OpenCloudOS-Kernel/drivers/net/ethernet/mellanox/mlx4/en_rx.c

1264 lines
34 KiB
C
Raw Normal View History

/*
* Copyright (c) 2007 Mellanox Technologies. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*
*/
#include <net/busy_poll.h>
#include <linux/bpf.h>
bpf: add initial bpf tracepoints This work adds a number of tracepoints to paths that are either considered slow-path or exception-like states, where monitoring or inspecting them would be desirable. For bpf(2) syscall, tracepoints have been placed for main commands when they succeed. In XDP case, tracepoint is for exceptions, that is, f.e. on abnormal BPF program exit such as unknown or XDP_ABORTED return code, or when error occurs during XDP_TX action and the packet could not be forwarded. Both have been split into separate event headers, and can be further extended. Worst case, if they unexpectedly should get into our way in future, they can also removed [1]. Of course, these tracepoints (like any other) can be analyzed by eBPF itself, etc. Example output: # ./perf record -a -e bpf:* sleep 10 # ./perf script sock_example 6197 [005] 283.980322: bpf:bpf_map_create: map type=ARRAY ufd=4 key=4 val=8 max=256 flags=0 sock_example 6197 [005] 283.980721: bpf:bpf_prog_load: prog=a5ea8fa30ea6849c type=SOCKET_FILTER ufd=5 sock_example 6197 [005] 283.988423: bpf:bpf_prog_get_type: prog=a5ea8fa30ea6849c type=SOCKET_FILTER sock_example 6197 [005] 283.988443: bpf:bpf_map_lookup_elem: map type=ARRAY ufd=4 key=[06 00 00 00] val=[00 00 00 00 00 00 00 00] [...] sock_example 6197 [005] 288.990868: bpf:bpf_map_lookup_elem: map type=ARRAY ufd=4 key=[01 00 00 00] val=[14 00 00 00 00 00 00 00] swapper 0 [005] 289.338243: bpf:bpf_prog_put_rcu: prog=a5ea8fa30ea6849c type=SOCKET_FILTER [1] https://lwn.net/Articles/705270/ Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-25 09:28:18 +08:00
#include <linux/bpf_trace.h>
#include <linux/mlx4/cq.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/mlx4/qp.h>
#include <linux/skbuff.h>
hlist: drop the node parameter from iterators I'm not sure why, but the hlist for each entry iterators were conceived list_for_each_entry(pos, head, member) The hlist ones were greedy and wanted an extra parameter: hlist_for_each_entry(tpos, pos, head, member) Why did they need an extra pos parameter? I'm not quite sure. Not only they don't really need it, it also prevents the iterator from looking exactly like the list iterator, which is unfortunate. Besides the semantic patch, there was some manual work required: - Fix up the actual hlist iterators in linux/list.h - Fix up the declaration of other iterators based on the hlist ones. - A very small amount of places were using the 'node' parameter, this was modified to use 'obj->member' instead. - Coccinelle didn't handle the hlist_for_each_entry_safe iterator properly, so those had to be fixed up manually. The semantic patch which is mostly the work of Peter Senna Tschudin is here: @@ iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host; type T; expression a,c,d,e; identifier b; statement S; @@ -T b; <+... when != b ( hlist_for_each_entry(a, - b, c, d) S | hlist_for_each_entry_continue(a, - b, c) S | hlist_for_each_entry_from(a, - b, c) S | hlist_for_each_entry_rcu(a, - b, c, d) S | hlist_for_each_entry_rcu_bh(a, - b, c, d) S | hlist_for_each_entry_continue_rcu_bh(a, - b, c) S | for_each_busy_worker(a, c, - b, d) S | ax25_uid_for_each(a, - b, c) S | ax25_for_each(a, - b, c) S | inet_bind_bucket_for_each(a, - b, c) S | sctp_for_each_hentry(a, - b, c) S | sk_for_each(a, - b, c) S | sk_for_each_rcu(a, - b, c) S | sk_for_each_from -(a, b) +(a) S + sk_for_each_from(a) S | sk_for_each_safe(a, - b, c, d) S | sk_for_each_bound(a, - b, c) S | hlist_for_each_entry_safe(a, - b, c, d, e) S | hlist_for_each_entry_continue_rcu(a, - b, c) S | nr_neigh_for_each(a, - b, c) S | nr_neigh_for_each_safe(a, - b, c, d) S | nr_node_for_each(a, - b, c) S | nr_node_for_each_safe(a, - b, c, d) S | - for_each_gfn_sp(a, c, d, b) S + for_each_gfn_sp(a, c, d) S | - for_each_gfn_indirect_valid_sp(a, c, d, b) S + for_each_gfn_indirect_valid_sp(a, c, d) S | for_each_host(a, - b, c) S | for_each_host_safe(a, - b, c, d) S | for_each_mesh_entry(a, - b, c, d) S ) ...+> [akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c] [akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c] [akpm@linux-foundation.org: checkpatch fixes] [akpm@linux-foundation.org: fix warnings] [akpm@linux-foudnation.org: redo intrusive kvm changes] Tested-by: Peter Senna Tschudin <peter.senna@gmail.com> Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Sasha Levin <sasha.levin@oracle.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Gleb Natapov <gleb@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 09:06:00 +08:00
#include <linux/rculist.h>
#include <linux/if_ether.h>
#include <linux/if_vlan.h>
#include <linux/vmalloc.h>
#include <linux/irq.h>
#if IS_ENABLED(CONFIG_IPV6)
#include <net/ip6_checksum.h>
#endif
#include "mlx4_en.h"
static int mlx4_alloc_page(struct mlx4_en_priv *priv,
struct mlx4_en_rx_alloc *frag,
gfp_t gfp)
{
struct page *page;
dma_addr_t dma;
page = alloc_page(gfp);
if (unlikely(!page))
return -ENOMEM;
dma = dma_map_page(priv->ddev, page, 0, PAGE_SIZE, priv->dma_dir);
if (unlikely(dma_mapping_error(priv->ddev, dma))) {
__free_page(page);
return -ENOMEM;
}
frag->page = page;
frag->dma = dma;
frag->page_offset = priv->rx_headroom;
return 0;
}
static int mlx4_en_alloc_frags(struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring *ring,
struct mlx4_en_rx_desc *rx_desc,
struct mlx4_en_rx_alloc *frags,
gfp_t gfp)
{
int i;
for (i = 0; i < priv->num_frags; i++, frags++) {
if (!frags->page) {
if (mlx4_alloc_page(priv, frags, gfp))
return -ENOMEM;
ring->rx_alloc_pages++;
}
rx_desc->data[i].addr = cpu_to_be64(frags->dma +
frags->page_offset);
}
return 0;
}
static void mlx4_en_free_frag(const struct mlx4_en_priv *priv,
struct mlx4_en_rx_alloc *frag)
{
if (frag->page) {
dma_unmap_page(priv->ddev, frag->dma,
PAGE_SIZE, priv->dma_dir);
__free_page(frag->page);
}
/* We need to clear all fields, otherwise a change of priv->log_rx_info
* could lead to see garbage later in frag->page.
