linux-ext4 - Reads atomic with respect to writes?

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: <8DD4E915-014E-4D2A-AC59-1347A375C5AA@mindspring.com>
Date:	Mon, 11 Apr 2011 10:24:22 -0700
From:	Michael Ubell <ubell@...dspring.com>
To:	linux-ext4@...r.kernel.org
Subject: Reads atomic with respect to writes?

POSIX states in http://pubs.opengroup.org/onlinepubs/9699919799/functions/read.html
	I/O is intended to be atomic to ordinary files and pipes and FIFOs. Atomic means that all the bytes from a single operation that started out together end up together, without interleaving from other I/O operations.

Is this supported by the ext4 implementation?

To be more specific:  If I call the read() API with a nbyte value 65536 (at an offset of a multiple of 65536) and at the same time a write() call of 65536 bytes at the same offset is happening will ext4 guarantee that the read will see either the 65536 bytes in the file before the write operation started or the 65536 bytes in the file after the write completed?

Unix file systems have always had this semantic and there are applications which depend on it.

--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html