On Wed, Feb 11, 2015 at 6:43 AM, Greg Kroah-Hartman
[off-list ref] wrote:

On Tue, Feb 10, 2015 at 12:22:41PM +0400, Konstantin Khlebnikov wrote:

quoted

I've found regression:

[  257.139907] ================================================
[  257.139909] [ BUG: lock held when returning to user space! ]
[  257.139912] 3.18.6-debug+ #161 Tainted: G     U
[  257.139914] ------------------------------------------------
[  257.139916] python/22843 is leaving the kernel with locks still held!
[  257.139918] 1 lock held by python/22843:
[  257.139920]  #0:  (&mm->mmap_sem){++++++}, at: [<ffffffff8104e4c2>]
__do_page_fault+0x162/0x570

upstream commit 7fb08eca45270d0ae86e1ad9d39c40b7a55d0190 must be backported too.

Ah, nice, I missed that one.  How did you test this?

I've catched hang on mmap_sem in some python self-test inside exherbo chroot.
With that patch test has finished successfully.

It seems the only way to tigger this is stack-overflow: for now VM_FAULT_SIGSEGV
is returned only if kernel cannot add guard page when stack expands.

thanks,

greg k-h

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>