*/
memset(frag, 0, sizeof(*frag));
}
static void mlx4_en_init_rx_desc(const struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring *ring, int index)
{
struct mlx4_en_rx_desc *rx_desc = ring->buf + ring->stride * index;
int possible_frags;
int i;
/* Set size and memtype fields */
for (i = 0; i < priv->num_frags; i++) {
rx_desc->data[i].byte_count =
cpu_to_be32(priv->frag_info[i].frag_size);
rx_desc->data[i].lkey = cpu_to_be32(priv->mdev->mr.key);
}
/* If the number of used fragments does not fill up the ring stride,
* remaining (unused) fragments must be padded with null address/size
* and a special memory key */
possible_frags = (ring->stride - sizeof(struct mlx4_en_rx_desc)) / DS_SIZE;
for (i = priv->num_frags; i < possible_frags; i++) {
rx_desc->data[i].byte_count = 0;
rx_desc->data[i].lkey = cpu_to_be32(MLX4_EN_MEMTYPE_PAD);
rx_desc->data[i].addr = 0;
}
}
static int mlx4_en_prepare_rx_desc(struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring *ring, int index,
gfp_t gfp)
{
struct mlx4_en_rx_desc *rx_desc = ring->buf +
(index << ring->log_stride);
struct mlx4_en_rx_alloc *frags = ring->rx_info +
(index << priv->log_rx_info);
if (likely(ring->page_cache.index > 0)) {
/* XDP uses a single page per frame */
if (!frags->page) {
ring->page_cache.index--;
frags->page = ring->page_cache.buf[ring->page_cache.index].page;
frags->dma = ring->page_cache.buf[ring->page_cache.index].dma;
}
frags->page_offset = XDP_PACKET_HEADROOM;
rx_desc->data[0].addr = cpu_to_be64(frags->dma +
XDP_PACKET_HEADROOM);
net/mlx4_en: add page recycle to prepare rx ring for tx support The mlx4 driver by default allocates order-3 pages for the ring to consume in multiple fragments. When the device has an xdp program, this behavior will prevent tx actions since the page must be re-mapped in TODEVICE mode, which cannot be done if the page is still shared. Start by making the allocator configurable based on whether xdp is running, such that order-0 pages are always used and never shared. Since this will stress the page allocator, add a simple page cache to each rx ring. Pages in the cache are left dma-mapped, and in drop-only stress tests the page allocator is eliminated from the perf report. Note that setting an xdp program will now require the rings to be reconfigured. Before: 26.91% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq 17.88% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags 6.00% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag 4.49% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist 3.21% swapper [kernel.vmlinux] [k] intel_idle 2.73% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem 2.57% swapper [mlx4_en] [k] mlx4_en_process_rx_cq After: 31.72% swapper [kernel.vmlinux] [k] intel_idle 8.79% swapper [mlx4_en] [k] mlx4_en_process_rx_cq 7.54% swapper [kernel.vmlinux] [k] poll_idle 6.36% swapper [mlx4_core] [k] mlx4_eq_int 4.21% swapper [kernel.vmlinux] [k] tasklet_action 4.03% swapper [kernel.vmlinux] [k] cpuidle_enter_state 3.43% swapper [mlx4_en] [k] mlx4_en_prepare_rx_desc 2.18% swapper [kernel.vmlinux] [k] native_irq_return_iret 1.37% swapper [kernel.vmlinux] [k] menu_select 1.09% swapper [kernel.vmlinux] [k] bpf_map_lookup_elem Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:52 +08:00
return 0;
}
return mlx4_en_alloc_frags(priv, ring, rx_desc, frags, gfp);
}
static bool mlx4_en_is_ring_empty(const struct mlx4_en_rx_ring *ring)
{
return ring->prod == ring->cons;
}
static inline void mlx4_en_update_rx_prod_db(struct mlx4_en_rx_ring *ring)
{
*ring->wqres.db.db = cpu_to_be32(ring->prod & 0xffff);
}
/* slow path */
static void mlx4_en_free_rx_desc(const struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring *ring,
int index)
{
struct mlx4_en_rx_alloc *frags;
int nr;
frags = ring->rx_info + (index << priv->log_rx_info);
for (nr = 0; nr < priv->num_frags; nr++) {
en_dbg(DRV, priv, "Freeing fragment:%d\n", nr);
mlx4_en_free_frag(priv, frags + nr);
}
}
/* Function not in fast-path */
static int mlx4_en_fill_rx_buffers(struct mlx4_en_priv *priv)
{
struct mlx4_en_rx_ring *ring;
int ring_ind;
int buf_ind;
int new_size;
for (buf_ind = 0; buf_ind < priv->prof->rx_ring_size; buf_ind++) {
for (ring_ind = 0; ring_ind < priv->rx_ring_num; ring_ind++) {
ring = priv->rx_ring[ring_ind];
if (mlx4_en_prepare_rx_desc(priv, ring,
ring->actual_size,
GFP_KERNEL | __GFP_COLD)) {
if (ring->actual_size < MLX4_EN_MIN_RX_SIZE) {
en_err(priv, "Failed to allocate enough rx buffers\n");
return -ENOMEM;
} else {
new_size = rounddown_pow_of_two(ring->actual_size);
en_warn(priv, "Only %d buffers allocated reducing ring size to %d\n",
ring->actual_size, new_size);
goto reduce_rings;
}
}
ring->actual_size++;
ring->prod++;
}
}
return 0;
reduce_rings:
for (ring_ind = 0; ring_ind < priv->rx_ring_num; ring_ind++) {
ring = priv->rx_ring[ring_ind];
while (ring->actual_size > new_size) {
ring->actual_size--;
ring->prod--;
mlx4_en_free_rx_desc(priv, ring, ring->actual_size);
}
}
return 0;
}
static void mlx4_en_free_rx_buf(struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring *ring)
{
int index;
en_dbg(DRV, priv, "Freeing Rx buf - cons:%d prod:%d\n",
ring->cons, ring->prod);
/* Unmap and free Rx buffers */
for (index = 0; index < ring->size; index++) {
en_dbg(DRV, priv, "Processing descriptor:%d\n", index);
mlx4_en_free_rx_desc(priv, ring, index);
}
ring->cons = 0;
ring->prod = 0;
}
void mlx4_en_set_num_rx_rings(struct mlx4_en_dev *mdev)
{
int i;
int num_of_eqs;
int num_rx_rings;
struct mlx4_dev *dev = mdev->dev;
mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) {
num_of_eqs = max_t(int, MIN_RX_RINGS,
min_t(int,
mlx4_get_eqs_per_port(mdev->dev, i),
DEF_RX_RINGS));
num_rx_rings = mlx4_low_memory_profile() ? MIN_RX_RINGS :
min_t(int, num_of_eqs, num_online_cpus());
mdev->profile.prof[i].rx_ring_num =
rounddown_pow_of_two(num_rx_rings);
}
}
int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring **pring,
u32 size, u16 stride, int node)
{
struct mlx4_en_dev *mdev = priv->mdev;
struct mlx4_en_rx_ring *ring;
int err = -ENOMEM;
int tmp;
ring = kzalloc_node(sizeof(*ring), GFP_KERNEL, node);
if (!ring) {
ring = kzalloc(sizeof(*ring), GFP_KERNEL);
if (!ring) {
en_err(priv, "Failed to allocate RX ring structure\n");
return -ENOMEM;
}
}
ring->prod = 0;
ring->cons = 0;
ring->size = size;
ring->size_mask = size - 1;
ring->stride = stride;
ring->log_stride = ffs(ring->stride) - 1;
ring->buf_size = ring->size * ring->stride + TXBB_SIZE;
tmp = size * roundup_pow_of_two(MLX4_EN_MAX_RX_FRAGS *
sizeof(struct mlx4_en_rx_alloc));
ring->rx_info = vzalloc_node(tmp, node);
if (!ring->rx_info) {
ring->rx_info = vzalloc(tmp);
if (!ring->rx_info) {
err = -ENOMEM;
goto err_ring;
}
}
en_dbg(DRV, priv, "Allocated rx_info ring at addr:%p size:%d\n",
ring->rx_info, tmp);
/* Allocate HW buffers on provided NUMA node */
set_dev_node(&mdev->dev->persist->pdev->dev, node);
err = mlx4_alloc_hwq_res(mdev->dev, &ring->wqres, ring->buf_size);
set_dev_node(&mdev->dev->persist->pdev->dev, mdev->dev->numa_node);
if (err)
goto err_info;
ring->buf = ring->wqres.buf.direct.buf;
ring->hwtstamp_rx_filter = priv->hwtstamp_config.rx_filter;
*pring = ring;
return 0;
err_info:
vfree(ring->rx_info);
ring->rx_info = NULL;
err_ring:
kfree(ring);
*pring = NULL;
return err;
}
int mlx4_en_activate_rx_rings(struct mlx4_en_priv *priv)
{
struct mlx4_en_rx_ring *ring;
int i;
int ring_ind;
int err;
int stride = roundup_pow_of_two(sizeof(struct mlx4_en_rx_desc) +
DS_SIZE * priv->num_frags);
for (ring_ind = 0; ring_ind < priv->rx_ring_num; ring_ind++) {
ring = priv->rx_ring[ring_ind];
ring->prod = 0;
ring->cons = 0;
ring->actual_size = 0;
ring->cqn = priv->rx_cq[ring_ind]->mcq.cqn;
ring->stride = stride;
if (ring->stride <= TXBB_SIZE) {
/* Stamp first unused send wqe */
__be32 *ptr = (__be32 *)ring->buf;
__be32 stamp = cpu_to_be32(1 << STAMP_SHIFT);
*ptr = stamp;
/* Move pointer to start of rx section */
ring->buf += TXBB_SIZE;
}
ring->log_stride = ffs(ring->stride) - 1;
ring->buf_size = ring->size * ring->stride;
memset(ring->buf, 0, ring->buf_size);
mlx4_en_update_rx_prod_db(ring);
/* Initialize all descriptors */
for (i = 0; i < ring->size; i++)
mlx4_en_init_rx_desc(priv, ring, i);
}
err = mlx4_en_fill_rx_buffers(priv);
if (err)
goto err_buffers;
for (ring_ind = 0; ring_ind < priv->rx_ring_num; ring_ind++) {
ring = priv->rx_ring[ring_ind];
ring->size_mask = ring->actual_size - 1;
mlx4_en_update_rx_prod_db(ring);
}
return 0;
err_buffers:
for (ring_ind = 0; ring_ind < priv->rx_ring_num; ring_ind++)
mlx4_en_free_rx_buf(priv, priv->rx_ring[ring_ind]);
ring_ind = priv->rx_ring_num - 1;
while (ring_ind >= 0) {
if (priv->rx_ring[ring_ind]->stride <= TXBB_SIZE)
priv->rx_ring[ring_ind]->buf -= TXBB_SIZE;
ring_ind--;
}
return err;
}
/* We recover from out of memory by scheduling our napi poll
* function (mlx4_en_process_cq), which tries to allocate
* all missing RX buffers (call to mlx4_en_refill_rx_buffers).
*/
void mlx4_en_recover_from_oom(struct mlx4_en_priv *priv)
{
int ring;
if (!priv->port_up)
return;
for (ring = 0; ring < priv->rx_ring_num; ring++) {
if (mlx4_en_is_ring_empty(priv->rx_ring[ring])) {
local_bh_disable();
napi_reschedule(&priv->rx_cq[ring]->napi);
local_bh_enable();
}
}
}
net/mlx4_en: add page recycle to prepare rx ring for tx support The mlx4 driver by default allocates order-3 pages for the ring to consume in multiple fragments. When the device has an xdp program, this behavior will prevent tx actions since the page must be re-mapped in TODEVICE mode, which cannot be done if the page is still shared. Start by making the allocator configurable based on whether xdp is running, such that order-0 pages are always used and never shared. Since this will stress the page allocator, add a simple page cache to each rx ring. Pages in the cache are left dma-mapped, and in drop-only stress tests the page allocator is eliminated from the perf report. Note that setting an xdp program will now require the rings to be reconfigured. Before: 26.91% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq 17.88% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags 6.00% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag 4.49% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist 3.21% swapper [kernel.vmlinux] [k] intel_idle 2.73% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem 2.57% swapper [mlx4_en] [k] mlx4_en_process_rx_cq After: 31.72% swapper [kernel.vmlinux] [k] intel_idle 8.79% swapper [mlx4_en] [k] mlx4_en_process_rx_cq 7.54% swapper [kernel.vmlinux] [k] poll_idle 6.36% swapper [mlx4_core] [k] mlx4_eq_int 4.21% swapper [kernel.vmlinux] [k] tasklet_action 4.03% swapper [kernel.vmlinux] [k] cpuidle_enter_state 3.43% swapper [mlx4_en] [k] mlx4_en_prepare_rx_desc 2.18% swapper [kernel.vmlinux] [k] native_irq_return_iret 1.37% swapper [kernel.vmlinux] [k] menu_select 1.09% swapper [kernel.vmlinux] [k] bpf_map_lookup_elem Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:52 +08:00
/* When the rx ring is running in page-per-packet mode, a released frame can go
* directly into a small cache, to avoid unmapping or touching the page
* allocator. In bpf prog performance scenarios, buffers are either forwarded
* or dropped, never converted to skbs, so every page can come directly from
* this cache when it is sized to be a multiple of the napi budget.
