Re: [PATCH] Re: zero-filed page on VOP_PUTPAGES

To: yamt%mwd.biglobe.ne.jp@localhost (YAMAMOTO Takashi)
Subject: Re: [PATCH] Re: zero-filed page on VOP_PUTPAGES
From: manu%netbsd.org@localhost (Emmanuel Dreyfus)
Date: Fri, 26 Aug 2011 05:21:15 +0200

YAMAMOTO Takashi <yamt%mwd.biglobe.ne.jp@localhost> wrote:

> ioflush (syncer) is not a kernel thread responsible for freeing memory.
> pagedaemon is.
> 
> i don't think syncer alone can deadlock with puffs.

Indeed it does not: there is at least a third thread involved. In a
typical case, ioflush sleeps awaiting for perfused/glusterfsd to finish
fsync, and perfused/glusterfsd is stuck in the kernel awaiting memory
allocation.

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu%netbsd.org@localhost

Follow-Ups:
- Re: [PATCH] Re: zero-filed page on VOP_PUTPAGES
  - From: YAMAMOTO Takashi

References:
- Re: [PATCH] Re: zero-filed page on VOP_PUTPAGES
  - From: YAMAMOTO Takashi

Prev by Date: Re: [PATCH] Re: zero-filed page on VOP_PUTPAGES
Next by Date: Re: deadlock on flt_noram5
Previous by Thread: Re: [PATCH] Re: zero-filed page on VOP_PUTPAGES
Next by Thread: Re: [PATCH] Re: zero-filed page on VOP_PUTPAGES
Indexes:

Home | Main Index | Thread Index | Old Index