Re: [git pull] more SLUB updates for 2.6.25

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Christoph Lameter <clameter@...>
Cc: Linus Torvalds <torvalds@...>, Andrew Morton <akpm@...>, LKML <linux-kernel@...>, <linux-mm@...>
Date: Friday, February 8, 2008 - 3:12 am

On Friday 08 February 2008 13:13, Christoph Lameter wrote:

Ah, good. I think it is always a good thing to be able to remove atomics.
They place quite a bit of burden on the CPU, especially x86 where it also
has implicit memory ordering semantics (although x86 can speculatively
get around much of the problem, it's obviously worse than no restriction)

Even if perhaps some cache coherency or timing quirk makes the non-atomic
version slower (all else being equal), then I'd still say that the non
atomic version should be preferred.

Thanks,
Nick

--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[git pull] more SLUB updates for 2.6.25, Christoph Lameter, (Thu Feb 7, 10:13 pm)
Re: [git pull] more SLUB updates for 2.6.25, Nick Piggin, (Fri Feb 8, 3:12 am)
Re: [git pull] more SLUB updates for 2.6.25, Eric Dumazet, (Fri Feb 8, 3:29 am)
Re: [git pull] more SLUB updates for 2.6.25, Andi Kleen, (Fri Feb 8, 10:58 am)
Re: [git pull] more SLUB updates for 2.6.25, Christoph Lameter, (Fri Feb 8, 4:10 am)
Re: [git pull] more SLUB updates for 2.6.25, Eric Dumazet, (Sat Feb 9, 7:25 am)
Re: [git pull] more SLUB updates for 2.6.25, Nick Piggin, (Fri Feb 8, 4:08 am)