*/
bool mlx4_en_rx_recycle(struct mlx4_en_rx_ring *ring,
struct mlx4_en_rx_alloc *frame)
{
struct mlx4_en_page_cache *cache = &ring->page_cache;
if (cache->index >= MLX4_EN_CACHE_SIZE)
return false;
cache->buf[cache->index].page = frame->page;
cache->buf[cache->index].dma = frame->dma;
cache->index++;
net/mlx4_en: add page recycle to prepare rx ring for tx support The mlx4 driver by default allocates order-3 pages for the ring to consume in multiple fragments. When the device has an xdp program, this behavior will prevent tx actions since the page must be re-mapped in TODEVICE mode, which cannot be done if the page is still shared. Start by making the allocator configurable based on whether xdp is running, such that order-0 pages are always used and never shared. Since this will stress the page allocator, add a simple page cache to each rx ring. Pages in the cache are left dma-mapped, and in drop-only stress tests the page allocator is eliminated from the perf report. Note that setting an xdp program will now require the rings to be reconfigured. Before: 26.91% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq 17.88% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags 6.00% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag 4.49% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist 3.21% swapper [kernel.vmlinux] [k] intel_idle 2.73% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem 2.57% swapper [mlx4_en] [k] mlx4_en_process_rx_cq After: 31.72% swapper [kernel.vmlinux] [k] intel_idle 8.79% swapper [mlx4_en] [k] mlx4_en_process_rx_cq 7.54% swapper [kernel.vmlinux] [k] poll_idle 6.36% swapper [mlx4_core] [k] mlx4_eq_int 4.21% swapper [kernel.vmlinux] [k] tasklet_action 4.03% swapper [kernel.vmlinux] [k] cpuidle_enter_state 3.43% swapper [mlx4_en] [k] mlx4_en_prepare_rx_desc 2.18% swapper [kernel.vmlinux] [k] native_irq_return_iret 1.37% swapper [kernel.vmlinux] [k] menu_select 1.09% swapper [kernel.vmlinux] [k] bpf_map_lookup_elem Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:52 +08:00
return true;
}
void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring **pring,
u32 size, u16 stride)
{
struct mlx4_en_dev *mdev = priv->mdev;
struct mlx4_en_rx_ring *ring = *pring;
struct bpf_prog *old_prog;
net/mlx4_en: protect ring->xdp_prog with rcu_read_lock Depending on the preempt mode, the bpf_prog stored in xdp_prog may be freed despite the use of call_rcu inside bpf_prog_put. The situation is possible when running in PREEMPT_RCU=y mode, for instance, since the rcu callback for destroying the bpf prog can run even during the bh handling in the mlx4 rx path. Several options were considered before this patch was settled on: Add a napi_synchronize loop in mlx4_xdp_set, which would occur after all of the rings are updated with the new program. This approach has the disadvantage that as the number of rings increases, the speed of update will slow down significantly due to napi_synchronize's msleep(1). Add a new rcu_head in bpf_prog_aux, to be used by a new bpf_prog_put_bh. The action of the bpf_prog_put_bh would be to then call bpf_prog_put later. Those drivers that consume a bpf prog in a bh context (like mlx4) would then use the bpf_prog_put_bh instead when the ring is up. This has the problem of complexity, in maintaining proper refcnts and rcu lists, and would likely be harder to review. In addition, this approach to freeing must be exclusive with other frees of the bpf prog, for instance a _bh prog must not be referenced from a prog array that is consumed by a non-_bh prog. The placement of rcu_read_lock in this patch is functionally the same as putting an rcu_read_lock in napi_poll. Actually doing so could be a potentially controversial change, but would bring the implementation in line with sk_busy_loop (though of course the nature of those two paths is substantially different), and would also avoid future copy/paste problems with future supporters of XDP. Still, this patch does not take that opinionated option. Testing was done with kernels in either PREEMPT_RCU=y or CONFIG_PREEMPT_VOLUNTARY=y+PREEMPT_RCU=n modes, with neither exhibiting any drawback. With PREEMPT_RCU=n, the extra call to rcu_read_lock did not show up in the perf report whatsoever, and with PREEMPT_RCU=y the overhead of rcu_read_lock (according to perf) was the same before/after. In the rx path, rcu_read_lock is eventually called for every packet from netif_receive_skb_internal, so the napi poll call's rcu_read_lock is easily amortized. v2: Remove extra rcu_read_lock in mlx4_en_process_rx_cq body Annotate xdp_prog with __rcu, and convert all usages to rcu_assign or rcu_dereference[_protected] as appropriate. Add explicit mutex lock around rcu_assign instead of xchg loop. Fixes: d576acf0a22 ("net/mlx4_en: add page recycle to prepare rx ring for tx support") Acked-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <alexei.starovoitov@gmail.com> Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-04 12:29:58 +08:00
old_prog = rcu_dereference_protected(
ring->xdp_prog,
lockdep_is_held(&mdev->state_lock));
if (old_prog)
bpf_prog_put(old_prog);
mlx4_free_hwq_res(mdev->dev, &ring->wqres, size * stride + TXBB_SIZE);
vfree(ring->rx_info);
ring->rx_info = NULL;
kfree(ring);
*pring = NULL;
}
void mlx4_en_deactivate_rx_ring(struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring *ring)
{
net/mlx4_en: add page recycle to prepare rx ring for tx support The mlx4 driver by default allocates order-3 pages for the ring to consume in multiple fragments. When the device has an xdp program, this behavior will prevent tx actions since the page must be re-mapped in TODEVICE mode, which cannot be done if the page is still shared. Start by making the allocator configurable based on whether xdp is running, such that order-0 pages are always used and never shared. Since this will stress the page allocator, add a simple page cache to each rx ring. Pages in the cache are left dma-mapped, and in drop-only stress tests the page allocator is eliminated from the perf report. Note that setting an xdp program will now require the rings to be reconfigured. Before: 26.91% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq 17.88% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags 6.00% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag 4.49% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist 3.21% swapper [kernel.vmlinux] [k] intel_idle 2.73% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem 2.57% swapper [mlx4_en] [k] mlx4_en_process_rx_cq After: 31.72% swapper [kernel.vmlinux] [k] intel_idle 8.79% swapper [mlx4_en] [k] mlx4_en_process_rx_cq 7.54% swapper [kernel.vmlinux] [k] poll_idle 6.36% swapper [mlx4_core] [k] mlx4_eq_int 4.21% swapper [kernel.vmlinux] [k] tasklet_action 4.03% swapper [kernel.vmlinux] [k] cpuidle_enter_state 3.43% swapper [mlx4_en] [k] mlx4_en_prepare_rx_desc 2.18% swapper [kernel.vmlinux] [k] native_irq_return_iret 1.37% swapper [kernel.vmlinux] [k] menu_select 1.09% swapper [kernel.vmlinux] [k] bpf_map_lookup_elem Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:52 +08:00
int i;
for (i = 0; i < ring->page_cache.index; i++) {
dma_unmap_page(priv->ddev, ring->page_cache.buf[i].dma,
PAGE_SIZE, priv->dma_dir);
put_page(ring->page_cache.buf[i].page);
net/mlx4_en: add page recycle to prepare rx ring for tx support The mlx4 driver by default allocates order-3 pages for the ring to consume in multiple fragments. When the device has an xdp program, this behavior will prevent tx actions since the page must be re-mapped in TODEVICE mode, which cannot be done if the page is still shared. Start by making the allocator configurable based on whether xdp is running, such that order-0 pages are always used and never shared. Since this will stress the page allocator, add a simple page cache to each rx ring. Pages in the cache are left dma-mapped, and in drop-only stress tests the page allocator is eliminated from the perf report. Note that setting an xdp program will now require the rings to be reconfigured. Before: 26.91% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq 17.88% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags 6.00% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag 4.49% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist 3.21% swapper [kernel.vmlinux] [k] intel_idle 2.73% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem 2.57% swapper [mlx4_en] [k] mlx4_en_process_rx_cq After: 31.72% swapper [kernel.vmlinux] [k] intel_idle 8.79% swapper [mlx4_en] [k] mlx4_en_process_rx_cq 7.54% swapper [kernel.vmlinux] [k] poll_idle 6.36% swapper [mlx4_core] [k] mlx4_eq_int 4.21% swapper [kernel.vmlinux] [k] tasklet_action 4.03% swapper [kernel.vmlinux] [k] cpuidle_enter_state 3.43% swapper [mlx4_en] [k] mlx4_en_prepare_rx_desc 2.18% swapper [kernel.vmlinux] [k] native_irq_return_iret 1.37% swapper [kernel.vmlinux] [k] menu_select 1.09% swapper [kernel.vmlinux] [k] bpf_map_lookup_elem Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:52 +08:00
