In 2.6.25 we added UDP mem accounting.
This unfortunatly added a penalty when a frame is transmitted, since
we have at TX completion time to call sock_wfree() to perform necessary
memory accounting. This calls sock_def_write_space() and utimately
scheduler if any thread is waiting on the socket.
Thread(s) waiting for an incoming frame was scheduled, then had to sleep
again as event was meaningless.
(All threads waiting on a socket are using same sk_sleep anchor)
This adds lot of extra wakeups and increases latencies, as noted
by Christoph Lameter, and slows down softirq handler.
Reference : http://marc.info/?l=linux-netdev&m=124060437012283&w=2
Fortunatly, Davide Libenzi recently added concept of keyed wakeups
into kernel, and particularly for sockets (see commit
37e5540b3c
epoll keyed wakeups: make sockets use keyed wakeups)
Davide goal was to optimize epoll, but this new wakeup infrastructure
can help non epoll users as well, if they care to setup an appropriate
handler.
This patch introduces new DEFINE_WAIT_FUNC() helper and uses it
in wait_for_packet(), so that only relevant event can wakeup a thread
blocked in this function.
Trace of function calls from bnx2 TX completion bnx2_poll_work() is :
__kfree_skb()
skb_release_head_state()
sock_wfree()
sock_def_write_space()
__wake_up_sync_key()
__wake_up_common()
receiver_wake_function() : Stops here since thread is waiting for an INPUT
Reported-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
'777c6c5 wait: prevent exclusive waiter starvation' made
__wake_up_common() global to be used from abort_exclusive_wait().
It was needed to do a wake-up with the waitqueue lock held while
passing down a key to the wake-up function.
Since '4ede816 epoll keyed wakeups: add __wake_up_locked_key() and
__wake_up_sync_key()' there is an appropriate wrapper for this case:
__wake_up_locked_key().
Use it here and make __wake_up_common() private to the scheduler
again.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1239720785-19661-1-git-send-email-hannes@cmpxchg.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Introduce new wakeup macros that allow passing an event mask to the wakeup
targets. They exactly mimic their non-_poll() counterpart, with the added
event mask passing capability. I did add only the ones currently
requested, avoiding the _nr() and _all() for the moment.
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: William Lee Irwin III <wli@movementarian.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patchset introduces wakeup hints for some of the most popular (from
epoll POV) devices, so that epoll code can avoid spurious wakeups on its
waiters.
The problem with epoll is that the callback-based wakeups do not, ATM,
carry any information about the events the wakeup is related to. So the
only choice epoll has (not being able to call f_op->poll() from inside the
callback), is to add the file* to a ready-list and resolve the real events
later on, at epoll_wait() (or its own f_op->poll()) time. This can cause
spurious wakeups, since the wake_up() itself might be for an event the
caller is not interested into.
The rate of these spurious wakeup can be pretty high in case of many
network sockets being monitored.
By allowing devices to report the events the wakeups refer to (at least
the two major classes - POLLIN/POLLOUT), we are able to spare useless
wakeups by proper handling inside the epoll's poll callback.
Epoll will have in any case to call f_op->poll() on the file* later on,
since the change to be done in order to have the full event set sent via
wakeup, is too invasive for the way our f_op->poll() system works (the
full event set is calculated inside the poll function - there are too many
of them to even start thinking the change - also poll/select would need
change too).
Epoll is changed in a way that both devices which send event hints, and
the ones that don't, are correctly handled. The former will gain some
efficiency though.
As a general rule for devices, would be to add an event mask by using
key-aware wakeup macros, when making up poll wait queues. I tested it
(together with the epoll's poll fix patch Andrew has in -mm) and wakeups
for the supported devices are correctly filtered.
Test program available here:
http://www.xmailserver.org/epoll_test.c
This patch:
Nothing revolutionary here. Just using the available "key" that our
wakeup core already support. The __wake_up_locked_key() was no brainer,
since both __wake_up_locked() and __wake_up_locked_key() are thin wrappers
around __wake_up_common().
The __wake_up_sync() function had a body, so the choice was between
borrowing the body for __wake_up_sync_key() and calling it from
__wake_up_sync(), or make an inline and calling it from both. I chose the
former since in most archs it all resolves to "mov $0, REG; jmp ADDR".
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: William Lee Irwin III <wli@movementarian.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With exclusive waiters, every process woken up through the wait queue must
ensure that the next waiter down the line is woken when it has finished.
Interruptible waiters don't do that when aborting due to a signal. And if
an aborting waiter is concurrently woken up through the waitqueue, noone
will ever wake up the next waiter.
This has been observed with __wait_on_bit_lock() used by
lock_page_killable(): the first contender on the queue was aborting when
the actual lock holder woke it up concurrently. The aborted contender
didn't acquire the lock and therefor never did an unlock followed by
waking up the next waiter.
