linux-kernel - Block I/O layer semantics

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

Message-ID: <AANLkTilYqkiOhcaq7iuDJ8pP5wDCj-rmj1VLuRCsuQ6v@mail.gmail.com>
Date:	Tue, 15 Jun 2010 15:30:30 +0300
From:	Thanos Makatos <tmakatos@...il.com>
To:	linux-kernel@...r.kernel.org
Subject: Block I/O layer semantics

Hello.

I want to learn whether there are exact semantics offered by the Linux
block I/O layer. Specifically, I would like to know the following:
1. Can a write request bypass a read request (both in flight) that
access the same sector number? Does the I/O scheduler has anything to
do with it? I suppose it is legal since both these requests are
concurrent, however I can't find it anywhere clearly documented. And,
as a consequence:
2. A read request must return the contents of the last completed
write, right? (Obviously this shouldn't happen, still need a
confirmation, through..)

Thank you.

P.S: Please CC me on your reply - I haven't subscribed to the list.

-- 
Thanos Makatos
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/