linux-kernel - Re: [PATCH] netfs: If didn't read new data then abandon retry

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: <2890361.1734079302@warthog.procyon.org.uk>
Date: Fri, 13 Dec 2024 08:41:42 +0000
From: David Howells <dhowells@...hat.com>
To: Lizhi Xu <lizhi.xu@...driver.com>
Cc: dhowells@...hat.com, asmadeus@...ewreck.org, brauner@...nel.org,
    ericvh@...nel.org, linux-kernel@...r.kernel.org, linux-mm@...ck.org,
    linux_oss@...debyte.com, lucho@...kov.net,
    syzbot+1fc6f64c40a9d143cfb6@...kaller.appspotmail.com,
    syzkaller-bugs@...glegroups.com, v9fs@...ts.linux.dev
Subject: Re: [PATCH] netfs: If didn't read new data then abandon retry

Lizhi Xu <lizhi.xu@...driver.com> wrote:

> >      (1) Replace NETFS_SREQ_NO_PROGRESS with NETFS_SREQ_MADE_PROGRESS and make
> >          the filesystem set it if it managed to read or write at least one byte
> >          of data.  Clear this bit before issuing a subrequest.
> Will there be conflicts when reading and writing use the same flag to mark?

No, because, at the moment, a read done by a write (e.g. RMW with crypto) or a
write done by a read (e.g. writing just-read data to the cache) are handled
with an additional request structure since the set regions involved may differ
(RMW only needs read the unmodified ends for example).

David