* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1745 commits)
dp83640: free packet queues on remove
dp83640: use proper function to free transmit time stamping packets
ipv6: Do not use routes from locally generated RAs
|PATCH net-next] tg3: add tx_dropped counter
be2net: don't create multiple RX/TX rings in multi channel mode
be2net: don't create multiple TXQs in BE2
be2net: refactor VF setup/teardown code into be_vf_setup/clear()
be2net: add vlan/rx-mode/flow-control config to be_setup()
net_sched: cls_flow: use skb_header_pointer()
ipv4: avoid useless call of the function check_peer_pmtu
TCP: remove TCP_DEBUG
net: Fix driver name for mdio-gpio.c
ipv4: tcp: fix TOS value in ACK messages sent from TIME_WAIT
rtnetlink: Add missing manual netlink notification in dev_change_net_namespaces
ipv4: fix ipsec forward performance regression
jme: fix irq storm after suspend/resume
route: fix ICMP redirect validation
net: hold sock reference while processing tx timestamps
tcp: md5: add more const attributes
Add ethtool -g support to virtio_net
...
Fix up conflicts in:
- drivers/net/Kconfig:
The split-up generated a trivial conflict with removal of a
stale reference to Documentation/networking/net-modules.txt.
Remove it from the new location instead.
- fs/sysfs/dir.c:
Fairly nasty conflicts with the sysfs rb-tree usage, conflicting
with Eric Biederman's changes for tagged directories.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (59 commits)
MAINTAINERS: linux-m32r is moderated for non-subscribers
linux@lists.openrisc.net is moderated for non-subscribers
Drop default from "DM365 codec select" choice
parisc: Kconfig: cleanup Kernel page size default
Kconfig: remove redundant CONFIG_ prefix on two symbols
cris: remove arch/cris/arch-v32/lib/nand_init.S
microblaze: add missing CONFIG_ prefixes
h8300: drop puzzling Kconfig dependencies
MAINTAINERS: microblaze-uclinux@itee.uq.edu.au is moderated for non-subscribers
tty: drop superfluous dependency in Kconfig
ARM: mxc: fix Kconfig typo 'i.MX51'
Fix file references in Kconfig files
aic7xxx: fix Kconfig references to READMEs
Fix file references in drivers/ide/
thinkpad_acpi: Fix printk typo 'bluestooth'
bcmring: drop commented out line in Kconfig
btmrvl_sdio: fix typo 'btmrvl_sdio_sd6888'
doc: raw1394: Trivial typo fix
CIFS: Don't free volume_info->UNC until we are entirely done with it.
treewide: Correct spelling of successfully in comments
...
In the rds_iw_mr_pool struct the free_pinned field keeps track of
memory pinned by free MRs. While this field is incremented properly
upon allocation, it is never decremented upon unmapping. This would
cause the rds_rdma module to crash the kernel upon unloading, by
triggering the BUG_ON in the rds_iw_destroy_mr_pool function.
This change keeps track of the MRs that become unpinned, so that
free_pinned can be decremented appropriately.
Signed-off-by: Jonathan Lallinger <jonathan@ogc.us>
Signed-off-by: Steve Wise <swise@ogc.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
The functionality of xlist and llist is almost same. This patch
replace xlist with llist to avoid code duplication.
Known issues: don't know how to test this, need special hardware?
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Andy Grover <andy.grover@oracle.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
We presently define all kinds of notifiers in notifier.h. This is not
necessary at all, since different subsystems use different notifiers, they
are almost non-related with each other.
This can also save much build time. Suppose I add a new netdevice event,
really I don't have to recompile all the source, just network related.
Without this patch, all the source will be recompiled.
I move the notify events near to their subsystem notifier registers, so
that they can be found more easily.
This patch:
It is not necessary to share the same notifier.h.
Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: David Miller <davem@davemloft.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Correct the syntax so that both array and pointer are const.
Signed-off-by: Greg Dietsche <Gregory.Dietsche@cuw.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since printk_ratelimit() shouldn't be used anymore (see comment in
include/linux/printk.h), replace it with printk_ratelimited()
Signed-off-by: Manuel Zerpies <manuel.f.zerpies@ww.stud.uni-erlangen.de>
Signed-off-by: David S. Miller <davem@conan.davemloft.net>
* remove interrupt.g inclusion from netdevice.h -- not needed
* fixup fallout, add interrupt.h and hardirq.h back where needed.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The RDMA CM currently infers the QP type from the port space selected
by the user. In the future (eg with RDMA_PS_IB or XRC), there may not
be a 1-1 correspondence between port space and QP type. For netlink
export of RDMA CM state, we want to export the QP type to userspace,
so it is cleaner to explicitly associate a QP type to an ID.
