On Fri, Jun 24, 2016 at 10:08:46AM +0100, Chris Wilson wrote:
> diff --git a/kernel/async.c b/kernel/async.c
> index d2edd6efec56..d0bcb7cc4884 100644
> --- a/kernel/async.c
> +++ b/kernel/async.c
> @@ -50,6 +50,7 @@ asynchronous and synchronous parts of the kernel.
>
> #include
> #include
>
On Fri, Jun 24, 2016 at 10:08:46AM +0100, Chris Wilson wrote:
> +struct kfence {
> + wait_queue_head_t wait;
> + unsigned long flags;
> + struct kref kref;
> + atomic_t pending;
> +};
> +#define KFENCE_CHECKED_BIT 0
> +
> +static void kfence_free(struct kref *kref)
> +{
> + s
On Fri, Jun 24, 2016 at 10:08:49AM +0100, Chris Wilson wrote:
> kfence_add_delay() is a convenience wrapper around
> hrtimer_start_range_ns() to provide a time source for a kfence graph.
Changelog could be greatly improved by telling us why we'd want this.
On Fri, Jun 24, 2016 at 10:08:47AM +0100, Chris Wilson wrote:
> @@ -151,7 +161,11 @@ static void kfence_free(struct kref *kref)
>
> WARN_ON(atomic_read(&fence->pending) > 0);
>
> - kfree(fence);
> + if (fence->flags) {
> + kfence_notify_t fn = (kfence_notify_t)fence->f
On Tue, Jul 30, 2013 at 10:13:41AM +0200, Maarten Lankhorst wrote:
> The check needs to be for > 1, because ctx->acquired is already incremented.
> This will prevent ww_mutex_lock_slow from returning -EDEADLK and not locking
> the mutex. It caused a lot of false gpu lockups on radeon with
> CONFIG_
On Thu, Sep 12, 2013 at 05:36:43PM +0200, Daniel Vetter wrote:
> The set_need_resched in i915_gem.c:i915_gem_fault can actually be
> removed. It was there to give the error handler a chance to sneak in
> and reset the hw/sw tracking when the gpu is dead. That hack goes back
> to the days when the l
On Thu, Sep 12, 2013 at 05:58:49PM +0200, Daniel Vetter wrote:
> On Thu, Sep 12, 2013 at 5:43 PM, Peter Zijlstra wrote:
> >> The one in ttm is just bonghits to shut up lockdep: ttm can recurse
> >> into it's own pagefault handler and then deadlock, the trylock just
>
no other
> reason for this just drop it.
>
> Reported-by: Peter Zijlstra
Thanks!
Acked-by: Peter Zijlstra
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
On Thu, Sep 12, 2013 at 05:35:43PM +0100, Chris Wilson wrote:
> Not quite, as it would be possible for the evil userspace to trigger a
> GPU hang that would cause the sane userspace to spin indefinitely
> waiting for the error recovery to kick in.
So with FIFOn+1 preempting FIFOn its a live-lock
> rip the reschedule-point out.
>
> Reported-by: Peter Zijlstra
Thanks!
Acked-by: Peter Zijlstra
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
On Thu, Sep 12, 2013 at 05:11:25PM +0200, Maarten Lankhorst wrote:
> Op 12-09-13 17:06, Peter Zijlstra schreef:
> > Hi Dave,
> >
> > So I'm poking around the preemption code and stumbled upon:
> >
> > drivers/gpu/drm/i915/i915_gem.c:set_n
Hi Dave,
So I'm poking around the preemption code and stumbled upon:
drivers/gpu/drm/i915/i915_gem.c:set_need_resched();
drivers/gpu/drm/ttm/ttm_bo_vm.c:set_need_resched();
drivers/gpu/drm/ttm/ttm_bo_vm.c:set_need_resched();
drivers/
On Fri, Sep 13, 2013 at 09:46:03AM +0200, Thomas Hellstrom wrote:
> >>if (!bo_tryreserve()) {
> >> up_read mmap_sem(); // Release the mmap_sem to avoid deadlocks.
