linux-kernel - Re: [PATCH 2/3] vfs: Add a super block operation to check for writeback errors

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <6c60d0e19af9a820aae15cd3477da561115c9852.camel@kernel.org>
Date:   Wed, 23 Dec 2020 07:44:10 -0500
From:   Jeff Layton <jlayton@...nel.org>
To:     Vivek Goyal <vgoyal@...hat.com>,
        Matthew Wilcox <willy@...radead.org>
Cc:     linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-unionfs@...r.kernel.org, amir73il@...il.com,
        sargun@...gun.me, miklos@...redi.hu, jack@...e.cz, neilb@...e.com,
        viro@...iv.linux.org.uk, hch@....de
Subject: Re: [PATCH 2/3] vfs: Add a super block operation to check for
 writeback errors

On Tue, 2020-12-22 at 11:25 -0500, Vivek Goyal wrote:
> On Tue, Dec 22, 2020 at 04:19:00PM +0000, Matthew Wilcox wrote:
> > On Mon, Dec 21, 2020 at 02:50:54PM -0500, Vivek Goyal wrote:
> > > -	ret2 = errseq_check_and_advance(&sb->s_wb_err, &f.file->f_sb_err);
> > > +	if (sb->s_op->errseq_check_advance)
> > > +		ret2 = sb->s_op->errseq_check_advance(sb, f.file);
> > 
> > What a terrible name for an fs operation.  You don't seem to be able
> > to distinguish between semantics and implementation.  How about
> > check_error()?
> 
> check_error() sounds better. I was not very happy with the name either.
> Thought of starting with something.
> 

Maybe report_error() ?

The same error won't be reported on the next call on the same fd. I
think the important point to make here is that this error must be
reported to syncfs() or something like it once you call this.

(In hindsight, I sort of wish I had done s/serrseq_set/errseq_record/
and s/errseq_check_and_advance/errseq_report/ when I initially did this,
if only to make the API a little less dependent on the implementation.)
-- 
Jeff Layton <jlayton@...nel.org>