Modify rdma_create_id() to allow the user to specify the QP type, and
use it to make our selections of datagram versus connected mode.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
As a preparation for removing ext2 non-atomic bit operations from
asm/bitops.h. This converts ext2 non-atomic bit operations to
little-endian bit operations.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Andy Grover <andy.grover@oracle.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
asm-generic/bitops/le.h is only intended to be included directly from
asm-generic/bitops/ext2-non-atomic.h or asm-generic/bitops/minix-le.h
which implements generic ext2 or minix bit operations.
This stops including asm-generic/bitops/le.h directly and use ext2
non-atomic bit operations instead.
It seems odd to use ext2_*_bit() on rds, but it will replaced with
__{set,clear,test}_bit_le() after introducing little endian bit operations
for all architectures. This indirect step is necessary to maintain
bisectability for some architectures which have their own little-endian
bit operations.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Andy Grover <andy.grover@oracle.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1480 commits)
bonding: enable netpoll without checking link status
xfrm: Refcount destination entry on xfrm_lookup
net: introduce rx_handler results and logic around that
bonding: get rid of IFF_SLAVE_INACTIVE netdev->priv_flag
bonding: wrap slave state work
net: get rid of multiple bond-related netdevice->priv_flags
bonding: register slave pointer for rx_handler
be2net: Bump up the version number
be2net: Copyright notice change. Update to Emulex instead of ServerEngines
e1000e: fix kconfig for crc32 dependency
netfilter ebtables: fix xt_AUDIT to work with ebtables
xen network backend driver
bonding: Improve syslog message at device creation time
bonding: Call netif_carrier_off after register_netdevice
bonding: Incorrect TX queue offset
net_sched: fix ip_tos2prio
xfrm: fix __xfrm_route_forward()
be2net: Fix UDP packet detected status in RX compl
Phonet: fix aligned-mode pipe socket buffer header reserve
netxen: support for GbE port settings
...
Fix up conflicts in drivers/staging/brcm80211/brcmsmac/wl_mac80211.c
with the staging updates.
* 'for-2.6.39' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: fix build failure introduced by s/freezeable/freezable/
workqueue: add system_freezeable_wq
rds/ib: use system_wq instead of rds_ib_fmr_wq
net/9p: replace p9_poll_task with a work
net/9p: use system_wq instead of p9_mux_wq
xfs: convert to alloc_workqueue()
reiserfs: make commit_wq use the default concurrency level
ocfs2: use system_wq instead of ocfs2_quota_wq
ext4: convert to alloc_workqueue()
scsi/scsi_tgt_lib: scsi_tgtd isn't used in memory reclaim path
scsi/be2iscsi,qla2xxx: convert to alloc_workqueue()
misc/iwmc3200top: use system_wq instead of dedicated workqueues
i2o: use alloc_workqueue() instead of create_workqueue()
acpi: kacpi*_wq don't need WQ_MEM_RECLAIM
fs/aio: aio_wq isn't used in memory reclaim path
input/tps6507x-ts: use system_wq instead of dedicated workqueue
cpufreq: use system_wq instead of dedicated workqueues
wireless/ipw2x00: use system_wq instead of dedicated workqueues
arm/omap: use system_wq in mailbox
workqueue: use WQ_MEM_RECLAIM instead of WQ_RESCUER
With cmwq, there's no reason to use dedicated rds_ib_fmr_wq - it's not
in the memory reclaim path and the maximum number of concurrent work
items is bound by the number of devices. Drop it and use system_wq
instead. This rds_ib_fmr_init/exit() noops. Both removed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Andy Grover <andy.grover@oracle.com>
Clean up some unused macros in net/*.
1. be left for code change. e.g. PGV_FROM_VMALLOC, PGV_FROM_VMALLOC, KMEM_SAFETYZONE.
2. never be used since introduced to kernel.
e.g. P9_RDMA_MAX_SGE, UTIL_CTRL_PKT_SIZE.