> >> bo_reserve(); // Wait for the BO to become available
> >> (interruptible)
> >> bo_unreserve();
On Fri, Sep 13, 2013 at 10:41:54AM +0200, Daniel Vetter wrote:
> On Fri, Sep 13, 2013 at 10:29 AM, Peter Zijlstra wrote:
> > On Fri, Sep 13, 2013 at 09:46:03AM +0200, Thomas Hellstrom wrote:
> >> >>if (!bo_tryreserve()) {
> >> >> up_read mmap_sem(); //
Just to let you know, I've not been ignoring all the ww_mutex stuff,
I've been ill this week. I'm slowly returning to feeling human again,
but am still fairly wrecked.
On Thu, Dec 01, 2016 at 03:06:45PM +0100, Nicolai Hähnle wrote:
> +++ b/kernel/locking/mutex.c
> @@ -350,7 +350,8 @@ ww_mutex_set_context_slowpath(struct ww_mutex *lock,
> * access and not reliable.
> */
> static noinline
> -bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *own
On Thu, Dec 01, 2016 at 03:06:47PM +0100, Nicolai Hähnle wrote:
> +++ b/include/linux/ww_mutex.h
> @@ -222,11 +222,7 @@ extern int __must_check
> __ww_mutex_lock_interruptible(struct ww_mutex *lock,
> */
> static inline int ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx
> *ctx)
>
On Thu, Dec 01, 2016 at 03:06:47PM +0100, Nicolai Hähnle wrote:
> @@ -640,10 +640,11 @@ __mutex_lock_common(struct mutex *lock, long state,
> unsigned int subclass,
> struct mutex_waiter waiter;
> unsigned long flags;
> bool first = false;
> - struct ww_mutex *ww;
> i
On Thu, Dec 01, 2016 at 03:06:48PM +0100, Nicolai Hähnle wrote:
> +static inline int __sched
> +__ww_mutex_add_waiter(struct mutex_waiter *waiter,
> + struct mutex *lock,
> + struct ww_acquire_ctx *ww_ctx)
> +{
> + struct mutex_waiter *cur;
> +
> + if (!
On Thu, Dec 01, 2016 at 03:06:48PM +0100, Nicolai Hähnle wrote:
> @@ -693,8 +748,12 @@ __mutex_lock_common(struct mutex *lock, long state,
> unsigned int subclass,
>* mutex_unlock() handing the lock off to us, do a trylock
>* before testing the error conditions to
On Tue, Dec 06, 2016 at 11:03:28AM -0500, Waiman Long wrote:
>
> The mutex_spin_on_owner() function was originally marked noinline
> because it could be a major consumer of CPU cycles in a contended lock.
> Having it shown separately in the perf output will help the users have a
> better understan
On Fri, Dec 16, 2016 at 03:19:43PM +0100, Nicolai Hähnle wrote:
> Hi Peter and Chris,
>
> (trying to combine the handoff discussion here)
>
> On 06.12.2016 17:55, Peter Zijlstra wrote:
> >On Thu, Dec 01, 2016 at 03:06:48PM +0100, Nicolai Hähnle wrote:
On Fri, Dec 16, 2016 at 03:19:43PM +0100, Nicolai Hähnle wrote:
> The concern about picking up a handoff that we didn't request is real,
> though it cannot happen in the first iteration. Perhaps this __mutex_trylock
> can be moved to the end of the loop? See below...
> >>@@ -728,7 +800,7 @@ __mu
On Fri, Dec 16, 2016 at 03:19:43PM +0100, Nicolai Hähnle wrote:
> >>@@ -716,7 +775,20 @@ __mutex_lock_common(struct mutex *lock, long state,
> >>unsigned int subclass,
> >>spin_unlock_mutex(&lock->wait_lock, flags);
> >>schedule_preempt_disabled();
> >>
> >>- if
On Fri, Dec 16, 2016 at 07:11:41PM +0100, Nicolai Hähnle wrote:
> mutex_optimistic_spin() already calls __mutex_trylock, and for the no-spin
> case, __mutex_unlock_slowpath() only calls wake_up_q() after releasing the
> wait_lock.
