From: Maarten Lankhorst <dev@lankhorst.se>
To: intel-xe@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org
Subject: [i915-rt v2 13/16] drm/i915/gt: Use spin_lock_irq() instead of local_irq_disable() + spin_lock()
Date: Tue, 16 Dec 2025 10:22:40 +0100 [thread overview]
Message-ID: <20251216092226.1777909-31-dev@lankhorst.se> (raw)
In-Reply-To: <20251216092226.1777909-18-dev@lankhorst.se>
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
execlists_dequeue() is invoked from a function which uses
local_irq_disable() to disable interrupts so the spin_lock() behaves
like spin_lock_irq().
This breaks PREEMPT_RT because local_irq_disable() + spin_lock() is not
the same as spin_lock_irq().
execlists_dequeue_irq() and execlists_dequeue() has each one caller
only. If intel_engine_cs::active::lock is acquired and released with the
_irq suffix then it behaves almost as if execlists_dequeue() would be
invoked with disabled interrupts. The difference is the last part of the
function which is then invoked with enabled interrupts.
I can't tell if this makes a difference. From looking at it, it might
work to move the last unlock at the end of the function as I didn't find
anything that would acquire the lock again.
Reported-by: Clark Williams <williams@redhat.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
---
.../drm/i915/gt/intel_execlists_submission.c | 17 +++++------------
1 file changed, 5 insertions(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
index 3df683b0402ad..948975e72d295 100644
--- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
@@ -1300,7 +1300,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
* and context switches) submission.
*/
- spin_lock(&sched_engine->lock);
+ spin_lock_irq(&sched_engine->lock);
/*
* If the queue is higher priority than the last
@@ -1400,7 +1400,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
* Even if ELSP[1] is occupied and not worthy
* of timeslices, our queue might be.
*/
- spin_unlock(&sched_engine->lock);
+ spin_unlock_irq(&sched_engine->lock);
return;
}
}
@@ -1426,7 +1426,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
if (last && !can_merge_rq(last, rq)) {
spin_unlock(&ve->base.sched_engine->lock);
- spin_unlock(&engine->sched_engine->lock);
+ spin_unlock_irq(&engine->sched_engine->lock);
return; /* leave this for another sibling */
}
@@ -1588,7 +1588,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
*/
sched_engine->queue_priority_hint = queue_prio(sched_engine);
i915_sched_engine_reset_on_empty(sched_engine);
- spin_unlock(&sched_engine->lock);
+ spin_unlock_irq(&sched_engine->lock);
/*
* We can skip poking the HW if we ended up with exactly the same set
@@ -1614,13 +1614,6 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
}
}
-static void execlists_dequeue_irq(struct intel_engine_cs *engine)
-{
- local_irq_disable(); /* Suspend interrupts across request submission */
- execlists_dequeue(engine);
- local_irq_enable(); /* flush irq_work (e.g. breadcrumb enabling) */
-}
-
static void clear_ports(struct i915_request **ports, int count)
{
memset_p((void **)ports, NULL, count);
@@ -2475,7 +2468,7 @@ static void execlists_submission_tasklet(struct tasklet_struct *t)
}
if (!engine->execlists.pending[0]) {
- execlists_dequeue_irq(engine);
+ execlists_dequeue(engine);
start_timeslice(engine);
}
--
2.51.0
next prev parent reply other threads:[~2025-12-16 9:22 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-16 9:22 [i915-rt v2 00/16] drm/i915/display: All patches to make PREEMPT_RT work on i915 + xe Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 01/16] drm/i915/display: Fix intel_lpe_audio_irq_handler for PREEMPT-RT Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 02/16] drm/i915/display: Make get_vblank_counter use intel_de_read_fw() Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 03/16] drm/i915/display: Use intel_de_write_fw in intel_pipe_fastset Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 04/16] drm/i915/display: Make set_pipeconf use the fw variants Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 05/16] drm/i915/display: Move vblank put until after critical section Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 06/16] drm/i915/display: Remove locking from intel_vblank_evade " Maarten Lankhorst
2025-12-16 20:15 ` [i915-rt v2.1] " Maarten Lankhorst
2025-12-22 18:14 ` [i915-rt v2 06/16] " kernel test robot
2025-12-16 9:22 ` [i915-rt v2 07/16] drm/i915/display: Handle vlv dsi workaround in scanline_in_safe_range too Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 08/16] drm/i915/display: Make icl_dsi_frame_update use _fw too Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 09/16] drm/i915/display: Enable interrupts earlier on PREEMPT_RT Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 10/16] drm/i915: Use preempt_disable/enable_rt() where recommended Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 11/16] PREEMPT_RT injection Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 12/16] drm/i915/display: Use intel_de_read_fw in colorops Maarten Lankhorst
2025-12-16 9:22 ` Maarten Lankhorst [this message]
2025-12-16 9:22 ` [i915-rt v2 14/16] drm/i915: Drop the irqs_disabled() check Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 15/16] drm/i915/guc: Consider also RCU depth in busy loop Maarten Lankhorst
2025-12-16 9:22 ` [i915-rt v2 16/16] Revert "drm/i915: Depend on !PREEMPT_RT." Maarten Lankhorst
2025-12-16 11:35 ` ✗ Fi.CI.BUILD: failure for drm/i915/display: All patches to make PREEMPT_RT work on i915 + xe. (rev2) Patchwork
2025-12-16 20:54 ` ✗ Fi.CI.BUILD: warning for drm/i915/display: All patches to make PREEMPT_RT work on i915 + xe. (rev3) Patchwork
2025-12-16 20:55 ` ✗ Fi.CI.DOCS: " Patchwork
2025-12-16 21:27 ` ✗ i915.CI.BAT: failure " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20251216092226.1777909-31-dev@lankhorst.se \
--to=dev@lankhorst.se \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox