Hi,

I guess (not 100% sure if that civer all cases) if gettid() == getpid() you
are on the main process thread.
Having said that I do not like this solution very much: It really looks to
me that we have a larger problem trying to keep these things out of ODP

Something for ARCH call, maybe?

Christophe.

On 23 May 2016 at 10:16, Yi He <[email protected]> wrote:

> Hi, Christophe
>
> Here I met a difficulty, if I unified the API into 
> odph_odpthread_set_affinity(),
> inside the function how can I determine whether the current context is a
> pthread or process? So I cannot decide to call sched_setaffinity() or
> pthread_setaffinity_np().
>
> thanks and best regards, Yi
>
>
>
> On 23 May 2016 at 14:53, Yi He <[email protected]> wrote:
>
>> Hi, Christophe
>>
>> Yes, I'll apply your series and send a new one later.
>>
>> Best Regards, Yi
>>
>> On 23 May 2016 at 14:33, Christophe Milard <[email protected]>
>> wrote:
>>
>>>
>>>
>>> On 20 May 2016 at 10:48, Yi He <[email protected]> wrote:
>>>
>>>> Set affinity to 1st available control cpu for all odp
>>>> validation programs in odp_cunit_common library.
>>>>
>>>> Signed-off-by: Yi He <[email protected]>
>>>> ---
>>>>  helper/include/odp/helper/linux.h         | 47 +++++++++++++++++++
>>>>  helper/linux.c                            | 32 +++++++++++++
>>>>  helper/test/thread.c                      | 76
>>>> +++++++++++++++++++++++++++++--
>>>>  test/validation/common/odp_cunit_common.c | 15 ++++--
>>>>  4 files changed, 164 insertions(+), 6 deletions(-)
>>>>
>>>> diff --git a/helper/include/odp/helper/linux.h
>>>> b/helper/include/odp/helper/linux.h
>>>> index e2dca35..fa815e1 100644
>>>> --- a/helper/include/odp/helper/linux.h
>>>> +++ b/helper/include/odp/helper/linux.h
>>>> @@ -84,6 +84,29 @@ int odph_linux_pthread_create(odph_linux_pthread_t
>>>> *pthread_tbl,
>>>>   */
>>>>  void odph_linux_pthread_join(odph_linux_pthread_t *thread_tbl, int
>>>> num);
>>>>
>>>> +/**
>>>> + * Set CPU affinity of the current thread
>>>> + *
>>>> + * CPU affinity determines the set of CPUs on which the thread is
>>>> + * eligible to run.
>>>> + *
>>>> + * @param cpuset        A bitmask lists the affinity CPU cores
>>>> + *
>>>> + * @return 0 on success, -1 on failure
>>>> + */
>>>> +int odph_linux_pthread_setaffinity(const odp_cpumask_t *cpuset);
>>>>
>>>
>>> odph_odpthread_set_affinity(), is I guess better, at least as long as we
>>> try to keep thread and processes together. Dropping the linux prefix makes
>>> also the code more portable (on some other OS, change provide new helpers
>>> and the app does hopefully not need to change)
>>>
>>> I guess you can apply the "running things in  process mode" patch series
>>> to see what I am after...
>>>
>>> This comment applies to all your function names, of course.
>>>
>>> Christophe
>>>
>>> +
>>>> +/**
>>>> + * Get CPU affinity of the current thread
>>>> + *
>>>> + * CPU affinity determines the set of CPUs on which the thread is
>>>> + * eligible to run.
>>>> + *
>>>> + * @param cpuset[out]   A bitmask lists the affinity CPU cores
>>>> + *
>>>> + * @return 0 on success, -1 on failure
>>>> + */
>>>> +int odph_linux_pthread_getaffinity(odp_cpumask_t *cpuset);
>>>>
>>>>  /**
>>>>   * Fork a process
>>>> @@ -134,6 +157,30 @@ int odph_linux_process_fork_n(odph_linux_process_t
>>>> *proc_tbl,
>>>>  int odph_linux_process_wait_n(odph_linux_process_t *proc_tbl, int num);
>>>>
>>>>  /**
>>>> + * Set CPU affinity of the current process
>>>> + *
>>>> + * CPU affinity determines the set of CPUs on which the process is
>>>> + * eligible to run.
