) { - unsigned long over; - - if (access_ok(VERIFY_READ, from, n)) { + if (likely(access_ok(VERIFY_READ, from, n))) { if (!__builtin_constant_p(n)) check_object_size(to, n, false); return __copy_tofrom_user((__force void __user *)to, from, n); } - if ((unsigned long)from < TASK_SIZE) { - over = (unsigned long)from + n - TASK_SIZE; - if (!__builtin_constant_p(n - over)) - check_object_size(to, n - over, false); - return __copy_tofrom_user((__force void __user *)to, from, - n - over) + over; - } + memset(to, 0, n); return n; } static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) { - unsigned long over; - if (access_ok(VERIFY_WRITE, to, n)) { if (!__builtin_constant_p(n)) check_object_size(from, n, true); return __copy_tofrom_user(to, (__force void __user *)from, n); } - if ((unsigned long)to < TASK_SIZE) { - over = (unsigned long)to + n - TASK_SIZE; - if (!__builtin_constant_p(n)) - check_object_size(from, n - over, true); - return __copy_tofrom_user(to, (__force void __user *)from, - n - over) + over; - } return n; } @@ -439,10 +422,6 @@ static inline unsigned long clear_user(void __user *addr, unsigned long size) might_fault(); if (likely(access_ok(VERIFY_WRITE, addr, size))) return __clear_user(addr, size); - if ((unsigned long)addr < TASK_SIZE) { - unsigned long over = (unsigned long)addr + size - TASK_SIZE; - return __clear_user(addr, size - over) + over; - } return size; } From fd2d2b191fe75825c4c7a6f12f3fef35aaed7dd7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 21 Aug 2016 22:00:54 -0400 Subject: [PATCH 215/271] s390: get_user() should zero on failure Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/s390/include/asm/uaccess.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h index 9b49cf1daa8f..2c5d292ff752 100644 --- a/arch/s390/include/asm/uaccess.h +++ b/arch/s390/include/asm/uaccess.h @@ -266,28 +266,28 @@ int __put_user_bad(void) __attribute__((noreturn)); __chk_user_ptr(ptr); \ switch (sizeof(*(ptr))) { \ case 1: { \ - unsigned char __x; \ + unsigned char __x = 0; \ __gu_err = __get_user_fn(&__x, ptr, \ sizeof(*(ptr))); \ (x) = *(__force __typeof__(*(ptr)) *) &__x; \ break; \ }; \ case 2: { \ - unsigned short __x; \ + unsigned short __x = 0; \ __gu_err = __get_user_fn(&__x, ptr, \ sizeof(*(ptr))); \ (x) = *(__force __typeof__(*(ptr)) *) &__x; \ break; \ }; \ case 4: { \ - unsigned int __x; \ + unsigned int __x = 0; \ __gu_err = __get_user_fn(&__x, ptr, \ sizeof(*(ptr))); \ (x) = *(__force __typeof__(*(ptr)) *) &__x; \ break; \ }; \ case 8: { \ - unsigned long long __x; \ + unsigned long long __x = 0; \ __gu_err = __get_user_fn(&__x, ptr, \ sizeof(*(ptr))); \ (x) = *(__force __typeof__(*(ptr)) *) &__x; \ From c2f18fa4cbb3ad92e033a24efa27583978ce9600 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 21 Aug 2016 22:13:39 -0400 Subject: [PATCH 216/271] score: fix __get_user/get_user * should zero on any failure * __get_user() should use __copy_from_user(), not copy_from_user() Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/score/include/asm/uaccess.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/score/include/asm/uaccess.h b/arch/score/include/asm/uaccess.h index 20a3591225cc..c5d43111a5cc 100644 --- a/arch/score/include/asm/uaccess.h +++ b/arch/score/include/asm/uaccess.h @@ -163,7 +163,7 @@ do { \ __get_user_asm(val, "lw", ptr); \ break; \ case 8: \ - if ((copy_from_user((void *)&val, ptr, 8)) == 0) \ + if (__copy_from_user((void *)&val, ptr, 8) == 0) \ __gu_err = 0; \ else \ __gu_err = -EFAULT; \ @@ -188,6 +188,8 @@ do { \ \ if (likely(access_ok(VERIFY_READ, __gu_ptr, size))) \ __get_user_common((x), size, __gu_ptr); \ + else \ + (x) = 0; \ \ __gu_err; \ }) @@ -201,6 +203,7 @@ do { \ "2:\n" \ ".section .fixup,\"ax\"\n" \ "3:li %0, %4\n" \ + "li %1, 0\n" \ "j 2b\n" \ ".previous\n" \ ".section __ex_table,\"a\"\n" \ From b615e3c74621e06cd97f86373ca90d43d6d998aa Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 21 Aug 2016 22:30:44 -0400 Subject: [PATCH 217/271] score: fix copy_from_user() and friends Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/score/include/asm/uaccess.h | 41 ++++++++++++++++---------------- 1 file changed, 20 insertions(+), 21 deletions(-) diff --git a/arch/score/include/asm/uaccess.h b/arch/score/include/asm/uaccess.h index c5d43111a5cc..01aec8ccde83 100644 --- a/arch/score/include/asm/uaccess.h +++ b/arch/score/include/asm/uaccess.h @@ -301,35 +301,34 @@ extern int __copy_tofrom_user(void *to, const void *from, unsigned long len); static inline unsigned long copy_from_user(void *to, const void *from, unsigned long len) { - unsigned long over; + unsigned long res = len; - if (access_ok(VERIFY_READ, from, len)) - return __copy_tofrom_user(to, from, len); + if (likely(access_ok(VERIFY_READ, from, len))) + res = __copy_tofrom_user(to, from, len); - if ((unsigned long)from < TASK_SIZE) { - over = (unsigned long)from + len - TASK_SIZE; - return __copy_tofrom_user(to, from, len - over) + over; - } - return len; + if (unlikely(res)) + memset(to + (len - res), 0, res); + + return res; } static inline unsigned long copy_to_user(void *to, const void *from, unsigned long len) { - unsigned long over; + if (likely(access_ok(VERIFY_WRITE, to, len))) + len = __copy_tofrom_user(to, from, len); - if (access_ok(VERIFY_WRITE, to, len)) - return __copy_tofrom_user(to, from, len); - - if ((unsigned long)to < TASK_SIZE) { - over = (unsigned long)to + len - TASK_SIZE; - return __copy_tofrom_user(to, from, len - over) + over; - } return len; } -#define __copy_from_user(to, from, len) \ - __copy_tofrom_user((to), (from), (len)) +static inline unsigned long +__copy_from_user(void *to, const void *from, unsigned long len) +{ + unsigned long left = __copy_tofrom_user(to, from, len); + if (unlikely(left)) + memset(to + (len - left), 0, left); + return left; +} #define __copy_to_user(to, from, len) \ __copy_tofrom_user((to), (from), (len)) @@ -343,17 +342,17 @@ __copy_to_user_inatomic(void *to, const void *from, unsigned long len) static inline unsigned long __copy_from_user_inatomic(void *to, const void *from, unsigned long len) { - return __copy_from_user(to, from, len); + return __copy_tofrom_user(to, from, len); } -#define __copy_in_user(to, from, len) __copy_from_user(to, from, len) +#define __copy_in_user(to, from, len) __copy_tofrom_user(to, from, len) static inline unsigned long copy_in_user(void *to, const void *from, unsigned long len) { if (access_ok(VERIFY_READ, from, len) && access_ok(VERFITY_WRITE, to, len)) - return copy_from_user(to, from, len); + return __copy_tofrom_user(to, from, len); } /* From c6852389228df9fb3067f94f3b651de2a7921b36 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 21 Aug 2016 23:33:47 -0400 Subject: [PATCH 218/271] sh64: failing __get_user() should zero It could be done in exception-handling bits in __get_user_b() et.al., but the surgery involved would take more knowledge of sh64 details than I have or _want_ to have. Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/sh/include/asm/uaccess_64.h | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/sh/include/asm/uaccess_64.h b/arch/sh/include/asm/uaccess_64.h index c01376c76b86..ca5073dd4596 100644 --- a/arch/sh/include/asm/uaccess_64.h +++ b/arch/sh/include/asm/uaccess_64.h @@ -24,6 +24,7 @@ #define __get_user_size(x,ptr,size,retval) \ do { \ retval = 0; \ + x = 0; \ switch (size) { \ case 1: \ retval = __get_user_asm_b((void *)&x, \ From 6e050503a150b2126620c1a1e9b3a368fcd51eac Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 21 Aug 2016 23:39:47 -0400 Subject: [PATCH 219/271] sh: fix copy_from_user() Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/sh/include/asm/uaccess.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/sh/include/asm/uaccess.h b/arch/sh/include/asm/uaccess.h index a49635c51266..92ade79ac427 100644 --- a/arch/sh/include/asm/uaccess.h +++ b/arch/sh/include/asm/uaccess.h @@ -151,7 +151,10 @@ copy_from_user(void *to, const void __user *from, unsigned long n) __kernel_size_t __copy_size = (__kernel_size_t) n; if (__copy_size && __access_ok(__copy_from, __copy_size)) - return __copy_user(to, from, __copy_size); + __copy_size = __copy_user(to, from, __copy_size); + + if (unlikely(__copy_size)) + memset(to + (n - __copy_size), 0, __copy_size); return __copy_size; } From 917400cecb4b52b5cde5417348322bb9c8272fa6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 22 Aug 2016 00:23:07 -0400 Subject: [PATCH 220/271] sparc32: fix copy_from_user() Cc: stable@vger.kernel.org Acked-by: David S. Miller Signed-off-by: Al Viro --- arch/sparc/include/asm/uaccess_32.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/sparc/include/asm/uaccess_32.h b/arch/sparc/include/asm/uaccess_32.h index 341a5a133f48..035c89b7df5d 100644 --- a/arch/sparc/include/asm/uaccess_32.h +++ b/arch/sparc/include/asm/uaccess_32.h @@ -269,8 +269,10 @@ static inline unsigned long copy_from_user(void *to, const void __user *from, un if (!__builtin_constant_p(n)) check_object_size(to, n, false); return __copy_user((__force void __user *) to, from, n); - } else + } else { + memset(to, 0, n); return n; + } } static inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) From 8f035983dd826d7e04f67b28acf8e2f08c347e41 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 9 Sep 2016 19:16:58 -0400 Subject: [PATCH 221/271] blackfin: fix copy_from_user() Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/blackfin/include/asm/uaccess.h | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/blackfin/include/asm/uaccess.h b/arch/blackfin/include/asm/uaccess.h index 12f5d6851bbc..0a2a70096d8b 100644 --- a/arch/blackfin/include/asm/uaccess.h +++ b/arch/blackfin/include/asm/uaccess.h @@ -171,11 +171,12 @@ static inline int bad_user_access_length(void) static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n) { - if (access_ok(VERIFY_READ, from, n)) + if (likely(access_ok(VERIFY_READ, from, n))) { memcpy(to, (const void __force *)from, n); - else - return n; - return 0; + return 0; + } + memset(to, 0, n); + return n; } static inline unsigned long __must_check From c90a3bc5061d57e7931a9b7ad14784e1a0ed497d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 9 Sep 2016 19:20:13 -0400 Subject: [PATCH 222/271] m32r: fix __get_user() Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/m32r/include/asm/uaccess.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/m32r/include/asm/uaccess.h b/arch/m32r/include/asm/uaccess.h index cac7014daef3..6f8982157a75 100644 --- a/arch/m32r/include/asm/uaccess.h +++ b/arch/m32r/include/asm/uaccess.h @@ -219,7 +219,7 @@ extern int fixup_exception(struct pt_regs *regs); #define __get_user_nocheck(x, ptr, size) \ ({ \ long __gu_err = 0; \ - unsigned long __gu_val; \ + unsigned long __gu_val = 0; \ might_fault(); \ __get_user_size(__gu_val, (ptr), (size), __gu_err); \ (x) = (__force __typeof__(*(ptr)))__gu_val; \ From d0cf385160c12abd109746cad1f13e3b3e8b50b8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 9 Sep 2016 19:22:34 -0400 Subject: [PATCH 223/271] microblaze: fix copy_from_user() Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/microblaze/include/asm/uaccess.h | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h index 331b0d35f89c..3a486d3b6f52 100644 --- a/arch/microblaze/include/asm/uaccess.h +++ b/arch/microblaze/include/asm/uaccess.h @@ -373,10 +373,13 @@ extern long __user_bad(void); static inline long copy_from_user(void *to, const void __user *from, unsigned long n) { + unsigned long res = n; might_fault(); - if (access_ok(VERIFY_READ, from, n)) - return __copy_from_user(to, from, n); - return n; + if (likely(access_ok(VERIFY_READ, from, n))) + res = __copy_from_user(to, from, n); + if (unlikely(res)) + memset(to + (n - res), 0, res); + return res; } #define __copy_to_user(to, from, n) \ From e98b9e37ae04562d52c96f46b3cf4c2e80222dc1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 9 Sep 2016 19:23:33 -0400 Subject: [PATCH 224/271] microblaze: fix __get_user() Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/microblaze/include/asm/uaccess.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h index 3a486d3b6f52..826676778094 100644 --- a/arch/microblaze/include/asm/uaccess.h +++ b/arch/microblaze/include/asm/uaccess.h @@ -227,7 +227,7 @@ extern long __user_bad(void); #define __get_user(x, ptr) \ ({ \ - unsigned long __gu_val; \ + unsigned long __gu_val = 0; \ /*unsigned long __gu_ptr = (unsigned long)(ptr);*/ \ long __gu_err; \ switch (sizeof(*(ptr))) { \ From 8630c32275bac2de6ffb8aea9d9b11663e7ad28e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 9 Sep 2016 19:28:23 -0400 Subject: [PATCH 225/271] avr32: fix copy_from_user() really ugly, but apparently avr32 compilers turns access_ok() into something so bad that they want it in assembler. Left that way, zeroing added in inline wrapper. Cc: stable@vger.kernel.org Signed-off-by: Al Viro --- arch/avr32/include/asm/uaccess.h | 11 ++++++++++- arch/avr32/kernel/avr32_ksyms.c | 2 +- arch/avr32/lib/copy_user.S | 4 ++-- 3 files changed, 13 insertions(+), 4 deletions(-) diff --git a/arch/avr32/include/asm/uaccess.h b/arch/avr32/include/asm/uaccess.h index 68cf638faf48..b1ec1fa06463 100644 --- a/arch/avr32/include/asm/uaccess.h +++ b/arch/avr32/include/asm/uaccess.h @@ -74,7 +74,7 @@ extern __kernel_size_t __copy_user(void *to, const void *from, extern __kernel_size_t copy_to_user(void __user *to, const void *from, __kernel_size_t n); -extern __kernel_size_t copy_from_user(void *to, const void __user *from, +extern __kernel_size_t ___copy_from_user(void *to, const void __user *from, __kernel_size_t n); static inline __kernel_size_t __copy_to_user(void __user *to, const void *from, @@ -88,6 +88,15 @@ static inline __kernel_size_t __copy_from_user(void *to, { return __copy_user(to, (const void __force *)from, n); } +static inline __kernel_size_t copy_from_user(void *to, + const void __user *from, + __kernel_size_t n) +{ + size_t res = ___copy_from_user(to, from, n); + if (unlikely(res)) + memset(to + (n - res), 0, res); + return res; +} #define __copy_to_user_inatomic __copy_to_user #define __copy_from_user_inatomic __copy_from_user diff --git a/arch/avr32/kernel/avr32_ksyms.c b/arch/avr32/kernel/avr32_ksyms.c index d93ead02daed..7c6cf14f0985 100644 --- a/arch/avr32/kernel/avr32_ksyms.c +++ b/arch/avr32/kernel/avr32_ksyms.c @@ -36,7 +36,7 @@ EXPORT_SYMBOL(copy_page); /* * Userspace access stuff. */ -EXPORT_SYMBOL(copy_from_user); +EXPORT_SYMBOL(___copy_from_user); EXPORT_SYMBOL(copy_to_user); EXPORT_SYMBOL(__copy_user); EXPORT_SYMBOL(strncpy_from_user); diff --git a/arch/avr32/lib/copy_user.S b/arch/avr32/lib/copy_user.S index ea59c04b07de..96a6de9d578f 100644 --- a/arch/avr32/lib/copy_user.S +++ b/arch/avr32/lib/copy_user.S @@ -25,11 +25,11 @@ .align 1 .global copy_from_user .type copy_from_user, @function -copy_from_user: +___copy_from_user: branch_if_kernel r8, __copy_user ret_if_privileged r8, r11, r10, r10 rjmp __copy_user - .size copy_from_user, . - copy_from_user + .size ___copy_from_user, . - ___copy_from_user .global copy_to_user .type copy_to_user, @function From 6eaed1665fc6864fbdbffcc6f43a7f5d012f3052 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 13 Sep 2016 16:40:24 +1000 Subject: [PATCH 226/271] powerpc/powernv: Fix the state of root PE The PE for root bus (root PE) can be removed because of PCI hot remove in EEH recovery path for fenced PHB error. We need update @phb->root_pe_populated accordingly so that the root PE can be populated again in forthcoming PCI hot add path. Also, the PE shouldn't be destroyed as it's global and reserved resource. Fixes: c5f7700bbd2e ("powerpc/powernv: Dynamically release PE") Reported-by: Frederic Barrat Signed-off-by: Gavin Shan Signed-off-by: Michael Ellerman --- arch/powerpc/platforms/powernv/pci-ioda.