There are cases where read_seqbegin_or_lock() needs to block irqs,
because the seqlock in question nests inside a lock that is also
be taken from irq context.
Add read_seqbegin_or_lock_irqsave() and done_seqretry_irqrestore(), which
are almost identical to read_seqbegin_or_lock() and done_seqretry().
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: prarit@redhat.com
Cc: oleg@redhat.com
Cc: sgruszka@redhat.com
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Trond Myklebust <trond.myklebust@primarydata.com>
Link: http://lkml.kernel.org/r/1410527535-9814-2-git-send-email-riel@redhat.com
[ Improved the readability of the code a bit. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull timer and time updates from Thomas Gleixner:
"A rather large update of timers, timekeeping & co
- Core timekeeping code is year-2038 safe now for 32bit machines.
Now we just need to fix all in kernel users and the gazillion of
user space interfaces which rely on timespec/timeval :)
- Better cache layout for the timekeeping internal data structures.
- Proper nanosecond based interfaces for in kernel users.
- Tree wide cleanup of code which wants nanoseconds but does hoops
and loops to convert back and forth from timespecs. Some of it
definitely belongs into the ugly code museum.
- Consolidation of the timekeeping interface zoo.
- A fast NMI safe accessor to clock monotonic for tracing. This is a
long standing request to support correlated user/kernel space
traces. With proper NTP frequency correction it's also suitable
for correlation of traces accross separate machines.
- Checkpoint/restart support for timerfd.
- A few NOHZ[_FULL] improvements in the [hr]timer code.
- Code move from kernel to kernel/time of all time* related code.
- New clocksource/event drivers from the ARM universe. I'm really
impressed that despite an architected timer in the newer chips SoC
manufacturers insist on inventing new and differently broken SoC
specific timers.
[ Ed. "Impressed"? I don't think that word means what you think it means ]
- Another round of code move from arch to drivers. Looks like most
of the legacy mess in ARM regarding timers is sorted out except for
a few obnoxious strongholds.
- The usual updates and fixlets all over the place"
* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (114 commits)
timekeeping: Fixup typo in update_vsyscall_old definition
clocksource: document some basic timekeeping concepts
timekeeping: Use cached ntp_tick_length when accumulating error
timekeeping: Rework frequency adjustments to work better w/ nohz
timekeeping: Minor fixup for timespec64->timespec assignment
ftrace: Provide trace clocks monotonic
timekeeping: Provide fast and NMI safe access to CLOCK_MONOTONIC
seqcount: Add raw_write_seqcount_latch()
seqcount: Provide raw_read_seqcount()
timekeeping: Use tk_read_base as argument for timekeeping_get_ns()
timekeeping: Create struct tk_read_base and use it in struct timekeeper
timekeeping: Restructure the timekeeper some more
clocksource: Get rid of cycle_last
clocksource: Move cycle_last validation to core code
clocksource: Make delta calculation a function
wireless: ath9k: Get rid of timespec conversions
drm: vmwgfx: Use nsec based interfaces
drm: i915: Use nsec based interfaces
timekeeping: Provide ktime_get_raw()
hangcheck-timer: Use ktime_get_ns()
...
For NMI safe access to clock monotonic we use the seqcount LSB as
index of a timekeeper array. The update sequence looks like this:
smp_wmb(); <- prior stores to a[1]
seq++;
smp_wmb(); <- seq increment before update of a[0]
update(a[0]);
smp_wmb(); <- update of a[0]
seq++;
smp_wmb(); <- seq increment before update of a[1]
update(a[1]);
To avoid open coded barriers, provide a helper function.
[ tglx: Split out of a combo patch against the first implementation of
the NMI safe accessor ]
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
raw_read_seqcount opens a read critical section of the given seqcount
without any lockdep checking and without checking or masking the
LSB. Calling code is responsible for handling that.
Preparatory patch to provide a NMI safe clock monotonic accessor
function.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
This commit reverts the addition of lockdep checking to raw_seqcount_begin
for the following reasons:
1) It violates the naming convention that raw_* functions should not
do lockdep checks (a convention that is also followed by the other
raw_*_seqcount_begin functions).
2) raw_seqcount_begin does not spin, so it can only be part of an ABBA
deadlock in very special circumstances (for instance if a lock
is held across the entire raw_seqcount_begin()+read_seqcount_retry()
loop while also being taken inside the write_seqcount protected area).