}
ring->page_cache.index = 0;
mlx4_en_free_rx_buf(priv, ring);
if (ring->stride <= TXBB_SIZE)
ring->buf -= TXBB_SIZE;
}
static int mlx4_en_complete_rx_desc(struct mlx4_en_priv *priv,
struct mlx4_en_rx_alloc *frags,
struct sk_buff *skb,
int length)
{
const struct mlx4_en_frag_info *frag_info = priv->frag_info;
unsigned int truesize = 0;
int nr, frag_size;
struct page *page;
dma_addr_t dma;
bool release;
/* Collect used fragments while replacing them in the HW descriptors */
for (nr = 0;; frags++) {
frag_size = min_t(int, length, frag_info->frag_size);
page = frags->page;
if (unlikely(!page))
goto fail;
dma = frags->dma;
dma_sync_single_range_for_cpu(priv->ddev, dma, frags->page_offset,
frag_size, priv->dma_dir);
__skb_fill_page_desc(skb, nr, page, frags->page_offset,
frag_size);
truesize += frag_info->frag_stride;
if (frag_info->frag_stride == PAGE_SIZE / 2) {
frags->page_offset ^= PAGE_SIZE / 2;
release = page_count(page) != 1 ||
page_is_pfmemalloc(page) ||
page_to_nid(page) != numa_mem_id();
} else {
u32 sz_align = ALIGN(frag_size, SMP_CACHE_BYTES);
frags->page_offset += sz_align;
release = frags->page_offset + frag_info->frag_size > PAGE_SIZE;
}
if (release) {
dma_unmap_page(priv->ddev, dma, PAGE_SIZE, priv->dma_dir);
frags->page = NULL;
} else {
page_ref_inc(page);
}
nr++;
length -= frag_size;
if (!length)
break;
frag_info++;
}
skb->truesize += truesize;
return nr;
fail:
while (nr > 0) {
nr--;
__skb_frag_unref(skb_shinfo(skb)->frags + nr);
}
return 0;
}
static void validate_loopback(struct mlx4_en_priv *priv, void *va)
{
const unsigned char *data = va + ETH_HLEN;
int i;
for (i = 0; i < MLX4_LOOPBACK_TEST_PAYLOAD; i++) {
if (data[i] != (unsigned char)i)
return;
}
/* Loopback found */
priv->loopback_ok = 1;
}
static void mlx4_en_refill_rx_buffers(struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring *ring)
{
u32 missing = ring->actual_size - (ring->prod - ring->cons);
/* Try to batch allocations, but not too much. */
if (missing < 8)
return;
do {
if (mlx4_en_prepare_rx_desc(priv, ring,
ring->prod & ring->size_mask,
GFP_ATOMIC | __GFP_COLD |
__GFP_MEMALLOC))
break;
ring->prod++;
} while (likely(--missing));
mlx4_en_update_rx_prod_db(ring);
}
/* When hardware doesn't strip the vlan, we need to calculate the checksum
* over it and add it to the hardware's checksum calculation
*/
static inline __wsum get_fixed_vlan_csum(__wsum hw_checksum,
struct vlan_hdr *vlanh)
{
return csum_add(hw_checksum, *(__wsum *)vlanh);
}
/* Although the stack expects checksum which doesn't include the pseudo
* header, the HW adds it. To address that, we are subtracting the pseudo
* header checksum from the checksum value provided by the HW.
*/
static int get_fixed_ipv4_csum(__wsum hw_checksum, struct sk_buff *skb,
struct iphdr *iph)
{
__u16 length_for_csum = 0;
__wsum csum_pseudo_header = 0;
__u8 ipproto = iph->protocol;
if (unlikely(ipproto == IPPROTO_SCTP))
return -1;
length_for_csum = (be16_to_cpu(iph->tot_len) - (iph->ihl << 2));
csum_pseudo_header = csum_tcpudp_nofold(iph->saddr, iph->daddr,
length_for_csum, ipproto, 0);
skb->csum = csum_sub(hw_checksum, csum_pseudo_header);
return 0;
}
#if IS_ENABLED(CONFIG_IPV6)
/* In IPv6 packets, besides subtracting the pseudo header checksum,
* we also compute/add the IP header checksum which
* is not added by the HW.
*/
static int get_fixed_ipv6_csum(__wsum hw_checksum, struct sk_buff *skb,
struct ipv6hdr *ipv6h)
{
__u8 nexthdr = ipv6h->nexthdr;
__wsum csum_pseudo_hdr = 0;
if (unlikely(nexthdr == IPPROTO_FRAGMENT ||
nexthdr == IPPROTO_HOPOPTS ||
nexthdr == IPPROTO_SCTP))
return -1;
hw_checksum = csum_add(hw_checksum, (__force __wsum)htons(nexthdr));
csum_pseudo_hdr = csum_partial(&ipv6h->saddr,
sizeof(ipv6h->saddr) + sizeof(ipv6h->daddr), 0);
csum_pseudo_hdr = csum_add(csum_pseudo_hdr, (__force __wsum)ipv6h->payload_len);
csum_pseudo_hdr = csum_add(csum_pseudo_hdr,
(__force __wsum)htons(nexthdr));
skb->csum = csum_sub(hw_checksum, csum_pseudo_hdr);
skb->csum = csum_add(skb->csum, csum_partial(ipv6h, sizeof(struct ipv6hdr), 0));
return 0;
}
#endif
static int check_csum(struct mlx4_cqe *cqe, struct sk_buff *skb, void *va,
netdev_features_t dev_features)
{
__wsum hw_checksum = 0;
void *hdr = (u8 *)va + sizeof(struct ethhdr);
hw_checksum = csum_unfold((__force __sum16)cqe->checksum);
if (cqe->vlan_my_qpn & cpu_to_be32(MLX4_CQE_CVLAN_PRESENT_MASK) &&
!(dev_features & NETIF_F_HW_VLAN_CTAG_RX)) {
hw_checksum = get_fixed_vlan_csum(hw_checksum, hdr);
hdr += sizeof(struct vlan_hdr);
}
if (cqe->status & cpu_to_be16(MLX4_CQE_STATUS_IPV4))
return get_fixed_ipv4_csum(hw_checksum, skb, hdr);
#if IS_ENABLED(CONFIG_IPV6)
if (cqe->status & cpu_to_be16(MLX4_CQE_STATUS_IPV6))
return get_fixed_ipv6_csum(hw_checksum, skb, hdr);
#endif
return 0;
}
int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int budget)
{
struct mlx4_en_priv *priv = netdev_priv(dev);
int factor = priv->cqe_factor;
struct mlx4_en_rx_ring *ring;
struct bpf_prog *xdp_prog;
int cq_ring = cq->ring;
bool doorbell_pending;
struct mlx4_cqe *cqe;
int polled = 0;
int index;
if (unlikely(!priv->port_up))
return 0;
if (unlikely(budget <= 0))
return polled;
ring = priv->rx_ring[cq_ring];
net/mlx4_en: protect ring->xdp_prog with rcu_read_lock Depending on the preempt mode, the bpf_prog stored in xdp_prog may be freed despite the use of call_rcu inside bpf_prog_put. The situation is possible when running in PREEMPT_RCU=y mode, for instance, since the rcu callback for destroying the bpf prog can run even during the bh handling in the mlx4 rx path. Several options were considered before this patch was settled on: Add a napi_synchronize loop in mlx4_xdp_set, which would occur after all of the rings are updated with the new program. This approach has the disadvantage that as the number of rings increases, the speed of update will slow down significantly due to napi_synchronize's msleep(1). Add a new rcu_head in bpf_prog_aux, to be used by a new bpf_prog_put_bh. The action of the bpf_prog_put_bh would be to then call bpf_prog_put later. Those drivers that consume a bpf prog in a bh context (like mlx4) would then use the bpf_prog_put_bh instead when the ring is up. This has the problem of complexity, in maintaining proper refcnts and rcu lists, and would likely be harder to review. In addition, this approach to freeing must be exclusive with other frees of the bpf prog, for instance a _bh prog must not be referenced from a prog array that is consumed by a non-_bh prog. The placement of rcu_read_lock in this patch is functionally the same as putting an rcu_read_lock in napi_poll. Actually doing so could be a potentially controversial change, but would bring the implementation in line with sk_busy_loop (though of course the nature of those two paths is substantially different), and would also avoid future copy/paste problems with future supporters of XDP. Still, this patch does not take that opinionated option. Testing was done with kernels in either PREEMPT_RCU=y or CONFIG_PREEMPT_VOLUNTARY=y+PREEMPT_RCU=n modes, with neither exhibiting any drawback. With PREEMPT_RCU=n, the extra call to rcu_read_lock did not show up in the perf report whatsoever, and with PREEMPT_RCU=y the overhead of rcu_read_lock (according to perf) was the same before/after. In the rx path, rcu_read_lock is eventually called for every packet from netif_receive_skb_internal, so the napi poll call's rcu_read_lock is easily amortized. v2: Remove extra rcu_read_lock in mlx4_en_process_rx_cq body Annotate xdp_prog with __rcu, and convert all usages to rcu_assign or rcu_dereference[_protected] as appropriate. Add explicit mutex lock around rcu_assign instead of xchg loop. Fixes: d576acf0a22 ("net/mlx4_en: add page recycle to prepare rx ring for tx support") Acked-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <alexei.starovoitov@gmail.com> Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-04 12:29:58 +08:00
