* [patch] drm,radeon,i915: use preempt_disable/enable_rt() where recommended.
2016-02-27 7:05 [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread() Mike Galbraith
@ 2016-02-27 7:09 ` Mike Galbraith
2016-02-27 7:34 ` [patch] locking/lglocks: use preempt_enable/disable_nort() in lg_double_lock/lg_double_unlock Mike Galbraith
` (3 subsequent siblings)
4 siblings, 0 replies; 9+ messages in thread
From: Mike Galbraith @ 2016-02-27 7:09 UTC (permalink / raw)
To: Sebastian Andrzej Siewior; +Cc: Thomas Gleixner, linux-rt-users
DRM folks identified the spots, so use them.
Signed-off-by: Mike Galbraith <umgwanakikbuti@gmail.com>
---
drivers/gpu/drm/i915/i915_irq.c | 2 ++
drivers/gpu/drm/radeon/radeon_display.c | 2 ++
2 files changed, 4 insertions(+)
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -812,6 +812,7 @@ static int i915_get_crtc_scanoutpos(stru
spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
/* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
+ preempt_disable_rt();
/* Get optional system timestamp before query. */
if (stime)
@@ -863,6 +864,7 @@ static int i915_get_crtc_scanoutpos(stru
*etime = ktime_get();
/* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */
+ preempt_enable_rt();
spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -1849,6 +1849,7 @@ int radeon_get_crtc_scanoutpos(struct dr
struct radeon_device *rdev = dev->dev_private;
/* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
+ preempt_disable_rt();
/* Get optional system timestamp before query. */
if (stime)
@@ -1941,6 +1942,7 @@ int radeon_get_crtc_scanoutpos(struct dr
*etime = ktime_get();
/* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */
+ preempt_enable_rt();
/* Decode into vertical and horizontal scanout position. */
*vpos = position & 0x1fff;
^ permalink raw reply [flat|nested] 9+ messages in thread* [patch] locking/lglocks: use preempt_enable/disable_nort() in lg_double_lock/lg_double_unlock
2016-02-27 7:05 [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread() Mike Galbraith
2016-02-27 7:09 ` [patch] drm,radeon,i915: use preempt_disable/enable_rt() where recommended Mike Galbraith
@ 2016-02-27 7:34 ` Mike Galbraith
2016-02-27 8:01 ` [patch] drm,i915: Use local_lock/unlock_irq() in intel_pipe_update_start/end() Mike Galbraith
` (2 subsequent siblings)
4 siblings, 0 replies; 9+ messages in thread
From: Mike Galbraith @ 2016-02-27 7:34 UTC (permalink / raw)
To: Sebastian Andrzej Siewior; +Cc: Thomas Gleixner, linux-rt-users
Let's not do that when snagging an rtmutex.
Signed-off-by: Mike Galbraith <umgwanakilbuti@gmail.com>
---
kernel/locking/lglock.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- a/kernel/locking/lglock.c
+++ b/kernel/locking/lglock.c
@@ -86,7 +86,7 @@ void lg_double_lock(struct lglock *lg, i
if (cpu2 < cpu1)
swap(cpu1, cpu2);
- preempt_disable();
+ preempt_disable_nort();
lock_acquire_shared(&lg->lock_dep_map, 0, 0, NULL, _RET_IP_);
lg_do_lock(per_cpu_ptr(lg->lock, cpu1));
lg_do_lock(per_cpu_ptr(lg->lock, cpu2));
@@ -97,7 +97,7 @@ void lg_double_unlock(struct lglock *lg,
lock_release(&lg->lock_dep_map, 1, _RET_IP_);
lg_do_unlock(per_cpu_ptr(lg->lock, cpu1));
lg_do_unlock(per_cpu_ptr(lg->lock, cpu2));
- preempt_enable();
+ preempt_enable_nort();
}
void lg_global_lock(struct lglock *lg)
^ permalink raw reply [flat|nested] 9+ messages in thread* [patch] drm,i915: Use local_lock/unlock_irq() in intel_pipe_update_start/end()
