On 06/17, Alpár Török wrote:
>
> The producer is a utrace monitored task. At any time there is an
> arbitrary number of producers. On sys_entry and sys_exit the producers
> do :
>
> slot = book_message_in_buffer(buf);
> if (slot < 0) {
> /* no slot can be alocated, producer must sleep */
> enque_producer_on_wait_list(list, producer);
> return UTRACE_STOP;
> }
I think Roland is right. Let's forget about utrace for the moment,
this code looks like
if (!CONDITION) {
set_task_state(TASK_XXX);
schedule();
}
and it can obviously race with
CONDITION = true;
wake_up_xxx();
I am wondering if there is another way to handle such a race...
Suppose we
- add UTRACE_STOP_XXX
- modify utrace_control(UTRACE_STOP_XXX) to do
mark_engine_wants_stop() and nothing more.
- finish_callback_report(UTRACE_STOP_XXX) does
spin_lock(&utrace->lock);
if (engine_wants_stop(engine))
action = UTRACE_STOP;
else
action = UTRACE_REUME;
spin_unlock(&utrace->lock);
Then, probably producer could do
utrace_control(current, UTRACE_STOP_XXX);
slot = book_message_in_buffer(buf);
if (slot < 0)
return UTRACE_STOP_XXX;
...
return UTRACE_RESUME;
Even better, UTRACE_STOP_XXX can live outside of UTRACE_RESUME_MASK.
Unlikely this makes sense, just a random thought.
Oleg.
