xfs
[Top] [All Lists]

Re: [PATCH] xfs: prevent stack overflows from page cache allocation

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH] xfs: prevent stack overflows from page cache allocation
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Mon, 28 Oct 2013 02:49:57 -0700
Cc: linux-fsdevel@xxxxxxxxxxxxxxx, Ben Myers <bpm@xxxxxxx>, xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <20131027090409.GA2797@dastard>
References: <1382585110-1796-1-git-send-email-david@xxxxxxxxxxxxx> <20131024084803.GA28144@xxxxxxxxxxxxx> <20131024103751.GS2797@dastard> <20131024154220.GA19055@xxxxxxxxxxxxx> <20131024164115.GU1935@xxxxxxx> <20131024212448.GW2797@dastard> <20131025112934.GA1585@xxxxxxxxxxxxx> <20131027090409.GA2797@dastard>
User-agent: Mutt/1.5.21 (2010-09-15)
Thanks for the explanation Dave.  I suspected a generic bug because
the two places mentioned are generic code.  Turns out only XFS and some
cluster filesystems are affected because other filesystems do not call
those under filesystem locks or i_mutex.

But it seem like AOP_FLAG_NOFS is indeed pointless and we should remove it,
in XFS as part of this patch and everywhere else later after talking
to the maintainers for the few other places using it.

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