On Sat, May 28, 2011 at 9:43 PM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
On Sat, 28 May 2011, Belisko Marek wrote:
On Sat, May 28, 2011 at 6:29 PM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
i agree that there should be a "put_device(&dev->dev);" statement as you show above. however, i still don't see how this can be just a stylistic improvement as you seem to suggest. based on the warning from the kernel source file, it would seem that you *must* do a put_device() in that situation -- it's not optional.
Sure you're right. You can send a patch to fix this problem. Good catch.
i didn't want to submit anything until i verified what correct code should look like. and it's not like that's the only example -- others are trivially easy to find, like this in drivers/media/video/bt8xx/bttv-gpio.c (line 97):
err = device_register(&sub->dev); if (0 != err) { kfree(sub); return err; } I'm little bit confused. If you look at device_add which is part of device_register if something in device_add failed it will always call put_device (in any error case) so why then is necessary to call it again when device_register return error? As you said on some places put_device is used on some not. Assume device_register in normal conditions don't fail so this code is not called anyway or?
that would seem to be incorrect as well, no?
rday
--
======================================================================== Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca
Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
regards, marek -- as simple and primitive as possible ------------------------------------------------- Marek Belisko - OPEN-NANDRA Freelance Developer Ruska Nova Ves 219 | Presov, 08005 Slovak Republic Tel: +421 915 052 184 skype: marekwhite icq: 290551086 web: http://open-nandra.com