mutex_optimistic_spin() is a no-op when !CONFIG_MUTEX_SPIN_ON_OWNE
On Fri, Dec 16, 2016 at 02:17:25PM +0100, Nicolai Hähnle wrote:
> On 06.12.2016 16:25, Peter Zijlstra wrote:
> >On Thu, Dec 01, 2016 at 03:06:47PM +0100, Nicolai Hähnle wrote:
> >
> >>@@ -640,10 +640,11 @@ __mutex_lock_common(struct mutex *lock, long state,
&
On Wed, Dec 21, 2016 at 07:46:33PM +0100, Nicolai Hähnle wrote:
> diff --git a/include/linux/ww_mutex.h b/include/linux/ww_mutex.h
> index a5960e5..b2eaaab 100644
> --- a/include/linux/ww_mutex.h
> +++ b/include/linux/ww_mutex.h
> @@ -186,11 +186,6 @@ static inline void ww_acquire_fini(struct ww_a
On Fri, Dec 23, 2016 at 12:16:03PM +0100, Nicolai Hähnle wrote:
> >For some reason this patch appears to make lib/locking-selftest.c really
> >unhappy.
> >
> >I get endless streams of:
> >
> >../lib/locking-selftest.c: In function âww_test_fail_acquireâ:
> >../lib/locking-selftest.c:1141:6: e
On Tue, Sep 29, 2015 at 10:55:39PM +0800, kbuild test robot wrote:
> Hi Robert,
>
> [auto build test results on v4.3-rc3 -- if it's inappropriate base, please
> ignore]
>
> config: i386-defconfig (attached as .config)
> reproduce:
> git checkout a1d59679ae8f3e7e7659e9723ae3fc69af2532e6
> # s
/me wonders what's up with partial Cc's today..
On Wed, Jun 13, 2018 at 09:47:44AM +0200, Thomas Hellstrom wrote:
> The current Wound-Wait mutex algorithm is actually not Wound-Wait but
> Wait-Die. Implement also Wound-Wait as a per-ww-class choice. Wound-Wait
> is, contrary to Wait-Die a preempt
On Wed, Jun 13, 2018 at 12:40:29PM +0200, Thomas Hellstrom wrote:
> On 06/13/2018 11:50 AM, Peter Zijlstra wrote:
> >
> > > +
> > > + lockdep_assert_held(&lock->wait_lock);
> > > +
> > > + if (owner && hold_ctx && __ww_c
On Wed, Jun 13, 2018 at 04:05:43PM +0200, Thomas Hellstrom wrote:
> In short, with Wait-Die (before the patch) it's the process _taking_ the
> contended lock that backs off if necessary. No preemption required. With
> Wound-Wait, it's the process _holding_ the contended lock that gets wounded
> (pr
On Thu, Jun 14, 2018 at 09:29:21AM +0200, Thomas Hellstrom wrote:
> __ww_mutex_wakeup_for_backoff(struct mutex *lock, struct ww_acquire_ctx
> *ww_ctx)
> {
> struct mutex_waiter *cur;
> + unsigned int is_wait_die = ww_ctx->ww_class->is_wait_die;
>
> lockdep_assert_held(&lock->w
On Thu, Jun 14, 2018 at 09:29:21AM +0200, Thomas Hellstrom wrote:
> +static bool __ww_mutex_wound(struct mutex *lock,
> + struct ww_acquire_ctx *ww_ctx,
> + struct ww_acquire_ctx *hold_ctx)
> +{
> + struct task_struct *owner = __mutex_owner(lock
On Thu, Jun 14, 2018 at 01:48:39PM +0200, Thomas Hellstrom wrote:
> The literature makes a distinction between "killed" and "wounded". In our
> context, "Killed" is when a transaction actually receives an -EDEADLK and
> needs to back off. "Wounded" is when someone (typically another transaction)
>
On Thu, Jun 14, 2018 at 03:43:04PM +0200, Thomas Hellstrom wrote:
> It's intended to be enforced by storing the algorithm choice in the
> WW_MUTEX_CLASS which must be common for an acquire context and the
> ww_mutexes it acquires. However, I don't think there is a check that that
> holds. I guess w
On Thu, Jun 14, 2018 at 06:43:40PM +0200, Thomas Hellstrom wrote:
> Overall, I think this looks fine. I'll just fix up the FLAG_WAITERS setting
> and affected comments and do some torture testing on it.