>>>> + *
>>>> + * @param cpuset        A bitmask lists the affinity CPU cores
>>>> + *
>>>> + * @return 0 on success, -1 on failure
>>>> + */
>>>> +int odph_linux_process_setaffinity(const odp_cpumask_t *cpuset);
>>>> +
>>>> +/**
>>>> + * Get CPU affinity of the current process
>>>> + *
>>>> + * CPU affinity determines the set of CPUs on which the process is
>>>> + * eligible to run.
>>>> + *
>>>> + * @param cpuset[out]   A bitmask lists the affinity CPU cores
>>>> + *
>>>> + * @return 0 on success, -1 on failure
>>>> + */
>>>> +int odph_linux_process_getaffinity(odp_cpumask_t *cpuset);
>>>> +
>>>> +/**
>>>>   * @}
>>>>   */
>>>>
>>>> diff --git a/helper/linux.c b/helper/linux.c
>>>> index 24e243b..6ce7e7d 100644
>>>> --- a/helper/linux.c
>>>> +++ b/helper/linux.c
>>>> @@ -114,6 +114,22 @@ void odph_linux_pthread_join(odph_linux_pthread_t
>>>> *thread_tbl, int num)
>>>>         }
>>>>  }
>>>>
>>>> +int odph_linux_pthread_setaffinity(const odp_cpumask_t *cpuset)
>>>> +{
>>>> +       const cpu_set_t *_cpuset = &cpuset->set;
>>>> +
>>>> +       return (0 == pthread_setaffinity_np(pthread_self(),
>>>> +               sizeof(cpu_set_t), _cpuset)) ? 0 : -1;
>>>> +}
>>>> +
>>>> +int odph_linux_pthread_getaffinity(odp_cpumask_t *cpuset)
>>>> +{
>>>> +       cpu_set_t *_cpuset = &cpuset->set;
>>>> +
>>>> +       return (0 == pthread_getaffinity_np(pthread_self(),
>>>> +               sizeof(cpu_set_t), _cpuset)) ? 0 : -1;
>>>> +}
>>>> +
>>>>  int odph_linux_process_fork_n(odph_linux_process_t *proc_tbl,
>>>>                               const odp_cpumask_t *mask,
>>>>                               const odph_linux_thr_params_t *thr_params)
>>>> @@ -236,3 +252,19 @@ int odph_linux_process_wait_n(odph_linux_process_t
>>>> *proc_tbl, int num)
>>>>
>>>>         return 0;
>>>>  }
>>>> +
>>>> +int odph_linux_process_setaffinity(const odp_cpumask_t *cpuset)
>>>> +{
>>>> +       const cpu_set_t *_cpuset = &cpuset->set;
>>>> +
>>>> +       return (0 == sched_setaffinity(0, /* pid zero means calling
>>>> process */
>>>> +               sizeof(cpu_set_t), _cpuset)) ? 0 : -1;
>>>> +}
>>>> +
>>>> +int odph_linux_process_getaffinity(odp_cpumask_t *cpuset)
>>>> +{
>>>> +       cpu_set_t *_cpuset = &cpuset->set;
>>>> +
>>>> +       return (0 == sched_getaffinity(0, /* pid zero means calling
>>>> process */
>>>> +               sizeof(cpu_set_t), _cpuset)) ? 0 : -1;
>>>> +}
>>>> diff --git a/helper/test/thread.c b/helper/test/thread.c
>>>> index b290753..97b4331 100644
>>>> --- a/helper/test/thread.c
>>>> +++ b/helper/test/thread.c
>>>> @@ -4,19 +4,51 @@
>>>>   * SPDX-License-Identifier:     BSD-3-Clause
>>>>   */
>>>>
>>>> +#include <string.h>
>>>> +
>>>>  #include <test_debug.h>
>>>>  #include <odp_api.h>
>>>>  #include <odp/helper/linux.h>
>>>>
>>>>  #define NUMBER_WORKERS 16
>>>> +
>>>> +/* delayed assertion after threads collection */
>>>> +static int worker_results[NUMBER_WORKERS];
>>>> +
>>>>  static void *worker_fn(void *arg TEST_UNUSED)
>>>>  {
