commit 6746eae4bbaddcc16b40efb33dab79210828b3ce upstream.
cti_enable_hw() and cti_disable_hw() are called from an atomic context
so shouldn't use runtime PM because it can result in a sleep when
communicating with firmware.
This can cause a hang when running the Perf Coresight tests or running
this command:
perf record -e cs_etm//u -- ls
With lock and scheduler debugging enabled the following is output:
coresight cti_sys0: cti_enable_hw -- dev:cti_sys0 parent: 20020000.cti
BUG: sleeping function called from invalid context at drivers/base/power/runtime.c:1151
in_atomic(): 1, irqs_disabled(): 128, non_block: 0, pid: 330, name: perf-exec
preempt_count: 2, expected: 0
RCU nest depth: 0, expected: 0
INFO: lockdep is turned off.
irq event stamp: 0
hardirqs last enabled at (0): [<0000000000000000>] 0x0
hardirqs last disabled at (0): [<ffff80000822b394>] copy_process+0xa0c/0x1948
softirqs last enabled at (0): [<ffff80000822b394>] copy_process+0xa0c/0x1948
softirqs last disabled at (0): [<0000000000000000>] 0x0
CPU: 3 PID: 330 Comm: perf-exec Not tainted 6.0.0-00053-g042116d99298 #7
Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform, BIOS EDK II Sep 13 2022
Call trace:
dump_backtrace+0x134/0x140
show_stack+0x20/0x58
dump_stack_lvl+0x8c/0xb8
dump_stack+0x18/0x34
__might_resched+0x180/0x228
__might_sleep+0x50/0x88
__pm_runtime_resume+0xac/0xb0
cti_enable+0x44/0x120
coresight_control_assoc_ectdev+0xc0/0x150
coresight_enable_path+0xb4/0x288
etm_event_start+0x138/0x170
etm_event_add+0x48/0x70
event_sched_in.isra.122+0xb4/0x280
merge_sched_in+0x1fc/0x3d0
visit_groups_merge.constprop.137+0x16c/0x4b0
ctx_sched_in+0x114/0x1f0
perf_event_sched_in+0x60/0x90
ctx_resched+0x68/0xb0
perf_event_exec+0x138/0x508
begin_new_exec+0x52c/0xd40
load_elf_binary+0x6b8/0x17d0
bprm_execve+0x360/0x7f8
do_execveat_common.isra.47+0x218/0x238
__arm64_sys_execve+0x48/0x60
invoke_syscall+0x4c/0x110
el0_svc_common.constprop.4+0xfc/0x120
do_el0_svc+0x34/0xc0
el0_svc+0x40/0x98
el0t_64_sync_handler+0x98/0xc0
el0t_64_sync+0x170/0x174
Fix the issue by removing the runtime PM calls completely. They are not
needed here because it must have already been done when building the
path for a trace.
Fixes: 835d722ba10a ("coresight: cti: Initial CoreSight CTI Driver")
Cc: stable <stable(a)kernel.org> # 5.10.x
Reported-by: Aishwarya TCV <Aishwarya.TCV(a)arm.com>
Reported-by: Cristian Marussi <Cristian.Marussi(a)arm.com>
Suggested-by: Suzuki K Poulose <suzuki.poulose(a)arm.com>
Reviewed-by: Mike Leach <mike.leach(a)linaro.org>
Tested-by: Mike Leach <mike.leach(a)linaro.org>
[ Fix build warnings ]
Signed-off-by: Suzuki K Poulose <suzuki.poulose(a)arm.com>
Link: https://lore.kernel.org/r/20221025131032.1149459-1-suzuki.poulose@arm.com
Signed-off-by: James Clark <james.clark(a)arm.com>
---
drivers/hwtracing/coresight/coresight-cti-core.c | 5 -----
1 file changed, 5 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c b/drivers/hwtracing/coresight/coresight-cti-core.c
index 0276700c246d..90270696206c 100644
--- a/drivers/hwtracing/coresight/coresight-cti-core.c
+++ b/drivers/hwtracing/coresight/coresight-cti-core.c
@@ -90,11 +90,9 @@ void cti_write_all_hw_regs(struct cti_drvdata *drvdata)
static int cti_enable_hw(struct cti_drvdata *drvdata)
{
struct cti_config *config = &drvdata->config;
- struct device *dev = &drvdata->csdev->dev;
unsigned long flags;
int rc = 0;
- pm_runtime_get_sync(dev->parent);
spin_lock_irqsave(&drvdata->spinlock, flags);
/* no need to do anything if enabled or unpowered*/
@@ -119,7 +117,6 @@ static int cti_enable_hw(struct cti_drvdata *drvdata)
/* cannot enable due to error */
cti_err_not_enabled:
spin_unlock_irqrestore(&drvdata->spinlock, flags);
- pm_runtime_put(dev->parent);
return rc;
}
@@ -153,7 +150,6 @@ static void cti_cpuhp_enable_hw(struct cti_drvdata *drvdata)
static int cti_disable_hw(struct cti_drvdata *drvdata)
{
struct cti_config *config = &drvdata->config;
- struct device *dev = &drvdata->csdev->dev;
spin_lock(&drvdata->spinlock);
@@ -174,7 +170,6 @@ static int cti_disable_hw(struct cti_drvdata *drvdata)
coresight_disclaim_device_unlocked(drvdata->base);
CS_LOCK(drvdata->base);
spin_unlock(&drvdata->spinlock);
- pm_runtime_put(dev->parent);
return 0;
/* not disabled this call */
--
2.25.1
Version 1.3.3 of OpenCSD is now released.
Fixes problems where bad input data to the CoreSight frame demuxer was
not being correctly handled, resulting in buffer misalignment and
possible read beyond the end of the buffers. (Github issues #49, #50,
#51)
Adds a test program to validate these fixes and other error checking
which has been improved / corrected.
Also fixed and issue in PTM decode which failed to extract the address
of a waypoint marker correctly under certain circumstances.(github
issue #48)
Adds build env for VS2022.
Regards
Mike
--
Mike Leach
Principal Engineer, ARM Ltd.
Manchester Design Centre. UK
