Hi Levi,

On 06/03/2025 09:47, Yeoreum Yun wrote:

In coresight-tmc drivers, tmc_drvdata->spinlock can be held during __schedule() by perf_event_task_sched_out()/in().

Since tmc_drvdata->spinlock type is spinlock_t and perf_event_task_sched_out()/in() is called after acquiring rq_lock, which is raw_spinlock_t (an unsleepable lock), this poses an issue in PREEMPT_RT kernel where spinlock_t is sleepable.

To address this, change type tmc_drvdata->spinlock in coresight-tmc drivers, which can be called by perf_event_task_sched_out()/in(), from spinlock_t to raw_spinlock_t.

Reviewed-by: James Clark james.clark@linaro.org Reviewed-by: Mike Leach mike.leach@linaro.org Signed-off-by: Yeoreum Yun yeoreum.yun@arm.com

Unfortunately, this still doesn't cover the current coresight next branch. I get build errors as below :

"[PATCH] coresight-tmc: change tmc_drvdata spinlock's type to" has style problems, please review. This is because, we merged the coresight panic trace support in the coresight next, on 21st Feb.

NOTE: If any of the errors are false positives, please report them to the maintainer, see CHECKPATCH in MAINTAINERS. CALL scripts/checksyscalls.sh CC [M] drivers/hwtracing/coresight/coresight-tmc-core.o CC [M] drivers/hwtracing/coresight/coresight-tmc-etf.o CC [M] drivers/hwtracing/coresight/coresight-tmc-etr.o CC [M] drivers/hwtracing/coresight/coresight-catu.o In file included from ./include/linux/mmzone.h:8, from ./include/linux/gfp.h:7, from ./include/linux/slab.h:16, from ./include/linux/resource_ext.h:11, from ./include/linux/acpi.h:13, from drivers/hwtracing/coresight/coresight-tmc-core.c:7: drivers/hwtracing/coresight/coresight-tmc-core.c: In function ‘tmc_crashdata_open’: drivers/hwtracing/coresight/coresight-tmc-core.c:361:20: error: passing argument 1 of ‘spinlock_check’ from incompatible pointer type [-Werror=incompatible-pointer-types] 361 | spin_lock_irqsave(&drvdata->spinlock, flags); | ^~~~~~~~~~~~~~~~~~ | | | raw_spinlock_t * {aka struct raw_spinlock *} ./include/linux/spinlock.h:244:34: note: in definition of macro ‘raw_spin_lock_irqsave’ 244 | flags = _raw_spin_lock_irqsave(lock); \ | ^~~~ drivers/hwtracing/coresight/coresight-tmc-core.c:361:2: note: in expansion of macro ‘spin_lock_irqsave’ 361 | spin_lock_irqsave(&drvdata->spinlock, flags); | ^~~~~~~~~~~~~~~~~ In file included from ./include/linux/mmzone.h:8, from ./include/linux/gfp.h:7, from ./include/linux/slab.h:16, from ./include/linux/resource_ext.h:11, from ./include/linux/acpi.h:13, from drivers/hwtracing/coresight/coresight-tmc-core.c:7: ./include/linux/spinlock.h:324:67: note: expected ‘spinlock_t *’ {aka ‘struct spinlock *’} but argument is of type ‘raw_spinlock_t *’ {aka ‘struct raw_spinlock *’} 324 | static __always_inline raw_spinlock_t *spinlock_check(spinlock_t *lock)

Cheers Suzuki