Thanks!
> Are you OK with adding the new feature and the cleanup in the same patch?
I suppose
On Mon, Sep 03, 2018 at 04:07:08PM +0200, Thomas Hellstrom wrote:
> Commit 08295b3b5bee ("Implement an algorithm choice for Wound-Wait
> mutexes") introduced a reference in the documentation to a function
> that was removed in an earlier commit.
>
> It also forgot to remove a call to debug_mutex_a
On Fri, Jun 15, 2018 at 02:08:27PM +0200, Thomas Hellstrom wrote:
> @@ -772,6 +856,25 @@ __ww_mutex_add_waiter(struct mutex_waiter *waiter,
> }
>
> list_add_tail(&waiter->list, pos);
> + if (__mutex_waiter_is_first(lock, waiter))
> + __mutex_set_flag(lock, MUTEX_FLAG_
linux/ww_mutex.h | 28 ++---
> kernel/locking/mutex.c| 202
> ++
> 3 files changed, 145 insertions(+), 97 deletions(-)
Acked-by: Peter Zijlstra (Intel)
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
edanne
> Cc: Greg Kroah-Hartman
> Cc: linux-...@vger.kernel.org
> Cc: linux-me...@vger.kernel.org
> Cc: linaro-mm-...@lists.linaro.org
> Co-authored-by: Peter Zijlstra
> Signed-off-by: Thomas Hellstrom
>
> ---
> Documentation/locking/ww-mutex-design.txt | 57 +
I suspect you want this through the DRM tree? Ingo are you OK with that?
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
On Tue, Jun 19, 2018 at 12:44:52PM +0200, Thomas Hellstrom wrote:
> On 06/19/2018 11:44 AM, Peter Zijlstra wrote:
> > On Tue, Jun 19, 2018 at 10:24:43AM +0200, Thomas Hellstrom wrote:
> > > From: Peter Ziljstra
> > >
> > > Make the WW mutex code more reada
On Mon, Jul 09, 2018 at 10:36:49AM +0200, Daniel Vetter wrote:
> Avoids complaints from gcc about ambiguous else clauses.
Is that a new thing? I'm fairly sure I've never seen it do that,
> Signed-off-by: Daniel Vetter
> Cc: Andrew Morton
> Cc: Peter Zijlstra
> ---
&g
On Mon, Jul 09, 2018 at 05:00:07PM +0200, Daniel Vetter wrote:
> On Mon, Jul 9, 2018 at 12:36 PM, Peter Zijlstra wrote:
> > On Mon, Jul 09, 2018 at 10:36:49AM +0200, Daniel Vetter wrote:
> >> #define for_each_node_with_cpus(node)\
> >>
On Mon, Jul 09, 2018 at 05:52:04PM +0200, Daniel Vetter wrote:
> for_each_something(foo)
> if (foo->bla)
> call_bla(foo);
> else
> call_default(foo);
>
> Totally contrived, but this complains. Liberally sprinkling {} also shuts
> up the compiler, but it's a
On Mon, Jul 09, 2018 at 05:52:04PM +0200, Daniel Vetter wrote:
> for_each_something(foo)
> if (foo->bla)
> call_bla(foo);
> else
> call_default(foo);
Note that the kernel coding style 'discourages' this style and would
like you to write:
for_each_so
On Mon, Dec 10, 2018 at 03:13:37PM +0100, Michal Hocko wrote:
> I do not see any scheduler guys Cced and it would be really great to get
> their opinion here.
>
> On Mon 10-12-18 11:36:39, Daniel Vetter wrote:
> > In some special cases we must not block, but there's not a
> > spinlock, preempt-off
On Mon, Dec 10, 2018 at 05:20:10PM +0100, Michal Hocko wrote:
> > OK, no real objections to the thing. Just so long we're all on the same
> > page as to what it does and doesn't do ;-)
>
> I am not really sure whether there are other potential users besides
> this one and whether the check as suc
On Mon, Dec 10, 2018 at 04:01:59PM +0100, Michal Hocko wrote:
> On Mon 10-12-18 15:47:11, Peter Zijlstra wrote:
> > On Mon, Dec 10, 2018 at 03:13:37PM +0100, Michal Hocko wrote:
> > > I do not see any scheduler guys Cced and it would be really great to get
> > > their op
On Tue, Jul 23, 2024 at 10:30:08AM -0500, Lucas De Marchi wrote:
> On Tue, Jul 23, 2024 at 09:03:25AM GMT, Tvrtko Ursulin wrote:
> >
> > On 22/07/2024 22:06, Lucas De Marchi wrote:
> > > Instead of calling perf_pmu_unregister() when unbinding, defer that to
> > > the destruction of i915 object. Si
On Fri, May 03, 2024 at 11:37:25AM +0200, Christian Brauner wrote:
> On Thu, May 02, 2024 at 05:41:23PM -0700, Kees Cook wrote:
> > On Fri, May 03, 2024 at 01:14:45AM +0100, Al Viro wrote:
> > > On Thu, May 02, 2024 at 05:10:18PM -0700, Kees Cook wrote:
> > >
> > > > But anyway, there needs to be
On Thu, Mar 15, 2018 at 11:31:57AM +0100, Daniel Vetter wrote:
> Is there any progress on getting cross-release enabled again?
Not yet, I'm still fighting the meltdown/spectre induced backlog.
We'll get to it eventually.
___
dri-devel mailing list
dri-d
On Tue, Feb 20, 2018 at 01:58:26PM +0100, Christian König wrote:
> amdgpu needs to verify if userspace sends us valid addresses and the simplest
> way of doing this is to check if the buffer object is locked with the ticket
> of the current submission.
>
> Clean up the access to the ww_mutex inter
On Tue, Feb 20, 2018 at 02:08:26PM +0100, Christian König wrote:
> Am 20.02.2018 um 13:35 schrieb Peter Zijlstra:
> > > +static inline bool ww_mutex_is_owned_by(struct ww_mutex *lock,
> > > + struct task_struct *task,
> > > +
This really should've been Cc'ed to me.
On Thu, Feb 15, 2018 at 03:19:42PM +0100, Christian König wrote:
> diff --git a/include/linux/ww_mutex.h b/include/linux/ww_mutex.h
> index 39fda195bf78..dd580db289e8 100644
> --- a/include/linux/ww_mutex.h
> +++ b/include/linux/ww_mutex.h
> @@ -358,4 +358,
On Tue, Feb 20, 2018 at 02:26:55PM +0100, Christian König wrote:
> > > +static inline bool ww_mutex_is_owned_by(struct ww_mutex *lock,
> > > + struct ww_acquire_ctx *ctx)
> > > +{
> > > + if (ctx)
> > > + return likely(READ_ONCE(lock->ctx) == ctx);
> > > + el
On Tue, Feb 20, 2018 at 03:34:07PM +0100, Christian König wrote:
> > OK, but neither case would in fact need the !ctx case right? That's just
> > there for completeness sake?
>
> Unfortunately not. TTM uses trylock to lock BOs which are about to be
> evicted to make room for all the BOs locked wit
On Tue, Feb 20, 2018 at 04:05:49PM +0100, Christian König wrote:
> Am 20.02.2018 um 15:54 schrieb Peter Zijlstra:
> > On Tue, Feb 20, 2018 at 03:34:07PM +0100, Christian König wrote:
> > > > OK, but neither case would in fact need the !ctx case right? That's just
>
On Thu, Jul 13, 2017 at 03:57:31PM +0200, Jiri Slaby wrote:
> Hmm, not that easy:
> [ +0.011001] BUG: key 8807573ed758 not in .data!