Signed-off-by: Shan Wei <shanwei@cn.fujitsu.com>
Acked-by: Sjur Braendeland <sjur.brandeland@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Changed Makefile to use <modules>-y instead of <modules>-objs
because -objs is deprecated and not mentioned in
Documentation/kbuild/makefiles.txt.
Also, use the ccflags-$ flag instead of EXTRA_CFLAGS because EXTRA_CFLAGS is
deprecated and should now be switched.
Last but not least, took out if-conditionals.
Signed-off-by: Tracey Dent <tdent48227@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In rds_cmsg_rdma_args(), the user-provided args->nr_local value is
restricted to less than UINT_MAX. This seems to need a tighter upper
bound, since the calculation of total iov_size can overflow, resulting
in a small sock_kmalloc() allocation. This would probably just result
in walking off the heap and crashing when calling rds_rdma_pages() with
a high count value. If it somehow doesn't crash here, then memory
corruption could occur soon after.
Signed-off-by: Dan Rosenberg <drosenberg@vsecurity.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The sgs allocation error path leaks the allocated message.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All the rds_tcp_connection objects are stored list, but when
being freed it should be removed from there.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The conn is removed from list in there and this requires
proper lock protection.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Even with the previous fix, we still are reading the iovecs once
to determine SGs needed, and then again later on. Preallocating
space for sg lists as part of rds_message seemed like a good idea
but it might be better to not do this. While working to redo that
code, this patch attempts to protect against userspace rewriting
the rds_iovec array between the first and second accesses.
The consequences of this would be either a too-small or too-large
sg list array. Too large is not an issue. This patch changes all
callers of message_alloc_sgs to handle running out of preallocated
sgs, and fail gracefully.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change rds_rdma_pages to take a passed-in rds_iovec array instead
of doing copy_from_user itself.
Change rds_cmsg_rdma_args to copy rds_iovec array once only. This
eliminates the possibility of userspace changing it after our
sanity checks.
Implement stack-based storage for small numbers of iovecs, based
on net/socket.c, to save an alloc in the extremely common case.
Although this patch reduces iovec copies in cmsg_rdma_args to 1,
we still do another one in rds_rdma_extra_size. Getting rid of
that one will be trickier, so it'll be a separate patch.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We don't need to set ret = 0 at the end -- it's initialized to 0.
Also, don't increment s_send_rdma stat if we're exiting with an
error.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rds_cmsg_rdma_args would still return success even if rds_rdma_pages
returned an error (or overflowed).
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As reported by Thomas Pollet, the rdma page counting can overflow. We
get the rdma sizes in 64-bit unsigned entities, but then limit it to
UINT_MAX bytes and shift them down to pages (so with a possible "+1" for
an unaligned address).
So each individual page count fits comfortably in an 'unsigned int' (not
even close to overflowing into signed), but as they are added up, they
might end up resulting in a signed return value. Which would be wrong.
Catch the case of tot_pages turning negative, and return the appropriate
error code.
Reported-by: Thomas Pollet <thomas.pollet@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The RDS protocol has lots of functions that should be
declared static. rds_message_get/add_version_extension is
removed since it defined but never used.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't try to "optimize" rds_page_copy_user() by using kmap_atomic() and
the unsafe atomic user mode accessor functions. It's actually slower
than the straightforward code on any reasonable modern CPU.
Back when the code was written (although probably not by the time it was
actually merged, though), 32-bit x86 may have been the dominant
architecture. And there kmap_atomic() can be a lot faster than kmap()
(unless you have very good locality, in which case the virtual address
caching by kmap() can overcome all the downsides).
But these days, x86-64 may not be more populous, but it's getting there
(and if you care about performance, it's definitely already there -
you'd have upgraded your CPU's already in the last few years). And on
x86-64, the non-kmap_atomic() version is faster, simply because the code
is simpler and doesn't have the "re-try page fault" case.
People with old hardware are not likely to care about RDS anyway, and
the optimization for the 32-bit case is simply buggy, since it doesn't
verify the user addresses properly.
Reported-by: Dan Rosenberg <drosenberg@vsecurity.com>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We have for each socket :
One spinlock (sk_slock.slock)
One rwlock (sk_callback_lock)
Possible scenarios are :
(A) (this is used in net/sunrpc/xprtsock.c)
read_lock(&sk->sk_callback_lock) (without blocking BH)
<BH>
spin_lock(&sk->sk_slock.slock);
...
read_lock(&sk->sk_callback_lock);
...
