On Fri, Oct 28, 2005 Andrew Morton wrote:It's not in our tree... This is what I can see in the MAINTAINERS file:
Alexey, please don't assume that everyone reads lkml.I Cc: the v4l people, maybe they can answer the question.
Johannes
Begin forwarded message:
Date: Sat, 29 Oct 2005 01:16:47 +0400
From: Alexey Dobriyan <adobriyan@xxxxxxxxx>
To: linux-kernel@xxxxxxxxxxxxxxx
Subject: zoran drivers: absense of locking?
I've tried to read random part of a tree and now scratching my head
with a question:
what protects the number and a list of registered codecs in
zoran drivers?
Example: drivers/media/video/zr36050.c:
/* amount of chips attached via this driver */
static int zr36050_codecs = 0;
Decremented in zr36050_unset().
Checked for maximum value, used and incremented in zr36050_setup().
[Assigment to 0 in zr36050_init_module is not needed. dprintk() in
zr36050_cleanup_module() should be converted to BUG_ON, so I'll ignore
them.]
zr36050_codecs
zr36050_unset() = struct videocodec::unset
zr36050_setup() = struct videocodec::setup
The only place where ->unset and ->setup methods are called is
drivers/media/video/videocodec.c:
zr36050_codecs
zr36050_unset()
videocodec_detach()
zr36050_setup()
videocodec_attach()
Both videocodec functions are exported.
No spinlocks or semaphores in sight.
Does anybody know what protects the list of registered codecs in zoran
drivers?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/