> [ +0.000103] DEBUG_LOCKS_WARN_ON(1)
> [ +0.12] [ cut here ]
> [ +0.000147] WARNING: CPU: 1 PID: 282 at
> ../kernel/locking/lock
gs no longer warn, they blow the box out of the water. I made a
> dinky testcase module (attached), and bisected to the real root
>
> 19d436268dde95389c616bb3819da73f0a8b28a8 is the first bad commit
> commit 19d436268dde95389c616bb3819da73f0a8b28a8
> Author: Peter Zijlstra
&g
On Fri, Jul 14, 2017 at 11:20:01AM -0400, Ilia Mirkin wrote:
> On Fri, Jul 14, 2017 at 11:19 AM, Tobias Klausmann
> wrote:
> > The conversion is a nice catch, but i'd like to have a bit more context, see
> > below!
> >
> > With a better description:
> >
> > Tobias Klausmann
>
> I don't think it
On Fri, Jul 14, 2017 at 05:58:18PM +0200, Mike Galbraith wrote:
> On Fri, 2017-07-14 at 17:50 +0200, Peter Zijlstra wrote:
> > Urgh, is for some mysterious reason the __bug_table section of modules
> > ending up in RO memory?
> >
> > I forever get lost in that link m
On Sat, Jul 15, 2017 at 11:53:28AM +0200, Daniel Vetter wrote:
> A more complete solution would be to do the mutex_init in the drm core
> only for legacy drivers, plus add it to each modern driver that still
> needs it, which would also give each its own lockdep key. Trying to do
> that dynamically
On Mon, Jul 17, 2017 at 05:06:42PM +0200, Daniel Vetter wrote:
> On Mon, Jul 17, 2017 at 11:35 AM, Peter Zijlstra wrote:
> > On Sat, Jul 15, 2017 at 11:53:28AM +0200, Daniel Vetter wrote:
> >> A more complete solution would be to do the mutex_init in the drm core
> >&
On Thu, Dec 07, 2017 at 11:08:49AM +0100, Daniel Vetter wrote:
> Since -rc1 we're hitting a bunch of lockdep splats using the new
> cross-release stuff around the 2 kthread completions. In all cases
> they are because totally independent uses of kthread are mixed up by
> lockdep into the same locki
On Thu, Dec 07, 2017 at 03:58:28PM +0100, Daniel Vetter wrote:
> [ 85.069417] gem_exec_captur/2810 is trying to acquire lock:
> [ 85.069419] ((completion)&self->parked){+.+.}, at: []
> kthread_park+0x3d/0x50
> [ 85.069426]
>but task is already holding lock:
> [ 85.069428
On Thu, Dec 07, 2017 at 09:56:57PM +0100, Daniel Vetter wrote:
> On Thu, Dec 07, 2017 at 08:57:09PM +0100, Peter Zijlstra wrote:
> > Is what it says I suppose. Now I don't know enough about that i915 code
> > to say if that breadcrumbs_signal thread can ever trigger a fau
ue to the many entry points this is slightly
> tedious.
>
> Cc: Tvrtko Ursulin
> Cc: Marta Lofstedt
> Cc: Byungchul Park
> Cc: Ingo Molnar
> Cc: Peter Zijlstra
> Cc: Tejun Heo
> Cc: Kees Cook
> Cc: Thomas Gleixner
> Cc: Shaohua Li
> Cc: Andrew Morton
&
On Fri, Dec 08, 2017 at 05:36:28PM +0100, Daniel Vetter wrote:
> Aside: Could/should we take some fake lockdep locks around these
> callbacks, since not all drivers call them from a hardirq context? Just to
> validate that everyone follows the contract.
What I typically do is use local_irq_save/r
On Wed, Dec 13, 2017 at 06:36:33PM +0100, Sebastian Andrzej Siewior wrote:
> +peterz
> context: http://www.spinics.net/lists/intel-gfx/msg149011.html
>
> On 2017-12-13 17:37:21 [+0200], Joonas Lahtinen wrote:
> > On Wed, 2017-12-13 at 16:06 +0100, Sebastian Andrzej Siewior wrote:
> > > On 2017-12-
On Tue, 2012-08-21 at 10:15 +0100, Alan Cox wrote:
> > So after much tracing with direct netconsole writes (printks
> > under console_lock not so useful), I think I found the race.