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index c16d790808f1..0c71a2ffb5be 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -3426,7 +3426,17 @@ static void pnv_ioda_release_pe(struct pnv_ioda_pe *pe) } } - pnv_ioda_free_pe(pe); + /* + * The PE for root bus can be removed because of hotplug in EEH + * recovery for fenced PHB error. We need to mark the PE dead so + * that it can be populated again in PCI hot add path. The PE + * shouldn't be destroyed as it's the global reserved resource. + */ + if (phb->ioda.root_pe_populated && + phb->ioda.root_pe_idx == pe->pe_number) + phb->ioda.root_pe_populated = false; + else + pnv_ioda_free_pe(pe); } static void pnv_pci_release_device(struct pci_dev *pdev) From 74712339a4fc0f4ddc710e6bca836a6b78b7d8de Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Thu, 25 Aug 2016 08:23:14 +0100 Subject: [PATCH 227/271] drm/i915: Restore lost "Initialized i915" welcome message MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A side effect of removing the midlayer from driver loading was the loss of a useful message announcing to userspace that i915 had successfully started, e.g.: [drm] Initialized i915 1.6.0 20160425 for 0000:00:02.0 on minor 0 Reported-by: Timo Aaltonen Signed-off-by: Chris Wilson Fixes: 8f460e2c78f2 ("drm/i915: Demidlayer driver loading") Cc: Daniel Vetter Cc: Ville Syrjälä Cc: drm-intel-fixes@lists.freedesktop.org Link: http://patchwork.freedesktop.org/patch/msgid/20160825072314.17402-1-chris@chris-wilson.co.uk Reviewed-by: Daniel Vetter (cherry picked from commit bc5ca47c0af4f949ba889e666b7da65569e36093) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_drv.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 95ddd56b89f0..5de36d8dcc68 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -1281,6 +1281,11 @@ int i915_driver_load(struct pci_dev *pdev, const struct pci_device_id *ent) intel_runtime_pm_enable(dev_priv); + /* Everything is in place, we can now relax! */ + DRM_INFO("Initialized %s %d.%d.%d %s for %s on minor %d\n", + driver.name, driver.major, driver.minor, driver.patchlevel, + driver.date, pci_name(pdev), dev_priv->drm.primary->index); + intel_runtime_pm_put(dev_priv); return 0; From 86dfb76cba284114cf586005cd943eeb6e4f328d Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Wed, 7 Sep 2016 17:42:31 -0700 Subject: [PATCH 228/271] Revert "drm/i915/psr: Make idle_frames sensible again" This reverts commit 1c80c25fb622973dd135878e98d172be20859049 Author: Daniel Vetter Date: Wed May 18 18:47:12 2016 +0200 drm/i915/psr: Make idle_frames sensible again There are panels that needs 4 idle frames before entering PSR, but VBT is unproperly set. Also lately it was identified that idle frame count calculated at HW can be off by 1, what makes the minimum of 2, at least. Without the current vbt+1 we are with the risk of having HW calculating 0 idle frames and entering PSR when it shouldn't. Regardless the lack of link training. [Jani: there is some disagreement on the explanation, but the commit regresses so revert it is.] References: http://marc.info/?i=20160904191153.GA2328@light.dominikbrodowski.net Cc: Dominik Brodowski Cc: Jani Nikula Cc: Daniel Vetter Signed-off-by: Rodrigo Vivi Fixes: 1c80c25fb622 ("drm/i915/psr: Make idle_frames sensible again") Cc: drm-intel-fixes@lists.freedesktop.org # v4.8-rc1+ Signed-off-by: Jani Nikula Link: http://patchwork.freedesktop.org/patch/msgid/1473295351-8766-1-git-send-email-rodrigo.vivi@intel.com (cherry picked from commit 40918e0bb81be02f507a941f8b2741f0dc1771b0) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_psr.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c index 2b0d1baf15b3..cf171b4b8c67 100644 --- a/drivers/gpu/drm/i915/intel_psr.c +++ b/drivers/gpu/drm/i915/intel_psr.c @@ -255,14 +255,14 @@ static void hsw_psr_enable_source(struct intel_dp *intel_dp) struct drm_i915_private *dev_priv = to_i915(dev); uint32_t max_sleep_time = 0x1f; - /* Lately it was identified that depending on panel idle frame count - * calculated at HW can be off by 1. So let's use what came - * from VBT + 1. - * There are also other cases where panel demands at least 4 - * but VBT is not being set. To cover these 2 cases lets use - * at least 5 when VBT isn't set to be on the safest side. + /* + * Let's respect VBT in case VBT asks a higher idle_frame value. + * Let's use 6 as the minimum to cover all known cases including + * the off-by-one issue that HW has in some cases. Also there are + * cases where sink should be able to train + * with the 5 or 6 idle patterns. */ - uint32_t idle_frames = dev_priv->vbt.psr.idle_frames + 1; + uint32_t idle_frames = max(6, dev_priv->vbt.psr.idle_frames); uint32_t val = EDP_PSR_ENABLE; val |= max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT; From ea54ff4008892b46c7a3e6bc8ab8aaec9d198639 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= Date: Tue, 13 Sep 2016 12:22:19 +0300 Subject: [PATCH 229/271] drm/i915: Ignore OpRegion panel type except on select machines MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Turns out commit a05628195a0d ("drm/i915: Get panel_type from OpRegion panel details") has regressed quite a few machines. So it looks like we can't use the panel type from OpRegion on all systems, and yet we absolutely must use it on some specific systems. Despite trying, I was unable to find any automagic way to determine if the OpRegion panel type is respectable or not. The only glimmer of hope I had was bit 8 in the SCIC response, but that turned out to not work either (it was always 0 on both types of systems). So, to fix the regressions without breaking the machine we know to need the OpRegion panel type, let's just add a quirk for this. Only specific machines known to require the OpRegion panel type will therefore use it. Everyone else will fall bck to the VBT panel type. The only known machine so far is a "Conrac GmbH IX45GM2". The PCI subsystem ID on this machine is just a generic 8086:2a42, so of no use. Instead we'll go with a DMI match. I suspect we can now also revert commit aeddda06c1a7 ("drm/i915: Ignore panel type from OpRegion on SKL") but let's leave that to a separate patch. v2: Do the DMI match in the opregion code directly, as dev_priv->quirks gets populated too late Cc: Rob Kramer Cc: Martin van Es Cc: Andrea Arcangeli Cc: Dave Airlie Cc: Marco Krüger Cc: Sean Greenslade Cc: Trudy Tective Cc: Robin Müller Cc: Alexander Kobel Cc: Alexey Shumitsky Cc: Emil Andersen Lauridsen Cc: oceans112@gmail.com Cc: James Hogan Cc: James Bottomley Cc: stable@vger.kernel.org References: https://lists.freedesktop.org/archives/intel-gfx/2016-August/105545.html References: https://lists.freedesktop.org/archives/dri-devel/2016-August/116888.html References: https://lists.freedesktop.org/archives/intel-gfx/2016-June/098826.html Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94825 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97060 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97443 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97363 Fixes: a05628195a0d ("drm/i915: Get panel_type from OpRegion panel details") Tested-by: Marco Krüger Tested-by: Alexey Shumitsky Tested-by: Sean Greenslade Tested-by: Emil Andersen Lauridsen Tested-by: Robin Müller Tested-by: oceans112@gmail.com Tested-by: Rob Kramer Signed-off-by: Ville Syrjälä Link: http://patchwork.freedesktop.org/patch/msgid/1473758539-21565-1-git-send-email-ville.syrjala@linux.intel.com References: http://patchwork.freedesktop.org/patch/msgid/1473602239-15855-1-git-send-email-adrienverge@gmail.com Acked-by: Jani Nikula (cherry picked from commit c8ebfad7a063fe665417fa0eeb0da7cfe987d8ed) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_opregion.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c index adca262d591a..7acbbbf97833 100644 --- a/drivers/gpu/drm/i915/intel_opregion.c +++ b/drivers/gpu/drm/i915/intel_opregion.c @@ -1047,6 +1047,23 @@ err_out: return err; } +static int intel_use_opregion_panel_type_callback(const struct dmi_system_id *id) +{ + DRM_INFO("Using panel type from OpRegion on %s\n", id->ident); + return 1; +} + +static const struct dmi_system_id intel_use_opregion_panel_type[] = { + { + .callback = intel_use_opregion_panel_type_callback, + .ident = "Conrac GmbH IX45GM2", + .matches = {DMI_MATCH(DMI_SYS_VENDOR, "Conrac GmbH"), + DMI_MATCH(DMI_PRODUCT_NAME, "IX45GM2"), + }, + }, + { } +}; + int intel_opregion_get_panel_type(struct drm_i915_private *dev_priv) { @@ -1072,6 +1089,16 @@ intel_opregion_get_panel_type(struct drm_i915_private *dev_priv) return -ENODEV; } + /* + * So far we know that some machined must use it, others must not use it. + * There doesn't seem to be any way to determine which way to go, except + * via a quirk list :( + */ + if (!dmi_check_system(intel_use_opregion_panel_type)) { + DRM_DEBUG_KMS("Ignoring OpRegion panel type (%d)\n", ret - 1); + return -ENODEV; + } + /* * FIXME On Dell XPS 13 9350 the OpRegion panel type (0) gives us * low vswing for eDP, whereas the VBT panel type (2) gives us normal From e5789608766113ca9c30d596d93ca7d5cbd8b461 Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Wed, 14 Sep 2016 14:22:07 +0300 Subject: [PATCH 230/271] mmc: omap_hsmmc: Initialize dma_slave_config to avoid random data It is wrong to use uninitialized dma_slave_config and configure only certain fields as the DMAengine driver might look at non initialized (random data) fields and tries to interpret it. Signed-off-by: Peter Ujfalusi Signed-off-by: Ulf Hansson --- drivers/mmc/host/omap_hsmmc.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index 24ebc9a8de89..5f2f24a7360d 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -1409,11 +1409,18 @@ static int omap_hsmmc_pre_dma_transfer(struct omap_hsmmc_host *host, static int omap_hsmmc_setup_dma_transfer(struct omap_hsmmc_host *host, struct mmc_request *req) { - struct dma_slave_config cfg; struct dma_async_tx_descriptor *tx; int ret = 0, i; struct mmc_data *data = req->data; struct dma_chan *chan; + struct dma_slave_config cfg = { + .src_addr = host->mapbase + OMAP_HSMMC_DATA, + .dst_addr = host->mapbase + OMAP_HSMMC_DATA, + .src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES, + .dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES, + .src_maxburst = data->blksz / 4, + .dst_maxburst = data->blksz / 4, + }; /* Sanity check: all the SG entries must be aligned by block size. */ for (i = 0; i < data->sg_len; i++) { @@ -1433,13 +1440,6 @@ static int omap_hsmmc_setup_dma_transfer(struct omap_hsmmc_host *host, chan = omap_hsmmc_get_dma_chan(host, data); - cfg.src_addr = host->mapbase + OMAP_HSMMC_DATA; - cfg.dst_addr = host->mapbase + OMAP_HSMMC_DATA; - cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES; - cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES; - cfg.src_maxburst = data->blksz / 4; - cfg.dst_maxburst = data->blksz / 4; - ret = dmaengine_slave_config(chan, &cfg); if (ret) return ret; From df804d5e27490151da1ce9f216031a31352203e6 Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Wed, 14 Sep 2016 14:21:54 +0300 Subject: [PATCH 231/271] mmc: omap: Initialize dma_slave_config to avoid random data in it's fields It is wrong to use uninitialized dma_slave_config and configure only certain fields as the DMAengine driver might look at non initialized (random data) fields and tries to interpret it. Signed-off-by: Peter Ujfalusi Signed-off-by: Ulf Hansson --- drivers/mmc/host/omap.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/drivers/mmc/host/omap.c b/drivers/mmc/host/omap.c index f23d65eb070d..be3c49fa7382 100644 --- a/drivers/mmc/host/omap.c +++ b/drivers/mmc/host/omap.c @@ -1016,14 +1016,16 @@ mmc_omap_prepare_data(struct mmc_omap_host *host, struct mmc_request *req) /* Only reconfigure if we have a different burst size */ if (*bp != burst) { - struct dma_slave_config cfg; - - cfg.src_addr = host->phys_base + OMAP_MMC_REG(host, DATA); - cfg.dst_addr = host->phys_base + OMAP_MMC_REG(host, DATA); - cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES; - cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES; - cfg.src_maxburst = burst; - cfg.dst_maxburst = burst; + struct dma_slave_config cfg = { + .src_addr = host->phys_base + + OMAP_MMC_REG(host, DATA), + .dst_addr = host->phys_base + + OMAP_MMC_REG(host, DATA), + .src_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES, + .dst_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES, + .src_maxburst = burst, + .dst_maxburst = burst, + }; if (dmaengine_slave_config(c, &cfg)) goto use_pio; From 6cfeaf5125d425043d44002d0a1a8a147be582bf Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Wed, 14 Sep 2016 11:00:26 +0100 Subject: [PATCH 232/271] cpu/hotplug: Include linux/types.h in linux/cpuhotplug.h The linux/cpuhotplug.h header makes use of the bool type, but wasn't including linux/types.h to ensure that type has been defined. Fix this by including linux/types.h in preparation for including linux/cpuhotplug.h in a file that doesn't do so already. Signed-off-by: Paul Burton Cc: linux-mips@linux-mips.org Cc: Richard Cochran Cc: Sebastian Andrzej Siewior Cc: Ralf Baechle Cc: Anna-Maria Gleixner Link: http://lkml.kernel.org/r/20160914100027.20945-1-paul.burton@imgtec.com Signed-off-by: Thomas Gleixner --- include/linux/cpuhotplug.