2016-02-27 7:05 [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread() Mike Galbraith
2016-02-27 7:09 ` [patch] drm,radeon,i915: use preempt_disable/enable_rt() where recommended Mike Galbraith
2016-02-27 7:34 ` [patch] locking/lglocks: use preempt_enable/disable_nort() in lg_double_lock/lg_double_unlock Mike Galbraith
@ 2016-02-27 8:01 ` Mike Galbraith
2016-02-27 9:04 ` [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread() Thomas Gleixner
2016-02-27 9:16 ` Thomas Gleixner
4 siblings, 0 replies; 9+ messages in thread
From: Mike Galbraith @ 2016-02-27 8:01 UTC (permalink / raw)
To: Sebastian Andrzej Siewior; +Cc: Thomas Gleixner, linux-rt-users
[ 8.014039] BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:918
[ 8.014041] in_atomic(): 0, irqs_disabled(): 1, pid: 78, name: kworker/u4:4
[ 8.014045] CPU: 1 PID: 78 Comm: kworker/u4:4 Not tainted 4.1.7-rt7 #5
[ 8.014055] Workqueue: events_unbound async_run_entry_fn
[ 8.014059] 0000000000000000 ffff880037153748 ffffffff815f32c9 0000000000000002
[ 8.014063] ffff88013a50e380 ffff880037153768 ffffffff815ef075 ffff8800372c06c8
[ 8.014066] ffff8800372c06c8 ffff880037153778 ffffffff8107c0b3 ffff880037153798
[ 8.014067] Call Trace:
[ 8.014074] [<ffffffff815f32c9>] dump_stack+0x4a/0x61
[ 8.014078] [<ffffffff815ef075>] ___might_sleep.part.93+0xe9/0xee
[ 8.014082] [<ffffffff8107c0b3>] ___might_sleep+0x53/0x80
[ 8.014086] [<ffffffff815f9064>] rt_spin_lock+0x24/0x50
[ 8.014090] [<ffffffff8109368b>] prepare_to_wait+0x2b/0xa0
[ 8.014152] [<ffffffffa016c04c>] intel_pipe_update_start+0x17c/0x300 [i915]
[ 8.014156] [<ffffffff81093b40>] ? prepare_to_wait_event+0x120/0x120
[ 8.014201] [<ffffffffa0158f36>] intel_begin_crtc_commit+0x166/0x1e0 [i915]
[ 8.014215] [<ffffffffa00c806d>] drm_atomic_helper_commit_planes+0x5d/0x1a0 [drm_kms_helper]
[ 8.014260] [<ffffffffa0171e9b>] intel_atomic_commit+0xab/0xf0 [i915]
[ 8.014288] [<ffffffffa00654c7>] drm_atomic_commit+0x37/0x60 [drm]
[ 8.014298] [<ffffffffa00c6fcd>] drm_atomic_helper_plane_set_property+0x8d/0xd0 [drm_kms_helper]
[ 8.014301] [<ffffffff815f77d9>] ? __ww_mutex_lock+0x39/0x40
[ 8.014319] [<ffffffffa0053b3d>] drm_mode_plane_set_obj_prop+0x2d/0x90 [drm]
[ 8.014328] [<ffffffffa00c8edb>] restore_fbdev_mode+0x6b/0xf0 [drm_kms_helper]
[ 8.014337] [<ffffffffa00cae49>] drm_fb_helper_restore_fbdev_mode_unlocked+0x29/0x80 [drm_kms_helper]
[ 8.014346] [<ffffffffa00caec2>] drm_fb_helper_set_par+0x22/0x50 [drm_kms_helper]
[ 8.014390] [<ffffffffa016dfba>] intel_fbdev_set_par+0x1a/0x60 [i915]
[ 8.014394] [<ffffffff81327dc4>] fbcon_init+0x4f4/0x580
[ 8.014398] [<ffffffff8139ef4c>] visual_init+0xbc/0x120
[ 8.014401] [<ffffffff813a1623>] do_bind_con_driver+0x163/0x330
[ 8.014405] [<ffffffff813a1b2c>] do_take_over_console+0x11c/0x1c0
[ 8.014408] [<ffffffff813236e3>] do_fbcon_takeover+0x63/0xd0
[ 8.014410] [<ffffffff81328965>] fbcon_event_notify+0x785/0x8d0
[ 8.014413] [<ffffffff8107c12d>] ? __might_sleep+0x4d/0x90
[ 8.014416] [<ffffffff810775fe>] notifier_call_chain+0x4e/0x80
[ 8.014419] [<ffffffff810779cd>] __blocking_notifier_call_chain+0x4d/0x70
[ 8.014422] [<ffffffff81077a06>] blocking_notifier_call_chain+0x16/0x20
[ 8.014425] [<ffffffff8132b48b>] fb_notifier_call_chain+0x1b/0x20
[ 8.014428] [<ffffffff8132d8fa>] register_framebuffer+0x21a/0x350
[ 8.014439] [<ffffffffa00cb164>] drm_fb_helper_initial_config+0x274/0x3e0 [drm_kms_helper]
