2022-04-07 20:19:38

by Shuah Khan

[permalink] [raw]
Subject: Re: [PATCH V2] testing/selftests/mqueue: Fix mq_perf_tests to free the allocated cpu set

On 4/7/22 12:40 PM, Athira Rajeev wrote:
> The selftest "mqueue/mq_perf_tests.c" use CPU_ALLOC to allocate
> CPU set. This cpu set is used further in pthread_attr_setaffinity_np
> and by pthread_create in the code. But in current code, allocated
> cpu set is not freed.
>
> Fix this issue by adding CPU_FREE in the "shutdown" function which
> is called in most of the error/exit path for the cleanup. Also add
> CPU_FREE in some of the error paths where shutdown is not called.
>
> Fixes: 7820b0715b6f ("tools/selftests: add mq_perf_tests")
> Signed-off-by: Athira Rajeev <[email protected]>
> ---
> Changelog:
> From v1 -> v2:
> Addressed review comment from Shuah Khan to add
> CPU_FREE in other exit paths where it is needed
>

Thank you. I have a couple of comments on making the error
paths simpler. Please see below.

> tools/testing/selftests/mqueue/mq_perf_tests.c | 10 +++++++++-
> 1 file changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/tools/testing/selftests/mqueue/mq_perf_tests.c b/tools/testing/selftests/mqueue/mq_perf_tests.c
> index b019e0b8221c..182434c7898d 100644
> --- a/tools/testing/selftests/mqueue/mq_perf_tests.c
> +++ b/tools/testing/selftests/mqueue/mq_perf_tests.c
> @@ -180,6 +180,9 @@ void shutdown(int exit_val, char *err_cause, int line_no)
> if (in_shutdown++)
> return;
>
> + /* Free the cpu_set allocated using CPU_ALLOC in main function */
> + CPU_FREE(cpu_set);
> +
> for (i = 0; i < num_cpus_to_pin; i++)
> if (cpu_threads[i]) {
> pthread_kill(cpu_threads[i], SIGUSR1);
> @@ -589,6 +592,7 @@ int main(int argc, char *argv[])
> cpu_set)) {
> fprintf(stderr, "Any given CPU may "
> "only be given once.\n");
> + CPU_FREE(cpu_set);

This could be done in a common error path handling.

> exit(1);
> } else
> CPU_SET_S(cpus_to_pin[cpu],
> @@ -607,6 +611,7 @@ int main(int argc, char *argv[])
> queue_path = malloc(strlen(option) + 2);
> if (!queue_path) {
> perror("malloc()");
> + CPU_FREE(cpu_set);

This could be done in a common error path handling.

> exit(1);
> }
> queue_path[0] = '/';
> @@ -619,6 +624,7 @@ int main(int argc, char *argv[])
> }
>
> if (continuous_mode && num_cpus_to_pin == 0) {
> + CPU_FREE(cpu_set);

This could be done in a common error path handling.

> fprintf(stderr, "Must pass at least one CPU to continuous "
> "mode.\n");
> poptPrintUsage(popt_context, stderr, 0);
> @@ -628,10 +634,12 @@ int main(int argc, char *argv[])
> cpus_to_pin[0] = cpus_online - 1;
> }
>
> - if (getuid() != 0)
> + if (getuid() != 0) {
> + CPU_FREE(cpu_set);
> ksft_exit_skip("Not running as root, but almost all tests "
> "require root in order to modify\nsystem settings. "
> "Exiting.\n");
> + }
>

Why not move this check before CPU_ALLOC and make this the very first
check in main()?