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h index 242bf530edfc..34bd80512a0c 100644 --- a/include/linux/cpuhotplug.h +++ b/include/linux/cpuhotplug.h @@ -1,6 +1,8 @@ #ifndef __CPUHOTPLUG_H #define __CPUHOTPLUG_H +#include + enum cpuhp_state { CPUHP_OFFLINE, CPUHP_CREATE_THREADS, From de75abbe0121a6c3c9c6b04c75300088e57ad1d5 Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Thu, 8 Sep 2016 11:48:28 +0200 Subject: [PATCH 233/271] arm/xen: fix SMP guests boot Commit 88e957d6e47f ("xen: introduce xen_vcpu_id mapping") broke SMP ARM guests on Xen. When FIFO-based event channels are in use (this is the default), evtchn_fifo_alloc_control_block() is called on CPU_UP_PREPARE event and this happens before we set up xen_vcpu_id mapping in xen_starting_cpu. Temporary fix the issue by setting direct Linux CPU id <-> Xen vCPU id mapping for all possible CPUs at boot. We don't currently support kexec/kdump on Xen/ARM so these ids always match. In future, we have several ways to solve the issue, e.g.: - Eliminate all hypercalls from CPU_UP_PREPARE, do them from the starting CPU. This can probably be done for both x86 and ARM and, if done, will allow us to get Xen's idea of vCPU id from CPUID/MPIDR on the starting CPU directly, no messing with ACPI/device tree required. - Save vCPU id information from ACPI/device tree on ARM and use it to initialize xen_vcpu_id mapping. This is the same trick we currently do on x86. Reported-by: Julien Grall Tested-by: Wei Chen Signed-off-by: Vitaly Kuznetsov Acked-by: Stefano Stabellini Signed-off-by: David Vrabel --- arch/arm/xen/enlighten.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/arch/arm/xen/enlighten.c b/arch/arm/xen/enlighten.c index 3d2cef6488ea..f193414d0f6f 100644 --- a/arch/arm/xen/enlighten.c +++ b/arch/arm/xen/enlighten.c @@ -170,9 +170,6 @@ static int xen_starting_cpu(unsigned int cpu) pr_info("Xen: initializing cpu%d\n", cpu); vcpup = per_cpu_ptr(xen_vcpu_info, cpu); - /* Direct vCPU id mapping for ARM guests. */ - per_cpu(xen_vcpu_id, cpu) = cpu; - info.mfn = virt_to_gfn(vcpup); info.offset = xen_offset_in_page(vcpup); @@ -330,6 +327,7 @@ static int __init xen_guest_init(void) { struct xen_add_to_physmap xatp; struct shared_info *shared_info_page = NULL; + int cpu; if (!xen_domain()) return 0; @@ -380,7 +378,8 @@ static int __init xen_guest_init(void) return -ENOMEM; /* Direct vCPU id mapping for ARM guests. */ - per_cpu(xen_vcpu_id, 0) = 0; + for_each_possible_cpu(cpu) + per_cpu(xen_vcpu_id, cpu) = cpu; xen_auto_xlat_grant_frames.count = gnttab_max_grant_frames(); if (xen_xlate_map_ballooned_pages(&xen_auto_xlat_grant_frames.pfn, From 7ccb8e633cfeb7969eba09bbf53346e746bb7f89 Mon Sep 17 00:00:00 2001 From: Fabian Frederick Date: Mon, 15 Aug 2016 19:34:44 +0200 Subject: [PATCH 234/271] ARM: multi_v7_defconfig: update XILINX_VDMA Commit fde57a7c4474 ("dmaengine: xilinx: Rename driver and config") renamed config XILINX_VDMA to config XILINX_DMA Update defconfig accordingly. Signed-off-by: Fabian Frederick Signed-off-by: Arnd Bergmann --- arch/arm/configs/multi_v7_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig index 2c8665cd9dc5..ea3566fb92e2 100644 --- a/arch/arm/configs/multi_v7_defconfig +++ b/arch/arm/configs/multi_v7_defconfig @@ -781,7 +781,7 @@ CONFIG_MXS_DMA=y CONFIG_DMA_BCM2835=y CONFIG_DMA_OMAP=y CONFIG_QCOM_BAM_DMA=y -CONFIG_XILINX_VDMA=y +CONFIG_XILINX_DMA=y CONFIG_DMA_SUN6I=y CONFIG_STAGING=y CONFIG_SENSORS_ISL29018=y From f2a89d3b2b85b90b05453872aaabfdb412a21a03 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 1 Aug 2016 10:54:16 +0100 Subject: [PATCH 235/271] arm64: dts: Fix broken architected timer interrupt trigger The ARM architected timer specification mandates that the interrupt associated with each timer is level triggered (which corresponds to the "counter >= comparator" condition). A number of DTs are being remarkably creative, declaring the interrupt to be edge triggered. A quick look at the TRM for the corresponding ARM CPUs clearly shows that this is wrong, and I've corrected those. For non-ARM designs (and in the absence of a publicly available TRM), I've made them active low as well, which can't be completely wrong as the GIC cannot disinguish between level low and level high. The respective maintainers are of course welcome to prove me wrong. While I was at it, I took the liberty to fix a couple of related issue, such as some spurious affinity bits on ThunderX, and their complete absence on ls1043a (both of which seem to be related to copy-pasting from other DTs). Acked-by: Duc Dang Acked-by: Carlo Caione Acked-by: Michal Simek Acked-by: Krzysztof Kozlowski Acked-by: Dinh Nguyen Acked-by: Masahiro Yamada Signed-off-by: Marc Zyngier Signed-off-by: Arnd Bergmann --- arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi | 8 ++++---- arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi | 8 ++++---- arch/arm64/boot/dts/apm/apm-storm.dtsi | 8 ++++---- arch/arm64/boot/dts/broadcom/ns2.dtsi | 8 ++++---- arch/arm64/boot/dts/cavium/thunder-88xx.dtsi | 8 ++++---- arch/arm64/boot/dts/exynos/exynos7.dtsi | 8 ++++---- arch/arm64/boot/dts/freescale/fsl-ls1043a.dtsi | 8 ++++---- arch/arm64/boot/dts/freescale/fsl-ls2080a.dtsi | 8 ++++---- arch/arm64/boot/dts/marvell/armada-ap806.dtsi | 8 ++++---- arch/arm64/boot/dts/socionext/uniphier-ph1-ld20.dtsi | 8 ++++---- arch/arm64/boot/dts/xilinx/zynqmp.dtsi | 8 ++++---- 11 files changed, 44 insertions(+), 44 deletions(-) diff --git a/arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi b/arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi index 445aa678f914..c2b9bcb0ef61 100644 --- a/arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi +++ b/arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi @@ -255,10 +255,10 @@ /* Local timer */ timer { compatible = "arm,armv8-timer"; - interrupts = <1 13 0xf01>, - <1 14 0xf01>, - <1 11 0xf01>, - <1 10 0xf01>; + interrupts = <1 13 0xf08>, + <1 14 0xf08>, + <1 11 0xf08>, + <1 10 0xf08>; }; timer0: timer0@ffc03000 { diff --git a/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi b/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi index e502c24b0ac7..bf6c8d051002 100644 --- a/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi @@ -102,13 +102,13 @@ timer { compatible = "arm,armv8-timer"; interrupts = , + (GIC_CPU_MASK_RAW(0xff) | IRQ_TYPE_LEVEL_LOW)>, , + (GIC_CPU_MASK_RAW(0xff) | IRQ_TYPE_LEVEL_LOW)>, , + (GIC_CPU_MASK_RAW(0xff) | IRQ_TYPE_LEVEL_LOW)>, ; + (GIC_CPU_MASK_RAW(0xff) | IRQ_TYPE_LEVEL_LOW)>; }; xtal: xtal-clk { diff --git a/arch/arm64/boot/dts/apm/apm-storm.dtsi b/arch/arm64/boot/dts/apm/apm-storm.dtsi index f1c2c713f9b0..c29dab9d1834 100644 --- a/arch/arm64/boot/dts/apm/apm-storm.dtsi +++ b/arch/arm64/boot/dts/apm/apm-storm.dtsi @@ -110,10 +110,10 @@ timer { compatible = "arm,armv8-timer"; - interrupts = <1 0 0xff01>, /* Secure Phys IRQ */ - <1 13 0xff01>, /* Non-secure Phys IRQ */ - <1 14 0xff01>, /* Virt IRQ */ - <1 15 0xff01>; /* Hyp IRQ */ + interrupts = <1 0 0xff08>, /* Secure Phys IRQ */ + <1 13 0xff08>, /* Non-secure Phys IRQ */ + <1 14 0xff08>, /* Virt IRQ */ + <1 15 0xff08>; /* Hyp IRQ */ clock-frequency = <50000000>; }; diff --git a/arch/arm64/boot/dts/broadcom/ns2.dtsi b/arch/arm64/boot/dts/broadcom/ns2.dtsi index f53b0955bfd3..d4a12fad8afd 100644 --- a/arch/arm64/boot/dts/broadcom/ns2.dtsi +++ b/arch/arm64/boot/dts/broadcom/ns2.dtsi @@ -88,13 +88,13 @@ timer { compatible = "arm,armv8-timer"; interrupts = , + IRQ_TYPE_LEVEL_LOW)>, , + IRQ_TYPE_LEVEL_LOW)>, , + IRQ_TYPE_LEVEL_LOW)>, ; + IRQ_TYPE_LEVEL_LOW)>; }; pmu { diff --git a/arch/arm64/boot/dts/cavium/thunder-88xx.dtsi b/arch/arm64/boot/dts/cavium/thunder-88xx.dtsi index 2eb9b225f0bc..04dc8a8d1539 100644 --- a/arch/arm64/boot/dts/cavium/thunder-88xx.dtsi +++ b/arch/arm64/boot/dts/cavium/thunder-88xx.dtsi @@ -354,10 +354,10 @@ timer { compatible = "arm,armv8-timer"; - interrupts = <1 13 0xff01>, - <1 14 0xff01>, - <1 11 0xff01>, - <1 10 0xff01>; + interrupts = <1 13 4>, + <1 14 4>, + <1 11 4>, + <1 10 4>; }; pmu { diff --git a/arch/arm64/boot/dts/exynos/exynos7.dtsi b/arch/arm64/boot/dts/exynos/exynos7.dtsi index ca663dfe5189..162831546e18 100644 --- a/arch/arm64/boot/dts/exynos/exynos7.dtsi +++ b/arch/arm64/boot/dts/exynos/exynos7.dtsi @@ -473,10 +473,10 @@ timer { compatible = "arm,armv8-timer"; - interrupts = <1 13 0xff01>, - <1 14 0xff01>, - <1 11 0xff01>, - <1 10 0xff01>; + interrupts = <1 13 0xff08>, + <1 14 0xff08>, + <1 11 0xff08>, + <1 10 0xff08>; }; pmu_system_controller: system-controller@105c0000 { diff --git a/arch/arm64/boot/dts/freescale/fsl-ls1043a.dtsi b/arch/arm64/boot/dts/freescale/fsl-ls1043a.dtsi index e669fbd7f9c3..a67e210e2019 100644 --- a/arch/arm64/boot/dts/freescale/fsl-ls1043a.dtsi +++ b/arch/arm64/boot/dts/freescale/fsl-ls1043a.dtsi @@ -119,10 +119,10 @@ timer { compatible = "arm,armv8-timer"; - interrupts = <1 13 0x1>, /* Physical Secure PPI */ - <1 14 0x1>, /* Physical Non-Secure PPI */ - <1 11 0x1>, /* Virtual PPI */ - <1 10 0x1>; /* Hypervisor PPI */ + interrupts = <1 13 0xf08>, /* Physical Secure PPI */ + <1 14 0xf08>, /* Physical Non-Secure PPI */ + <1 11 0xf08>, /* Virtual PPI */ + <1 10 0xf08>; /* Hypervisor PPI */ }; pmu { diff --git a/arch/arm64/boot/dts/freescale/fsl-ls2080a.dtsi b/arch/arm64/boot/dts/freescale/fsl-ls2080a.dtsi index 21023a388c29..e3b6034ea5d9 100644 --- a/arch/arm64/boot/dts/freescale/fsl-ls2080a.dtsi +++ b/arch/arm64/boot/dts/freescale/fsl-ls2080a.dtsi @@ -191,10 +191,10 @@ timer { compatible = "arm,armv8-timer"; - interrupts = <1 13 0x8>, /* Physical Secure PPI, active-low */ - <1 14 0x8>, /* Physical Non-Secure PPI, active-low */ - <1 11 0x8>, /* Virtual PPI, active-low */ - <1 10 0x8>; /* Hypervisor PPI, active-low */ + interrupts = <1 13 4>, /* Physical Secure PPI, active-low */ + <1 14 4>, /* Physical Non-Secure PPI, active-low */ + <1 11 4>, /* Virtual PPI, active-low */ + <1 10 4>; /* Hypervisor PPI, active-low */ }; pmu { diff --git a/arch/arm64/boot/dts/marvell/armada-ap806.dtsi b/arch/arm64/boot/dts/marvell/armada-ap806.dtsi index eab1a42fb934..c2a6745f168c 100644 --- a/arch/arm64/boot/dts/marvell/armada-ap806.dtsi +++ b/arch/arm64/boot/dts/marvell/armada-ap806.dtsi @@ -122,10 +122,10 @@ timer { compatible = "arm,armv8-timer"; - interrupts = , - , - , - ; + interrupts = , + , + , + ; }; odmi: odmi@300000 { diff --git a/arch/arm64/boot/dts/socionext/uniphier-ph1-ld20.dtsi b/arch/arm64/boot/dts/socionext/uniphier-ph1-ld20.dtsi index c223915f0907..d73bdc8c9115 100644 --- a/arch/arm64/boot/dts/socionext/uniphier-ph1-ld20.dtsi +++ b/arch/arm64/boot/dts/socionext/uniphier-ph1-ld20.dtsi @@ -129,10 +129,10 @@ timer { compatible = "arm,armv8-timer"; - interrupts = <1 13 0xf01>, - <1 14 0xf01>, - <1 11 0xf01>, - <1 10 0xf01>; + interrupts = <1 13 4>, + <1 14 4>, + <1 11 4>, + <1 10 4>; }; soc { diff --git a/arch/arm64/boot/dts/xilinx/zynqmp.dtsi b/arch/arm64/boot/dts/xilinx/zynqmp.dtsi index e595f22e7e4b..3e2e51fbd2bc 100644 --- a/arch/arm64/boot/dts/xilinx/zynqmp.dtsi +++ b/arch/arm64/boot/dts/xilinx/zynqmp.dtsi @@ -65,10 +65,10 @@ timer { compatible = "arm,armv8-timer"; interrupt-parent = <&gic>; - interrupts = <1 13 0xf01>, - <1 14 0xf01>, - <1 11 0xf01>, - <1 10 0xf01>; + interrupts = <1 13 0xf08>, + <1 14 0xf08>, + <1 11 0xf08>, + <1 10 0xf08>; }; amba_apu { From 29bf282dec94f6015a675c007614cb29563f1c18 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 6 Sep 2016 16:34:01 +1000 Subject: [PATCH 236/271] powerpc/powernv: Detach from PE on releasing PCI device The PCI hotplug can be part of EEH error recovery. The @pdn and the device's PE number aren't removed and added afterwords. The PE number in @pdn should be set to an invalid one. Otherwise, the PE's device count is decreased on removing devices while failing to be increased on adding devices. It leads to unbalanced PE's device count and make normal PCI hotplug path broken. Fixes: c5f7700bbd2e ("powerpc/powernv: Dynamically release PE") Signed-off-by: Gavin Shan Signed-off-by: Michael Ellerman --- arch/powerpc/platforms/powernv/pci-ioda.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index 0c71a2ffb5be..da5da11a6223 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -3452,7 +3452,17 @@ static void pnv_pci_release_device(struct pci_dev *pdev) if (!pdn || pdn->pe_number == IODA_INVALID_PE) return; + /* + * PCI hotplug can happen as part of EEH error recovery. The @pdn + * isn't removed and added afterwards in this scenario. We should + * set the PE number in @pdn to an invalid one. Otherwise, the PE's + * device count is decreased on removing devices while failing to + * be increased on adding devices. It leads to unbalanced PE's device + * count and eventually make normal PCI hotplug path broken. + */ pe = &phb->ioda.pe_array[pdn->pe_number]; + pdn->pe_number = IODA_INVALID_PE; + WARN_ON(--pe->device_count < 0); if (pe->device_count == 0) pnv_ioda_release_pe(pe); From ed7d9a1d7da6fe7b1c7477dc70e95051583fd60c Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Thu, 15 Sep 2016 17:03:06 +1000 Subject: [PATCH 237/271] powerpc/powernv/pci: Fix missed TCE invalidations that should fallback to OPAL In commit f0228c413011 ("powerpc/powernv/pci: Fallback to OPAL for TCE invalidations"), we added logic to fallback to OPAL for doing TCE invalidations if we can't do it in Linux. Ben sent a v2 of the patch, containing these additional call sites, but I had already applied v1 and didn't notice. So fix them now. Fixes: f0228c413011 ("powerpc/powernv/pci: Fallback to OPAL for TCE invalidations") Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman --- arch/powerpc/platforms/powernv/pci-ioda.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index da5da11a6223..bc0c91e84ca0 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -2217,7 +2217,7 @@ static long pnv_pci_ioda2_set_window(struct iommu_table_group *table_group, pnv_pci_link_table_and_group(phb->hose->node, num, tbl, &pe->table_group); - pnv_pci_phb3_tce_invalidate_pe(pe); + pnv_pci_ioda2_tce_invalidate_pe(pe); return 0; } @@ -2355,7 +2355,7 @@ static long pnv_pci_ioda2_unset_window(struct iommu_table_group *table_group, if (ret) pe_warn(pe, "Unmapping failed, ret = %ld\n", ret); else - pnv_pci_phb3_tce_invalidate_pe(pe); + pnv_pci_ioda2_tce_invalidate_pe(pe); pnv_pci_unlink_table_and_group(table_group->tables[num], table_group); From cecf62352aee2b4fe114aafd1b8c5f265a4243ce Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Thu, 15 Sep 2016 11:22:33 +0300 Subject: [PATCH 238/271] perf/x86/intel: Don't disable "intel_bts" around "intel" event batching At the moment, intel_bts events get disabled from intel PMU's disable callback, which includes event scheduling transactions of said PMU, which have nothing to do with intel_bts events. We do want to keep intel_bts events off inside the PMI handler to avoid filling up their buffer too soon. This patch moves intel_bts enabling/disabling directly to the PMI handler. Reported-by: Vince Weaver Signed-off-by: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160915082233.11065-1-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/core.