/* Protect accesses to: ring->xdp_prog, priv->mac_hash list */
rcu_read_lock();
xdp_prog = rcu_dereference(ring->xdp_prog);
doorbell_pending = 0;
/* We assume a 1:1 mapping between CQEs and Rx descriptors, so Rx
* descriptor offset can be deduced from the CQE index instead of
* reading 'cqe->index' */
index = cq->mcq.cons_index & ring->size_mask;
cqe = mlx4_en_get_cqe(cq->buf, index, priv->cqe_size) + factor;
/* Process all completed CQEs */
while (XNOR(cqe->owner_sr_opcode & MLX4_CQE_OWNER_MASK,
cq->mcq.cons_index & cq->size)) {
struct mlx4_en_rx_alloc *frags;
enum pkt_hash_types hash_type;
struct sk_buff *skb;
unsigned int length;
int ip_summed;
void *va;
int nr;
frags = ring->rx_info + (index << priv->log_rx_info);
va = page_address(frags[0].page) + frags[0].page_offset;
prefetchw(va);
/*
* make sure we read the CQE after we read the ownership bit
*/
dma_rmb();
/* Drop packet on bad receive or bad checksum */
if (unlikely((cqe->owner_sr_opcode & MLX4_CQE_OPCODE_MASK) ==
MLX4_CQE_OPCODE_ERROR)) {
en_err(priv, "CQE completed in error - vendor syndrom:%d syndrom:%d\n",
((struct mlx4_err_cqe *)cqe)->vendor_err_syndrome,
((struct mlx4_err_cqe *)cqe)->syndrome);
goto next;
}
if (unlikely(cqe->badfcs_enc & MLX4_CQE_BAD_FCS)) {
en_dbg(RX_ERR, priv, "Accepted frame with bad FCS\n");
goto next;
}
/* Check if we need to drop the packet if SRIOV is not enabled
* and not performing the selftest or flb disabled
*/
if (priv->flags & MLX4_EN_FLAG_RX_FILTER_NEEDED) {
const struct ethhdr *ethh = va;
dma_addr_t dma;
/* Get pointer to first fragment since we haven't
* skb yet and cast it to ethhdr struct
*/
dma = frags[0].dma + frags[0].page_offset;
dma_sync_single_for_cpu(priv->ddev, dma, sizeof(*ethh),
DMA_FROM_DEVICE);
if (is_multicast_ether_addr(ethh->h_dest)) {
struct mlx4_mac_entry *entry;
struct hlist_head *bucket;
unsigned int mac_hash;
/* Drop the packet, since HW loopback-ed it */
mac_hash = ethh->h_source[MLX4_EN_MAC_HASH_IDX];
bucket = &priv->mac_hash[mac_hash];
hlist: drop the node parameter from iterators I'm not sure why, but the hlist for each entry iterators were conceived list_for_each_entry(pos, head, member) The hlist ones were greedy and wanted an extra parameter: hlist_for_each_entry(tpos, pos, head, member) Why did they need an extra pos parameter? I'm not quite sure. Not only they don't really need it, it also prevents the iterator from looking exactly like the list iterator, which is unfortunate. Besides the semantic patch, there was some manual work required: - Fix up the actual hlist iterators in linux/list.h - Fix up the declaration of other iterators based on the hlist ones. - A very small amount of places were using the 'node' parameter, this was modified to use 'obj->member' instead. - Coccinelle didn't handle the hlist_for_each_entry_safe iterator properly, so those had to be fixed up manually. The semantic patch which is mostly the work of Peter Senna Tschudin is here: @@ iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host; type T; expression a,c,d,e; identifier b; statement S; @@ -T b; <+... when != b ( hlist_for_each_entry(a, - b, c, d) S | hlist_for_each_entry_continue(a, - b, c) S | hlist_for_each_entry_from(a, - b, c) S | hlist_for_each_entry_rcu(a, - b, c, d) S | hlist_for_each_entry_rcu_bh(a, - b, c, d) S | hlist_for_each_entry_continue_rcu_bh(a, - b, c) S | for_each_busy_worker(a, c, - b, d) S | ax25_uid_for_each(a, - b, c) S | ax25_for_each(a, - b, c) S | inet_bind_bucket_for_each(a, - b, c) S | sctp_for_each_hentry(a, - b, c) S | sk_for_each(a, - b, c) S | sk_for_each_rcu(a, - b, c) S | sk_for_each_from -(a, b) +(a) S + sk_for_each_from(a) S | sk_for_each_safe(a, - b, c, d) S | sk_for_each_bound(a, - b, c) S | hlist_for_each_entry_safe(a, - b, c, d, e) S | hlist_for_each_entry_continue_rcu(a, - b, c) S | nr_neigh_for_each(a, - b, c) S | nr_neigh_for_each_safe(a, - b, c, d) S | nr_node_for_each(a, - b, c) S | nr_node_for_each_safe(a, - b, c, d) S | - for_each_gfn_sp(a, c, d, b) S + for_each_gfn_sp(a, c, d) S | - for_each_gfn_indirect_valid_sp(a, c, d, b) S + for_each_gfn_indirect_valid_sp(a, c, d) S | for_each_host(a, - b, c) S | for_each_host_safe(a, - b, c, d) S | for_each_mesh_entry(a, - b, c, d) S ) ...+> [akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c] [akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c] [akpm@linux-foundation.org: checkpatch fixes] [akpm@linux-foundation.org: fix warnings] [akpm@linux-foudnation.org: redo intrusive kvm changes] Tested-by: Peter Senna Tschudin <peter.senna@gmail.com> Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Sasha Levin <sasha.levin@oracle.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Gleb Natapov <gleb@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 09:06:00 +08:00
hlist_for_each_entry_rcu(entry, bucket, hlist) {
if (ether_addr_equal_64bits(entry->mac,
net/mlx4_en: protect ring->xdp_prog with rcu_read_lock Depending on the preempt mode, the bpf_prog stored in xdp_prog may be freed despite the use of call_rcu inside bpf_prog_put. The situation is possible when running in PREEMPT_RCU=y mode, for instance, since the rcu callback for destroying the bpf prog can run even during the bh handling in the mlx4 rx path. Several options were considered before this patch was settled on: Add a napi_synchronize loop in mlx4_xdp_set, which would occur after all of the rings are updated with the new program. This approach has the disadvantage that as the number of rings increases, the speed of update will slow down significantly due to napi_synchronize's msleep(1). Add a new rcu_head in bpf_prog_aux, to be used by a new bpf_prog_put_bh. The action of the bpf_prog_put_bh would be to then call bpf_prog_put later. Those drivers that consume a bpf prog in a bh context (like mlx4) would then use the bpf_prog_put_bh instead when the ring is up. This has the problem of complexity, in maintaining proper refcnts and rcu lists, and would likely be harder to review. In addition, this approach to freeing must be exclusive with other frees of the bpf prog, for instance a _bh prog must not be referenced from a prog array that is consumed by a non-_bh prog. The placement of rcu_read_lock in this patch is functionally the same as putting an rcu_read_lock in napi_poll. Actually doing so could be a potentially controversial change, but would bring the implementation in line with sk_busy_loop (though of course the nature of those two paths is substantially different), and would also avoid future copy/paste problems with future supporters of XDP. Still, this patch does not take that opinionated option. Testing was done with kernels in either PREEMPT_RCU=y or CONFIG_PREEMPT_VOLUNTARY=y+PREEMPT_RCU=n modes, with neither exhibiting any drawback. With PREEMPT_RCU=n, the extra call to rcu_read_lock did not show up in the perf report whatsoever, and with PREEMPT_RCU=y the overhead of rcu_read_lock (according to perf) was the same before/after. In the rx path, rcu_read_lock is eventually called for every packet from netif_receive_skb_internal, so the napi poll call's rcu_read_lock is easily amortized. v2: Remove extra rcu_read_lock in mlx4_en_process_rx_cq body Annotate xdp_prog with __rcu, and convert all usages to rcu_assign or rcu_dereference[_protected] as appropriate. Add explicit mutex lock around rcu_assign instead of xchg loop. Fixes: d576acf0a22 ("net/mlx4_en: add page recycle to prepare rx ring for tx support") Acked-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <alexei.starovoitov@gmail.com> Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-04 12:29:58 +08:00
ethh->h_source))
goto next;
}
}
}
if (unlikely(priv->validate_loopback)) {
validate_loopback(priv, va);
goto next;
}
/*
* Packet is OK - process it.
*/
length = be32_to_cpu(cqe->byte_cnt);
length -= ring->fcs_del;
/* A bpf program gets first chance to drop the packet. It may
* read bytes but not past the end of the frag.
