linux-kernel - Re: Out of order read() completion and buffer filling beyond returned amount

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] [day] [month] [year] [list]

Message-ID: <YeZrWoQY/3dKZHfT@infradead.org>
Date:   Mon, 17 Jan 2022 23:25:14 -0800
From:   Christoph Hellwig <hch@...radead.org>
To:     Matthew Wilcox <willy@...radead.org>
Cc:     Linus Torvalds <torvalds@...ux-foundation.org>,
        David Howells <dhowells@...hat.com>,
        Alexander Viro <viro@...iv.linux.org.uk>,
        Anna Schumaker <anna.schumaker@...app.com>,
        Dave Wysochanski <dwysocha@...hat.com>,
        Dominique Martinet <asmadeus@...ewreck.org>,
        Jeff Layton <jlayton@...nel.org>,
        Latchesar Ionkov <lucho@...kov.net>,
        Marc Dionne <marc.dionne@...istor.com>,
        Omar Sandoval <osandov@...ndov.com>,
        Shyam Prasad N <nspmangalore@...il.com>,
        Steve French <sfrench@...ba.org>,
        Trond Myklebust <trondmy@...merspace.com>,
        Peter Zijlstra <peterz@...radead.org>,
        ceph-devel@...r.kernel.org, linux-afs@...ts.infradead.org,
        linux-cachefs@...hat.com, CIFS <linux-cifs@...r.kernel.org>,
        linux-fsdevel <linux-fsdevel@...r.kernel.org>,
        Linux-MM <linux-mm@...ck.org>,
        "open list:NFS, SUNRPC, AND..." <linux-nfs@...r.kernel.org>,
        v9fs-developer@...ts.sourceforge.net,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: Out of order read() completion and buffer filling beyond
 returned amount

On Mon, Jan 17, 2022 at 01:30:05PM +0000, Matthew Wilcox wrote:
> I think for DIO, you're sacrificing the entire buffer with any filesystem.
> If the underlying file is split across multiple drives, or is even
> just fragmented on a single drive, we'll submit multiple BIOs which
> will complete independently (even for SCSI which writes sequentially;
> never mind NVMe which can DMA blocks asynchronously).  It might be
> more apparent in a networking situation where errors are more common,
> but it's always been a possibility since Linux introduced DIO.

Yes.  Probably because of that we also never allow short reads or writes
due to I/O errrors but always fail the whole I/O.