linux-kernel - PG_locked vs PG

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: <c83b914b0707060257g31cd84ex4f4afc57d0bbd865@mail.gmail.com>
Date:	Fri, 6 Jul 2007 15:27:18 +0530
From:	"sridhar lanka" <slanka64@...il.com>
To:	linux-kernel@...r.kernel.org
Subject: PG_locked vs PG_writeback

I see that in 2.6 kernels, page lock (PG_locked) is cleared and
PG_writeback is set
when the page is submitted for i/o.  Refer __block_write_full_page().

However the header file says that

" During disk I/O, PG_locked is used. This bit is set before I/O and
 * reset when I/O completes. page_waitqueue(page) is a wait queue of all tasks
 * waiting for the I/O on this page to complete.
"

Is the comment correct?

Does this mean that a page can be written to (dirtied) while the same page
is being written to by the low level driver. From what i can see PG_locked is
only ensuring that 2 threads don't read the same page but does not gate writes
or flushes of the page (partially). On the other hand  PG_writeback is gating
another write if the same page is already being flushed.

appreciate help.
-
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/