*/
if (xdp_prog) {
struct xdp_buff xdp;
dma_addr_t dma;
void *orig_data;
u32 act;
dma = frags[0].dma + frags[0].page_offset;
dma_sync_single_for_cpu(priv->ddev, dma,
priv->frag_info[0].frag_size,
DMA_FROM_DEVICE);
xdp.data_hard_start = va - frags[0].page_offset;
xdp.data = va;
bpf: add meta pointer for direct access This work enables generic transfer of metadata from XDP into skb. The basic idea is that we can make use of the fact that the resulting skb must be linear and already comes with a larger headroom for supporting bpf_xdp_adjust_head(), which mangles xdp->data. Here, we base our work on a similar principle and introduce a small helper bpf_xdp_adjust_meta() for adjusting a new pointer called xdp->data_meta. Thus, the packet has a flexible and programmable room for meta data, followed by the actual packet data. struct xdp_buff is therefore laid out that we first point to data_hard_start, then data_meta directly prepended to data followed by data_end marking the end of packet. bpf_xdp_adjust_head() takes into account whether we have meta data already prepended and if so, memmove()s this along with the given offset provided there's enough room. xdp->data_meta is optional and programs are not required to use it. The rationale is that when we process the packet in XDP (e.g. as DoS filter), we can push further meta data along with it for the XDP_PASS case, and give the guarantee that a clsact ingress BPF program on the same device can pick this up for further post-processing. Since we work with skb there, we can also set skb->mark, skb->priority or other skb meta data out of BPF, thus having this scratch space generic and programmable allows for more flexibility than defining a direct 1:1 transfer of potentially new XDP members into skb (it's also more efficient as we don't need to initialize/handle each of such new members). The facility also works together with GRO aggregation. The scratch space at the head of the packet can be multiple of 4 byte up to 32 byte large. Drivers not yet supporting xdp->data_meta can simply be set up with xdp->data_meta as xdp->data + 1 as bpf_xdp_adjust_meta() will detect this and bail out, such that the subsequent match against xdp->data for later access is guaranteed to fail. The verifier treats xdp->data_meta/xdp->data the same way as we treat xdp->data/xdp->data_end pointer comparisons. The requirement for doing the compare against xdp->data is that it hasn't been modified from it's original address we got from ctx access. It may have a range marking already from prior successful xdp->data/xdp->data_end pointer comparisons though. Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 08:25:51 +08:00
xdp_set_data_meta_invalid(&xdp);
xdp.data_end = xdp.data + length;
orig_data = xdp.data;
act = bpf_prog_run_xdp(xdp_prog, &xdp);
if (xdp.data != orig_data) {
length = xdp.data_end - xdp.data;
frags[0].page_offset = xdp.data -
xdp.data_hard_start;
va = xdp.data;
}
switch (act) {
case XDP_PASS:
break;
case XDP_TX:
if (likely(!mlx4_en_xmit_frame(ring, frags, priv,
length, cq_ring,
&doorbell_pending))) {
frags[0].page = NULL;
goto next;
}
bpf: add initial bpf tracepoints This work adds a number of tracepoints to paths that are either considered slow-path or exception-like states, where monitoring or inspecting them would be desirable. For bpf(2) syscall, tracepoints have been placed for main commands when they succeed. In XDP case, tracepoint is for exceptions, that is, f.e. on abnormal BPF program exit such as unknown or XDP_ABORTED return code, or when error occurs during XDP_TX action and the packet could not be forwarded. Both have been split into separate event headers, and can be further extended. Worst case, if they unexpectedly should get into our way in future, they can also removed [1]. Of course, these tracepoints (like any other) can be analyzed by eBPF itself, etc. Example output: # ./perf record -a -e bpf:* sleep 10 # ./perf script sock_example 6197 [005] 283.980322: bpf:bpf_map_create: map type=ARRAY ufd=4 key=4 val=8 max=256 flags=0 sock_example 6197 [005] 283.980721: bpf:bpf_prog_load: prog=a5ea8fa30ea6849c type=SOCKET_FILTER ufd=5 sock_example 6197 [005] 283.988423: bpf:bpf_prog_get_type: prog=a5ea8fa30ea6849c type=SOCKET_FILTER sock_example 6197 [005] 283.988443: bpf:bpf_map_lookup_elem: map type=ARRAY ufd=4 key=[06 00 00 00] val=[00 00 00 00 00 00 00 00] [...] sock_example 6197 [005] 288.990868: bpf:bpf_map_lookup_elem: map type=ARRAY ufd=4 key=[01 00 00 00] val=[14 00 00 00 00 00 00 00] swapper 0 [005] 289.338243: bpf:bpf_prog_put_rcu: prog=a5ea8fa30ea6849c type=SOCKET_FILTER [1] https://lwn.net/Articles/705270/ Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-25 09:28:18 +08:00
trace_xdp_exception(dev, xdp_prog, act);
goto xdp_drop_no_cnt; /* Drop on xmit failure */
default:
bpf_warn_invalid_xdp_action(act);
case XDP_ABORTED:
bpf: add initial bpf tracepoints This work adds a number of tracepoints to paths that are either considered slow-path or exception-like states, where monitoring or inspecting them would be desirable. For bpf(2) syscall, tracepoints have been placed for main commands when they succeed. In XDP case, tracepoint is for exceptions, that is, f.e. on abnormal BPF program exit such as unknown or XDP_ABORTED return code, or when error occurs during XDP_TX action and the packet could not be forwarded. Both have been split into separate event headers, and can be further extended. Worst case, if they unexpectedly should get into our way in future, they can also removed [1]. Of course, these tracepoints (like any other) can be analyzed by eBPF itself, etc. Example output: # ./perf record -a -e bpf:* sleep 10 # ./perf script sock_example 6197 [005] 283.980322: bpf:bpf_map_create: map type=ARRAY ufd=4 key=4 val=8 max=256 flags=0 sock_example 6197 [005] 283.980721: bpf:bpf_prog_load: prog=a5ea8fa30ea6849c type=SOCKET_FILTER ufd=5 sock_example 6197 [005] 283.988423: bpf:bpf_prog_get_type: prog=a5ea8fa30ea6849c type=SOCKET_FILTER sock_example 6197 [005] 283.988443: bpf:bpf_map_lookup_elem: map type=ARRAY ufd=4 key=[06 00 00 00] val=[00 00 00 00 00 00 00 00] [...] sock_example 6197 [005] 288.990868: bpf:bpf_map_lookup_elem: map type=ARRAY ufd=4 key=[01 00 00 00] val=[14 00 00 00 00 00 00 00] swapper 0 [005] 289.338243: bpf:bpf_prog_put_rcu: prog=a5ea8fa30ea6849c type=SOCKET_FILTER [1] https://lwn.net/Articles/705270/ Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-25 09:28:18 +08:00
trace_xdp_exception(dev, xdp_prog, act);
case XDP_DROP:
ring->xdp_drop++;
xdp_drop_no_cnt:
goto next;
}
}
ring->bytes += length;
ring->packets++;
skb = napi_get_frags(&cq->napi);
if (unlikely(!skb))
goto next;
if (unlikely(ring->hwtstamp_rx_filter == HWTSTAMP_FILTER_ALL)) {
u64 timestamp = mlx4_en_get_cqe_ts(cqe);
mlx4_en_fill_hwtstamps(priv->mdev, skb_hwtstamps(skb),
timestamp);
}
skb_record_rx_queue(skb, cq_ring);
if (likely(dev->features & NETIF_F_RXCSUM)) {
if (cqe->status & cpu_to_be16(MLX4_CQE_STATUS_TCP |
MLX4_CQE_STATUS_UDP)) {
if ((cqe->status & cpu_to_be16(MLX4_CQE_STATUS_IPOK)) &&
cqe->checksum == cpu_to_be16(0xffff)) {
bool l2_tunnel = (dev->hw_enc_features & NETIF_F_RXCSUM) &&
(cqe->vlan_my_qpn & cpu_to_be32(MLX4_CQE_L2_TUNNEL));
ip_summed = CHECKSUM_UNNECESSARY;
hash_type = PKT_HASH_TYPE_L4;
if (l2_tunnel)
skb->csum_level = 1;
ring->csum_ok++;
} else {
goto csum_none;
}
} else {
if (priv->flags & MLX4_EN_FLAG_RX_CSUM_NON_TCP_UDP &&
(cqe->status & cpu_to_be16(MLX4_CQE_STATUS_IPV4 |
MLX4_CQE_STATUS_IPV6))) {
if (check_csum(cqe, skb, va, dev->features)) {
goto csum_none;
} else {
ip_summed = CHECKSUM_COMPLETE;
hash_type = PKT_HASH_TYPE_L3;
ring->csum_complete++;
}
} else {
goto csum_none;
}
}
} else {
csum_none:
ip_summed = CHECKSUM_NONE;
hash_type = PKT_HASH_TYPE_L3;
ring->csum_none++;
}
skb->ip_summed = ip_summed;
if (dev->features & NETIF_F_RXHASH)
skb_set_hash(skb,
be32_to_cpu(cqe->immed_rss_invalid),
hash_type);
if ((cqe->vlan_my_qpn &
cpu_to_be32(MLX4_CQE_CVLAN_PRESENT_MASK)) &&
(dev->features & NETIF_F_HW_VLAN_CTAG_RX))
__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
be16_to_cpu(cqe->sl_vid));
else if ((cqe->vlan_my_qpn &
cpu_to_be32(MLX4_CQE_SVLAN_PRESENT_MASK)) &&
(dev->features & NETIF_F_HW_VLAN_STAG_RX))
__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021AD),
be16_to_cpu(cqe->sl_vid));
nr = mlx4_en_complete_rx_desc(priv, frags, skb, length);
if (likely(nr)) {
skb_shinfo(skb)->nr_frags = nr;
skb->len = length;
skb->data_len = length;
napi_gro_frags(&cq->napi);
} else {
skb->vlan_tci = 0;
skb_clear_hash(skb);
}
next:
++cq->mcq.cons_index;
index = (cq->mcq.cons_index) & ring->size_mask;
cqe = mlx4_en_get_cqe(cq->buf, index, priv->cqe_size) + factor;
if (unlikely(++polled == budget))
break;
}
net/mlx4_en: protect ring->xdp_prog with rcu_read_lock Depending on the preempt mode, the bpf_prog stored in xdp_prog may be freed despite the use of call_rcu inside bpf_prog_put. The situation is possible when running in PREEMPT_RCU=y mode, for instance, since the rcu callback for destroying the bpf prog can run even during the bh handling in the mlx4 rx path. Several options were considered before this patch was settled on: Add a napi_synchronize loop in mlx4_xdp_set, which would occur after all of the rings are updated with the new program. This approach has the disadvantage that as the number of rings increases, the speed of update will slow down significantly due to napi_synchronize's msleep(1). Add a new rcu_head in bpf_prog_aux, to be used by a new bpf_prog_put_bh. The action of the bpf_prog_put_bh would be to then call bpf_prog_put later. Those drivers that consume a bpf prog in a bh context (like mlx4) would then use the bpf_prog_put_bh instead when the ring is up. This has the problem of complexity, in maintaining proper refcnts and rcu lists, and would likely be harder to review. In addition, this approach to freeing must be exclusive with other frees of the bpf prog, for instance a _bh prog must not be referenced from a prog array that is consumed by a non-_bh prog. The placement of rcu_read_lock in this patch is functionally the same as putting an rcu_read_lock in napi_poll. Actually doing so could be a potentially controversial change, but would bring the implementation in line with sk_busy_loop (though of course the nature of those two paths is substantially different), and would also avoid future copy/paste problems with future supporters of XDP. Still, this patch does not take that opinionated option. Testing was done with kernels in either PREEMPT_RCU=y or CONFIG_PREEMPT_VOLUNTARY=y+PREEMPT_RCU=n modes, with neither exhibiting any drawback. With PREEMPT_RCU=n, the extra call to rcu_read_lock did not show up in the perf report whatsoever, and with PREEMPT_RCU=y the overhead of rcu_read_lock (according to perf) was the same before/after. In the rx path, rcu_read_lock is eventually called for every packet from netif_receive_skb_internal, so the napi poll call's rcu_read_lock is easily amortized. v2: Remove extra rcu_read_lock in mlx4_en_process_rx_cq body Annotate xdp_prog with __rcu, and convert all usages to rcu_assign or rcu_dereference[_protected] as appropriate. Add explicit mutex lock around rcu_assign instead of xchg loop. Fixes: d576acf0a22 ("net/mlx4_en: add page recycle to prepare rx ring for tx support") Acked-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <alexei.starovoitov@gmail.com> Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-04 12:29:58 +08:00
rcu_read_unlock();
if (likely(polled)) {
if (doorbell_pending) {
priv->tx_cq[TX_XDP][cq_ring]->xdp_busy = true;
mlx4_en_xmit_doorbell(priv->tx_ring[TX_XDP][cq_ring]);
}
mlx4_cq_set_ci(&cq->mcq);
wmb(); /* ensure HW sees CQ consumer before we post new buffers */
ring->cons = cq->mcq.cons_index;
}
AVG_PERF_COUNTER(priv->pstats.rx_coal_avg, polled);
mlx4_en_refill_rx_buffers(priv, ring);
return polled;
}
void mlx4_en_rx_irq(struct mlx4_cq *mcq)
{
struct mlx4_en_cq *cq = container_of(mcq, struct mlx4_en_cq, mcq);
struct mlx4_en_priv *priv = netdev_priv(cq->dev);
if (likely(priv->port_up))
napi_schedule_irqoff(&cq->napi);
else
mlx4_en_arm_cq(priv, cq);
}
/* Rx CQ polling - called by NAPI */
int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget)
{
struct mlx4_en_cq *cq = container_of(napi, struct mlx4_en_cq, napi);
struct net_device *dev = cq->dev;
struct mlx4_en_priv *priv = netdev_priv(dev);
struct mlx4_en_cq *xdp_tx_cq = NULL;
bool clean_complete = true;
int done;
if (priv->tx_ring_num[TX_XDP]) {
xdp_tx_cq = priv->tx_cq[TX_XDP][cq->ring];
if (xdp_tx_cq->xdp_busy) {
clean_complete = mlx4_en_process_tx_cq(dev, xdp_tx_cq,
budget);
xdp_tx_cq->xdp_busy = !clean_complete;
}
}
done = mlx4_en_process_rx_cq(dev, cq, budget);
/* If we used up all the quota - we're probably not done yet... */
if (done == budget || !clean_complete) {
const struct cpumask *aff;
struct irq_data *idata;
int cpu_curr;
/* in case we got here because of !clean_complete */
done = budget;
INC_PERF_COUNTER(priv->pstats.napi_quota);
cpu_curr = smp_processor_id();
idata = irq_desc_get_irq_data(cq->irq_desc);
aff = irq_data_get_affinity_mask(idata);
if (likely(cpumask_test_cpu(cpu_curr, aff)))
return budget;
/* Current cpu is not according to smp_irq_affinity -
* probably affinity changed. Need to stop this NAPI
* poll, and restart it on the right CPU.
* Try to avoid returning a too small value (like 0),
* to not fool net_rx_action() and its netdev_budget
*/
if (done)
done--;
}
/* Done for now */
if (likely(napi_complete_done(napi, done)))
mlx4_en_arm_cq(priv, cq);
return done;
}
void mlx4_en_calc_rx_buf(struct net_device *dev)
{
struct mlx4_en_priv *priv = netdev_priv(dev);
int eff_mtu = MLX4_EN_EFF_MTU(dev->mtu);
int i = 0;
net/mlx4_en: add page recycle to prepare rx ring for tx support The mlx4 driver by default allocates order-3 pages for the ring to consume in multiple fragments. When the device has an xdp program, this behavior will prevent tx actions since the page must be re-mapped in TODEVICE mode, which cannot be done if the page is still shared. Start by making the allocator configurable based on whether xdp is running, such that order-0 pages are always used and never shared. Since this will stress the page allocator, add a simple page cache to each rx ring. Pages in the cache are left dma-mapped, and in drop-only stress tests the page allocator is eliminated from the perf report. Note that setting an xdp program will now require the rings to be reconfigured. Before: 26.91% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq 17.88% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags 6.00% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag 4.49% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist 3.21% swapper [kernel.vmlinux] [k] intel_idle 2.73% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem 2.57% swapper [mlx4_en] [k] mlx4_en_process_rx_cq After: 31.72% swapper [kernel.vmlinux] [k] intel_idle 8.79% swapper [mlx4_en] [k] mlx4_en_process_rx_cq 7.54% swapper [kernel.vmlinux] [k] poll_idle 6.36% swapper [mlx4_core] [k] mlx4_eq_int 4.21% swapper [kernel.vmlinux] [k] tasklet_action 4.03% swapper [kernel.vmlinux] [k] cpuidle_enter_state 3.43% swapper [mlx4_en] [k] mlx4_en_prepare_rx_desc 2.18% swapper [kernel.vmlinux] [k] native_irq_return_iret 1.37% swapper [kernel.vmlinux] [k] menu_select 1.09% swapper [kernel.vmlinux] [k] bpf_map_lookup_elem Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:52 +08:00
/* bpf requires buffers to be set up as 1 packet per page.
* This only works when num_frags == 1.
*/
if (priv->tx_ring_num[TX_XDP]) {
priv->frag_info[0].frag_size = eff_mtu;
/* This will gain efficient xdp frame recycling at the
* expense of more costly truesize accounting
net/mlx4_en: add page recycle to prepare rx ring for tx support The mlx4 driver by default allocates order-3 pages for the ring to consume in multiple fragments. When the device has an xdp program, this behavior will prevent tx actions since the page must be re-mapped in TODEVICE mode, which cannot be done if the page is still shared. Start by making the allocator configurable based on whether xdp is running, such that order-0 pages are always used and never shared. Since this will stress the page allocator, add a simple page cache to each rx ring. Pages in the cache are left dma-mapped, and in drop-only stress tests the page allocator is eliminated from the perf report. Note that setting an xdp program will now require the rings to be reconfigured. Before: 26.91% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq 17.88% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags 6.00% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag 4.49% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist 3.21% swapper [kernel.vmlinux] [k] intel_idle 2.73% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem 2.57% swapper [mlx4_en] [k] mlx4_en_process_rx_cq After: 31.72% swapper [kernel.vmlinux] [k] intel_idle 8.79% swapper [mlx4_en] [k] mlx4_en_process_rx_cq 7.54% swapper [kernel.vmlinux] [k] poll_idle 6.36% swapper [mlx4_core] [k] mlx4_eq_int 4.21% swapper [kernel.vmlinux] [k] tasklet_action 4.03% swapper [kernel.vmlinux] [k] cpuidle_enter_state 3.43% swapper [mlx4_en] [k] mlx4_en_prepare_rx_desc 2.18% swapper [kernel.vmlinux] [k] native_irq_return_iret 1.37% swapper [kernel.vmlinux] [k] menu_select 1.09% swapper [kernel.vmlinux] [k] bpf_map_lookup_elem Signed-off-by: Brenden Blanco <bblanco@plumgrid.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:52 +08:00
*/
priv->frag_info[0].frag_stride = PAGE_SIZE;
priv->dma_dir = PCI_DMA_BIDIRECTIONAL;
priv->rx_headroom = XDP_PACKET_HEADROOM;
i = 1;
} else {
int frag_size_max = 2048, buf_size = 0;