[ 8.014483] [<ffffffffa016f1cb>] intel_fbdev_initial_config+0x1b/0x20 [i915]
[ 8.014486] [<ffffffff8107912c>] async_run_entry_fn+0x4c/0x160
[ 8.014490] [<ffffffff81070ffa>] process_one_work+0x14a/0x470
[ 8.014493] [<ffffffff81071489>] worker_thread+0x169/0x4c0
[ 8.014496] [<ffffffff81071320>] ? process_one_work+0x470/0x470
[ 8.014499] [<ffffffff81076606>] kthread+0xc6/0xe0
[ 8.014502] [<ffffffff81070000>] ? queue_work_on+0x80/0x110
[ 8.014506] [<ffffffff81076540>] ? kthread_worker_fn+0x1c0/0x1c0
Signed-off-by: Mike Galbraith <umgwanakikbuti@gmail.com>
---
drivers/gpu/drm/i915/intel_sprite.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -38,6 +38,7 @@
#include "intel_drv.h"
#include <drm/i915_drm.h>
#include "i915_drv.h"
+#include <linux/locallock.h>
static bool
format_is_yuv(uint32_t format)
@@ -64,6 +65,8 @@ static int usecs_to_scanlines(const stru
1000 * adjusted_mode->crtc_htotal);
}
+static DEFINE_LOCAL_IRQ_LOCK(pipe_update_lock);
+
/**
* intel_pipe_update_start() - start update of a set of display registers
* @crtc: the crtc of which the registers are going to be updated
@@ -96,7 +99,7 @@ void intel_pipe_update_start(struct inte
min = vblank_start - usecs_to_scanlines(adjusted_mode, 100);
max = vblank_start - 1;
- local_irq_disable();
+ local_lock_irq(pipe_update_lock);
if (min <= 0 || max <= 0)
return;
@@ -126,11 +129,11 @@ void intel_pipe_update_start(struct inte
break;
}
- local_irq_enable();
+ local_unlock_irq(pipe_update_lock);
timeout = schedule_timeout(timeout);
- local_irq_disable();
+ local_lock_irq(pipe_update_lock);
}
finish_wait(wq, &wait);
@@ -164,7 +167,7 @@ void intel_pipe_update_end(struct intel_
trace_i915_pipe_update_end(crtc, end_vbl_count, scanline_end);
- local_irq_enable();
+ local_unlock_irq(pipe_update_lock);
if (crtc->debug.start_vbl_count &&
crtc->debug.start_vbl_count != end_vbl_count) {
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread()
2016-02-27 7:05 [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread() Mike Galbraith
` (2 preceding siblings ...)
2016-02-27 8:01 ` [patch] drm,i915: Use local_lock/unlock_irq() in intel_pipe_update_start/end() Mike Galbraith
@ 2016-02-27 9:04 ` Thomas Gleixner
2016-02-27 9:14 ` Thomas Gleixner
2016-02-27 9:16 ` Thomas Gleixner
4 siblings, 1 reply; 9+ messages in thread
From: Thomas Gleixner @ 2016-02-27 9:04 UTC (permalink / raw)
To: Mike Galbraith; +Cc: Sebastian Andrzej Siewior, linux-rt-users
On Sat, 27 Feb 2016, Mike Galbraith wrote:
> [ 0.221225] ------------[ cut here ]------------
> [ 0.221241] WARNING: CPU: 1 PID: 69 at kernel/sched/core.c:7733 __might_sleep+0x7f/0x90()
> [ 0.221253] do not call blocking ops when !TASK_RUNNING; state=1 set at [<ffffffff8106edc4>] rescuer_thread+0x64/0x340
> [ 0.221263] Modules linked in:
> [ 0.221272] CPU: 1 PID: 69 Comm: khelper Not tainted 4.0.0-rt4-lockdep #71
> [ 0.221273] Hardware name: IBM System x3550 M3 -[7944K3G]-/69Y5698 , BIOS -[D6E150AUS-1.10]- 12/15/2010
> [ 0.221280] ffffffff81a2269f ffff8801789a7cc8 ffffffff81605182 0000000000000002
> [ 0.221282] ffff8801789a7d18 ffff8801789a7d08 ffffffff81053cda ffff8801789a7d48
> [ 0.221284] ffffffff81a23b44 0000000000000394 0000000000000000 0000000000000000
> [ 0.221284] Call Trace:
> [ 0.221288] [<ffffffff81605182>] dump_stack+0x4f/0x9d
> [ 0.221291] [<ffffffff81053cda>] warn_slowpath_common+0x8a/0xe0
> [ 0.221293] [<ffffffff81053d76>] warn_slowpath_fmt+0x46/0x50
