[GitHub] [incubator-nuttx] xiaoxiang781216 commented on pull request #2989: Fixes race condition in event wait logic of SDMMC driver.

[GitBox <gi...@apache.org>]

xiaoxiang781216 commented on pull request #2989:
URL: https://github.com/apache/incubator-nuttx/pull/2989#issuecomment-792203932


   > 2. The second is an issue @davids5 says he has some recollection of @xiaoxiang781216 identifying awhile back. Basically it's that enter_critical_section's are not being respected when calling into the watchdog in tickless mode. A task switch causes interrupts to be re-enabled and thus breaks the critical section. This sched switch and re-enabling of interrupts is actually the reason the watchdog wasn't getting started until after the the watchdog was started. Wrapping the wd_start in sched_lock() as a test fixed the issue. @xiaoxiang781216 Do you recall this? Is there an issue somewhere we could discuss solutions?
   > 
   
   Yes, here is the issue: https://github.com/apache/incubator-nuttx/issues/1138
   Basically, enter_critical_section should stop any schedule until the thread initiatively give up the time slice by wait/sleep and should postone the schedule of the high priority thread waked up by the current thread. 


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org