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 2cbde2f449aa..4c9a79b9cd69 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -1730,9 +1730,11 @@ static __initconst const u64 knl_hw_cache_extra_regs * disabled state if called consecutively. * * During consecutive calls, the same disable value will be written to related - * registers, so the PMU state remains unchanged. hw.state in - * intel_bts_disable_local will remain PERF_HES_STOPPED too in consecutive - * calls. + * registers, so the PMU state remains unchanged. + * + * intel_bts events don't coexist with intel PMU's BTS events because of + * x86_add_exclusive(x86_lbr_exclusive_lbr); there's no need to keep them + * disabled around intel PMU's event batching etc, only inside the PMI handler. */ static void __intel_pmu_disable_all(void) { @@ -1742,8 +1744,6 @@ static void __intel_pmu_disable_all(void) if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask)) intel_pmu_disable_bts(); - else - intel_bts_disable_local(); intel_pmu_pebs_disable_all(); } @@ -1771,8 +1771,7 @@ static void __intel_pmu_enable_all(int added, bool pmi) return; intel_pmu_enable_bts(event->hw.config); - } else - intel_bts_enable_local(); + } } static void intel_pmu_enable_all(int added) @@ -2073,6 +2072,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) */ if (!x86_pmu.late_ack) apic_write(APIC_LVTPC, APIC_DM_NMI); + intel_bts_disable_local(); __intel_pmu_disable_all(); handled = intel_pmu_drain_bts_buffer(); handled += intel_bts_interrupt(); @@ -2172,6 +2172,7 @@ done: /* Only restore PMU state when it's active. See x86_pmu_disable(). */ if (cpuc->enabled) __intel_pmu_enable_all(0, true); + intel_bts_enable_local(); /* * Only unmask the NMI after the overflow counters From a6805884e263e82d9fb87bd5f39ad4bb38cde246 Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Wed, 17 Aug 2016 13:44:50 +0300 Subject: [PATCH 239/271] ARM: keystone: defconfig: Fix USB configuration Simply enabling CONFIG_KEYSTONE_USB_PHY doesn't work anymore as it depends on CONFIG_NOP_USB_XCEIV. We need to enable that as well. This fixes USB on Keystone boards from v4.8-rc1 onwards. Signed-off-by: Roger Quadros Acked-by: Santosh Shilimkar Signed-off-by: Arnd Bergmann --- arch/arm/configs/keystone_defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/configs/keystone_defconfig b/arch/arm/configs/keystone_defconfig index 71b42e66488a..78cd2f197e01 100644 --- a/arch/arm/configs/keystone_defconfig +++ b/arch/arm/configs/keystone_defconfig @@ -161,6 +161,7 @@ CONFIG_USB_MON=y CONFIG_USB_XHCI_HCD=y CONFIG_USB_STORAGE=y CONFIG_USB_DWC3=y +CONFIG_NOP_USB_XCEIV=y CONFIG_KEYSTONE_USB_PHY=y CONFIG_NEW_LEDS=y CONFIG_LEDS_CLASS=y From b0eaf4506f5f95d15d6731d72c0ddf4a2179eefa Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Wed, 14 Sep 2016 23:39:12 +0200 Subject: [PATCH 240/271] kvm: x86: correctly reset dest_map->vector when restoring LAPIC state MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When userspace sends KVM_SET_LAPIC, KVM schedules a check between the vCPU's IRR and ISR and the IOAPIC redirection table, in order to re-establish the IOAPIC's dest_map (the list of CPUs servicing the real-time clock interrupt with the corresponding vectors). However, __rtc_irq_eoi_tracking_restore_one was forgetting to set dest_map->vectors. Because of this, the IOAPIC did not process the real-time clock interrupt EOI, ioapic->rtc_status.pending_eoi got stuck at a non-zero value, and further RTC interrupts were reported to userspace as coalesced. Fixes: 9e4aabe2bb3454c83dac8139cf9974503ee044db Fixes: 4d99ba898dd0c521ca6cdfdde55c9b58aea3cb3d Cc: stable@vger.kernel.org Cc: Joerg Roedel Cc: David Gilbert Reviewed-by: Radim Krčmář Signed-off-by: Paolo Bonzini --- arch/x86/kvm/ioapic.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/arch/x86/kvm/ioapic.c b/arch/x86/kvm/ioapic.c index 5f42d038fcb4..c7220ba94aa7 100644 --- a/arch/x86/kvm/ioapic.c +++ b/arch/x86/kvm/ioapic.c @@ -109,6 +109,7 @@ static void __rtc_irq_eoi_tracking_restore_one(struct kvm_vcpu *vcpu) { bool new_val, old_val; struct kvm_ioapic *ioapic = vcpu->kvm->arch.vioapic; + struct dest_map *dest_map = &ioapic->rtc_status.dest_map; union kvm_ioapic_redirect_entry *e; e = &ioapic->redirtbl[RTC_GSI]; @@ -117,16 +118,17 @@ static void __rtc_irq_eoi_tracking_restore_one(struct kvm_vcpu *vcpu) return; new_val = kvm_apic_pending_eoi(vcpu, e->fields.vector); - old_val = test_bit(vcpu->vcpu_id, ioapic->rtc_status.dest_map.map); + old_val = test_bit(vcpu->vcpu_id, dest_map->map); if (new_val == old_val) return; if (new_val) { - __set_bit(vcpu->vcpu_id, ioapic->rtc_status.dest_map.map); + __set_bit(vcpu->vcpu_id, dest_map->map); + dest_map->vectors[vcpu->vcpu_id] = e->fields.vector; ioapic->rtc_status.pending_eoi++; } else { - __clear_bit(vcpu->vcpu_id, ioapic->rtc_status.dest_map.map); + __clear_bit(vcpu->vcpu_id, dest_map->map); ioapic->rtc_status.pending_eoi--; rtc_status_pending_eoi_check_valid(ioapic); } From 1c109fabbd51863475cd12ac206bdd249aee35af Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 15 Sep 2016 02:35:29 +0100 Subject: [PATCH 241/271] fix minor infoleak in get_user_ex() get_user_ex(x, ptr) should zero x on failure. It's not a lot of a leak (at most we are leaking uninitialized 64bit value off the kernel stack, and in a fairly constrained situation, at that), but the fix is trivial, so... Cc: stable@vger.kernel.org Signed-off-by: Al Viro [ This sat in different branch from the uaccess fixes since mid-August ] Signed-off-by: Linus Torvalds --- arch/x86/include/asm/uaccess.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h index e3af86f58eaf..2131c4ce7d8a 100644 --- a/arch/x86/include/asm/uaccess.h +++ b/arch/x86/include/asm/uaccess.h @@ -433,7 +433,11 @@ do { \ #define __get_user_asm_ex(x, addr, itype, rtype, ltype) \ asm volatile("1: mov"itype" %1,%"rtype"0\n" \ "2:\n" \ - _ASM_EXTABLE_EX(1b, 2b) \ + ".section .fixup,\"ax\"\n" \ + "3:xor"itype" %"rtype"0,%"rtype"0\n" \ + " jmp 2b\n" \ + ".previous\n" \ + _ASM_EXTABLE_EX(1b, 3b) \ : ltype(x) : "m" (__m(addr))) #define __put_user_nocheck(x, ptr, size) \ From 5297e0f0fe13305a1fc7f01986be0dccd063d57a Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Wed, 14 Sep 2016 20:20:00 -0700 Subject: [PATCH 242/271] vfs: fix return type of ioctl_file_dedupe_range All the VFS functions in the dedupe ioctl path return int status, so the ioctl handler ought to as well. Found by Coverity, CID 1350952. Signed-off-by: Darrick J. Wong Signed-off-by: Linus Torvalds --- fs/ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ioctl.c b/fs/ioctl.c index 0f56deb24ce6..26aba0942a98 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -568,7 +568,7 @@ static int ioctl_fsthaw(struct file *filp) return thaw_super(sb); } -static long ioctl_file_dedupe_range(struct file *file, void __user *arg) +static int ioctl_file_dedupe_range(struct file *file, void __user *arg) { struct file_dedupe_range __user *argp = arg; struct file_dedupe_range *same = NULL; From b71dbf1032f546bf3efd60fb5d9d0cefd200a508 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Wed, 14 Sep 2016 20:20:44 -0700 Subject: [PATCH 243/271] vfs: cap dedupe request structure size at PAGE_SIZE Kirill A Shutemov reports that the kernel doesn't try to cap dest_count in any way, and uses the number to allocate kernel memory. This causes high order allocation warnings in the kernel log if someone passes in a big enough value. We should clamp the allocation at PAGE_SIZE to avoid stressing the VM. The two existing users of the dedupe ioctl never send more than 120 requests, so we can safely clamp dest_range at PAGE_SIZE, because with 4k pages we can handle up to 127 dedupe candidates. Given the max extent length of 16MB, we can end up doing 2GB of IO which is plenty. [ Note: the "offsetof()" can't overflow, because 'count' is just a 16-bit integer. That's not obvious in the limited context of the patch, so I'm noting it here because it made me go look. - Linus ] Reported-by: "Kirill A. Shutemov" Signed-off-by: Darrick J. Wong Signed-off-by: Linus Torvalds --- fs/ioctl.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/ioctl.c b/fs/ioctl.c index 26aba0942a98..c415668c86d4 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -582,6 +582,10 @@ static int ioctl_file_dedupe_range(struct file *file, void __user *arg) } size = offsetof(struct file_dedupe_range __user, info[count]); + if (size > PAGE_SIZE) { + ret = -ENOMEM; + goto out; + } same = memdup_user(argp, size); if (IS_ERR(same)) { From 22f6b4d34fcf039c63a94e7670e0da24f8575a5a Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Fri, 16 Sep 2016 00:31:22 +0200 Subject: [PATCH 244/271] aio: mark AIO pseudo-fs noexec This ensures that do_mmap() won't implicitly make AIO memory mappings executable if the READ_IMPLIES_EXEC personality flag is set. Such behavior is problematic because the security_mmap_file LSM hook doesn't catch this case, potentially permitting an attacker to bypass a W^X policy enforced by SELinux. I have tested the patch on my machine. To test the behavior, compile and run this: #define _GNU_SOURCE #include #include #include #include #include #include #include int main(void) { personality(READ_IMPLIES_EXEC); aio_context_t ctx = 0; if (syscall(__NR_io_setup, 1, &ctx)) err(1, "io_setup"); char cmd[1000]; sprintf(cmd, "cat /proc/%d/maps | grep -F '/[aio]'", (int)getpid()); system(cmd); return 0; } In the output, "rw-s" is good, "rwxs" is bad. Signed-off-by: Jann Horn Signed-off-by: Linus Torvalds --- fs/aio.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/fs/aio.c b/fs/aio.c index fb8e45b88cd4..4fe81d1c60f9 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -239,7 +239,12 @@ static struct dentry *aio_mount(struct file_system_type *fs_type, static const struct dentry_operations ops = { .d_dname = simple_dname, }; - return mount_pseudo(fs_type, "aio:", NULL, &ops, AIO_RING_MAGIC); + struct dentry *root = mount_pseudo(fs_type, "aio:", NULL, &ops, + AIO_RING_MAGIC); + + if (!IS_ERR(root)) + root->d_sb->s_iflags |= SB_I_NOEXEC; + return root; } /* aio_setup From 778935778c3b88e5152a88765850009006ef2e32 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 26 Apr 2016 10:42:25 -0700 Subject: [PATCH 245/271] PM / runtime: Use _rcuidle for runtime suspend tracepoints Further testing with false negatives suppressed by commit 293e2421fe25 ("rcu: Remove superfluous versions of rcu_read_lock_sched_held()") identified a few more unprotected uses of RCU from the idle loop. Because RCU actively ignores idle-loop code (for energy-efficiency reasons, among other things), using RCU from the idle loop can result in too-short grace periods, in turn resulting in arbitrary misbehavior. The affected function is rpm_suspend(). The resulting lockdep-RCU splat is as follows: ------------------------------------------------------------------------ Warning from omap3 =============================== [ INFO: suspicious RCU usage. ] 4.6.0-rc5-next-20160426+ #1112 Not tainted ------------------------------- include/trace/events/rpm.h:63 suspicious rcu_dereference_check() usage! other info that might help us debug this: RCU used illegally from idle CPU! rcu_scheduler_active = 1, debug_locks = 0 RCU used illegally from extended quiescent state! 1 lock held by swapper/0/0: #0: (&(&dev->power.lock)->rlock){-.-...}, at: [] __pm_runtime_suspend+0x54/0x84 stack backtrace: CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.6.0-rc5-next-20160426+ #1112 Hardware name: Generic OMAP36xx (Flattened Device Tree) [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [] (show_stack) from [] (dump_stack+0xb0/0xe4) [] (dump_stack) from [] (rpm_suspend+0x604/0x7e4) [] (rpm_suspend) from [] (__pm_runtime_suspend+0x64/0x84) [] (__pm_runtime_suspend) from [] (omap2_gpio_prepare_for_idle+0x5c/0x70) [] (omap2_gpio_prepare_for_idle) from [] (omap_sram_idle+0x140/0x244) [] (omap_sram_idle) from [] (omap3_enter_idle_bm+0xfc/0x1ec) [] (omap3_enter_idle_bm) from [] (cpuidle_enter_state+0x80/0x3d4) [] (cpuidle_enter_state) from [] (cpu_startup_entry+0x198/0x3a0) [] (cpu_startup_entry) from [] (start_kernel+0x354/0x3c8) [] (start_kernel) from [<8000807c>] (0x8000807c) ------------------------------------------------------------------------ Reported-by: Tony Lindgren Signed-off-by: Paul E. McKenney Tested-by: Tony Lindgren Tested-by: Guenter Roeck [ rjw: Subject ] Signed-off-by: Rafael J. Wysocki --- drivers/base/power/runtime.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 17995fadebd7..82a081ea4317 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -419,7 +419,7 @@ static int rpm_suspend(struct device *dev, int rpmflags) struct device *parent = NULL; int retval; - trace_rpm_suspend(dev, rpmflags); + trace_rpm_suspend_rcuidle(dev, rpmflags); repeat: retval = rpm_check_suspend_allowed(dev); @@ -549,7 +549,7 @@ static int rpm_suspend(struct device *dev, int rpmflags) } out: - trace_rpm_return_int(dev, _THIS_IP_, retval); + trace_rpm_return_int_rcuidle(dev, _THIS_IP_, retval); return retval; From 95f60084acbcee6c466256cf26eb52191fad9edc Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Thu, 15 Sep 2016 18:13:50 +0300 Subject: [PATCH 246/271] perf/x86/intel/pt: Fix an off-by-one in address filter configuration PT address filter configuration requires that a range is specified by its first and last address, but at the moment we're obtaining the end of the range by adding user specified size to its start, which is off by one from what it actually needs to be. Fix this and make sure that zero-sized filters don't pass the filter validation. Reported-by: Adrian Hunter Signed-off-by: Alexander Shishkin Acked-by: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: stable@vger.kernel.org # v4.7 Cc: stable@vger.kernel.org#v4.7 Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160915151352.21306-2-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/pt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c index 04bb5fb5a8d7..5ec0100e3fc6 100644 --- a/arch/x86/events/intel/pt.c +++ b/arch/x86/events/intel/pt.c @@ -1081,7 +1081,7 @@ static int pt_event_addr_filters_validate(struct list_head *filters) list_for_each_entry(filter, filters, entry) { /* PT doesn't support single address triggers */ - if (!filter->range) + if (!filter->range || !filter->size) return -EOPNOTSUPP; if (!filter->inode && !kernel_ip(filter->offset)) @@ -1111,7 +1111,7 @@ static void pt_event_addr_filters_sync(struct perf_event *event) } else { /* apply the offset */ msr_a = filter->offset + offs[range]; - msr_b = filter->size + msr_a; + msr_b = filter->size + msr_a - 1; } filters->filter[range].msr_a = msr_a; From ddfdad991e55b65c1cc4ee29502f6dceee04455a Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Thu, 15 Sep 2016 18:13:51 +0300 Subject: [PATCH 247/271] perf/x86/intel/pt: Fix kernel address filter's offset validation The kernel_ip() filter is used mostly by the DS/LBR code to look at the branch addresses, but Intel PT also uses it to validate the address filter offsets for kernel addresses, for which it is not sufficient: supplying something in bits 64:48 that's not a sign extension of the lower address bits (like 0xf00d000000000000) throws a #GP. This patch adds address validation for the user supplied kernel filters. Reported-by: Adrian Hunter Signed-off-by: Alexander Shishkin Acked-by: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: stable@vger.kernel.org # v4.7 Cc: stable@vger.kernel.org#v4.7 Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160915151352.21306-3-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/pt.