On 08/22/2011 10:25 AM, Frank Mehnert wrote:
Larry,
On Sunday 14 August 2011 03:39:52 Larry Finger wrote:
Kmemleak shows the following leak in the VB kernel module for version 4.1.0
r73009 (x86_64):
unreferenced object 0xffff88006baf1d00 (size 128):
comm "VirtualBox", pid 549, jiffies 4314134921 (age 30845.876s)
hex dump (first 32 bytes):
19 01 09 18 00 00 00 80 58 00 00 00 58 00 00 00 ........X...X...
03 01 92 18 0f 32 9c 65 2e 2e 00 00 ad 4e ad de .....2.e.....N..
backtrace:
[<ffffffff81122d77>] create_object+0x127/0x2b0
[<ffffffff813580b1>] kmemleak_alloc+0x21/0x50
[<ffffffff8111cd33>] __kmalloc+0x113/0x1c0
[<ffffffffa04c21ca>] rtR0MemAllocEx+0x5a/0x160 [vboxdrv]
[<ffffffffa04bff03>] VBoxHost_RTMemAllocTag+0x13/0x30 [vboxdrv]
[<ffffffffa04c4cfb>] VBoxHost_RTSemFastMutexCreate+0x1b/0x90 [vboxdrv]
[<ffffffffa04d7ff4>] g_abExecMemory+0x28f4/0x180000 [vboxdrv]
[<ffffffffa04eaef9>] g_abExecMemory+0x157f9/0x180000 [vboxdrv]
[<ffffffffa04bd619>] supdrvIOCtl+0x11b9/0x2820 [vboxdrv]
[<ffffffffa04b8435>] VBoxDrvLinuxIOCtl+0x135/0x200 [vboxdrv]
[<ffffffff81137de6>] do_vfs_ioctl+0x96/0x550
[<ffffffff811382ea>] sys_ioctl+0x4a/0x80
[<ffffffff8136befb>] system_call_fastpath+0x16/0x1b
[<ffffffffffffffff>] 0xffffffffffffffff
I have not yet tried to find the source of the leak.
is this really a leak? I see similar kmemleak complaints if a VM is started
but these complaints vanish as soon as the VM is terminated. Also, I also
don't see that the number of the complaints is growing.
It very well may be a false positive. I'm running kmemleak to debug various
wireless drivers and noted a couple of those in the logs. If it disappears after
VB is closed, then kmemleak just couldn't find the reference to it. I'll keep an
eye on it and report again if the leak is still there after closing VB.
Larry
_______________________________________________
vbox-dev mailing list
[email protected]
http://vbox.innotek.de/mailman/listinfo/vbox-dev