3) It is causing false positives with some existing callers, and there
is no non-lockdep alternative for those callers to use.
None of the three existing callers (__d_lookup_rcu, netdev_get_name, and
the NFS state code) appear to use the function in a manner that is ABBA
deadlock prone.
Fixes: 1ca7d67cf5: seqcount: Add lockdep functionality to seqcount/seqlock
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: John Stultz <john.stultz@linaro.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Waiman Long <Waiman.Long@hp.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/CAHQdGtRR6SvEhXiqWo24hoUh9AU9cL82Z8Z-d8-7u951F_d+5g@mail.gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Linus disliked the _no_lockdep() naming, so instead
use the more-consistent raw_* prefix to the non-lockdep
enabled seqcount methods.
This also adds raw_ methods for the write operations
as well, which will be utilized in a following patch.
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Krzysztof Hałasa <khalasa@piap.pl>
Cc: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Cc: Willy Tarreau <w@1wt.eu>
Link: http://lkml.kernel.org/r/1388704274-5278-1-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently seqlocks and seqcounts don't support lockdep.
After running across a seqcount related deadlock in the timekeeping
code, I used a less-refined and more focused variant of this patch
to narrow down the cause of the issue.
This is a first-pass attempt to properly enable lockdep functionality
on seqlocks and seqcounts.
Since seqcounts are used in the vdso gettimeofday code, I've provided
non-lockdep accessors for those needs.
I've also handled one case where there were nested seqlock writers
and there may be more edge cases.
Comments and feedback would be appreciated!
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Li Zefan <lizefan@huawei.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Link: http://lkml.kernel.org/r/1381186321-4906-3-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The sequence lock (seqlock) was originally designed for the cases where
the readers do not need to block the writers by making the readers retry
the read operation when the data change.
Since then, the use cases have been expanded to include situations where
a thread does not need to change the data (effectively a reader) at all
but have to take the writer lock because it can't tolerate changes to
the protected structure. Some examples are the d_path() function and
the getcwd() syscall in fs/dcache.c where the functions take the writer
lock on rename_lock even though they don't need to change anything in
the protected data structure at all. This is inefficient as a reader is
now blocking other sequence number reading readers from moving forward
by pretending to be a writer.
This patch tries to eliminate this inefficiency by introducing a new
type of locking reader to the seqlock locking mechanism. This new
locking reader will try to take an exclusive lock preventing other
writers and locking readers from going forward. However, it won't
affect the progress of the other sequence number reading readers as the
sequence number won't be changed.
Signed-off-by: Waiman Long <Waiman.Long@hp.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
No point in having different implementations for the same
thing. Change the macro mess to inline functions where possible.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The normal read_seqcount_begin() function will wait for any current
writers to exit their critical region by looping until the sequence
count is even.
That "wait for sequence count to stabilize" is the right thing to do if
the read-locker will just retry the whole operation on contention: no
point in doing a potentially expensive reader sequence if we know at the
beginning that we'll just end up re-doing it all.
HOWEVER. Some users don't actually retry the operation, but instead
will abort and do the operation with proper locking. So the sequence
count case may be the optimistic quick case, but in the presense of
writers you may want to do full locking in order to guarantee forward
progress. The prime example of this would be the RCU name lookup.
And in that case, you may well be better off without the "retry early",
and are in a rush to instead get to the failure handling. Thus this
"raw" interface that just returns the sequence number without testing it
- it just forces the low bit to zero so that read_seqcount_retry() will
always fail such a "active concurrent writer" scenario.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We really need to use a ACCESS_ONCE() on the sequence value read in
__read_seqcount_begin(), because otherwise the compiler might end up
reloading the value in between the test and the return of it. As a
result, it might end up returning an odd value (which means that a write
is in progress).
If the reader is then fast enough that that odd value is still the
current one when the read_seqcount_retry() is done, we might end up with
a "successful" read sequence, even despite the concurrent write being
active.
In practice this probably never really happens - there just isn't
anything else going on around the read of the sequence count, and the
common case is that we end up having a read barrier immediately
afterwards.