/* should not happen, right ? */
if (eff_mtu > PAGE_SIZE + (MLX4_EN_MAX_RX_FRAGS - 1) * 2048)
frag_size_max = PAGE_SIZE;
while (buf_size < eff_mtu) {
int frag_stride, frag_size = eff_mtu - buf_size;
int pad, nb;
if (i < MLX4_EN_MAX_RX_FRAGS - 1)
frag_size = min(frag_size, frag_size_max);
priv->frag_info[i].frag_size = frag_size;
frag_stride = ALIGN(frag_size, SMP_CACHE_BYTES);
/* We can only pack 2 1536-bytes frames in on 4K page
* Therefore, each frame would consume more bytes (truesize)
*/
nb = PAGE_SIZE / frag_stride;
pad = (PAGE_SIZE - nb * frag_stride) / nb;
pad &= ~(SMP_CACHE_BYTES - 1);
priv->frag_info[i].frag_stride = frag_stride + pad;
buf_size += frag_size;
i++;
}
priv->dma_dir = PCI_DMA_FROMDEVICE;
priv->rx_headroom = 0;
}
priv->num_frags = i;
priv->rx_skb_size = eff_mtu;
priv->log_rx_info = ROUNDUP_LOG2(i * sizeof(struct mlx4_en_rx_alloc));
en_dbg(DRV, priv, "Rx buffer scatter-list (effective-mtu:%d num_frags:%d):\n",
eff_mtu, priv->num_frags);
for (i = 0; i < priv->num_frags; i++) {
en_dbg(DRV,
priv,
" frag:%d - size:%d stride:%d\n",
i,
priv->frag_info[i].frag_size,
priv->frag_info[i].frag_stride);
}
}
/* RSS related functions */
static int mlx4_en_config_rss_qp(struct mlx4_en_priv *priv, int qpn,
struct mlx4_en_rx_ring *ring,
enum mlx4_qp_state *state,
struct mlx4_qp *qp)
{
struct mlx4_en_dev *mdev = priv->mdev;
struct mlx4_qp_context *context;
int err = 0;
context = kmalloc(sizeof(*context), GFP_KERNEL);
if (!context)
return -ENOMEM;
err = mlx4_qp_alloc(mdev->dev, qpn, qp);
if (err) {
en_err(priv, "Failed to allocate qp #%x\n", qpn);
goto out;
}
qp->event = mlx4_en_sqp_event;
memset(context, 0, sizeof(*context));
mlx4_en_fill_qp_context(priv, ring->actual_size, ring->stride, 0, 0,
qpn, ring->cqn, -1, context);
context->db_rec_addr = cpu_to_be64(ring->wqres.db.dma);
/* Cancel FCS removal if FW allows */
if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP) {
context->param3 |= cpu_to_be32(1 << 29);
if (priv->dev->features & NETIF_F_RXFCS)
ring->fcs_del = 0;
else
ring->fcs_del = ETH_FCS_LEN;
} else
ring->fcs_del = 0;
err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, context, qp, state);
if (err) {
mlx4_qp_remove(mdev->dev, qp);
mlx4_qp_free(mdev->dev, qp);
}
mlx4_en_update_rx_prod_db(ring);
out:
kfree(context);
return err;
}
int mlx4_en_create_drop_qp(struct mlx4_en_priv *priv)
{
int err;
u32 qpn;
net/mlx4: Add A0 hybrid steering A0 hybrid steering is a form of high performance flow steering. By using this mode, mlx4 cards use a fast limited table based steering, in order to enable fast steering of unicast packets to a QP. In order to implement A0 hybrid steering we allocate resources from different zones: (1) General range (2) Special MAC-assigned QPs [RSS, Raw-Ethernet] each has its own region. When we create a rss QP or a raw ethernet (A0 steerable and BF ready) QP, we try hard to allocate the QP from range (2). Otherwise, we try hard not to allocate from this range. However, when the system is pushed to its limits and one needs every resource, the allocator uses every region it can. Meaning, when we run out of raw-eth qps, the allocator allocates from the general range (and the special-A0 area is no longer active). If we run out of RSS qps, the mechanism tries to allocate from the raw-eth QP zone. If that is also exhausted, the allocator will allocate from the general range (and the A0 region is no longer active). Note that if a raw-eth qp is allocated from the general range, it attempts to allocate the range such that bits 6 and 7 (blueflame bits) in the QP number are not set. When the feature is used in SRIOV, the VF has to notify the PF what kind of QP attributes it needs. In order to do that, along with the "Eth QP blueflame" bit, we reserve a new "A0 steerable QP". According to the combination of these bits, the PF tries to allocate a suitable QP. In order to maintain backward compatibility (with older PFs), the PF notifies which QP attributes it supports via QUERY_FUNC_CAP command. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 16:57:57 +08:00
err = mlx4_qp_reserve_range(priv->mdev->dev, 1, 1, &qpn,
MLX4_RESERVE_A0_QP,
MLX4_RES_USAGE_DRIVER);
if (err) {
en_err(priv, "Failed reserving drop qpn\n");
return err;
}
err = mlx4_qp_alloc(priv->mdev->dev, qpn, &priv->drop_qp);
if (err) {
en_err(priv, "Failed allocating drop qp\n");
mlx4_qp_release_range(priv->mdev->dev, qpn, 1);
return err;
}
return 0;
}
void mlx4_en_destroy_drop_qp(struct mlx4_en_priv *priv)
{
u32 qpn;
qpn = priv->drop_qp.qpn;
mlx4_qp_remove(priv->mdev->dev, &priv->drop_qp);
mlx4_qp_free(priv->mdev->dev, &priv->drop_qp);
mlx4_qp_release_range(priv->mdev->dev, qpn, 1);
}
/* Allocate rx qp's and configure them according to rss map */
int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv)
{
struct mlx4_en_dev *mdev = priv->mdev;
struct mlx4_en_rss_map *rss_map = &priv->rss_map;
struct mlx4_qp_context context;
struct mlx4_rss_context *rss_context;
int rss_rings;
void *ptr;
u8 rss_mask = (MLX4_RSS_IPV4 | MLX4_RSS_TCP_IPV4 | MLX4_RSS_IPV6 |
MLX4_RSS_TCP_IPV6);
int i, qpn;
int err = 0;
int good_qps = 0;
u8 flags;
en_dbg(DRV, priv, "Configuring rss steering\n");
flags = priv->rx_ring_num == 1 ? MLX4_RESERVE_A0_QP : 0;
err = mlx4_qp_reserve_range(mdev->dev, priv->rx_ring_num,
priv->rx_ring_num,
&rss_map->base_qpn, flags,
MLX4_RES_USAGE_DRIVER);
if (err) {
en_err(priv, "Failed reserving %d qps\n", priv->rx_ring_num);
return err;
}
for (i = 0; i < priv->rx_ring_num; i++) {
qpn = rss_map->base_qpn + i;
err = mlx4_en_config_rss_qp(priv, qpn, priv->rx_ring[i],
&rss_map->state[i],
&rss_map->qps[i]);
if (err)
goto rss_err;
++good_qps;
}
if (priv->rx_ring_num == 1) {
rss_map->indir_qp = &rss_map->qps[0];
priv->base_qpn = rss_map->indir_qp->qpn;
en_info(priv, "Optimized Non-RSS steering\n");
return 0;
}
rss_map->indir_qp = kzalloc(sizeof(*rss_map->indir_qp), GFP_KERNEL);
if (!rss_map->indir_qp) {
err = -ENOMEM;
goto rss_err;
}
/* Configure RSS indirection qp */
err = mlx4_qp_alloc(mdev->dev, priv->base_qpn, rss_map->indir_qp);
if (err) {
en_err(priv, "Failed to allocate RSS indirection QP\n");
goto rss_err;
}
rss_map->indir_qp->event = mlx4_en_sqp_event;
mlx4_en_fill_qp_context(priv, 0, 0, 0, 1, priv->base_qpn,
priv->rx_ring[0]->cqn, -1, &context);
if (!priv->prof->rss_rings || priv->prof->rss_rings > priv->rx_ring_num)
rss_rings = priv->rx_ring_num;
else
rss_rings = priv->prof->rss_rings;
ptr = ((void *) &context) + offsetof(struct mlx4_qp_context, pri_path)
+ MLX4_RSS_OFFSET_IN_QPC_PRI_PATH;
rss_context = ptr;
rss_context->base_qpn = cpu_to_be32(ilog2(rss_rings) << 24 |
(rss_map->base_qpn));
rss_context->default_qpn = cpu_to_be32(rss_map->base_qpn);
if (priv->mdev->profile.udp_rss) {
rss_mask |= MLX4_RSS_UDP_IPV4 | MLX4_RSS_UDP_IPV6;
rss_context->base_qpn_udp = rss_context->default_qpn;
}
if (mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) {
en_info(priv, "Setting RSS context tunnel type to RSS on inner headers\n");
rss_mask |= MLX4_RSS_BY_INNER_HEADERS;
}
rss_context->flags = rss_mask;
rss_context->hash_fn = MLX4_RSS_HASH_TOP;
if (priv->rss_hash_fn == ETH_RSS_HASH_XOR) {
rss_context->hash_fn = MLX4_RSS_HASH_XOR;
} else if (priv->rss_hash_fn == ETH_RSS_HASH_TOP) {
rss_context->hash_fn = MLX4_RSS_HASH_TOP;
memcpy(rss_context->rss_key, priv->rss_key,
MLX4_EN_RSS_KEY_SIZE);
} else {
en_err(priv, "Unknown RSS hash function requested\n");
err = -EINVAL;
goto indir_err;
}
err = mlx4_qp_to_ready(mdev->dev, &priv->res.mtt, &context,
rss_map->indir_qp, &rss_map->indir_state);
if (err)
goto indir_err;
return 0;
indir_err:
mlx4_qp_modify(mdev->dev, NULL, rss_map->indir_state,
MLX4_QP_STATE_RST, NULL, 0, 0, rss_map->indir_qp);
mlx4_qp_remove(mdev->dev, rss_map->indir_qp);
mlx4_qp_free(mdev->dev, rss_map->indir_qp);
kfree(rss_map->indir_qp);
rss_map->indir_qp = NULL;
rss_err:
for (i = 0; i < good_qps; i++) {
mlx4_qp_modify(mdev->dev, NULL, rss_map->state[i],
MLX4_QP_STATE_RST, NULL, 0, 0, &rss_map->qps[i]);
mlx4_qp_remove(mdev->dev, &rss_map->qps[i]);
mlx4_qp_free(mdev->dev, &rss_map->qps[i]);
}
mlx4_qp_release_range(mdev->dev, rss_map->base_qpn, priv->rx_ring_num);
return err;
}
void mlx4_en_release_rss_steer(struct mlx4_en_priv *priv)
{
struct mlx4_en_dev *mdev = priv->mdev;
struct mlx4_en_rss_map *rss_map = &priv->rss_map;
int i;
if (priv->rx_ring_num > 1) {
mlx4_qp_modify(mdev->dev, NULL, rss_map->indir_state,
MLX4_QP_STATE_RST, NULL, 0, 0,
rss_map->indir_qp);
mlx4_qp_remove(mdev->dev, rss_map->indir_qp);
mlx4_qp_free(mdev->dev, rss_map->indir_qp);
kfree(rss_map->indir_qp);
rss_map->indir_qp = NULL;
}
for (i = 0; i < priv->rx_ring_num; i++) {
mlx4_qp_modify(mdev->dev, NULL, rss_map->state[i],
MLX4_QP_STATE_RST, NULL, 0, 0, &rss_map->qps[i]);
mlx4_qp_remove(mdev->dev, &rss_map->qps[i]);
mlx4_qp_free(mdev->dev, &rss_map->qps[i]);
}
mlx4_qp_release_range(mdev->dev, rss_map->base_qpn, priv->rx_ring_num);
}