(B)
write_lock_bh(&sk->sk_callback_lock)
stuff
write_unlock_bh(&sk->sk_callback_lock)
(C)
spin_lock_bh(&sk->sk_slock)
...
write_lock_bh(&sk->sk_callback_lock)
stuff
write_unlock_bh(&sk->sk_callback_lock)
spin_unlock_bh(&sk->sk_slock)
This (C) case conflicts with (A) :
CPU1 [A] CPU2 [C]
read_lock(callback_lock)
<BH> spin_lock_bh(slock)
<wait to spin_lock(slock)>
<wait to write_lock_bh(callback_lock)>
We have one problematic (C) use case in inet_csk_listen_stop() :
local_bh_disable();
bh_lock_sock(child); // spin_lock_bh(&sk->sk_slock)
WARN_ON(sock_owned_by_user(child));
...
sock_orphan(child); // write_lock_bh(&sk->sk_callback_lock)
lockdep is not happy with this, as reported by Tetsuo Handa
It seems only way to deal with this is to use read_lock_bh(callbacklock)
everywhere.
Thanks to Jarek for pointing a bug in my first attempt and suggesting
this solution.
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Tested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Jarek Poplawski <jarkao2@gmail.com>
Tested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is basically just a cleanup. IRQs were disabled on the previous
line so we don't need to do it again here. In the current code IRQs
would get turned on one line earlier than intended.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the original code if the copy_from_user() fails in rds_rdma_pages()
then the error handling fails and we get a stack trace from kmalloc().
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add two CMSGs for masked versions of cswp and fadd. args
struct modified to use a union for different atomic op type's
arguments. Change IB to do masked atomic ops. Atomic op type
in rds_message similarly unionized.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
This prints the constant identifier for work completion status and rdma
cm event types, like we already do for IB event types.
A core string array helper is added that each string type uses.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
Nothing was canceling the send and receive work that might have been
queued as a conn was being destroyed.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
rds_conn_shutdown() can return before the connection is shut down when
it encounters an existing state that it doesn't understand. This lets
rds_conn_destroy() then start tearing down the conn from under paths
that are still using it.
It's more reliable the shutdown work and wait for krdsd to complete the
shutdown callback. This stopped some hangs I was seeing where krdsd was
trying to shut down a freed conn.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
Right now there's nothing to stop the various paths that use
rs->rs_transport from racing with rmmod and executing freed transport
code. The simple fix is to have binding to a transport also hold a
reference to the transport's module, removing this class of races.
We already had an unused t_owner field which was set for the modular
transports and which wasn't set for the built-in loop transport.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
rs_transport is now also used by the rdma paths once the socket is
bound. We don't need this stale comment to tell us what cscope can.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
rds_conn_destroy() can race with all other modifications of the
rds_conn_count but it was modifying the count without locking.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
The RDS IB device list wasn't protected by any locking. Traversal in
both the get_mr and FMR flushing paths could race with additon and
removal.
List manipulation is done with RCU primatives and is protected by the
write side of a rwsem. The list traversal in the get_mr fast path is
protected by a rcu read critical section. The FMR list traversal is
more problematic because it can block while traversing the list. We
protect this with the read side of the rwsem.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
It's nice to not have to go digging in the code to see which event
occurred. It's easy to throw together a quick array that maps the ib
event enums to their strings. I didn't see anything in the stack that
does this translation for us, but I also didn't look very hard.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
Flushing FMRs is somewhat expensive, and is currently kicked off when
the interrupt handler notices that we are getting low. The result of
this is that FMR flushing only happens from the interrupt cpus.
This spreads the load more effectively by triggering flushes just before
we allocate a new FMR.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We're seeing bugs today where IB connection shutdown clears the send
ring while the tasklet is processing completed sends. Implementation
details cause this to dereference a null pointer. Shutdown needs to
wait for send completion to stop before tearing down the connection. We
can't simply wait for the ring to empty because it may contain
unsignaled sends that will never be processed.
This patch tracks the number of signaled sends that we've posted and
waits for them to complete. It also makes sure that the tasklet has
finished executing.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
We are *definitely* counting cycles as closely as DaveM, so
ensure hwcache alignment for our recv ring control structs.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
The recv refill path was leaking fragments because the recv event handler had
marked a ring element as free without freeing its frag. This was happening
because it wasn't processing receives when the conn wasn't marked up or
connecting, as can be the case if it races with rmmod.
