If the event for which an AUX area is about to be allocated, does
not support setting up an AUX area, rb_alloc_aux() return -ENOTSUPP.
This error condition is being returned unfiltered to the user space,
and, for example, the perf tools fails with:
failed to mmap with 524 (INTERNAL ERROR: strerror_r(524, 0x3fff497a1c8, 512)=22)
This error can be easily seen with "perf record -m 128,256 -e cpu-clock".
The 524 error code maps to -ENOTSUPP (in rb_alloc_aux()). The -ENOTSUPP
error code shall be only used within the kernel. So the correct error
code would then be -EOPNOTSUPP.
With this commit, the perf tool then reports:
failed to mmap with 95 (Operation not supported)
which is more clear.
Cc: Pu Hou <[email protected]>
Cc: Thomas-Mich Richter <[email protected]>
Signed-off-by: Hendrik Brueckner <[email protected]>
---
kernel/events/ring_buffer.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
index 2831480..ee97196 100644
--- a/kernel/events/ring_buffer.c
+++ b/kernel/events/ring_buffer.c
@@ -580,7 +580,7 @@ int rb_alloc_aux(struct ring_buffer *rb, struct perf_event *event,
int ret = -ENOMEM, max_order = 0;
if (!has_aux(event))
- return -ENOTSUPP;
+ return -EOPNOTSUPP;
if (event->pmu->capabilities & PERF_PMU_CAP_AUX_NO_SG) {
/*
--
1.8.3.1
Hendrik Brueckner <[email protected]> writes:
> If the event for which an AUX area is about to be allocated, does
> not support setting up an AUX area, rb_alloc_aux() return -ENOTSUPP.
>
> This error condition is being returned unfiltered to the user space,
> and, for example, the perf tools fails with:
>
> failed to mmap with 524 (INTERNAL ERROR: strerror_r(524, 0x3fff497a1c8, 512)=22)
>
> This error can be easily seen with "perf record -m 128,256 -e cpu-clock".
>
> The 524 error code maps to -ENOTSUPP (in rb_alloc_aux()). The -ENOTSUPP
> error code shall be only used within the kernel. So the correct error
> code would then be -EOPNOTSUPP.
>
> With this commit, the perf tool then reports:
>
> failed to mmap with 95 (Operation not supported)
>
> which is more clear.
Acked-by: Alexander Shishkin <[email protected]>
Curious as to why does the tool allow this.
Regards,
--
Alex
Commit-ID: 8a1898db51a3390241cd5fae267dc8aaa9db0f8b
Gitweb: http://git.kernel.org/tip/8a1898db51a3390241cd5fae267dc8aaa9db0f8b
Author: Hendrik Brueckner <[email protected]>
AuthorDate: Tue, 20 Jun 2017 12:26:39 +0200
Committer: Ingo Molnar <[email protected]>
CommitDate: Wed, 21 Jun 2017 11:58:30 +0200
perf/aux: Correct return code of rb_alloc_aux() if !has_aux(ev)
If the event for which an AUX area is about to be allocated, does
not support setting up an AUX area, rb_alloc_aux() return -ENOTSUPP.
This error condition is being returned unfiltered to the user space,
and, for example, the perf tools fails with:
failed to mmap with 524 (INTERNAL ERROR: strerror_r(524, 0x3fff497a1c8, 512)=22)
This error can be easily seen with "perf record -m 128,256 -e cpu-clock".
The 524 error code maps to -ENOTSUPP (in rb_alloc_aux()). The -ENOTSUPP
error code shall be only used within the kernel. So the correct error
code would then be -EOPNOTSUPP.
With this commit, the perf tool then reports:
failed to mmap with 95 (Operation not supported)
which is more clear.
Signed-off-by: Hendrik Brueckner <[email protected]>
Acked-by: Alexander Shishkin <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Pu Hou <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Thomas-Mich Richter <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
---
kernel/events/ring_buffer.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
index 2831480..ee97196 100644
--- a/kernel/events/ring_buffer.c
+++ b/kernel/events/ring_buffer.c
@@ -580,7 +580,7 @@ int rb_alloc_aux(struct ring_buffer *rb, struct perf_event *event,
int ret = -ENOMEM, max_order = 0;
if (!has_aux(event))
- return -ENOTSUPP;
+ return -EOPNOTSUPP;
if (event->pmu->capabilities & PERF_PMU_CAP_AUX_NO_SG) {
/*