Re: [PATCH 1/3] sched, timer: Remove usages of ACCESS_ONCE in the scheduler

From: Steven Rostedt
Date: Wed Apr 15 2015 - 15:17:09 EST

On Wed, 15 Apr 2015 11:49:09 -0700
Jason Low <jason.low2@xxxxxx> wrote:

> So I'll keep the READ_ONCE nested inside WRITE_ONCE for the purpose of
> this patch since this patch is a conversion from ACCESS_ONCE, but yes,
> if the original purpose of ACCESS_ONCE was to do an atomic increment,
> then the ACCESS_ONCE doesn't help with that.

For the purpose of this patch, I think it's fine, as being more
paranoid is better than not being paranoid enough.

But this has shined light onto whether it is needed or not, and we
should figure that out in the not so far future.

-- Steve
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at