Le 12/10/2017 à 18:25, Bartlomiej Zolnierkiewicz a écrit :
[ added dri-devel ML to cc: ]
On Tuesday, September 12, 2017 07:39:30 AM Christophe JAILLET wrote:
If 'dmam_alloc_attrs()' fails, we must go through the error handling code,
as done elsewhere in this function. Otherwise, there is a resource leak.
Signed-off-by: Christophe JAILLET <christophe.jaillet@xxxxxxxxxx>
---
I'm also puzzled by the 'framebuffer_alloc()' call a few lines above.
'ret' is known to be 0 at this point. I guess that -ENOMEM should also be
returned.
Yes, moreover the "failed:" error path is incomplete (please take
a look at au1200fb_drv_remove() for comparison) and needs to be fixed.
Could you please take care of it?
I will try, but be indulgent :)
My understanding of the code is that they are several issues all related
to this error handling path (double free, missing memory release,
invalid irq release...)
I will try to sort it out, but my first tries will likely be incomplete
and/or broken.
CJ
--
To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