>
> Direct netconsole write would be a useful patch to have mainline I think
> 8)
could we make that use the earlyp
On Tue, 2012-08-21 at 16:40 +1000, Dave Airlie wrote:
> So after much tracing with direct netconsole writes (printks
> under console_lock not so useful)
I always use earlyprintk on serial..
--
Live Security Virtual Conf
On Mon, 2012-09-24 at 14:17 +0200, Peter Zijlstra wrote:
> On Tue, 2012-09-18 at 01:03 +0200, Daniel Vetter wrote:
> > - In the printk code there's a special trylock, only used to kick off
> > the logbuffer printk'ing in console_unlock. But all that happens
> >
On Mon, 2012-09-24 at 14:54 +0200, Daniel Vetter wrote:
> I've read through the patches and I'm hoping you don't volunteer me to
> pick these up ... ;-)
Worth a try, right? :-)
> But there doesn't seem to be anything that would
> get worse through this lockdep annotation patch, right?
No indee
On Tue, 2012-09-18 at 01:03 +0200, Daniel Vetter wrote:
> - In the printk code there's a special trylock, only used to kick off
> the logbuffer printk'ing in console_unlock. But all that happens
> while lockdep is disable (since printk does a few other evil
> tricks). So no issue there, eithe
On Thu, 2013-02-28 at 11:25 +0100, Maarten Lankhorst wrote:
> +struct ticket_mutex {
> + struct mutex base;
> + atomic_long_t reservation_id;
> +};
I'm not sure this is a good name, esp. due to the potential confusion
vs the ticket spinlocks we have.
_
On Thu, 2013-02-28 at 11:25 +0100, Maarten Lankhorst wrote:
> +Reservation type mutexes
> +struct ticket_mutex {
> +extern int __must_check _mutex_reserve_lock(struct ticket_mutex *lock,
That's two different names and two different forms of one (for a total
of 3 variants) for the same scheme.
F
On Thu, 2013-02-28 at 11:25 +0100, Maarten Lankhorst wrote:
> +The algorithm that TTM came up with for dealing with this problem is
> +quite simple. For each group of buffers (execbuf) that need to be
> +locked, the caller would be assigned a unique reservation_id, from a
> +global counter. In ca
On Thu, 2013-02-28 at 11:25 +0100, Maarten Lankhorst wrote:
> +mutex_reserve_lock_slow and mutex_reserve_lock_intr_slow:
> + Similar to mutex_reserve_lock, except it won't backoff with
> -EAGAIN.
> + This is useful when mutex_reserve_lock failed with -EAGAIN, and you
> + unreserved all reservati
On Tue, 2013-04-02 at 16:57 +0200, Maarten Lankhorst wrote:
> Hey,
>
> Thanks for reviewing.
Only partway through so far :-)
> Op 02-04-13 13:00, Peter Zijlstra schreef:
> > On Thu, 2013-02-28 at 11:25 +0100, Maarten Lankhorst wrote:
> >> +Reservation type mutexe
I'm sorry, this email ended up quite a bit longer than I had hoped for;
please bear with me.
On Tue, 2013-04-02 at 18:59 +0200, Peter Zijlstra wrote:
> struct ww_mutex; /* wound/wait */
>
> int mutex_wound_lock(struct ww_mutex *); /* returns -EDEADLK */
> int mutex
On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> Hm, I guess your aim with the TASK_DEADLOCK wakeup is to bound the
> wait
> times of older task.
No, imagine the following:
struct ww_mutex A, B;
struct mutex C;
task-O task-Y task-X
A
B
On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> The trick with the current code is that the oldest task
> will never see an -EAGAIN ever and hence is guaranteed to make forward
> progress. If the task is really unlucky though it might be forced to
> wait
> for a younger task for every ww_
On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> The thing is now that you're not expected to hold these locks for a
> long
> time - if you need to synchronously stall while holding a lock
> performance
> will go down the gutters anyway. And since most current
> gpus/co-processors
> still
On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> Another big reason for having a start/end marker like you've describe
> is
> lockdep support.
Yeah, I saw how you did that.. but there's other ways of making it work
too, you could for instance create a new validation state for this type
of
On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> I'm a bit confused about the different classes you're talking about.
