Bug 74831

Summary: Hibernate failing on my hardware
Product: Power Management Reporter: Fabian (faerbit)
Component: Hibernation/SuspendAssignee: Lan Tianyu (tianyu.lan)
Status: CLOSED DUPLICATE    
Severity: high CC: faerbit, tianyu.lan
Priority: P1    
Hardware: x86-64   
OS: Linux   
Kernel Version: 3.14 Subsystem:
Regression: No Bisected commit-id:
Attachments: PKGBUILD I use to build the kernel

Description Fabian 2014-04-25 17:00:59 UTC
Since the upgrade to kernel 3.14 I am unable to hibernate.
I am using Archlinux.
I am hibernating with "systemctl hibernate".
It always fails to freeze device 00:08:

journalctl tells me:
kernel: serial 00:08: disable failed
kernel: dpm_run_callback(): pnp_bus_freeze+0x0/0x20 returns -5
kernel: PM: Device 00:08 failed to freeze: error -5

So I did a kernel bisect and came up with this commit:
https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=202317a573b20d77a9abb7c16a3fd5b40cef3d9d

My mainboard is an ASUS M4N98TD Evo.
Comment 1 Lan Tianyu 2014-04-28 02:45:26 UTC
Could you check whether this bug is the same as bug 74371?
Comment 2 Fabian 2014-04-28 14:24:02 UTC
How do I do this? It seems to be caused by the same git commit.
Comment 3 Fabian 2014-05-01 17:20:33 UTC
Created attachment 134621 [details]
PKGBUILD I use to build the kernel

I am really unsure on how to proceed.

I tried this(https://bugzilla.kernel.org/attachment.cgi?id=134481) patch from the bug you linked. To do that I had to use the linux-pm repo, because otherwise the patch couldn't be applied. It seemed to work. To verify this was the solution I checked out the commit I bisected in the linux-pm repo and tried that. Good thing I did because strangely in the linux-pm repo this commit seemed to be working. I did some further testing to maybe do another git bisect. Because I have no idea what is happening I just post what seems to be working and what not:
[repo]/[tag]/[status]
linux-stable/v3.13/working
linux-stable/v3.14/not working
linux-stable/commit 202317a573b20d77a9abb7c16a3fd5b40cef3d9d/ not working
     (seems to be the first bad commit in the linux-stable repo; result of my git bisect)
linux-pm/commit 202317a573b20d77a9abb7c16a3fd5b40cef3d9d/working
linux-pm/pm+acpi-3.14-rc7/not working
linux-pm/pm+acpi-3.13-rc8/not working

This doesn't make any sense to me. Maybe it is worth talking about how I compile my kernel since the issue is eventually not linked to the code.
I use "make localmodconfig" to keep my compile time somewhat reasonable.
"make localmodconfig" asks a lot of question on what to enable and what not. I don't know how to answer them, so I just press enter until there are no more questions. Is this a bad idea?
I attached my PKGBUILD (I am using Archlinux) with which I build my kernel. The linux-pm.tar is just a tar of the linux-pm repo (I haven't found an easier way to do a git bisect). I don't think I did anything stupid there since it is based on a PKGBUILD many people use everyday. But since I have no idea what is going on I included this.

To conclude:
Please advice on how to properly do a git bisect to find the commit causing this bug:
The bug seems to be 100% reproducable with the 3.14-Arch kernels which are based on the tar downloads from kernel.org
How should I compile my kernel to rule out mistakes?
On which repo should I do my git bisect?

Last thought: Is a git bisect really a good idea, with the problem I am facing?
Comment 4 Lan Tianyu 2014-05-12 01:34:53 UTC
(In reply to Fabian from comment #3)
> Created attachment 134621 [details]
> PKGBUILD I use to build the kernel
> 
> I am really unsure on how to proceed.
> 
> I tried this(https://bugzilla.kernel.org/attachment.cgi?id=134481) patch
> from the bug you linked. To do that I had to use the linux-pm repo, because
> otherwise the patch couldn't be applied. It seemed to work.

The patch resolves your issue and so I think it's the same one as the bug 74371.

About the git bisect, you can reference the following link.
http://wiki.gentoo.org/wiki/Kernel_git-bisect

*** This bug has been marked as a duplicate of bug 74371 ***