>>>> -       /* depend on the odp helper to call odp_init_local */
>>>> +       odp_cpumask_t workers, affinity;
>>>> +       /* save the thread result for delayed assertion */
>>>> +       int *result = &worker_results[odp_cpu_id() % NUMBER_WORKERS];
>>>>
>>>> +       /* depend on the odp helper to call odp_init_local */
>>>>         printf("Worker thread on CPU %d\n", odp_cpu_id());
>>>>
>>>> -       /* depend on the odp helper to call odp_term_local */
>>>> +       odp_cpumask_zero(&workers);
>>>> +       odp_cpumask_zero(&affinity);
>>>> +
>>>> +       odp_cpumask_default_worker(&workers, NUMBER_WORKERS);
>>>> +
>>>> +       /* verify affinity works */
>>>> +       if (odph_linux_pthread_getaffinity(&affinity) != 0) {
>>>> +               printf("Read worker thread affinity failed %d.\n",
>>>> +                       odp_cpu_id());
>>>> +               *result = -1;
>>>> +       } else if (!odp_cpumask_isset(&workers,
>>>> +                       odp_cpumask_first(&affinity))) {
>>>> +               printf("Verify worker thread affinity failed %d.\n",
>>>> +                       odp_cpu_id());
>>>> +               *result = -1;
>>>> +       }
>>>> +
>>>> +       /* verify API is not broken */
>>>> +       if (odph_linux_pthread_setaffinity(&affinity) != 0) {
>>>> +               printf("Re-configure worker thread affinity failed
>>>> %d.\n",
>>>> +                       odp_cpu_id());
>>>> +               *result = -1;
>>>> +       }
>>>>
>>>> +       /* depend on the odp helper to call odp_term_local */
>>>>         return 0;
>>>>  }
>>>>
>>>> @@ -24,7 +56,7 @@ static void *worker_fn(void *arg TEST_UNUSED)
>>>>  int main(int argc TEST_UNUSED, char *argv[] TEST_UNUSED)
>>>>  {
>>>>         odph_linux_pthread_t thread_tbl[NUMBER_WORKERS];
>>>> -       odp_cpumask_t cpu_mask;
>>>> +       odp_cpumask_t cpu_mask, cpuset;
>>>>         int num_workers;
>>>>         int cpu;
>>>>         char cpumaskstr[ODP_CPUMASK_STR_SIZE];
>>>> @@ -41,6 +73,36 @@ int main(int argc TEST_UNUSED, char *argv[]
>>>> TEST_UNUSED)
>>>>                 exit(EXIT_FAILURE);
>>>>         }
>>>>
>>>> +       /* reset all worker thread results to success */
>>>> +       memset(worker_results, 0, sizeof(worker_results));
>>>> +       odp_cpumask_zero(&cpu_mask);
>>>> +       odp_cpumask_zero(&cpuset);
>>>> +
>>>> +       /* allocate the 1st available control cpu to main process */
>>>> +       if (odp_cpumask_default_control(&cpu_mask, 1) != 1) {
>>>> +               LOG_ERR("Allocate main process affinity failed.\n");
>>>> +               exit(EXIT_FAILURE);
>>>> +       }
>>>> +       if (odph_linux_process_setaffinity(&cpu_mask) != 0) {
>>>> +               LOG_ERR("Set main process affinify (%d) failed.\n",
>>>> +                       odp_cpumask_first(&cpu_mask));
>>>> +               exit(EXIT_FAILURE);
>>>> +       }
>>>> +       /* read back affinity to verify */
>>>> +       if ((odph_linux_process_getaffinity(&cpuset) != 0) ||
>>>> +               !odp_cpumask_equal(&cpu_mask, &cpuset)) {