Two observations support always processing receives in the callback.
First, buildup should only post receives, thus triggering recv event handler
calls, once it has built up all the state to handle them. Teardown should
destroy the CQ and drain the ring before tearing down the state needed to
process recvs. Both appear to be true today.
Second, this test was fundamentally racy. There is nothing to stop rmmod and
connection destruction from swooping in the moment after the conn state was
sampled but before real receive procesing starts.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
We were seeing very nasty bugs due to fundamental assumption the current code
makes about concurrent work struct processing. The code simpy isn't able to
handle concurrent connection shutdown work function execution today, for
example, which is very much possible once a multi-threaded krdsd was
introduced. The problem compounds as additional work structs are added to the
mix.
krdsd is no longer perforance critical now that send and receive posting and
FMR flushing are done elsewhere, so the safest fix is to move back to the
single threaded krdsd that the current code was built around.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
This patch moves the FMR flushing work in to its own mult-threaded work queue.
This is to maintain performance in preparation for returning the main krdsd
work queue back to a single threaded work queue to avoid deep-rooted
concurrency bugs.
This is also good because it further separates FMRs, which might be removed
some day, from the rest of the code base.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
IB connections were not being destroyed during rmmod.
First, recently IB device removal callback was changed to disconnect
connections that used the removing device rather than destroying them. So
connections with devices during rmmod were not being destroyed.
Second, rds_ib_destroy_nodev_conns() was being called before connections are
disassociated with devices. It would almost never find connections in the
nodev list.
We first get rid of rds_ib_destroy_conns(), which is no longer called, and
refactor the existing caller into the main body of the function and get rid of
the list and lock wrappers.
Then we call rds_ib_destroy_nodev_conns() *after* ib_unregister_client() has
removed the IB device from all the conns and put the conns on the nodev list.
The result is that IB connections are destroyed by rmmod.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
The RDS IB client removal callback can queue work to drop the final reference
to an IB device. We have to make sure that this function has returned before
we complete rmmod or the work threads can try to execute freed code.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
Using a delayed work queue helps us make sure a healthy number of FMRs
have queued up over the limit. It makes for a large improvement in RDMA
iops.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
FRM allocation and recycling is performance critical and fairly lock
intensive. The current code has a per connection lock that all
processes bang on and it becomes a major bottleneck on large systems.
This changes things to use a number of cmpxchg based lists instead,
allowing us to go through the whole FMR lifecycle without locking inside
RDS.
Zach Brown pointed out that our usage of cmpxchg for xlist removal is
racey if someone manages to remove and add back an FMR struct into the list
while another CPU can see the FMR's address at the head of the list.
The second CPU might assume the list hasn't changed when in fact any
number of operations might have happened in between the deletion and
reinsertion.
This commit maintains a per cpu count of CPUs that are currently
in xlist removal, and establishes a grace period to make sure that
nobody can see an entry we have just removed from the list.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
rds_send_xmit() was changed to hold an interrupt masking spinlock instead of a
mutex so that it could be called from the IB receive tasklet path. This broke
the TCP transport because its xmit method can block and masks and unmasks
interrupts.
This patch serializes callers to rds_send_xmit() with a simple bit instead of
the current spinlock or previous mutex. This enables rds_send_xmit() to be
called from any context and to call functions which block. Getting rid of the
c_send_lock exposes the bare c_lock acquisitions which are changed to block
interrupts.
A waitqueue is added so that rds_conn_shutdown() can wait for callers to leave
rds_send_xmit() before tearing down partial send state. This lets us get rid
of c_senders.
rds_send_xmit() is changed to check the conn state after acquiring the
RDS_IN_XMIT bit to resolve races with the shutdown path. Previously both
worked with the conn state and then the lock in the same order, allowing them
to race and execute the paths concurrently.
rds_send_reset() isn't racing with rds_send_xmit() now that rds_conn_shutdown()
properly ensures that rds_send_xmit() can't start once the conn state has been
changed. We can remove its previous use of the spinlock.
