[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCHES]

To: Ingo Molnar <mingo@chiara.csoma.elte.hu>
Subject: Re: [PATCHES]
From: Manfred Spraul <manfreds@colorfullife.com>
Date: Mon, 24 May 1999 19:20:41 +0200
CC: "Eric W. Biederman" <ebiederm+eric@ccr.net>, Linus Torvalds <torvalds@transmeta.com>, linux-kernel@vger.rutgers.edu, linux-mm@kvack.org
References: <Pine.LNX.3.96.990523171206.21583A-100000@chiara.csoma.elte.hu>
Reply-To: masp0008@stud.uni-sb.de
Sender: owner-linux-mm@kvack.org

Ingo Molnar wrote:
> on my box the page cache is already completely parallel on SMP,
> we drop the kernel lock on entry into page-cache routines and
> re-lock it only if we call filesystem-specific code or
> buffer-cache code.

How have you called the 'release_kernel_lock()' function?

I found several lengthy operations in the kernel which
should also release the kernel lock.
(the slowest: clear_page() when called by get_free_page(GFP_WAIT))

--
	Manfred

--
To unsubscribe, send a message with 'unsubscribe linux-mm my@address'
in the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://humbolt.geo.uu.nl/Linux-MM/

References:
- Re: [PATCHES]
  - From: Ingo Molnar <mingo@chiara.csoma.elte.hu>

Prev by Date: Re: [PATCH] cache large files in the page cache
Next by Date: Re: Q: PAGE_CACHE_SIZE?
Prev by thread: Re: [PATCHES]
Next by thread: kernel_lock() profiling results
Index(es):
- Date
- Thread