xfs
[Top] [All Lists]

Re: [PATCH] xfstests: fix async io error handling in fsx

To: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Subject: Re: [PATCH] xfstests: fix async io error handling in fsx
From: Felix Blyakher <felixb@xxxxxxx>
Date: Mon, 30 Mar 2009 13:59:38 -0500
Cc: xfs@xxxxxxxxxxx
In-reply-to: <20090330185624.GA21753@xxxxxxxxxxxxx>
References: <1238391050-14211-1-git-send-email-felixb@xxxxxxx> <20090330175451.GA1041@xxxxxxxxxxxxx> <2466E954-2122-4FEB-B3DA-8F06AA22BAB7@xxxxxxx> <20090330185624.GA21753@xxxxxxxxxxxxx>

On Mar 30, 2009, at 1:56 PM, Christoph Hellwig wrote:

On Mon, Mar 30, 2009 at 01:53:21PM -0500, Felix Blyakher wrote:
+               /*
+                * The b0rked libaio defines event.res as signed.
+                * However the kernel strucuture has it unsigned,
+                * and it's used to pass negated error value.
+                * Till the library is fixed use the temp var.
+                */

This comment seems backwards to the patch description and the actual
code.

Hmm, not really.
If the libaio library would be correct (do you know whom to talk
to and where to send a patch), the code would look like this:

Yeah, but shouldn't the comment read:

        /*
         * The b0rked libaio defines event.res as *unsigned*.
         * However the kernel strucuture has it *signed*,

Ohh, I was blind twice.
Sure, you're right. Thanks for pointing it out.

Felix


         * and it's used to pass negated error value.
         * Till the library is fixed use the temp var.
         */

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