Finally, c_send_generation is redundant. Callers can race to test the c_flags
bit by simply retrying instead of racing to test the c_send_generation atomic.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
conn->c_lock is acquired in interrupt context. rds_conn_message_info() is
called from user context and was acquiring c_lock without blocking interrupts,
leading to possible deadlocks.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
rds_send_acked_before() wasn't blocking interrupts when acquiring c_lock from
user context but nothing calls it. Rather than fix its use of c_lock we just
remove the function.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
When prefilling the rds frags, we end up doing a lot of allocations.
We're not in atomic context here, and so there's no reason to dip into
atomic reserves. This changes the prefills to use masks that allow
waiting.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch is based heavily on an initial patch by Chris Mason.
Instead of freeing slab memory and pages, it keeps them, and
funnels them back to be reused.
The lock minimization strategy uses xchg and cmpxchg atomic ops
for manipulation of pointers to list heads. We anchor the lists with a
pointer to a list_head struct instead of a static list_head struct.
We just have to carefully use the existing primitives with
the difference between a pointer and a static head struct.
For example, 'list_empty()' means that our anchor pointer points to a list with
a single item instead of meaning that our static head element doesn't point to
any list items.
Original patch by Chris, with significant mods and fixes by Andy and Zach.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: Zach Brown <zach.brown@oracle.com>
All it does is call unmap_sg(), so just call that directly.
The comment above unmap_page also may be incorrect, so we
shouldn't hold on to it, either.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
refill_one() should never be called on a recv struct that
doesn't need a new r_frag allocated. Add a WARN and remove
conditional around r_frag alloc code.
Also, add a comment to explain why r_ibinc may or may not
need refilling.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Instead of splitting up a page into RDS_FRAG_SIZE chunks
ourselves, ask rds_page_remainder_alloc() to do it. While it
is possible PAGE_SIZE > FRAG_SIZE, on x86en it isn't, so having
duplicate "carve up a page into buffers" code seems excessive.
The other modification this spawns is the use of a single
struct scatterlist in rds_page_frag instead of a bare page ptr.
This causes verbosity to increase in some places, and decrease
in others.
Finally, I decided to unify the lifetimes and alloc/free of
rds_page_frag and its page. This is a nice simplification in itself,
but will be extra-nice once we come to adding cmason's recycling
patch.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Currently IB device removal destroys connections which are associated with the
device. This prevents connections from being re-established when replacement
devices are added.
Instead we'll queue shutdown work on the connections as their devices are
removed. When we see that devices are added we triger connection attempts on
all connections that don't currently have a device.
The result is that RDS sockets can resume device-independent work (bcopy, not
RDMA) across IB device removal and restoration.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
A few paths had the same block of code to queue a connection's connect work if
it was in the right state. Let's move this in to a helper function.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
The RDS IB client .remove callback used to free the rds_ibdev for the given
device unconditionally. This could race other users of the struct. This patch
adds refcounting so that we only free the rds_ibdev once all of its users are
done.
Many rds_ibdev users are tied to connections. We give the connection a
reference and change these users to reference the device in the connection
instead of looking it up in the IB client data. The only user of the IB client
data remaining is the first lookup of the device as connections are built up.
Incrementing the reference count of a device found in the IB client data could
race with final freeing so we use an RCU grace period to make sure that freeing
won't happen until those lookups are done.
MRs need the rds_ibdev to get at the pool that they're freed in to. They exist
outside a connection and many MRs can reference different devices from one
socket, so it was natural to have each MR hold a reference. MR refs can be
dropped from interrupt handlers and final device teardown can block so we push
it off to a work struct. Pool teardown had to be fixed to cancel its pending
work instead of deadlocking waiting for all queued work, including itself, to
finish.
MRs get their reference from the global device list, which gets a reference.
It is left unprotected by locks and remains racy. A simple global lock would
be a significant bottleneck. More scalable (complicated) locking should be
done carefully in a later patch.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
rds_ib_xmit_rdma() was calling ib_get_client_data() to get at the rds_ibdevice
just to get the max_sge for the transmit. This patch instead has it get it
directly off the rds_ibdev which is stored on the connection.
The current code won't free the rds_ibdev until all the IB connections that use
it are freed. So it's safe to reference the rds_ibdev this way. In the future
it also makes it easier to support proper reference counting of the rds_ibdev
struct.
