xfs
[Top] [All Lists]

Re: Linux 2.4.4

To: Russell Cattelan <cattelan@xxxxxxxxxxx>
Subject: Re: Linux 2.4.4
From: Jens Axboe <axboe@xxxxxxx>
Date: Mon, 30 Apr 2001 22:47:44 +0200
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <3AEDC6AB.3A0FD947@xxxxxxxxxxx>; from cattelan@xxxxxxxxxxx on Mon, Apr 30, 2001 at 04:10:20PM -0400
References: <4.3.2.7.2.20010428140403.04271ec0@xxxxxxxxxxxxx> <3AEAE40A.8E940486@xxxxxxxxxxx> <4.3.2.7.2.20010428221754.04278bd8@xxxxxxxxxxxxx> <20010428230127.A1753@xxxxxxxxxxxxxxxxxxx> <20010430181324.A22446@xxxxxxxxxxxxxxxxxx> <3AED9741.57322D14@xxxxxxxxxxx> <3AEDC6AB.3A0FD947@xxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
On Mon, Apr 30 2001, Russell Cattelan wrote:
> Ok after thinking about the problem for a bit I realized what
> happened.  The kiobuf allocation routines switched from using slab
> allocated memory to vm allocated memory... (I'm not sure why this was
> done) but since XFS calls alloc_kiobuf from an interrupt thread, and
> since vmalloc cannot be called from and interrupt thread it was
> deadlocking a  processor.

If you ever called alloc_kiovec from interrupt context, then it was
buggy all along (hint: SLAB_KERNEL). Using slab or other doesn't matter.

-- 
Jens Axboe


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