So the code sequence in which gcc might decide to reaload from memory is
small, and there's no reason to believe it would ever actually do the
reload. But if the compiler ever were to decide to do so, it would be
incredibly annoying to debug. Let's just make sure.
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It uses cpu_relax(), and so needs <asm/processor.h>
Without this patch, I see:
CC arch/mn10300/kernel/asm-offsets.s
In file included from include/linux/time.h:8,
from include/linux/timex.h:56,
from include/linux/sched.h:57,
from arch/mn10300/kernel/asm-offsets.c:7:
include/linux/seqlock.h: In function 'read_seqbegin':
include/linux/seqlock.h:91: error: implicit declaration of function 'cpu_relax'
whilst building asb2364_defconfig on MN10300.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
All static seqlock should be initialized with the lockdep friendly
__SEQLOCK_UNLOCKED() macro.
Remove legacy SEQLOCK_UNLOCKED() macro.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David Miller <davem@davemloft.net>
Link: http://lkml.kernel.org/r/%3C1306238888.3026.31.camel%40edumazet-laptop%3E
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Move the smp_rmb after cpu_relax loop in read_seqlock and add
ACCESS_ONCE to make sure the test and return are consistent.
A multi-threaded core in the lab didn't like the update
from 2.6.35 to 2.6.36, to the point it would hang during
boot when multiple threads were active. Bisection showed
af5ab277de (clockevents:
Remove the per cpu tick skew) as the culprit and it is
supported with stack traces showing xtime_lock waits including
tick_do_update_jiffies64 and/or update_vsyscall.
Experimentation showed the combination of cpu_relax and smp_rmb
was significantly slowing the progress of other threads sharing
the core, and this patch is effective in avoiding the hang.
A theory is the rmb is affecting the whole core while the
cpu_relax is causing a resource rebalance flush, together they
cause an interfernce cadance that is unbroken when the seqlock
reader has interrupts disabled.
At first I was confused why the refactor in
3c22cd5709 (kernel: optimise
seqlock) didn't affect this patch application, but after some
study that affected seqcount not seqlock. The new seqcount was
not factored back into the seqlock. I defer that the future.
While the removal of the timer interrupt offset created
contention for the xtime lock while a cpu does the
additonal work to update the system clock, the seqlock
implementation with the tight rmb spin loop goes back much
further, and is just waiting for the right trigger.
Cc: <stable@vger.kernel.org>
Signed-off-by: Milton Miller <miltonm@bga.com>
Cc: <linuxppc-dev@lists.ozlabs.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Anton Blanchard <anton@samba.org>
Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Link: http://lkml.kernel.org/r/%3Cseqlock-rmb%40mdm.bga.com%3E
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Add branch annotations for seqlock read fastpath, and introduce
__read_seqcount_begin and __read_seqcount_end functions, that can avoid the
smp_rmb() if used carefully. These will be used by store-free path walking
algorithm performance is critical and seqlocks are in use.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Thomas Gleixner debugged a particularly ugly seqlock related livelock:
do not process the seq-read section if we know it beforehand that the
test at the end of the section will fail ...
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Correct mis-spellings of "algorithm", "appear", "consistent" and
(shame, shame) "kernel".
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
seqlock_init() needs to use spin_lock_init() for dynamic locks, so that
lockdep is notified about the presence of a new lock.
(this is a fallout of the recent networking merge, which started using
the so-far unused seqlock_init() API.)
This fix solves the following lockdep-internal warning on current -git:
INFO: trying to register non-static key.
the code is fine but needs lockdep annotation.
turning off the locking correctness validator.
__lock_acquire+0x10c/0x9f9
lock_acquire+0x56/0x72
_spin_lock+0x35/0x42
neigh_destroy+0x9d/0x12e
neigh_periodic_timer+0x10a/0x15c
run_timer_softirq+0x126/0x18e
__do_softirq+0x6b/0xe6
do_softirq+0x64/0xd2
ksoftirqd+0x82/0x138
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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>
In vsyscall function do_vgettimeofday(), some functions are declared as
inlined, which is a hint for gcc to compile the function inlined but it
not forced. Sometimes compiler does not compile the function as
inlined, so here inline is replaced by __always_inline prefix.
It does not happen in gcc compiler actually, but it possibly happens.
Signed-off-by: bibo mao <bibo.mao@intel.com>
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!
