[RFC,PATCH 0/2] cputimers/proc: can't we make /proc/pid/stat less accurate?

Previous thread: Re: [PATCH 2/7] drivers:staging: Kconfig, Makefile for TI's ST ldisc by Pavan Savoy on Wednesday, March 24, 2010 - 1:47 pm. (1 message)

Next thread: [RFC,PATCH 1/2] cputimers/proc: do_task_stat()->task_times() can race with getrusage() by Oleg Nesterov on Wednesday, March 24, 2010 - 1:45 pm. (15 messages)
From: Oleg Nesterov
Date: Wednesday, March 24, 2010 - 1:45 pm

Mostly to provoke the discussion and untested, please comment.

In short: I do not actually know if /proc/pid/stat should be "really"
accurate wrt cpu accounting. I mean, I don't know how much /bin/top
and friends depend on xtime monotonicity.

However, it is sad that /proc/pid/stat iterates over ->thread_group
_twice_ under ->siglock.

Any thoughts?

Oleg.

--

Previous thread: Re: [PATCH 2/7] drivers:staging: Kconfig, Makefile for TI's ST ldisc by Pavan Savoy on Wednesday, March 24, 2010 - 1:47 pm. (1 message)

Next thread: [RFC,PATCH 1/2] cputimers/proc: do_task_stat()->task_times() can race with getrusage() by Oleg Nesterov on Wednesday, March 24, 2010 - 1:45 pm. (15 messages)