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c index 5ec0100e3fc6..1f94963a283b 100644 --- a/arch/x86/events/intel/pt.c +++ b/arch/x86/events/intel/pt.c @@ -1074,6 +1074,11 @@ static void pt_addr_filters_fini(struct perf_event *event) event->hw.addr_filters = NULL; } +static inline bool valid_kernel_ip(unsigned long ip) +{ + return virt_addr_valid(ip) && kernel_ip(ip); +} + static int pt_event_addr_filters_validate(struct list_head *filters) { struct perf_addr_filter *filter; @@ -1084,7 +1089,7 @@ static int pt_event_addr_filters_validate(struct list_head *filters) if (!filter->range || !filter->size) return -EOPNOTSUPP; - if (!filter->inode && !kernel_ip(filter->offset)) + if (!filter->inode && !valid_kernel_ip(filter->offset)) return -EINVAL; if (++range > pt_cap_get(PT_CAP_num_address_ranges)) From 1155bafcb79208abc6ae234c6e135ac70607755c Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Thu, 15 Sep 2016 18:13:52 +0300 Subject: [PATCH 248/271] perf/x86/intel/pt: Do validate the size of a kernel address filter Right now, the kernel address filters in PT are prone to integer overflow that may happen in adding filter's size to its offset to obtain the end of the range. Such an overflow would also throw a #GP in the PT event configuration path. Fix this by explicitly validating the result of this calculation. Reported-by: Adrian Hunter Signed-off-by: Alexander Shishkin Acked-by: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: stable@vger.kernel.org # v4.7 Cc: stable@vger.kernel.org#v4.7 Cc: vince@deater.net Link: http://lkml.kernel.org/r/20160915151352.21306-4-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/pt.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c index 1f94963a283b..861a7d9cb60f 100644 --- a/arch/x86/events/intel/pt.c +++ b/arch/x86/events/intel/pt.c @@ -1089,8 +1089,13 @@ static int pt_event_addr_filters_validate(struct list_head *filters) if (!filter->range || !filter->size) return -EOPNOTSUPP; - if (!filter->inode && !valid_kernel_ip(filter->offset)) - return -EINVAL; + if (!filter->inode) { + if (!valid_kernel_ip(filter->offset)) + return -EINVAL; + + if (!valid_kernel_ip(filter->offset + filter->size)) + return -EINVAL; + } if (++range > pt_cap_get(PT_CAP_num_address_ranges)) return -EOPNOTSUPP; From 080fe0b790ad438fc1b61621dac37c1964ce7f35 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Wed, 24 Aug 2016 14:12:08 +0100 Subject: [PATCH 249/271] perf/x86/amd: Make HW_CACHE_REFERENCES and HW_CACHE_MISSES measure L2 While the Intel PMU monitors the LLC when perf enables the HW_CACHE_REFERENCES and HW_CACHE_MISSES events, these events monitor L1 instruction cache fetches (0x0080) and instruction cache misses (0x0081) on the AMD PMU. This is extremely confusing when monitoring the same workload across Intel and AMD machines, since parameters like, $ perf stat -e cache-references,cache-misses measure completely different things. Instead, make the AMD PMU measure instruction/data cache and TLB fill requests to the L2 and instruction/data cache and TLB misses in the L2 when HW_CACHE_REFERENCES and HW_CACHE_MISSES are enabled, respectively. That way the events measure unified caches on both platforms. Signed-off-by: Matt Fleming Acked-by: Peter Zijlstra Cc: Cc: Borislav Petkov Cc: Linus Torvalds Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1472044328-21302-1-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- arch/x86/events/amd/core.c | 4 ++-- arch/x86/kvm/pmu_amd.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/x86/events/amd/core.c b/arch/x86/events/amd/core.c index e07a22bb9308..f5f4b3fbbbc2 100644 --- a/arch/x86/events/amd/core.c +++ b/arch/x86/events/amd/core.c @@ -119,8 +119,8 @@ static const u64 amd_perfmon_event_map[PERF_COUNT_HW_MAX] = { [PERF_COUNT_HW_CPU_CYCLES] = 0x0076, [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, - [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0080, - [PERF_COUNT_HW_CACHE_MISSES] = 0x0081, + [PERF_COUNT_HW_CACHE_REFERENCES] = 0x077d, + [PERF_COUNT_HW_CACHE_MISSES] = 0x077e, [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c2, [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c3, [PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x00d0, /* "Decoder empty" event */ diff --git a/arch/x86/kvm/pmu_amd.c b/arch/x86/kvm/pmu_amd.c index 39b91127ef07..cd944435dfbd 100644 --- a/arch/x86/kvm/pmu_amd.c +++ b/arch/x86/kvm/pmu_amd.c @@ -23,8 +23,8 @@ static struct kvm_event_hw_type_mapping amd_event_mapping[] = { [0] = { 0x76, 0x00, PERF_COUNT_HW_CPU_CYCLES }, [1] = { 0xc0, 0x00, PERF_COUNT_HW_INSTRUCTIONS }, - [2] = { 0x80, 0x00, PERF_COUNT_HW_CACHE_REFERENCES }, - [3] = { 0x81, 0x00, PERF_COUNT_HW_CACHE_MISSES }, + [2] = { 0x7d, 0x07, PERF_COUNT_HW_CACHE_REFERENCES }, + [3] = { 0x7e, 0x07, PERF_COUNT_HW_CACHE_MISSES }, [4] = { 0xc2, 0x00, PERF_COUNT_HW_BRANCH_INSTRUCTIONS }, [5] = { 0xc3, 0x00, PERF_COUNT_HW_BRANCH_MISSES }, [6] = { 0xd0, 0x00, PERF_COUNT_HW_STALLED_CYCLES_FRONTEND }, From ed3d6d0ac036bcff6e4649a6f8fb987bb4e34444 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Fri, 16 Sep 2016 09:24:44 -0500 Subject: [PATCH 250/271] usb: musb: Fix tusb6010 compile error on blackfin We have CONFIG_BLACKFIN ifdef redefining all musb registers in musb_regs.h and tusb6010.h is never included causing a build error with blackfin-allmodconfig and COMPILE_TEST. Let's fix the issue by not building tusb6010 if CONFIG_BLACKFIN is selected. Reported-by: kbuild test robot Signed-off-by: Tony Lindgren Signed-off-by: Bin Liu Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/musb/Kconfig b/drivers/usb/musb/Kconfig index 886526b5fcdd..73cfa13fc0dc 100644 --- a/drivers/usb/musb/Kconfig +++ b/drivers/usb/musb/Kconfig @@ -87,7 +87,7 @@ config USB_MUSB_DA8XX config USB_MUSB_TUSB6010 tristate "TUSB6010" depends on HAS_IOMEM - depends on ARCH_OMAP2PLUS || COMPILE_TEST + depends on (ARCH_OMAP2PLUS || COMPILE_TEST) && !BLACKFIN depends on NOP_USB_XCEIV = USB_MUSB_HDRC # both built-in or both modules config USB_MUSB_OMAP2PLUS From 08c5cd37480f59ea39682f4585d92269be6b1424 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Fri, 16 Sep 2016 10:24:26 -0400 Subject: [PATCH 251/271] USB: change bInterval default to 10 ms Some full-speed mceusb infrared transceivers contain invalid endpoint descriptors for their interrupt endpoints, with bInterval set to 0. In the past they have worked out okay with the mceusb driver, because the driver sets the bInterval field in the descriptor to 1, overwriting whatever value may have been there before. However, this approach was never sanctioned by the USB core, and in fact it does not work with xHCI controllers, because they use the bInterval value that was present when the configuration was installed. Currently usbcore uses 32 ms as the default interval if the value in the endpoint descriptor is invalid. It turns out that these IR transceivers don't work properly unless the interval is set to 10 ms or below. To work around this mceusb problem, this patch changes the endpoint-descriptor parsing routine, making the default interval value be 10 ms rather than 32 ms. Signed-off-by: Alan Stern Tested-by: Wade Berrier CC: Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/config.c | 28 +++++++++++++++++----------- 1 file changed, 17 insertions(+), 11 deletions(-) diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c index 15ce4ab11688..a2d90aca779f 100644 --- a/drivers/usb/core/config.c +++ b/drivers/usb/core/config.c @@ -240,8 +240,10 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, int inum, memcpy(&endpoint->desc, d, n); INIT_LIST_HEAD(&endpoint->urb_list); - /* Fix up bInterval values outside the legal range. Use 32 ms if no - * proper value can be guessed. */ + /* + * Fix up bInterval values outside the legal range. + * Use 10 or 8 ms if no proper value can be guessed. + */ i = 0; /* i = min, j = max, n = default */ j = 255; if (usb_endpoint_xfer_int(d)) { @@ -250,13 +252,15 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, int inum, case USB_SPEED_SUPER_PLUS: case USB_SPEED_SUPER: case USB_SPEED_HIGH: - /* Many device manufacturers are using full-speed + /* + * Many device manufacturers are using full-speed * bInterval values in high-speed interrupt endpoint - * descriptors. Try to fix those and fall back to a - * 32 ms default value otherwise. */ + * descriptors. Try to fix those and fall back to an + * 8-ms default value otherwise. + */ n = fls(d->bInterval*8); if (n == 0) - n = 9; /* 32 ms = 2^(9-1) uframes */ + n = 7; /* 8 ms = 2^(7-1) uframes */ j = 16; /* @@ -271,10 +275,12 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, int inum, } break; default: /* USB_SPEED_FULL or _LOW */ - /* For low-speed, 10 ms is the official minimum. + /* + * For low-speed, 10 ms is the official minimum. * But some "overclocked" devices might want faster - * polling so we'll allow it. */ - n = 32; + * polling so we'll allow it. + */ + n = 10; break; } } else if (usb_endpoint_xfer_isoc(d)) { @@ -282,10 +288,10 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, int inum, j = 16; switch (to_usb_device(ddev)->speed) { case USB_SPEED_HIGH: - n = 9; /* 32 ms = 2^(9-1) uframes */ + n = 7; /* 8 ms = 2^(7-1) uframes */ break; default: /* USB_SPEED_FULL */ - n = 6; /* 32 ms = 2^(6-1) frames */ + n = 4; /* 8 ms = 2^(4-1) frames */ break; } } From 5e102b3b4fa944815aead89e63eec2a35069a07b Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Fri, 2 Sep 2016 23:46:53 +0300 Subject: [PATCH 252/271] IB/rxe: fix GFP_KERNEL in spinlock context There is skb_clone(skb, GFP_KERNEL) in spinlock context in rxe_rcv_mcast_pkt(). Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Acked-by: Moni Shoua Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rxe/rxe_recv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/sw/rxe/rxe_recv.c b/drivers/infiniband/sw/rxe/rxe_recv.c index 3d464c23e08b..144d2f129fcd 100644 --- a/drivers/infiniband/sw/rxe/rxe_recv.c +++ b/drivers/infiniband/sw/rxe/rxe_recv.c @@ -312,7 +312,7 @@ static void rxe_rcv_mcast_pkt(struct rxe_dev *rxe, struct sk_buff *skb) * make a copy of the skb to post to the next qp */ skb_copy = (mce->qp_list.next != &mcg->qp_list) ? - skb_clone(skb, GFP_KERNEL) : NULL; + skb_clone(skb, GFP_ATOMIC) : NULL; pkt->qp = qp; rxe_add_ref(qp); From 344bacca8cd811809fc33a249f2738ab757d327f Mon Sep 17 00:00:00 2001 From: Alex Vesker Date: Mon, 12 Sep 2016 09:55:28 +0300 Subject: [PATCH 253/271] IB/ipoib: Don't allow MC joins during light MC flush This fix solves a race between light flush and on the fly joins. Light flush doesn't set the device to down and unset IPOIB_OPER_UP flag, this means that if while flushing we have a MC join in progress and the QP was attached to BC MGID we can have a mismatches when re-attaching a QP to the BC MGID. The light flush would set the broadcast group to NULL causing an on the fly join to rejoin and reattach to the BC MCG as well as adding the BC MGID to the multicast list. The flush process would later on remove the BC MGID and detach it from the QP. On the next flush the BC MGID is present in the multicast list but not found when trying to detach it because of the previous double attach and single detach. [18332.714265] ------------[ cut here ]------------ [18332.717775] WARNING: CPU: 6 PID: 3767 at drivers/infiniband/core/verbs.c:280 ib_dealloc_pd+0xff/0x120 [ib_core] ... [18332.775198] Hardware name: Red Hat KVM, BIOS Bochs 01/01/2011 [18332.779411] 0000000000000000 ffff8800b50dfbb0 ffffffff813fed47 0000000000000000 [18332.784960] 0000000000000000 ffff8800b50dfbf0 ffffffff8109add1 0000011832f58300 [18332.790547] ffff880226a596c0 ffff880032482000 ffff880032482830 ffff880226a59280 [18332.796199] Call Trace: [18332.798015] [] dump_stack+0x63/0x8c [18332.801831] [] __warn+0xd1/0xf0 [18332.805403] [] warn_slowpath_null+0x1d/0x20 [18332.809706] [] ib_dealloc_pd+0xff/0x120 [ib_core] [18332.814384] [] ipoib_transport_dev_cleanup+0xfc/0x1d0 [ib_ipoib] [18332.820031] [] ipoib_ib_dev_cleanup+0x98/0x110 [ib_ipoib] [18332.825220] [] ipoib_dev_cleanup+0x2d8/0x550 [ib_ipoib] [18332.830290] [] ipoib_uninit+0x2f/0x40 [ib_ipoib] [18332.834911] [] rollback_registered_many+0x1aa/0x2c0 [18332.839741] [] rollback_registered+0x31/0x40 [18332.844091] [] unregister_netdevice_queue+0x48/0x80 [18332.848880] [] ipoib_vlan_delete+0x1fb/0x290 [ib_ipoib] [18332.853848] [] delete_child+0x7d/0xf0 [ib_ipoib] [18332.858474] [] dev_attr_store+0x18/0x30 [18332.862510] [] sysfs_kf_write+0x3a/0x50 [18332.866349] [] kernfs_fop_write+0x120/0x170 [18332.870471] [] __vfs_write+0x28/0xe0 [18332.874152] [] ? percpu_down_read+0x1f/0x50 [18332.878274] [] vfs_write+0xa2/0x1a0 [18332.881896] [] SyS_write+0x46/0xa0 [18332.885632] [] do_syscall_64+0x57/0xb0 [18332.889709] [] entry_SYSCALL64_slow_path+0x25/0x25 [18332.894727] ---[ end trace 09ebbe31f831ef17 ]--- Fixes: ee1e2c82c245 ("IPoIB: Refresh paths instead of flushing them on SM change events") Signed-off-by: Alex Vesker Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/ipoib/ipoib_ib.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index dc6d241b9406..be11d5d5b8c1 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -1161,8 +1161,17 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, } if (level == IPOIB_FLUSH_LIGHT) { + int oper_up; ipoib_mark_paths_invalid(dev); + /* Set IPoIB operation as down to prevent races between: + * the flush flow which leaves MCG and on the fly joins + * which can happen during that time. mcast restart task + * should deal with join requests we missed. + */ + oper_up = test_and_clear_bit(IPOIB_FLAG_OPER_UP, &priv->flags); ipoib_mcast_dev_flush(dev); + if (oper_up) + set_bit(IPOIB_FLAG_OPER_UP, &priv->flags); ipoib_flush_ah(dev); } From e5ac40cd66c2f3cd11bc5edc658f012661b16347 Mon Sep 17 00:00:00 2001 From: Alex Vesker Date: Mon, 12 Sep 2016 19:16:18 +0300 Subject: [PATCH 254/271] IB/mlx4: Fix incorrect MC join state bit-masking on SR-IOV Because of an incorrect bit-masking done on the join state bits, when handling a join request we failed to detect a difference between the group join state and the request join state when joining as send only full member (0x8). This caused the MC join request not to be sent. This issue is relevant only when SRIOV is enabled and SM supports send only full member. This fix separates scope bits and join states bits a nibble each. Fixes: b9c5d6a64358 ('IB/mlx4: Add multicast group (MCG) paravirtualization for SR-IOV') Signed-off-by: Alex Vesker Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx4/mcg.