Re: [BISECTED, REGRESSION] Successful resume from suspend but freezes after I/O

[Yinghai Lu <yinghai@kernel.org>]

Volker Lanz wrote:
> On Monday 07 December 2009 19:24:02 Yinghai Lu wrote:
>> Volker Lanz wrote:
>>> Hi,
>>>
>>> updating to my distro's new 2.6.31 kernel on an x86_64 quad core machine
>>> with 6 GB of RAM I noticed resuming from suspend still worked as before,
>>> but the machine will now reproducably freeze (have to hard reset)
>>> afterwards as soon as I do something disk I/O heavy, though the problem
>>> is probably not related to disk activity at all.
>>>
>>> A current mainline 2.6.32 checkout shows the same behaviour.
>>>
>>> I git-bisected the problem to this commit:
>>>
>>>
>>> -------------------------------------------------------------------------
>>> ---- commit 78a8b35bc7abf8b8333d6f625e08c0f7cc1c3742
>>> Author: Yinghai Lu <yinghai@kernel.org>
>>> Date:   Thu Mar 12 22:36:01 2009 -0700
>>>
>>>     x86: make e820_update_range() handle small range update
>>>
>>>     Impact: enhance e820 code to handle more cases
>>>
>>>     Try to handle new range which could be covered by one entry.
>>>
>>>     Signed-off-by: Yinghai Lu <yinghai@kernel.org>
>>>     Cc: jbeulich@novell.com
>>>     LKML-Reference: <49B9F0C1.10402@kernel.org>
>>>     Signed-off-by: Ingo Molnar <mingo@elte.hu>
>>> -------------------------------------------------------------------------
>>> ----
>>>
>>>
>>> A kernel built from this revision does not boot, so the first booting
>>> kernel to show the problem actually seems to be:
>>>
>>>
>>> -------------------------------------------------------------------------
>>> ---- commit 6d7942dc2a70a7e74c352107b150265602671588
>>> Author: Yinghai Lu <yinghai@kernel.org>
>>> Date:   Sat Mar 14 14:32:41 2009 -0700
>>>
>>>     x86: fix 64k corruption-check
>>>
>>>     Impact: fix boot crash
>>>
>>>     Need to exit early if the addr is far above 64k.
>>>
>>>     The crash got exposed by:
>>>
>>>       78a8b35: x86: make e820_update_range() handle small range update
>>>
>>>     Signed-off-by: Yinghai Lu <yinghai@kernel.org>
>>>     Cc: <stable@kernel.org>
>>>     LKML-Reference: <49BC2279.2030101@kernel.org>
>>>     Signed-off-by: Ingo Molnar <mingo@elte.hu>
>>> -------------------------------------------------------------------------
>>> ----
>>>
>>>
>>> The last kernel to work without problems thus seems to be this one:
>>>
>>>
>>> -------------------------------------------------------------------------
>>> ---- commit 773e673de27297d07d852e7e9bfd1a695cae1da2
>>> Author: Yinghai Lu <yinghai@kernel.org>
>>> Date:   Thu Mar 12 21:35:18 2009 -0700
>>>
>>>     x86: fix e820_update_range()
>>>
>>>     Impact: fix left range size on head
>>>
>>>     | commit 5c0e6f035df983210e4d22213aed624ced502d3d
>>>     |    x86: fix code paths used by update_mptable
>>>     |    Impact: fix crashes under Xen due to unrobust e820 code
>>>
>>>     fixes one e820 bug, but introduces another bug.
>>>
>>>     Need to update size for left range at first in case it is header.
>>>
>>>     also add __e820_add_region take more parameter.
>>>
>>>     Signed-off-by: Yinghai Lu <yinghai@kernel.org>
>>>     Cc: jbeulich@novell.com
>>>     LKML-Reference: <49B9E286.502@kernel.org>
>>>     Signed-off-by: Ingo Molnar <mingo@elte.hu>
>>> -------------------------------------------------------------------------
>>> ----
>>>
>>>
>>> The problem is 100% reproducable on this machine: Resuming and then
>>> copying /usr/ to $HOME will freeze after a few hundred MB have been
>>> copied. Earlier kernels worked fine for the last couple of months.
>>>
>>> What additional information is required to help diagnose and hopefully
>>> fix the problem?
>> whole boot log with CONFIG_PCI_DEBUG and debug on command line.
> 
> Here it is. It's huge, I hope you were expecting that...

and the one with current tip?

http://people.redhat.com/mingo/tip.git/readme.txt

YH