Add abort_exclusive_wait() which removes the process' wait descriptor from
the waitqueue, iff still queued, or wakes up the next waiter otherwise.
It does so under the waitqueue lock. Racing with a wake up means the
aborting process is either already woken (removed from the queue) and will
wake up the next waiter, or it will remove itself from the queue and the
concurrent wake up will apply to the next waiter after it.
Use abort_exclusive_wait() in __wait_event_interruptible_exclusive() and
__wait_on_bit_lock() when they were interrupted by other means than a wake
up through the queue.
[akpm@linux-foundation.org: coding-style fixes]
Reported-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Mentored-by: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Chuck Lever <cel@citi.umich.edu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org> ["after some testing"]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
is_sync_wait() is used to distinguish between sync and async waits.
Basically sync waits are the ones initialized with init_waitqueue_entry()
and async ones with init_waitqueue_func_entry(). The sync/async
distinction is used only in prepare_to_wait[_exclusive]() and its only
function is to skip setting the current task state if the wait is async.
This has a few problems.
* No one uses it. None of func_entry users use prepare_to_wait()
functions, so the code path never gets executed.
* The distinction is bogus. Maybe back when func_entry is used only
by aio but it's now also used by epoll and in future possibly by 9p
and poll/select.
* Taking @state as argument and ignoring it silenly depending on how
@wait is initialized is just a bad error-prone API.
* It prevents func_entry waits from using wait->private for no good
reason.
This patch kills is_sync_wait() and the associated code paths from
prepare_to_wait[_exclusive](). As there was no user of these code paths,
this patch doesn't cause any behavior difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
FASTCALL() is always expanded to empty, remove it.
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
On Sat, 2008-01-05 at 13:35 -0800, Davide Libenzi wrote:
> I remember I talked with Arjan about this time ago. Basically, since 1)
> you can drop an epoll fd inside another epoll fd 2) callback-based wakeups
> are used, you can see a wake_up() from inside another wake_up(), but they
> will never refer to the same lock instance.
> Think about:
>
> dfd = socket(...);
> efd1 = epoll_create();
> efd2 = epoll_create();
> epoll_ctl(efd1, EPOLL_CTL_ADD, dfd, ...);
> epoll_ctl(efd2, EPOLL_CTL_ADD, efd1, ...);
>
> When a packet arrives to the device underneath "dfd", the net code will
> issue a wake_up() on its poll wake list. Epoll (efd1) has installed a
> callback wakeup entry on that queue, and the wake_up() performed by the
> "dfd" net code will end up in ep_poll_callback(). At this point epoll
> (efd1) notices that it may have some event ready, so it needs to wake up
> the waiters on its poll wait list (efd2). So it calls ep_poll_safewake()
> that ends up in another wake_up(), after having checked about the
> recursion constraints. That are, no more than EP_MAX_POLLWAKE_NESTS, to
> avoid stack blasting. Never hit the same queue, to avoid loops like:
>
> epoll_ctl(efd2, EPOLL_CTL_ADD, efd1, ...);
> epoll_ctl(efd3, EPOLL_CTL_ADD, efd2, ...);
> epoll_ctl(efd4, EPOLL_CTL_ADD, efd3, ...);
> epoll_ctl(efd1, EPOLL_CTL_ADD, efd4, ...);
>
> The code "if (tncur->wq == wq || ..." prevents re-entering the same
> queue/lock.
Since the epoll code is very careful to not nest same instance locks
allow the recursion.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
clean up the sleep_on() APIs:
- do not use fastcall
- replace fragile macro magic with proper inline functions
Signed-off-by: Ingo Molnar <mingo@elte.hu>
allyesconfig vmlinux size delta:
text data bss dec filename
20736884 6073834 3075176 29885894 vmlinux.before
20721009 6073966 3075176 29870151 vmlinux.after
~18 bytes per callsite, 15K of text size (~0.1%) saved.
(as an added bonus this also removes a lockdep annotation.)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Create one lock class for all waitqueue locks in the kernel. Has no effect on
non-lockdep kernels.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Locking init improvement:
- introduce and use __SPIN_LOCK_UNLOCKED for array initializations,
to pass in the name string of locks, used by debugging
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix more include file problems that surfaced since I submitted the previous
fix-missing-includes.patch. This should now allow not to include sched.h
from module.h, which is done by a followup patch.
Signed-off-by: Tim Schmielau <tim@physik3.uni-rostock.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
In the upcoming aio_down patch, it is useful to store a private data
pointer in the kiocb's wait_queue. Since we provide our own wake up
function and do not require the task_struct pointer, it makes sense to
convert the task pointer into a generic private pointer.
Signed-off-by: Benjamin LaHaise <benjamin.c.lahaise@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use LIST_HEAD_INIT rather than doing it by hand in DEFINE_WAIT.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
