6.13-stable review patch. If anyone has any objections, please let me know.
------------------
From: Krzysztof Karas <krzysztof.karas@xxxxxxxxx>
commit e49477f7f78598295551d486ecc7f020d796432e upstream.
spin_lock/unlock() functions used in interrupt contexts could
result in a deadlock, as seen in GitLab issue #13399,
which occurs when interrupt comes in while holding a lock.
Try to remedy the problem by saving irq state before spin lock
acquisition.
v2: add irqs' state save/restore calls to all locks/unlocks in
signal_irq_work() execution (Maciej)
v3: use with spin_lock_irqsave() in guc_lrc_desc_unpin() instead
of other lock/unlock calls and add Fixes and Cc tags (Tvrtko);
change title and commit message
Fixes: 2f2cc53b5fe7 ("drm/i915/guc: Close deregister-context race against CT-loss")
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13399
Signed-off-by: Krzysztof Karas <krzysztof.karas@xxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx> # v6.9+
Reviewed-by: Maciej Patelczyk <maciej.patelczyk@xxxxxxxxx>
Reviewed-by: Andi Shyti <andi.shyti@xxxxxxxxxxxxxxx>
Signed-off-by: Andi Shyti <andi.shyti@xxxxxxxxxxxxxxx>
Link: https://patchwork.freedesktop.org/patch/msgid/pusppq5ybyszau2oocboj3mtj5x574gwij323jlclm5zxvimmu@mnfg6odxbpsv
(cherry picked from commit c088387ddd6482b40f21ccf23db1125e8fa4af7e)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
@@ -3425,10 +3425,10 @@ static inline int guc_lrc_desc_unpin(str
*/
ret = deregister_context(ce, ce->guc_id.id);
if (ret) {
- spin_lock(&ce->guc_state.lock);
+ spin_lock_irqsave(&ce->guc_state.lock, flags);
set_context_registered(ce);
clr_context_destroyed(ce);
- spin_unlock(&ce->guc_state.lock);
+ spin_unlock_irqrestore(&ce->guc_state.lock, flags);
/*
* As gt-pm is awake at function entry, intel_wakeref_put_async merely decrements
* the wakeref immediately but per function spec usage call this after unlock.