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c index 8f7ad07915b0..097bfcc4ee99 100644 --- a/drivers/infiniband/hw/mlx4/mcg.c +++ b/drivers/infiniband/hw/mlx4/mcg.c @@ -489,7 +489,7 @@ static u8 get_leave_state(struct mcast_group *group) if (!group->members[i]) leave_state |= (1 << i); - return leave_state & (group->rec.scope_join_state & 7); + return leave_state & (group->rec.scope_join_state & 0xf); } static int join_group(struct mcast_group *group, int slave, u8 join_mask) @@ -564,8 +564,8 @@ static void mlx4_ib_mcg_timeout_handler(struct work_struct *work) } else mcg_warn_group(group, "DRIVER BUG\n"); } else if (group->state == MCAST_LEAVE_SENT) { - if (group->rec.scope_join_state & 7) - group->rec.scope_join_state &= 0xf8; + if (group->rec.scope_join_state & 0xf) + group->rec.scope_join_state &= 0xf0; group->state = MCAST_IDLE; mutex_unlock(&group->lock); if (release_group(group, 1)) @@ -605,7 +605,7 @@ static int handle_leave_req(struct mcast_group *group, u8 leave_mask, static int handle_join_req(struct mcast_group *group, u8 join_mask, struct mcast_req *req) { - u8 group_join_state = group->rec.scope_join_state & 7; + u8 group_join_state = group->rec.scope_join_state & 0xf; int ref = 0; u16 status; struct ib_sa_mcmember_data *sa_data = (struct ib_sa_mcmember_data *)req->sa_mad.data; @@ -690,8 +690,8 @@ static void mlx4_ib_mcg_work_handler(struct work_struct *work) u8 cur_join_state; resp_join_state = ((struct ib_sa_mcmember_data *) - group->response_sa_mad.data)->scope_join_state & 7; - cur_join_state = group->rec.scope_join_state & 7; + group->response_sa_mad.data)->scope_join_state & 0xf; + cur_join_state = group->rec.scope_join_state & 0xf; if (method == IB_MGMT_METHOD_GET_RESP) { /* successfull join */ @@ -710,7 +710,7 @@ process_requests: req = list_first_entry(&group->pending_list, struct mcast_req, group_list); sa_data = (struct ib_sa_mcmember_data *)req->sa_mad.data; - req_join_state = sa_data->scope_join_state & 0x7; + req_join_state = sa_data->scope_join_state & 0xf; /* For a leave request, we will immediately answer the VF, and * update our internal counters. The actual leave will be sent From baa0be7026e2f7d1d40bfd45909044169e9e3c68 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Mon, 12 Sep 2016 19:16:19 +0300 Subject: [PATCH 255/271] IB/mlx4: Fix code indentation in QP1 MAD flow The indentation in the QP1 GRH flow in procedure build_mlx_header is really confusing. Fix it, in preparation for a commit which touches this code. Fixes: 1ffeb2eb8be9 ("IB/mlx4: SR-IOV IB context objects and proxy/tunnel SQP support") Signed-off-by: Jack Morgenstein Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx4/qp.c | 36 +++++++++++++++++---------------- 1 file changed, 19 insertions(+), 17 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 768085f59566..e398c04903fa 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -2493,24 +2493,26 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_ud_wr *wr, sqp->ud_header.grh.flow_label = ah->av.ib.sl_tclass_flowlabel & cpu_to_be32(0xfffff); sqp->ud_header.grh.hop_limit = ah->av.ib.hop_limit; - if (is_eth) + if (is_eth) { memcpy(sqp->ud_header.grh.source_gid.raw, sgid.raw, 16); - else { - if (mlx4_is_mfunc(to_mdev(ib_dev)->dev)) { - /* When multi-function is enabled, the ib_core gid - * indexes don't necessarily match the hw ones, so - * we must use our own cache */ - sqp->ud_header.grh.source_gid.global.subnet_prefix = - to_mdev(ib_dev)->sriov.demux[sqp->qp.port - 1]. - subnet_prefix; - sqp->ud_header.grh.source_gid.global.interface_id = - to_mdev(ib_dev)->sriov.demux[sqp->qp.port - 1]. - guid_cache[ah->av.ib.gid_index]; - } else - ib_get_cached_gid(ib_dev, - be32_to_cpu(ah->av.ib.port_pd) >> 24, - ah->av.ib.gid_index, - &sqp->ud_header.grh.source_gid, NULL); + } else { + if (mlx4_is_mfunc(to_mdev(ib_dev)->dev)) { + /* When multi-function is enabled, the ib_core gid + * indexes don't necessarily match the hw ones, so + * we must use our own cache + */ + sqp->ud_header.grh.source_gid.global.subnet_prefix = + to_mdev(ib_dev)->sriov.demux[sqp->qp.port - 1]. + subnet_prefix; + sqp->ud_header.grh.source_gid.global.interface_id = + to_mdev(ib_dev)->sriov.demux[sqp->qp.port - 1]. + guid_cache[ah->av.ib.gid_index]; + } else { + ib_get_cached_gid(ib_dev, + be32_to_cpu(ah->av.ib.port_pd) >> 24, + ah->av.ib.gid_index, + &sqp->ud_header.grh.source_gid, NULL); + } } memcpy(sqp->ud_header.grh.destination_gid.raw, ah->av.ib.dgid, 16); From 8ec07bf8a8b57d6c58927a16a0a22c0115cf2855 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Mon, 12 Sep 2016 19:16:20 +0300 Subject: [PATCH 256/271] IB/mlx4: Use correct subnet-prefix in QP1 mads under SR-IOV When sending QP1 MAD packets which use a GRH, the source GID (which consists of the 64-bit subnet prefix, and the 64 bit port GUID) must be included in the packet GRH. For SR-IOV, a GID cache is used, since the source GID needs to be the slave's source GID, and not the Hypervisor's GID. This cache also included a subnet_prefix. Unfortunately, the subnet_prefix field in the cache was never initialized (to the default subnet prefix 0xfe80::0). As a result, this field remained all zeroes. Therefore, when SR-IOV was active, all QP1 packets which included a GRH had a source GID subnet prefix of all-zeroes. However, the subnet-prefix should initially be 0xfe80::0 (the default subnet prefix). In addition, if OpenSM modifies a port's subnet prefix, the new subnet prefix must be used in the GRH when sending QP1 packets. To fix this we now initialize the subnet prefix in the SR-IOV GID cache to the default subnet prefix. We update the cached value if/when OpenSM modifies the port's subnet prefix. We take this cached value when sending QP1 packets when SR-IOV is active. Note that the value is stored as an atomic64. This eliminates any need for locking when the subnet prefix is being updated. Note also that we depend on the FW generating the "port management change" event for tracking subnet-prefix changes performed by OpenSM. If running early FW (before 2.9.4630), subnet prefix changes will not be tracked (but the default subnet prefix still will be stored in the cache; therefore users who do not modify the subnet prefix will not have a problem). IF there is a need for such tracking also for early FW, we will add that capability in a subsequent patch. Fixes: 1ffeb2eb8be9 ("IB/mlx4: SR-IOV IB context objects and proxy/tunnel SQP support") Signed-off-by: Jack Morgenstein Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx4/mad.c | 23 +++++++++++++++++++++++ drivers/infiniband/hw/mlx4/mlx4_ib.h | 2 +- drivers/infiniband/hw/mlx4/qp.c | 5 +++-- 3 files changed, 27 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c index 9c2e53d28f98..0f21c3a25552 100644 --- a/drivers/infiniband/hw/mlx4/mad.c +++ b/drivers/infiniband/hw/mlx4/mad.c @@ -1128,6 +1128,27 @@ void handle_port_mgmt_change_event(struct work_struct *work) /* Generate GUID changed event */ if (changed_attr & MLX4_EQ_PORT_INFO_GID_PFX_CHANGE_MASK) { + if (mlx4_is_master(dev->dev)) { + union ib_gid gid; + int err = 0; + + if (!eqe->event.port_mgmt_change.params.port_info.gid_prefix) + err = __mlx4_ib_query_gid(&dev->ib_dev, port, 0, &gid, 1); + else + gid.global.subnet_prefix = + eqe->event.port_mgmt_change.params.port_info.gid_prefix; + if (err) { + pr_warn("Could not change QP1 subnet prefix for port %d: query_gid error (%d)\n", + port, err); + } else { + pr_debug("Changing QP1 subnet prefix for port %d. old=0x%llx. new=0x%llx\n", + port, + (u64)atomic64_read(&dev->sriov.demux[port - 1].subnet_prefix), + be64_to_cpu(gid.global.subnet_prefix)); + atomic64_set(&dev->sriov.demux[port - 1].subnet_prefix, + be64_to_cpu(gid.global.subnet_prefix)); + } + } mlx4_ib_dispatch_event(dev, port, IB_EVENT_GID_CHANGE); /*if master, notify all slaves*/ if (mlx4_is_master(dev->dev)) @@ -2202,6 +2223,8 @@ int mlx4_ib_init_sriov(struct mlx4_ib_dev *dev) if (err) goto demux_err; dev->sriov.demux[i].guid_cache[0] = gid.global.interface_id; + atomic64_set(&dev->sriov.demux[i].subnet_prefix, + be64_to_cpu(gid.global.subnet_prefix)); err = alloc_pv_object(dev, mlx4_master_func_num(dev->dev), i + 1, &dev->sriov.sqps[i]); if (err) diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h index 7c5832ede4bd..686ab48ff644 100644 --- a/drivers/infiniband/hw/mlx4/mlx4_ib.h +++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h @@ -448,7 +448,7 @@ struct mlx4_ib_demux_ctx { struct workqueue_struct *wq; struct workqueue_struct *ud_wq; spinlock_t ud_lock; - __be64 subnet_prefix; + atomic64_t subnet_prefix; __be64 guid_cache[128]; struct mlx4_ib_dev *dev; /* the following lock protects both mcg_table and mcg_mgid0_list */ diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index e398c04903fa..7fb9629bd12b 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -2502,8 +2502,9 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_ud_wr *wr, * we must use our own cache */ sqp->ud_header.grh.source_gid.global.subnet_prefix = - to_mdev(ib_dev)->sriov.demux[sqp->qp.port - 1]. - subnet_prefix; + cpu_to_be64(atomic64_read(&(to_mdev(ib_dev)->sriov. + demux[sqp->qp.port - 1]. + subnet_prefix))); sqp->ud_header.grh.source_gid.global.interface_id = to_mdev(ib_dev)->sriov.demux[sqp->qp.port - 1]. guid_cache[ah->av.ib.gid_index]; From 69d269d38910e697e461ec5677368f57d2046cbe Mon Sep 17 00:00:00 2001 From: Kamal Heib Date: Mon, 12 Sep 2016 19:16:22 +0300 Subject: [PATCH 257/271] IB/mlx4: Diagnostic HW counters are not supported in slave mode Modify the mlx4_ib_diag_counters() to avoid the following error in the hypervisor when the slave tries to query the hardware counters in SR-IOV mode. mlx4_core 0000:81:00.0: Unknown command:0x30 accepted from slave:1 Fixes: 3f85f2aaabf7 ("IB/mlx4: Add diagnostic hardware counters") Signed-off-by: Kamal Heib Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx4/main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 2af44c2de262..87ba9bca4181 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -2202,6 +2202,9 @@ static int mlx4_ib_alloc_diag_counters(struct mlx4_ib_dev *ibdev) bool per_port = !!(ibdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_DIAG_PER_PORT); + if (mlx4_is_slave(ibdev->dev)) + return 0; + for (i = 0; i < MLX4_DIAG_COUNTERS_TYPES; i++) { /* i == 1 means we are building port counters */ if (i && !per_port) From 7fae6655a0c897875bd34501ec092232b526d3e4 Mon Sep 17 00:00:00 2001 From: Noa Osherovich Date: Mon, 12 Sep 2016 19:16:23 +0300 Subject: [PATCH 258/271] IB/mlx5: Enable MAD_IFC commands for IB ports only MAD_IFC command is supported only for physical functions (PF) and when physical port is IB. The proposed fix enforces it. Fixes: d603c809ef91 ("IB/mlx5: Fix decision on using MAD_IFC") Reported-by: David Chang Signed-off-by: Noa Osherovich Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/main.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 8150ea372c53..0480b643fb9c 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -288,7 +288,9 @@ __be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev, u8 port_num, static int mlx5_use_mad_ifc(struct mlx5_ib_dev *dev) { - return !MLX5_CAP_GEN(dev->mdev, ib_virt); + if (MLX5_CAP_GEN(dev->mdev, port_type) == MLX5_CAP_PORT_TYPE_IB) + return !MLX5_CAP_GEN(dev->mdev, ib_virt); + return 0; } enum { From ee3da804ad1b1dd4c766199a6e8443542b0aaaef Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 12 Sep 2016 19:16:24 +0300 Subject: [PATCH 259/271] IB/mlx5: Set source mac address in FTE Set the source mac address in the FTE when L2 specification is provided. Fixes: 038d2ef87572 ('IB/mlx5: Add flow steering support') Signed-off-by: Maor Gottlieb Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/main.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 0480b643fb9c..e19537cf44ab 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -1430,6 +1430,13 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, dmac_47_16), ib_spec->eth.val.dst_mac); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + smac_47_16), + ib_spec->eth.mask.src_mac); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + smac_47_16), + ib_spec->eth.val.src_mac); + if (ib_spec->eth.mask.vlan_tag) { MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, vlan_tag, 1); From dfdd6158ca2c8838313470c5ab79243d04da70b2 Mon Sep 17 00:00:00 2001 From: Yonatan Cohen Date: Wed, 7 Sep 2016 14:04:04 +0300 Subject: [PATCH 260/271] IB/rxe: Fix kernel panic in udp_setup_tunnel Disable creation of a UDP socket for ipv6 when CONFIG_IPV6 is not enabeld. Since udp_sock_create6() returns 0 when CONFIG_IPV6 is not set [ 46.888632] IP: [] setup_udp_tunnel_sock+0x6/0x4f [ 46.891355] *pdpt = 0000000000000000 *pde = f000ff53f000ff53 [ 46.893918] Oops: 0002 [#1] PREEMPT [ 46.896014] CPU: 0 PID: 1 Comm: swapper Not tainted 4.7.0-rc4-00001-g8700e3e #1 [ 46.900280] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Debian-1.8.2-1 04/01/2014 [ 46.904905] task: cf06c040 ti: cf05e000 task.ti: cf05e000 [ 46.907854] EIP: 0060:[] EFLAGS: 00210246 CPU: 0 [ 46.911137] EIP is at setup_udp_tunnel_sock+0x6/0x4f [ 46.914070] EAX: 00000044 EBX: 00000001 ECX: cf05fef0 EDX: ca8142e0 [ 46.917236] ESI: c2c4505b EDI: cf05fef0 EBP: cf05fed0 ESP: cf05fed0 [ 46.919836] DS: 007b ES: 007b FS: 0000 GS: 00e0 SS: 0068 [ 46.922046] CR0: 80050033 CR2: 000001fc CR3: 02cec000 CR4: 000006b0 [ 46.924550] Stack: [ 46.926014] cf05ff10 c1fd4657 ca8142e0 0000000a 00000000 00000000 0000b712 00000008 [ 46.931274] 00000000 6bb5bd01 c1fd48de 00000000 00000000 cf05ff1c 00000000 00000000 [ 46.936122] cf05ff1c c1fd4bdf 00000000 cf05ff28 c2c4507b ffffffff cf05ff88 c2bf1c74 [ 46.942350] Call Trace: [ 46.944403] [] rxe_setup_udp_tunnel+0x8f/0x99 [ 46.947689] [] ? net_to_rxe+0x4e/0x4e [ 46.950567] [] rxe_net_init+0xe/0xa4 [ 46.953147] [] rxe_module_init+0x20/0x4c [ 46.955448] [] do_one_initcall+0x89/0x113 [ 46.957797] [] ? set_debug_rodata+0xf/0xf [ 46.959966] [] ? kernel_init_freeable+0xbe/0x15b [ 46.962262] [] kernel_init_freeable+0xde/0x15b [ 46.964418] [] kernel_init+0x8/0xd0 [ 46.966618] [] ret_from_kernel_thread+0xe/0x24 [ 46.969592] [] ? rest_init+0x6f/0x6f Fixes: 8700e3e7c485 ("Soft RoCE driver") Signed-off-by: Yonatan Cohen Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rxe/rxe.c | 25 +++++++++++-- drivers/infiniband/sw/rxe/rxe_net.c | 57 +++++++++++++---------------- drivers/infiniband/sw/rxe/rxe_net.h | 5 ++- 3 files changed, 52 insertions(+), 35 deletions(-) diff --git a/drivers/infiniband/sw/rxe/rxe.c b/drivers/infiniband/sw/rxe/rxe.c index 55f0e8f0ca79..ddd59270ff6d 100644 --- a/drivers/infiniband/sw/rxe/rxe.c +++ b/drivers/infiniband/sw/rxe/rxe.c @@ -362,15 +362,34 @@ static int __init rxe_module_init(void) return err; } - err = rxe_net_init(); + err = rxe_net_ipv4_init(); if (err) { - pr_err("rxe: unable to init\n"); + pr_err("rxe: unable to init ipv4 tunnel\n"); rxe_cache_exit(); - return err; + goto exit; } + + err = rxe_net_ipv6_init(); + if (err) { + pr_err("rxe: unable to init ipv6 tunnel\n"); + rxe_cache_exit(); + goto exit; + } + + err = register_netdevice_notifier(&rxe_net_notifier); + if (err) { + pr_err("rxe: Failed to rigister netdev notifier\n"); + goto exit; + } + pr_info("rxe: loaded\n"); return 0; + +exit: + rxe_release_udp_tunnel(recv_sockets.sk4); + rxe_release_udp_tunnel(recv_sockets.sk6); + return err; } static void __exit rxe_module_exit(void) diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index 0b8d2ea8b41d..eedf2f1cafdf 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -275,9 +275,10 @@ static struct socket *rxe_setup_udp_tunnel(struct net *net, __be16 port, return sock; } -static void rxe_release_udp_tunnel(struct socket *sk) +void rxe_release_udp_tunnel(struct socket *sk) { - udp_tunnel_sock_release(sk); + if (sk) + udp_tunnel_sock_release(sk); } static void prepare_udp_hdr(struct sk_buff *skb, __be16 src_port, @@ -658,51 +659,45 @@ out: return NOTIFY_OK; } -static struct notifier_block rxe_net_notifier = { +struct notifier_block rxe_net_notifier = { .notifier_call = rxe_notify, }; -int rxe_net_init(void) +int rxe_net_ipv4_init(void) { - int err; + spin_lock_init(&dev_list_lock); + + recv_sockets.sk4 = rxe_setup_udp_tunnel(&init_net, + htons(ROCE_V2_UDP_DPORT), false); + if (IS_ERR(recv_sockets.sk4)) { + recv_sockets.sk4 = NULL; + pr_err("rxe: Failed to create IPv4 UDP tunnel\n"); + return -1; + } + + return 0; +} + +int rxe_net_ipv6_init(void) +{ +#if IS_ENABLED(CONFIG_IPV6) spin_lock_init(&dev_list_lock); recv_sockets.sk6 = rxe_setup_udp_tunnel(&init_net, - htons(ROCE_V2_UDP_DPORT), true); + htons(ROCE_V2_UDP_DPORT), true); if (IS_ERR(recv_sockets.sk6)) { recv_sockets.sk6 = NULL; pr_err("rxe: Failed to create IPv6 UDP tunnel\n"); return -1; } - - recv_sockets.sk4 = rxe_setup_udp_tunnel(&init_net, - htons(ROCE_V2_UDP_DPORT), false); - if (IS_ERR(recv_sockets.sk4)) { - rxe_release_udp_tunnel(recv_sockets.sk6); - recv_sockets.sk4 = NULL; - recv_sockets.sk6 = NULL; - pr_err("rxe: Failed to create IPv4 UDP tunnel\n"); - return -1; - } - - err = register_netdevice_notifier(&rxe_net_notifier); - if (err) { - rxe_release_udp_tunnel(recv_sockets.sk6); - rxe_release_udp_tunnel(recv_sockets.sk4); - pr_err("rxe: Failed to rigister netdev notifier\n"); - } - - return err; +#endif + return 0; } void rxe_net_exit(void) { - if (recv_sockets.sk6) - rxe_release_udp_tunnel(recv_sockets.sk6); - - if (recv_sockets.sk4) - rxe_release_udp_tunnel(recv_sockets.sk4); - + rxe_release_udp_tunnel(recv_sockets.sk6); + rxe_release_udp_tunnel(recv_sockets.sk4); unregister_netdevice_notifier(&rxe_net_notifier); } diff --git a/drivers/infiniband/sw/rxe/rxe_net.h b/drivers/infiniband/sw/rxe/rxe_net.h index 7b06f76d16cc..0daf7f09e5b5 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.h +++ b/drivers/infiniband/sw/rxe/rxe_net.h @@ -44,10 +44,13 @@ struct rxe_recv_sockets { }; extern struct rxe_recv_sockets recv_sockets; +extern struct notifier_block rxe_net_notifier; +void rxe_release_udp_tunnel(struct socket *sk); struct rxe_dev *rxe_net_add(struct net_device *ndev); -int rxe_net_init(void); +int rxe_net_ipv4_init(void); +int rxe_net_ipv6_init(void); void rxe_net_exit(void); #endif /* RXE_NET_H */ From 908948877bbbb240f67d7eb27d3a392beb394a07 Mon Sep 17 00:00:00 2001 From: Yonatan Cohen Date: Wed, 7 Sep 2016 14:04:05 +0300 Subject: [PATCH 261/271] IB/rxe: Fix duplicate atomic request handling When handling ack for atomic opcodes like "fetch&add" or "cmp&swp", the method send_atomic_ack() saves the ack before sending it, in case it gets lost and never reach the requester. In which case the method duplicate_request() will need to find it using the duplicated request.psn. But send_atomic_ack() used a wrong psn value and thus the above ack was never found. This fix uses the ack.psn to locate the ack in case its needed. This fix also copies the ack packet to the skb's control buffer since duplicate_request() will need it when calling rxe_xmit_packet() Fixes: 8700e3e7c485 ("Soft RoCE driver") Signed-off-by: Yonatan Cohen Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rxe/rxe_resp.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index ebb03b46e2ad..3e0f0f2baace 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -972,11 +972,13 @@ static int send_atomic_ack(struct rxe_qp *qp, struct rxe_pkt_info *pkt, free_rd_atomic_resource(qp, res); rxe_advance_resp_resource(qp); + memcpy(SKB_TO_PKT(skb), &ack_pkt, sizeof(skb->cb)); + res->type = RXE_ATOMIC_MASK; res->atomic.skb = skb; - res->first_psn = qp->resp.psn; - res->last_psn = qp->resp.psn; - res->cur_psn = qp->resp.psn; + res->first_psn = ack_pkt.psn; + res->last_psn = ack_pkt.psn; + res->cur_psn = ack_pkt.psn; rc = rxe_xmit_packet(rxe, qp, &ack_pkt, skb_copy); if (rc) { @@ -1116,8 +1118,7 @@ static enum resp_states duplicate_request(struct rxe_qp *qp, rc = RESPST_CLEANUP; goto out; } - bth_set_psn(SKB_TO_PKT(skb_copy), - qp->resp.psn - 1); + /* Resend the result. */ rc = rxe_xmit_packet(to_rdev(qp->ibqp.device), qp, pkt, skb_copy); From 3050b99850247695cb07a5c15265afcc08bcf400 Mon Sep 17 00:00:00 2001 From: Yonatan Cohen Date: Wed, 7 Sep 2016 14:04:06 +0300 Subject: [PATCH 262/271] IB/rxe: Fix race condition between requester and completer rxe_requester() is sending a pkt with rxe_xmit_packet() and then calls rxe_update() to update the wqe and qp's psn values. But sometimes the response is received before the requester had time to update the wqe in which case the completer acts on errornous wqe values. This fix updates the wqe and qp before actually sending the request and rolls back when xmit fails. Fixes: 8700e3e7c485 ("Soft RoCE driver") Signed-off-by: Yonatan Cohen Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rxe/rxe_req.c | 57 ++++++++++++++++++++++------- 1 file changed, 44 insertions(+), 13 deletions(-) diff --git a/drivers/infiniband/sw/rxe/rxe_req.c b/drivers/infiniband/sw/rxe/rxe_req.c index 33b2d9d77021..13a848a518e8 100644 --- a/drivers/infiniband/sw/rxe/rxe_req.c +++ b/drivers/infiniband/sw/rxe/rxe_req.c @@ -511,24 +511,21 @@ static int fill_packet(struct rxe_qp *qp, struct rxe_send_wqe *wqe, } static void update_wqe_state(struct rxe_qp *qp, - struct rxe_send_wqe *wqe, - struct rxe_pkt_info *pkt, - enum wqe_state *prev_state) + struct rxe_send_wqe *wqe, + struct rxe_pkt_info *pkt) { - enum wqe_state prev_state_ = wqe->state; - if (pkt->mask & RXE_END_MASK) { if (qp_type(qp) == IB_QPT_RC) wqe->state = wqe_state_pending; } else { wqe->state = wqe_state_processing; } - - *prev_state = prev_state_; } -static void update_state(struct rxe_qp *qp, struct rxe_send_wqe *wqe, - struct rxe_pkt_info *pkt, int payload) +static void update_wqe_psn(struct rxe_qp *qp, + struct rxe_send_wqe *wqe, + struct rxe_pkt_info *pkt, + int payload) { /* number of packets left to send including current one */ int num_pkt = (wqe->dma.resid + payload + qp->mtu - 1) / qp->mtu; @@ -546,10 +543,35 @@ static void update_state(struct rxe_qp *qp, struct rxe_send_wqe *wqe, qp->req.psn = (wqe->first_psn + num_pkt) & BTH_PSN_MASK; else qp->req.psn = (qp->req.psn + 1) & BTH_PSN_MASK; +} +static void save_state(struct rxe_send_wqe *wqe, + struct rxe_qp *qp, + struct rxe_send_wqe *rollback_wqe, + struct rxe_qp *rollback_qp) +{ + rollback_wqe->state = wqe->state; + rollback_wqe->first_psn = wqe->first_psn; + rollback_wqe->last_psn = wqe->last_psn; + rollback_qp->req.psn = qp->req.psn; +} + +static void rollback_state(struct rxe_send_wqe *wqe, + struct rxe_qp *qp, + struct rxe_send_wqe *rollback_wqe, + struct rxe_qp *rollback_qp) +{ + wqe->state = rollback_wqe->state; + wqe->first_psn = rollback_wqe->first_psn; + wqe->last_psn = rollback_wqe->last_psn; + qp->req.psn = rollback_qp->req.psn; +} + +static void update_state(struct rxe_qp *qp, struct rxe_send_wqe *wqe, + struct rxe_pkt_info *pkt, int payload) +{ qp->req.opcode = pkt->opcode; - if (pkt->mask & RXE_END_MASK) qp->req.wqe_index = next_index(qp->sq.queue, qp->req.wqe_index); @@ -571,7 +593,8 @@ int rxe_requester(void *arg) int mtu; int opcode; int ret; - enum wqe_state prev_state; + struct rxe_qp rollback_qp; + struct rxe_send_wqe rollback_wqe; next_wqe: if (unlikely(!qp->valid || qp->req.state == QP_STATE_ERROR)) @@ -688,13 +711,21 @@ next_wqe: goto err; } - update_wqe_state(qp, wqe, &pkt, &prev_state); + /* + * To prevent a race on wqe access between requester and completer, + * wqe members state and psn need to be set before calling + * rxe_xmit_packet(). + * Otherwise, completer might initiate an unjustified retry flow. + */ + save_state(wqe, qp, &rollback_wqe, &rollback_qp); + update_wqe_state(qp, wqe, &pkt); + update_wqe_psn(qp, wqe, &pkt, payload); ret = rxe_xmit_packet(to_rdev(qp->ibqp.device), qp, &pkt, skb); if (ret) { qp->need_req_skb = 1; kfree_skb(skb); - wqe->state = prev_state; + rollback_state(wqe, qp, &rollback_wqe, &rollback_qp); if (ret == -EAGAIN) { rxe_run_task(&qp->req.task, 1); From c1cc72cb6ff95cbe02ed747866576dccab4a5911 Mon Sep 17 00:00:00 2001 From: Yonatan Cohen Date: Wed, 7 Sep 2016 14:04:07 +0300 Subject: [PATCH 263/271] IB/rxe: Fix kmem_cache leak Decrement qp reference when handling error path in completer to prevent kmem_cache leak. Fixes: 8700e3e7c485 ("Soft RoCE driver") Signed-off-by: Yonatan Cohen Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rxe/rxe_comp.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/infiniband/sw/rxe/rxe_comp.c b/drivers/infiniband/sw/rxe/rxe_comp.c index 36f67de44095..1c59ef2c67aa 100644 --- a/drivers/infiniband/sw/rxe/rxe_comp.c +++ b/drivers/infiniband/sw/rxe/rxe_comp.c @@ -689,7 +689,14 @@ int rxe_completer(void *arg) qp->req.need_retry = 1; rxe_run_task(&qp->req.task, 1); } + + if (pkt) { + rxe_drop_ref(pkt->qp); + kfree_skb(skb); + } + goto exit; + } else { wqe->status = IB_WC_RETRY_EXC_ERR; state = COMPST_ERROR; @@ -716,6 +723,12 @@ int rxe_completer(void *arg) case COMPST_ERROR: do_complete(qp, wqe); rxe_qp_error(qp); + + if (pkt) { + rxe_drop_ref(pkt->qp); + kfree_skb(skb); + } + goto exit; } } From e4618d40eb3dc1a6d1f55f7150ea25bb23ab410a Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Fri, 9 Sep 2016 08:15:37 +0100 Subject: [PATCH 264/271] IB/rdmavt: Don't vfree a kzalloc'ed memory region The userspace memory region 'mr' is allocated with kzalloc in __rvt_alloc_mr however it is incorrectly being freed with vfree in __rvt_free_mr. Fix this by using kfree to free it. Signed-off-by: Colin Ian King Reviewed-by: Leon Romanovsky Acked-by: Dennis Dalessandro Signed-off-by: Doug Ledford --- drivers/infiniband/sw/rdmavt/mr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/sw/rdmavt/mr.c b/drivers/infiniband/sw/rdmavt/mr.c index 80c4b6b401b8..46b64970058e 100644 --- a/drivers/infiniband/sw/rdmavt/mr.c +++ b/drivers/infiniband/sw/rdmavt/mr.c @@ -294,7 +294,7 @@ static void __rvt_free_mr(struct rvt_mr *mr) { rvt_deinit_mregion(&mr->mr); rvt_free_lkey(&mr->mr); - vfree(mr); + kfree(mr); } /** From d4690f1e1cdabb4d61207b6787b1605a0dc0aeab Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 16 Sep 2016 00:11:45 +0100 Subject: [PATCH 265/271] fix iov_iter_fault_in_readable() ... by turning it into what used to be multipages counterpart Cc: stable@vger.kernel.org Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/uio.h | 2 +- lib/iov_iter.c | 24 ++---------------------- 2 files changed, 3 insertions(+), 23 deletions(-) diff --git a/include/linux/uio.h b/include/linux/uio.h index 1b5d1cd796e2..75b4aaf31a9d 100644 --- a/include/linux/uio.h +++ b/include/linux/uio.h @@ -76,7 +76,7 @@ size_t iov_iter_copy_from_user_atomic(struct page *page, struct iov_iter *i, unsigned long offset, size_t bytes); void iov_iter_advance(struct iov_iter *i, size_t bytes); int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes); -int iov_iter_fault_in_multipages_readable(struct iov_iter *i, size_t bytes); +#define iov_iter_fault_in_multipages_readable iov_iter_fault_in_readable size_t iov_iter_single_seg_count(const struct iov_iter *i); size_t copy_page_to_iter(struct page *page, size_t offset, size_t bytes, struct iov_iter *i); diff --git a/lib/iov_iter.c b/lib/iov_iter.c index 9e8c7386b3a0..7e3138cfc8c9 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -290,26 +290,6 @@ done: return wanted - bytes; } -/* - * Fault in the first iovec of the given iov_iter, to a maximum length - * of bytes. Returns 0 on success, or non-zero if the memory could not be - * accessed (ie. because it is an invalid address). - * - * writev-intensive code may want this to prefault several iovecs -- that - * would be possible (callers must not rely on the fact that _only_ the - * first iovec will be faulted with the current implementation). - */ -int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes) -{ - if (!(i->type & (ITER_BVEC|ITER_KVEC))) { - char __user *buf = i->iov->iov_base + i->iov_offset; - bytes = min(bytes, i->iov->iov_len - i->iov_offset); - return fault_in_pages_readable(buf, bytes); - } - return 0; -} -EXPORT_SYMBOL(iov_iter_fault_in_readable); - /* * Fault in one or more iovecs of the given iov_iter, to a maximum length of * bytes. For each iovec, fault in each page that constitutes the iovec. @@ -317,7 +297,7 @@ EXPORT_SYMBOL(iov_iter_fault_in_readable); * Return 0 on success, or non-zero if the memory could not be accessed (i.e. * because it is an invalid address). */ -int iov_iter_fault_in_multipages_readable(struct iov_iter *i, size_t bytes) +int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes) { size_t skip = i->iov_offset; const struct iovec *iov; @@ -334,7 +314,7 @@ int iov_iter_fault_in_multipages_readable(struct iov_iter *i, size_t bytes) } return 0; } -EXPORT_SYMBOL(iov_iter_fault_in_multipages_readable); +EXPORT_SYMBOL(iov_iter_fault_in_readable); void iov_iter_init(struct iov_iter *i, int direction, const struct iovec *iov, unsigned long nr_segs, From 65c0044ca8d7c7bbccae37f0ff2972f0210e9f41 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sat, 17 Sep 2016 07:52:49 -0700 Subject: [PATCH 266/271] avr32: fix 'undefined reference to `___copy_from_user' avr32 builds fail with: arch/avr32/kernel/built-in.o: In function `arch_ptrace': (.text+0x650): undefined reference to `___copy_from_user' arch/avr32/kernel/built-in.o:(___ksymtab+___copy_from_user+0x0): undefined reference to `___copy_from_user' kernel/built-in.o: In function `proc_doulongvec_ms_jiffies_minmax': (.text+0x5dd8): undefined reference to `___copy_from_user' kernel/built-in.o: In function `proc_dointvec_minmax_sysadmin': sysctl.c:(.text+0x6174): undefined reference to `___copy_from_user' kernel/built-in.o: In function `ptrace_has_cap': ptrace.c:(.text+0x69c0): undefined reference to `___copy_from_user' kernel/built-in.o:ptrace.c:(.text+0x6b90): more undefined references to `___copy_from_user' follow Fixes: 8630c32275ba ("avr32: fix copy_from_user()") Cc: Al Viro Acked-by: Havard Skinnemoen Acked-by: Hans-Christian Noren Egtvedt Signed-off-by: Guenter Roeck --- arch/avr32/lib/copy_user.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/avr32/lib/copy_user.S b/arch/avr32/lib/copy_user.S index 96a6de9d578f..075373471da1 100644 --- a/arch/avr32/lib/copy_user.S +++ b/arch/avr32/lib/copy_user.S @@ -23,8 +23,8 @@ */ .text .align 1 - .global copy_from_user - .type copy_from_user, @function + .global ___copy_from_user + .type ___copy_from_user, @function ___copy_from_user: branch_if_kernel r8, __copy_user ret_if_privileged r8, r11, r10, r10 From 8e4b72054f554967827e18be1de0e8122e6efc04 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sat, 17 Sep 2016 12:57:24 -0700 Subject: [PATCH 267/271] openrisc: fix the fix of copy_from_user() Since commit acb2505d0119 ("openrisc: fix copy_from_user()"), copy_from_user() returns the number of bytes requested, not the number of bytes not copied. Cc: Al Viro Fixes: acb2505d0119 ("openrisc: fix copy_from_user()") Signed-off-by: Guenter Roeck --- arch/openrisc/include/asm/uaccess.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/openrisc/include/asm/uaccess.h b/arch/openrisc/include/asm/uaccess.h index cbad29b5a131..5cc6b4f1b795 100644 --- a/arch/openrisc/include/asm/uaccess.h +++ b/arch/openrisc/include/asm/uaccess.h @@ -276,7 +276,7 @@ copy_from_user(void *to, const void *from, unsigned long n) unsigned long res = n; if (likely(access_ok(VERIFY_READ, from, n))) - n = __copy_tofrom_user(to, from, n); + res = __copy_tofrom_user(to, from, n); if (unlikely(res)) memset(to + (n - res), 0, res); return res; From 3be7988674ab33565700a37b210f502563d932e6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 18 Sep 2016 17:27:41 -0700 Subject: [PATCH 268/271] Linux 4.8-rc7 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 1a8c8ddcb22f..74e22c2f408b 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 8 SUBLEVEL = 0 -EXTRAVERSION = -rc6 +EXTRAVERSION = -rc7 NAME = Psychotic Stoned Sheep # *DOCUMENTATION* From 1984e075915cbae65336a99b1879865080d8e55e Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 19 Sep 2016 09:49:27 +0100 Subject: [PATCH 269/271] genirq: Skip chained interrupt trigger setup if type is IRQ_TYPE_NONE There is no point in trying to configure the trigger of a chained interrupt if no trigger information has been configured. At best this is ignored, and at the worse this confuses the underlying irqchip (which is likely not to handle such a thing), and unnecessarily alarms the user. Only apply the configuration if type is not IRQ_TYPE_NONE. Fixes: 1e12c4a9393b ("genirq: Correctly configure the trigger on chained interrupts") Reported-and-tested-by: Geert Uytterhoeven Signed-off-by: Marc Zyngier Link: https://lkml.kernel.org/r/CAMuHMdVW1eTn20=EtYcJ8hkVwohaSuH_yQXrY2MGBEvZ8fpFOg@mail.gmail.com Link: http://lkml.kernel.org/r/1474274967-15984-1-git-send-email-marc.zyngier@arm.com Signed-off-by: Thomas Gleixner --- kernel/irq/chip.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index 637389088b3f..26ba5654d9d5 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -820,6 +820,8 @@ __irq_do_set_handler(struct irq_desc *desc, irq_flow_handler_t handle, desc->name = name; if (handle != handle_bad_irq && is_chained) { + unsigned int type = irqd_get_trigger_type(&desc->irq_data); + /* * We're about to start this interrupt immediately, * hence the need to set the trigger configuration. @@ -828,8 +830,10 @@ __irq_do_set_handler(struct irq_desc *desc, irq_flow_handler_t handle, * chained interrupt. Reset it immediately because we * do know better. */ - __irq_set_trigger(desc, irqd_get_trigger_type(&desc->irq_data)); - desc->handle_irq = handle; + if (type != IRQ_TYPE_NONE) { + __irq_set_trigger(desc, type); + desc->handle_irq = handle; + } irq_settings_set_noprobe(desc); irq_settings_set_norequest(desc); From 727653d6ce7103b245eb8041f55dd5885f4c3289 Mon Sep 17 00:00:00 2001 From: James Morse Date: Mon, 19 Sep 2016 18:29:15 +0100 Subject: [PATCH 270/271] irqchip/gicv3: Silence noisy DEBUG_PER_CPU_MAPS warning gic_raise_softirq() walks the list of cpus using for_each_cpu(), it calls gic_compute_target_list() which advances the iterator by the number of CPUs in the cluster. If gic_compute_target_list() reaches the last CPU it leaves the iterator pointing at the last CPU. This means the next time round the for_each_cpu() loop cpumask_next() will be called with an invalid CPU. This triggers a warning when built with CONFIG_DEBUG_PER_CPU_MAPS: [ 3.077738] GICv3: CPU1: found redistributor 1 region 0:0x000000002f120000 [ 3.077943] CPU1: Booted secondary processor [410fd0f0] [ 3.078542] ------------[ cut here ]------------ [ 3.078746] WARNING: CPU: 1 PID: 0 at ../include/linux/cpumask.h:121 gic_raise_softirq+0x12c/0x170 [ 3.078812] Modules linked in: [ 3.078869] [ 3.078930] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.8.0-rc5+ #5188 [ 3.078994] Hardware name: Foundation-v8A (DT) [ 3.079059] task: ffff80087a1a0080 task.stack: ffff80087a19c000 [ 3.079145] PC is at gic_raise_softirq+0x12c/0x170 [ 3.079226] LR is at gic_raise_softirq+0xa4/0x170 [ 3.079296] pc : [] lr : [] pstate: 200001c9 [ 3.081139] Call trace: [ 3.081202] Exception stack(0xffff80087a19fbe0 to 0xffff80087a19fd10) [ 3.082269] [] gic_raise_softirq+0x12c/0x170 [ 3.082354] [] smp_send_reschedule+0x34/0x40 [ 3.082433] [] resched_curr+0x50/0x88 [ 3.082512] [] check_preempt_curr+0x60/0xd0 [ 3.082593] [] ttwu_do_wakeup+0x20/0xe8 [ 3.082672] [] ttwu_do_activate+0x90/0xc0 [ 3.082753] [] try_to_wake_up+0x224/0x370 [ 3.082836] [] default_wake_function+0x10/0x18 [ 3.082920] [] __wake_up_common+0x5c/0xa0 [ 3.083003] [] __wake_up_locked+0x14/0x20 [ 3.083086] [] complete+0x40/0x60 [ 3.083168] [] secondary_start_kernel+0x15c/0x1d0 [ 3.083240] [<00000000808911a4>] 0x808911a4 [ 3.113401] Detected PIPT I-cache on CPU2 Avoid updating the iterator if the next call to cpumask_next() would cause the for_each_cpu() loop to exit. There is no change to gic_raise_softirq()'s behaviour, (cpumask_next()s eventual call to _find_next_bit() will return early as start >= nbits), this patch just silences the warning. Fixes: 021f653791ad ("irqchip: gic-v3: Initial support for GICv3") Signed-off-by: James Morse Acked-by: Marc Zyngier Cc: linux-arm-kernel@lists.infradead.org Cc: Jason Cooper Link: http://lkml.kernel.org/r/1474306155-3303-1-git-send-email-james.morse@arm.com Signed-off-by: Thomas Gleixner --- drivers/irqchip/irq-gic-v3.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index ede5672ab34d..da6c0ba61d4f 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -548,7 +548,7 @@ static int gic_starting_cpu(unsigned int cpu) static u16 gic_compute_target_list(int *base_cpu, const struct cpumask *mask, unsigned long cluster_id) { - int cpu = *base_cpu; + int next_cpu, cpu = *base_cpu; unsigned long mpidr = cpu_logical_map(cpu); u16 tlist = 0; @@ -562,9 +562,10 @@ static u16 gic_compute_target_list(int *base_cpu, const struct cpumask *mask, tlist |= 1 << (mpidr & 0xf); - cpu = cpumask_next(cpu, mask); - if (cpu >= nr_cpu_ids) + next_cpu = cpumask_next(cpu, mask); + if (next_cpu >= nr_cpu_ids) goto out; + cpu = next_cpu; mpidr = cpu_logical_map(cpu); From e875bd66dfb68f4e898e9a43ef42858c504a7f23 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Tue, 13 Sep 2016 17:53:35 +0100 Subject: [PATCH 271/271] irqchip/mips-gic: Fix local interrupts Since the device hierarchy domain was added by commit c98c1822ee13 ("irqchip/mips-gic: Add device hierarchy domain"), GIC local interrupts have been broken. Users attempting to setup a per-cpu local IRQ, for example the GIC timer clock events code in drivers/clocksource/mips-gic-timer.c, the setup_percpu_irq function would refuse with -EINVAL because the GIC irqchip driver never called irq_set_percpu_devid so the IRQ_PER_CPU_DEVID flag was never set for the IRQ. This happens because irq_set_percpu_devid was being called from the gic_irq_domain_map function which is no longer called. Doing only that runs into further problems because gic_dev_domain_alloc set the struct irq_chip for all interrupts, local or shared, to gic_level_irq_controller despite that only being suitable for shared interrupts. The typical outcome of this is that gic_level_irq_controller callback functions are called for local interrupts, and then hwirq number calculations overflow & the driver ends up attempting to access some invalid register with an address calculated from an invalid hwirq number. Best case scenario is that this then leads to a bus error. This is fixed by abstracting the setup of the hwirq & chip to a new function gic_setup_dev_chip which is used by both the root GIC IRQ domain & the device domain. Finally, decoding local interrupts failed because gic_dev_domain_alloc only called irq_domain_alloc_irqs_parent for shared interrupts. Local ones were therefore never associated with hwirqs in the root GIC IRQ domain and the virq in gic_handle_local_int would always be 0. This is fixed by calling irq_domain_alloc_irqs_parent unconditionally & having gic_irq_domain_alloc handle both local & shared interrupts, which is easy due to the aforementioned abstraction of chip setup into gic_setup_dev_chip. This fixes use of the MIPS GIC timer for clock events, which has been broken since c98c1822ee13 ("irqchip/mips-gic: Add device hierarchy domain") but hadn't been noticed due to a silent fallback to the MIPS coprocessor 0 count/compare clock events device. Fixes: c98c1822ee13 ("irqchip/mips-gic: Add device hierarchy domain") Signed-off-by: Paul Burton Cc: linux-mips@linux-mips.org Cc: Jason Cooper Cc: Qais Yousef Cc: stable@vger.kernel.org Cc: Marc Zyngier Link: http://lkml.kernel.org/r/20160913165335.31389-1-paul.burton@imgtec.com Signed-off-by: Thomas Gleixner --- drivers/irqchip/irq-mips-gic.c | 105 ++++++++++++++++----------------- 1 file changed, 50 insertions(+), 55 deletions(-) diff --git a/drivers/irqchip/irq-mips-gic.c b/drivers/irqchip/irq-mips-gic.c index 83f498393a7f..6185696405d5 100644 --- a/drivers/irqchip/irq-mips-gic.c +++ b/drivers/irqchip/irq-mips-gic.c @@ -638,27 +638,6 @@ static int gic_local_irq_domain_map(struct irq_domain *d, unsigned int virq, if (!gic_local_irq_is_routable(intr)) return -EPERM; - /* - * HACK: These are all really percpu interrupts, but the rest - * of the MIPS kernel code does not use the percpu IRQ API for - * the CP0 timer and performance counter interrupts. - */ - switch (intr) { - case GIC_LOCAL_INT_TIMER: - case GIC_LOCAL_INT_PERFCTR: - case GIC_LOCAL_INT_FDC: - irq_set_chip_and_handler(virq, - &gic_all_vpes_local_irq_controller, - handle_percpu_irq); - break; - default: - irq_set_chip_and_handler(virq, - &gic_local_irq_controller, - handle_percpu_devid_irq); - irq_set_percpu_devid(virq); - break; - } - spin_lock_irqsave(&gic_lock, flags); for (i = 0; i < gic_vpes; i++) { u32 val = GIC_MAP_TO_PIN_MSK | gic_cpu_pin; @@ -724,16 +703,42 @@ static int gic_shared_irq_domain_map(struct irq_domain *d, unsigned int virq, return 0; } -static int gic_irq_domain_map(struct irq_domain *d, unsigned int virq, - irq_hw_number_t hw) +static int gic_setup_dev_chip(struct irq_domain *d, unsigned int virq, + unsigned int hwirq) { - if (GIC_HWIRQ_TO_LOCAL(hw) < GIC_NUM_LOCAL_INTRS) - return gic_local_irq_domain_map(d, virq, hw); + struct irq_chip *chip; + int err; - irq_set_chip_and_handler(virq, &gic_level_irq_controller, - handle_level_irq); + if (hwirq >= GIC_SHARED_HWIRQ_BASE) { + err = irq_domain_set_hwirq_and_chip(d, virq, hwirq, + &gic_level_irq_controller, + NULL); + } else { + switch (GIC_HWIRQ_TO_LOCAL(hwirq)) { + case GIC_LOCAL_INT_TIMER: + case GIC_LOCAL_INT_PERFCTR: + case GIC_LOCAL_INT_FDC: + /* + * HACK: These are all really percpu interrupts, but + * the rest of the MIPS kernel code does not use the + * percpu IRQ API for them. + */ + chip = &gic_all_vpes_local_irq_controller; + irq_set_handler(virq, handle_percpu_irq); + break; - return gic_shared_irq_domain_map(d, virq, hw, 0); + default: + chip = &gic_local_irq_controller; + irq_set_handler(virq, handle_percpu_devid_irq); + irq_set_percpu_devid(virq); + break; + } + + err = irq_domain_set_hwirq_and_chip(d, virq, hwirq, + chip, NULL); + } + + return err; } static int gic_irq_domain_alloc(struct irq_domain *d, unsigned int virq, @@ -744,15 +749,12 @@ static int gic_irq_domain_alloc(struct irq_domain *d, unsigned int virq, int cpu, ret, i; if (spec->type == GIC_DEVICE) { - /* verify that it doesn't conflict with an IPI irq */ - if (test_bit(spec->hwirq, ipi_resrv)) + /* verify that shared irqs don't conflict with an IPI irq */ + if ((spec->hwirq >= GIC_SHARED_HWIRQ_BASE) && + test_bit(GIC_HWIRQ_TO_SHARED(spec->hwirq), ipi_resrv)) return -EBUSY; - hwirq = GIC_SHARED_TO_HWIRQ(spec->hwirq); - - return irq_domain_set_hwirq_and_chip(d, virq, hwirq, - &gic_level_irq_controller, - NULL); + return gic_setup_dev_chip(d, virq, spec->hwirq); } else { base_hwirq = find_first_bit(ipi_resrv, gic_shared_intrs); if (base_hwirq == gic_shared_intrs) { @@ -821,7 +823,6 @@ int gic_irq_domain_match(struct irq_domain *d, struct device_node *node, } static const struct irq_domain_ops gic_irq_domain_ops = { - .map = gic_irq_domain_map, .alloc = gic_irq_domain_alloc, .free = gic_irq_domain_free, .match = gic_irq_domain_match, @@ -852,29 +853,20 @@ static int gic_dev_domain_alloc(struct irq_domain *d, unsigned int virq, struct irq_fwspec *fwspec = arg; struct gic_irq_spec spec = { .type = GIC_DEVICE, - .hwirq = fwspec->param[1], }; int i, ret; - bool is_shared = fwspec->param[0] == GIC_SHARED; - if (is_shared) { - ret = irq_domain_alloc_irqs_parent(d, virq, nr_irqs, &spec); - if (ret) - return ret; - } + if (fwspec->param[0] == GIC_SHARED) + spec.hwirq = GIC_SHARED_TO_HWIRQ(fwspec->param[1]); + else + spec.hwirq = GIC_LOCAL_TO_HWIRQ(fwspec->param[1]); + + ret = irq_domain_alloc_irqs_parent(d, virq, nr_irqs, &spec); + if (ret) + return ret; for (i = 0; i < nr_irqs; i++) { - irq_hw_number_t hwirq; - - if (is_shared) - hwirq = GIC_SHARED_TO_HWIRQ(spec.hwirq + i); - else - hwirq = GIC_LOCAL_TO_HWIRQ(spec.hwirq + i); - - ret = irq_domain_set_hwirq_and_chip(d, virq + i, - hwirq, - &gic_level_irq_controller, - NULL); + ret = gic_setup_dev_chip(d, virq + i, spec.hwirq + i); if (ret) goto error; } @@ -896,7 +888,10 @@ void gic_dev_domain_free(struct irq_domain *d, unsigned int virq, static void gic_dev_domain_activate(struct irq_domain *domain, struct irq_data *d) { - gic_shared_irq_domain_map(domain, d->irq, d->hwirq, 0); + if (GIC_HWIRQ_TO_LOCAL(d->hwirq) < GIC_NUM_LOCAL_INTRS) + gic_local_irq_domain_map(domain, d->irq, d->hwirq); + else + gic_shared_irq_domain_map(domain, d->irq, d->hwirq, 0); } static struct irq_domain_ops gic_dev_domain_ops = {