Re: [RFC] New locking primitive for 2.5

From: Alexander Viro (viro@math.psu.edu)
Date: Fri Feb 08 2002 - 14:10:25 EST


On Fri, 8 Feb 2002, Ingo Molnar wrote:

> > I'm going to send ext2-without-BKL patches to Linus - tonight or
> > tomorrow. I really wonder what effect that would have on the things.
>
> oh, that is a really cool thing!
>
> llseek() is unrelated, and i think pretty gross. Is there any other reason
> to llseek()'s i_sem usage other than the 64-bit atomic update of the file
> offset value? We can do lockless, SMP-correct 64-bit updates on x86 pretty
> easily.

Umm... Wait a second. You've seen the problems on ->i_sem variant
of llseek()? My apologies - I've misparsed you.

I seriously suspect that BKL-for-lseek() will be good enough once we
kill BKL in ext2_get_block() and friends. IOW, I doubt that
generic_file_lseek() showing up in BKL contention is the real
problem...

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Fri Feb 15 2002 - 21:00:21 EST