> Since
> the ticket queue is currently a global counter there's only one class
> of
> ww_mutexes.
Right, so that's not something that's going to fly.. we need to support
On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> We've discussed this approach of using (rt-prio, age) instead of just
> age
> to determine the the "oldness" of a task for deadlock-breaking with
> -EAGAIN. The problem is that through PI boosting or normal rt-prio
> changes
> while tasks ar
On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> Well, it was a good read and I'm rather happy that we agree on the
> ww_ctx
> thing (whatever it's called in the end), even though we have slightly
> different reasons for it.
Yeah, I tried various weirdness to get out from under it, but th
On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> We do add some form of owner tracking by storing the reservation
> ticket
> of the current holder into every ww_mutex. So when task-Y in your
> above
> example tries to acquire lock A it notices that it's behind in the
> global
> queue and i
On Thu, 2013-04-04 at 18:56 +0200, Daniel Vetter wrote:
> On Thu, Apr 4, 2013 at 3:31 PM, Daniel Vetter wrote:
> >> In this case when O blocks Y isn't actually blocked, so our
> >> TASK_DEADLOCK wakeup doesn't actually achieve anything.
> >>
> >> This means we also have to track (task) state so th
On Tue, 2013-04-09 at 18:42 -0400, Steven Rostedt wrote:
> What about setting an age as soon as it starts the process
> of grabbing one of these locks? And it keeps the age until it
> successfully grabs and releases all the locks again. It wont reset if
> it
> had to drop the locks and start over.
> Are there any issues left? I included the patch you wrote for injecting
> -EDEADLK too
> in my tree. The overwhelming silence makes me think there are either none, or
> nobody cared enough to review it. :(
It didn't manage to reach my inbox it seems,.. I can only find a debug
patch in this thre
On Wed, May 22, 2013 at 01:47:42PM +0200, Maarten Lankhorst wrote:
> Op 22-05-13 13:37, Peter Zijlstra schreef:
> >> Are there any issues left? I included the patch you wrote for injecting
> >> -EDEADLK too
> >> in my tree. The overwhelming silence makes me think t
On Wed, May 22, 2013 at 01:18:14PM +0200, Maarten Lankhorst wrote:
Lacking the actual msg atm, I'm going to paste in here...
> Subject: [PATCH v3 2/3] mutex: add support for wound/wait style locks, v3
> From: Maarten Lankhorst
>
> Changes since RFC patch v1:
> - Updated to use atomic_l
On Wed, May 22, 2013 at 07:24:38PM +0200, Maarten Lankhorst wrote:
> >> +- Functions to only acquire a single w/w mutex, which results in the
> >> exact same
> >> + semantics as a normal mutex. These functions have the _single postfix.
> > This is missing rationale.
> trylock_single is useful wh
On Wed, May 22, 2013 at 07:24:38PM +0200, Maarten Lankhorst wrote:
> >> +static inline void ww_acquire_init(struct ww_acquire_ctx *ctx,
> >> + struct ww_class *ww_class)
> >> +{
> >> + ctx->task = current;
> >> + do {
> >> + ctx->stamp = atomic_long_inc_return
On Wed, May 22, 2013 at 06:49:04PM +0200, Daniel Vetter wrote:
> - _slow functions can check whether all acquire locks have been
> released and whether the caller is indeed blocking on the contending
> lock. Not doing so could either result in needless spinning instead of
> blocking (when blocking
On Mon, May 27, 2013 at 10:26:39AM +0200, Maarten Lankhorst wrote:
> Op 27-05-13 10:00, Peter Zijlstra schreef:
> > On Wed, May 22, 2013 at 07:24:38PM +0200, Maarten Lankhorst wrote:
> >>>> +- Functions to only acquire a single w/w mutex, which results in the
> >>
On Mon, May 27, 2013 at 12:01:50PM +0200, Maarten Lankhorst wrote:
> > Again, early.. monday.. would a trylock, even if successful still need
> > the ctx?
> No ctx for trylock is supported. You can still do a trylock while
> holding a context, but the mutex won't be a part of the context.
> Normal
101 - 200 of 259 matches
Mail list logo