> [ 0.221295] [<ffffffff8106edc4>] ? rescuer_thread+0x64/0x340
> [ 0.221296] [<ffffffff8106edc4>] ? rescuer_thread+0x64/0x340
> [ 0.221298] [<ffffffff810794ef>] __might_sleep+0x7f/0x90
> [ 0.221299] [<ffffffff8106ed60>] ? worker_thread+0x4d0/0x4d0
> [ 0.221301] [<ffffffff8160b4d4>] rt_spin_lock+0x24/0x70
> [ 0.221304] [<ffffffff8107fddd>] ? migrate_disable+0x7d/0xe0
> [ 0.221305] [<ffffffff8106ee09>] rescuer_thread+0xa9/0x340
> [ 0.221307] [<ffffffff8106ed60>] ? worker_thread+0x4d0/0x4d0
> [ 0.221308] [<ffffffff8106ed60>] ? worker_thread+0x4d0/0x4d0
> [ 0.221311] [<ffffffff81073721>] kthread+0xe1/0x100
> [ 0.221312] [<ffffffff8160b1e0>] ? _raw_spin_unlock_irq+0x30/0x70
> [ 0.221315] [<ffffffff81073640>] ? __kthread_parkme+0xa0/0xa0
> [ 0.221317] [<ffffffff8160be88>] ret_from_fork+0x58/0x90
> [ 0.221319] [<ffffffff81073640>] ? __kthread_parkme+0xa0/0xa0
> [ 0.221320] ---[ end trace 0000000000000001 ]---
Hmm. That looks wrong. Spinlocks are state preserving on RT. So I need to
figure out why that thingy triggers here.
Thanks,
tglx
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread()
2016-02-27 7:05 [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread() Mike Galbraith
` (3 preceding siblings ...)
2016-02-27 9:04 ` [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread() Thomas Gleixner
@ 2016-02-27 9:16 ` Thomas Gleixner
2016-02-27 12:11 ` Mike Galbraith
4 siblings, 1 reply; 9+ messages in thread
From: Thomas Gleixner @ 2016-02-27 9:16 UTC (permalink / raw)
To: Mike Galbraith; +Cc: Sebastian Andrzej Siewior, linux-rt-users
On Sat, 27 Feb 2016, Mike Galbraith wrote:
> [ 0.221272] CPU: 1 PID: 69 Comm: khelper Not tainted 4.0.0-rt4-lockdep #71
Can to verify that against some less random rt version?
Thanks,
tglx
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread()
2016-02-27 9:16 ` Thomas Gleixner
@ 2016-02-27 12:11 ` Mike Galbraith
2016-02-27 13:48 ` Mike Galbraith
0 siblings, 1 reply; 9+ messages in thread
From: Mike Galbraith @ 2016-02-27 12:11 UTC (permalink / raw)
To: Thomas Gleixner; +Cc: Sebastian Andrzej Siewior, linux-rt-users
On Sat, 2016-02-27 at 10:16 +0100, Thomas Gleixner wrote:
> On Sat, 27 Feb 2016, Mike Galbraith wrote:
> > [ 0.221272] CPU: 1 PID: 69 Comm: khelper Not tainted 4.0.0-rt4
> -lockdep #71
>
> Can to verify that against some less random rt version?
Yeah, I'll remove the patch from 4.4-rt and see if it still happens.
Patch has been moldering for quite a while.
-Mike
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [patch] rt,workqueue: Fix blocking call when !TASK_RUNNING splat in rescuer_thread()
2016-02-27 12:11 ` Mike Galbraith
@ 2016-02-27 13:48 ` Mike Galbraith
0 siblings, 0 replies; 9+ messages in thread
From: Mike Galbraith @ 2016-02-27 13:48 UTC (permalink / raw)
To: Thomas Gleixner; +Cc: Sebastian Andrzej Siewior, linux-rt-users
On Sat, 2016-02-27 at 13:11 +0100, Mike Galbraith wrote:
> On Sat, 2016-02-27 at 10:16 +0100, Thomas Gleixner wrote:
> > On Sat, 27 Feb 2016, Mike Galbraith wrote:
> > > [ 0.221272] CPU: 1 PID: 69 Comm: khelper Not tainted 4.0.0-rt4
> > -lockdep #71
> >
> > Can to verify that against some less random rt version?
>
> Yeah, I'll remove the patch from 4.4-rt and see if it still happens.
> Patch has been moldering for quite a while.
Hohum, elves and gremlins at play I suppose. I think it's safe to say
oh, never mind to that one.
-Mike
^ permalink raw reply [flat|nested] 9+ messages in thread