>>>> +               odp_cpumask_to_str(&cpuset,
>>>> +                       cpumaskstr, sizeof(cpumaskstr));
>>>> +
>>>> +               LOG_ERR("Verify main process affinity failed: "
>>>> +                       "set(%d) read(%s).\n",
>>>> +                       odp_cpumask_first(&cpu_mask), cpumaskstr);
>>>> +               exit(EXIT_FAILURE);
>>>> +       }
>>>> +
>>>> +       odp_cpumask_zero(&cpuset);
>>>> +       odp_cpumask_zero(&cpu_mask);
>>>> +
>>>>         /* discover how many threads this system can support */
>>>>         num_workers = odp_cpumask_default_worker(&cpu_mask,
>>>> NUMBER_WORKERS);
>>>>         if (num_workers < NUMBER_WORKERS) {
>>>> @@ -73,6 +135,14 @@ int main(int argc TEST_UNUSED, char *argv[]
>>>> TEST_UNUSED)
>>>>         odph_linux_pthread_create(&thread_tbl[0], &cpu_mask,
>>>> &thr_params);
>>>>         odph_linux_pthread_join(thread_tbl, num_workers);
>>>>
>>>> +       /* assert all worker thread results */
>>>> +       for (cpu = 0; cpu < num_workers; cpu++) {
>>>> +               if (worker_results[cpu] < 0) {
>>>> +                       LOG_ERR("Worker thread %d failed.\n", cpu);
>>>> +                       exit(EXIT_FAILURE);
>>>> +               }
>>>> +       }
>>>> +
>>>>         if (odp_term_local()) {
>>>>                 LOG_ERR("Error: ODP local term failed.\n");
>>>>                 exit(EXIT_FAILURE);
>>>> diff --git a/test/validation/common/odp_cunit_common.c
>>>> b/test/validation/common/odp_cunit_common.c
>>>> index 2712abe..535a9f9 100644
>>>> --- a/test/validation/common/odp_cunit_common.c
>>>> +++ b/test/validation/common/odp_cunit_common.c
>>>> @@ -329,9 +329,18 @@ int odp_cunit_update(odp_suiteinfo_t testsuites[])
>>>>  int odp_cunit_register(odp_suiteinfo_t testsuites[])
>>>>  {
>>>>         /* call test executable init hook, if any */
>>>> -       if (global_init_term.global_init_ptr &&
>>>> -           ((*global_init_term.global_init_ptr)(&instance) != 0))
>>>> -               return -1;
>>>> +       if (global_init_term.global_init_ptr) {
>>>> +               if ((*global_init_term.global_init_ptr)(&instance) ==
>>>> 0) {
>>>> +                       odp_cpumask_t cpuset;
>>>> +
>>>> +                       odp_cpumask_zero(&cpuset);
>>>> +                       /* set main process affinity after ODP
>>>> initialization */
>>>> +                       if (1 == odp_cpumask_default_control(&cpuset,
>>>> 1))
>>>> +                               odph_linux_process_setaffinity(&cpuset);
>>>> +               } else {
>>>> +                       return -1;
>>>> +               }
>>>> +       }
>>>>
>>>>         CU_set_error_action(CUEA_ABORT);
>>>>
>>>> --
>>>> 1.9.1
>>>>
>>>> _______________________________________________
>>>> lng-odp mailing list
>>>> [email protected]
>>>> https://lists.linaro.org/mailman/listinfo/lng-odp
>>>>
>>>
>>>
>>
>
_______________________________________________
lng-odp mailing list
[email protected]
https://lists.linaro.org/mailman/listinfo/lng-odp

Reply via email to