Bug 48001

Summary:	vaio z1 stalls on reboot (regression between 3.4 and 3.5)
Product:	ACPI	Reporter:	Michael Ketslah (thezogg)
Component:	Power-Off	Assignee:	Aaron Lu (aaron.lu)
Status:	CLOSED PATCH_ALREADY_AVAILABLE
Severity:	normal	CC:	aaron.lu, alan, john
Priority:	P1
Hardware:	All
OS:	Linux
Kernel Version:	3.11	Subsystem:
Regression:	Yes	Bisected commit-id:

Description Michael Ketslah 2012-09-27 12:48:50 UTC

After update to 3.5.0 and version above i can turn off laptop with poweroff or power button, but when i try to reboot (via command line) it stalls at very end.
And freezes, as well as i can hear FAN is going mad and laptop is getting pretty warm. I think it might be related to https://bugzilla.kernel.org/show_bug.cgi?id=45281 as i think at 3.5.0 somy-modules were updated, though i don't have problems described as there and as well i have those builtin and not built as modules:

$grep -i sony /usr/src/linux/.config
CONFIG_IR_SONY_DECODER=y
CONFIG_HID_SONY=y
CONFIG_SONY_LAPTOP=y
CONFIG_SONYPI_COMPAT=y

Any other information i can provide?

P.S. Wasn't sure with category as i'm not 100% sure what is the problem

Comment 1 Alan 2012-09-30 13:25:02 UTC

Can you build a kernel without the sony modules and see if it then behaves, and work back until you've got a minimal kernel or found it.  I don't have any idea either so the best beginning is probably to start trying to simplify the causes to suspect.

Comment 2 Michael Ketslah 2012-10-07 13:41:26 UTC

Sorry for delay. I tried and the result is the same. But as i left it like that for about a half minute, minute it finished restarting. But i had no such performance on 3.4.x it restarted normally, so it's not due software changes, but some changes in kernel. Btw 3.6.0 is the same result.

Comment 3 John Hughes 2014-03-07 10:31:47 UTC

Bug still present as of 3.11

Comment 4 Aaron Lu 2014-03-12 06:04:48 UTC

Since this is a regression, please use git bisect to find the offending commit.

Comment 5 John Hughes 2014-03-12 11:47:44 UTC

Oh how boring.

<mario>Here we go!</mario>

john@russia:/usr/local/src/linux$ git bisect start
john@russia:/usr/local/src/linux$ git bisect bad v3.5
john@russia:/usr/local/src/linux$ git bisect good v3.4
Bisecting: 6023 revisions left to test after this (roughly 13 steps)
[59cd358a7a5b2f6b61faa01dae6cfda3830ac62a] Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent

Comment 6 John Hughes 2014-03-12 14:43:41 UTC

Since this bug doesn't exist in 3.14.9-rc6 I wonder if it's worth the effort of doing the git bisect?

(I feel like such a dummy for not testing that first).

Comment 7 Aaron Lu 2014-03-13 01:21:29 UTC

If there is no such problem on latest upstream kernel, the bisect is not necessary. I'll close the bug.