xfs
[Top] [All Lists]

Re: XFS memory allocation deadlock in 2.6.38

To: "'Christoph Hellwig'" <hch@xxxxxxxxxxxxx>
Subject: Re: XFS memory allocation deadlock in 2.6.38
From: Johannes Weiner <hannes@xxxxxxxxxxx>
Date: Tue, 29 Mar 2011 21:39:07 +0200
Cc: Sean Noonan <Sean.Noonan@xxxxxxxxxxxx>, "'Michel Lespinasse'" <walken@xxxxxxxxxx>, "'linux-kernel@xxxxxxxxxxxxxxx'" <linux-kernel@xxxxxxxxxxxxxxx>, Martin Bligh <Martin.Bligh@xxxxxxxxxxxx>, Trammell Hudson <Trammell.Hudson@xxxxxxxxxxxx>, Christos Zoulas <Christos.Zoulas@xxxxxxxxxxxx>, "'linux-xfs@xxxxxxxxxxx'" <linux-xfs@xxxxxxxxxxx>, Stephen Degler <Stephen.Degler@xxxxxxxxxxxx>, "'linux-mm@xxxxxxxxx'" <linux-mm@xxxxxxxxx>
In-reply-to: <20110329192434.GA10536@xxxxxxxxxxxxx>
References: <081DDE43F61F3D43929A181B477DCA95639B52FD@xxxxxxxxxxxxxxxxxxxx> <081DDE43F61F3D43929A181B477DCA95639B5327@xxxxxxxxxxxxxxxxxxxx> <20110324174311.GA31576@xxxxxxxxxxxxx> <AANLkTikwwRm6FHFtEdUg54NvmKdswQw-NPH5dtq1mXBK@xxxxxxxxxxxxxx> <081DDE43F61F3D43929A181B477DCA95639B5349@xxxxxxxxxxxxxxxxxxxx> <BANLkTin0jJevStg5P2hqsLbqMzo3o30sYg@xxxxxxxxxxxxxx> <081DDE43F61F3D43929A181B477DCA95639B534E@xxxxxxxxxxxxxxxxxxxx> <081DDE43F61F3D43929A181B477DCA95639B5359@xxxxxxxxxxxxxxxxxxxx> <20110329192434.GA10536@xxxxxxxxxxxxx>
On Tue, Mar 29, 2011 at 03:24:34PM -0400, 'Christoph Hellwig' wrote:
> Can you check if the brute force patch below helps?  If it does I
> still need to refine it a bit, but it could be that we are doing
> an allocation under an xfs lock that could recurse back into the
> filesystem.  We have a per-process flag to disable that for normal
> kmalloc allocation, but we lost it for vmalloc in the commit you
> bisected the regression to.
> 
> 
> Index: xfs/fs/xfs/linux-2.6/kmem.h
> ===================================================================
> --- xfs.orig/fs/xfs/linux-2.6/kmem.h  2011-03-29 21:16:58.039224236 +0200
> +++ xfs/fs/xfs/linux-2.6/kmem.h       2011-03-29 21:17:08.368223598 +0200
> @@ -63,7 +63,7 @@ static inline void *kmem_zalloc_large(si
>  {
>       void *ptr;
>  
> -     ptr = vmalloc(size);
> +     ptr = __vmalloc(size, GFP_NOFS | __GFP_HIGHMEM, PAGE_KERNEL);
>       if (ptr)
>               memset(ptr, 0, size);
>       return ptr;

Note that vmalloc is currently broken in that it does a GFP_KERNEL
allocation if it has to allocate page table pages, even when invoked
with GFP_NOFS:

        http://marc.info/?l=linux-mm&m=128942194520631&w=4

<Prev in Thread] Current Thread [Next in Thread>