On Thu, Jul 25, 2013 at 01:18:31PM +0100, Catalin Marinas wrote:
On 21 July 2013 16:11, Alexandra N. Kossovsky
[off-list ref] wrote:
quoted
I am running linux-3.10.0 with kmemleak and see following warnings
in /sys/kernel/debug/kmemleak:
unreferenced object 0xffff880216fcfe00 (size 512):
comm "swapper/0", pid 1, jiffies 4294895429 (age 1415.320s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 aa aa aa aa aa aa aa aa ................
55 55 55 55 55 55 55 55 ff ff ff ff ff ff ff ff UUUUUUUU........
backtrace:
[<ffffffff813e415c>] kmemleak_alloc+0x21/0x3e
[<ffffffff8111c17f>] kmemleak_alloc_recursive.constprop.57+0x16/0x18
[<ffffffff8111e63b>] __kmalloc+0xf9/0x144
[<ffffffff8123d9cf>] fb_alloc_cmap_gfp+0x47/0xe1
[<ffffffff8123da77>] fb_alloc_cmap+0xe/0x10
[<ffffffff81aff40a>] efifb_probe+0x3e9/0x48f
[<ffffffff812c566f>] platform_drv_probe+0x34/0x5e
[<ffffffff812c3e6d>] driver_probe_device+0x98/0x1b4
[<ffffffff812c3fd7>] __driver_attach+0x4e/0x6f
[<ffffffff812c25bf>] bus_for_each_dev+0x57/0x8a
[<ffffffff812c3984>] driver_attach+0x19/0x1b
[<ffffffff812c362b>] bus_add_driver+0xde/0x201
[<ffffffff812c453f>] driver_register+0x8c/0x110
[<ffffffff812c510d>] platform_driver_register+0x41/0x43
[<ffffffff812c5127>] platform_driver_probe+0x18/0x8a
[<ffffffff81aff002>] efifb_init+0x276/0x295
Does the efifb driver has any way to deallocate the cmap? I don't see
any explicit call to fb_dealloc_cmap() apart from the error handling.
My theory is that the efifb driver gets deregistered via
do_remove_conflicting_framebuffers(). I'm not familiar with this code,
just commenting from a kmemleak perspective.
You're right, that's the typical deregister path. I'll follow up with a
patch to test.
--
Peter