With this change the other places where CPU_FREE is added right before
exit(1). Something like this:

err_code:
CPU_FREE(cpu_set);
exit(code)

> max_msgs = fopen(MAX_MSGS, "r+");
> max_msgsize = fopen(MAX_MSGSIZE, "r+");
>

thanks,
-- Shuah


2022-04-08 07:51:56

by Athira Rajeev

[permalink] [raw]
Subject: Re: [PATCH V2] testing/selftests/mqueue: Fix mq_perf_tests to free the allocated cpu set



> On 08-Apr-2022, at 12:31 AM, Shuah Khan <[email protected]> wrote:
>
> On 4/7/22 12:40 PM, Athira Rajeev wrote:
>> The selftest "mqueue/mq_perf_tests.c" use CPU_ALLOC to allocate
>> CPU set. This cpu set is used further in pthread_attr_setaffinity_np
>> and by pthread_create in the code. But in current code, allocated
>> cpu set is not freed.
>> Fix this issue by adding CPU_FREE in the "shutdown" function which
>> is called in most of the error/exit path for the cleanup. Also add
>> CPU_FREE in some of the error paths where shutdown is not called.
>> Fixes: 7820b0715b6f ("tools/selftests: add mq_perf_tests")
>> Signed-off-by: Athira Rajeev <[email protected]>
>> ---
>> Changelog:
>> From v1 -> v2:
>> Addressed review comment from Shuah Khan to add
>> CPU_FREE in other exit paths where it is needed
>
> Thank you. I have a couple of comments on making the error
> paths simpler. Please see below.
>
>> tools/testing/selftests/mqueue/mq_perf_tests.c | 10 +++++++++-
>> 1 file changed, 9 insertions(+), 1 deletion(-)
>> diff --git a/tools/testing/selftests/mqueue/mq_perf_tests.c b/tools/testing/selftests/mqueue/mq_perf_tests.c
>> index b019e0b8221c..182434c7898d 100644
>> --- a/tools/testing/selftests/mqueue/mq_perf_tests.c
>> +++ b/tools/testing/selftests/mqueue/mq_perf_tests.c
>> @@ -180,6 +180,9 @@ void shutdown(int exit_val, char *err_cause, int line_no)
>> if (in_shutdown++)
>> return;
>> + /* Free the cpu_set allocated using CPU_ALLOC in main function */
>> + CPU_FREE(cpu_set);
>> +
>> for (i = 0; i < num_cpus_to_pin; i++)
>> if (cpu_threads[i]) {
>> pthread_kill(cpu_threads[i], SIGUSR1);
>> @@ -589,6 +592,7 @@ int main(int argc, char *argv[])
>> cpu_set)) {
>> fprintf(stderr, "Any given CPU may "
>> "only be given once.\n");
>> + CPU_FREE(cpu_set);
>
> This could be done in a common error path handling.
>
>> exit(1);
>> } else
>> CPU_SET_S(cpus_to_pin[cpu],
>> @@ -607,6 +611,7 @@ int main(int argc, char *argv[])
>> queue_path = malloc(strlen(option) + 2);
>> if (!queue_path) {
>> perror("malloc()");
>> + CPU_FREE(cpu_set);
>
> This could be done in a common error path handling.
>
>> exit(1);
>> }
>> queue_path[0] = '/';
>> @@ -619,6 +624,7 @@ int main(int argc, char *argv[])
>> }
>> if (continuous_mode && num_cpus_to_pin == 0) {
>> + CPU_FREE(cpu_set);
>
> This could be done in a common error path handling.
>
>> fprintf(stderr, "Must pass at least one CPU to continuous "
>> "mode.\n");
>> poptPrintUsage(popt_context, stderr, 0);
>> @@ -628,10 +634,12 @@ int main(int argc, char *argv[])
>> cpus_to_pin[0] = cpus_online - 1;
>> }
>> - if (getuid() != 0)
>> + if (getuid() != 0) {
>> + CPU_FREE(cpu_set);
>> ksft_exit_skip("Not running as root, but almost all tests "
>> "require root in order to modify\nsystem settings. "
>> "Exiting.\n");
>> + }
>>
>
> Why not move this check before CPU_ALLOC and make this the very first
> check in main()?
>
> With this change the other places where CPU_FREE is added right before
> exit(1). Something like this:
>
> err_code:
> CPU_FREE(cpu_set);
> exit(code)
Hi Shuah

Thanks for the comments. Addressed these in V3

Athira
>
>> max_msgs = fopen(MAX_MSGS, "r+");
>> max_msgsize = fopen(MAX_MSGSIZE, "r+");
>
> thanks,
> -- Shuah