On 2018-06-09 15:34, Andy Shevchenko wrote:
On Sat, Jun 9, 2018 at 7:58 PM, <okaya@xxxxxxxxxxxxxx> wrote:
On 2018-06-09 12:38, Anton Vasilyev wrote:
If rtsx_probe fails to allocate dev->chip, then NULL pointer
dereference occurs at rtsx_release_resources().
Patch adds checks chip on NULL before its dereference at
rtsx_release_resources and passing with dereference inside
rtsx_release_chip.
Found by Linux Driver Verification project (linuxtesting.org).
I think you should bail out if dev->chip is null rather than adding
conditiinals.
I'm wondering if it's false positive. At which circumstances that may
happen?
Only if dev->chip allocation fails. Code tries to cleanup prior
resources by calling clean_everything() function which ends up in
rtsx_release_resources()
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