As an additional bonus, this gets rid of the performance hit of calling in to
the IB stack to look up the rds_ibdev. The current implementation in the IB
stack acquires an interrupt blocking spinlock to protect the registration of
client callback data.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
rds_ib_cm_handle_connect() could return without unlocking the c_conn_lock if
rds_setup_qp() failed. Rather than adding another imbalanced mutex_unlock() to
this error path we only unlock the mutex once as we exit the function, reducing
the likelyhood of making this same mistake in the future. We remove the
previous mulitple return sites, leaving one unambigious return path.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
This makes sure we have the proper number of references in
rds_ib_xmit_atomic and rds_ib_xmit_rdma. We also consistently
drop references the same way for all message types as the IOs end.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The connection hash was almost entirely RCU ready, this
just makes the final couple of changes to use RCU instead
of spinlocks for everything.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The RDS send_xmit code was trying to get fancy with message
counting and was dropping the final reference on the RDMA messages
too early. This resulted in memory corruption and oopsen.
The fix here is to always add a ref as the parts of the message passes
through rds_send_xmit, and always drop a ref as the parts of the message
go through completion handling.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This is the first in a long line of patches that tries to fix races
between RDS connection shutdown and RDS traffic.
Here we are maintaining a count of active senders to make sure
the connection doesn't go away while they are using it.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The RDS bind lookups are somewhat expensive in terms of CPU
time and locking overhead. This commit changes them into a
faster RCU based hash tree instead of the rbtrees they were using
before.
On large NUMA systems it is a significant improvement.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Allocate send/recv rings in memory that is node-local to the HCA.
This significantly helps performance.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
rds_ib_get_device is called very often as we turn an
ip address into a corresponding device structure. It currently
take a global spinlock as it walks different lists to find active
devices.
This commit changes the lists over to RCU, which isn't very complex
because they are not updated very often at all.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This removes a global waitqueue used to wait for rds messages
and replaces it with a waitqueue inside the rds_message struct.
The global waitqueue turns into a global lock and significantly
bottlenecks operations on large machines.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The bind_lock is almost entirely readonly, but it gets
hammered during normal operations and is a major bottleneck.
This commit changes it to an rwlock, which takes it from 80%
of the system time on a big numa machine down to much lower
numbers.
A better fix would involve RCU, which is done in a later commit
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Update comments to reflect changes in previous commit.
Keeping as separate commits due to different authorship.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
rds_send_xmit is required to loop around after it releases the lock
because someone else could done a trylock, found someone working on the
list and backed off.
But, once we drop our lock, it is possible that someone else does come
in and make progress on the list. We should detect this and not loop
around if another process is actually working on the list.
This patch adds a generation counter that is bumped every time we
get the lock and do some send work. If the retry notices someone else
has bumped the generation counter, it does not need to loop around and
continue working.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Andy Grover <andy.grover@oracle.com>
The purpose of the send quota was really to give fairness
when different connections were all using the same
workq thread to send backlogged msgs -- they could only send
so many before another connection could make progress.
Now that each connection is pushing the backlog from its
completion handler, they are all guaranteed to make progress
and the quota isn't needed any longer.
A thread *will* have to send all previously queued data, as well
as any further msgs placed on the queue while while c_send_lock
was held. In a pathological case a single process can get
roped into doing this for long periods while other threads
get off free. But, since it can only do this until the transport
reports full, this is a bounded scenario.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Now that rds_send_xmit() does not block, we can call it directly
instead of going through the helper thread.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
rds_sendmsg() is calling the send worker function to
send the just-queued datagrams, presumably because it wants
the behavior where anything not sent will re-call the send
worker. We now ensure all queued datagrams are sent by retrying
from the send completion handler, so this isn't needed any more.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
rds_message_put() cannot be called with irqs off, so move it after
irqs are re-enabled.
Spinlocks throughout the function do not to use _irqsave because
the lock of c_send_lock at top already disabled irqs.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
This change allows us to call rds_send_xmit() from a tasklet,
which is crucial to our new operating model.
* Change c_send_lock to a spinlock
* Update stats fields "sem_" to "_lock"
* Remove unneeded rds_conn_is_sending()
About locking between shutdown and send -- send checks if the
connection is up. Shutdown puts the connection into
DISCONNECTING. After this, all threads entering send will exit
immediately. However, a thread could be *in* send_xmit(), so
shutdown acquires the c_send_lock to ensure everyone is out
before proceeding with connection shutdown.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Linus Torvalds