Bug 11347
Summary: | last commit to include/generic-asm/rtc.h is making my machine to freeze | ||
---|---|---|---|
Product: | Timers | Reporter: | Gilles Grandou (gilles) |
Component: | Realtime Clock | Assignee: | Gilles Grandou (gilles) |
Status: | CLOSED PATCH_ALREADY_AVAILABLE | ||
Severity: | high | CC: | akpm, bunk, hwanderer, john.stultz |
Priority: | P1 | ||
Hardware: | All | ||
OS: | Linux | ||
Kernel Version: | 2.6.25, 2.6.26, 2.6.27-rc3 | Subsystem: | |
Regression: | Yes | Bisected commit-id: |
Description
Gilles Grandou
2008-08-15 02:24:52 UTC
There are many unspecific (kernel and debian) bug reports, which I think have the same reason. Something related to BIOS/RTC/HPET ?
Because there is nothing in the logs in time of freeze it is hard to give more information. After I use the Ingo Molnar's netconsole I got on a other computer the log message "rtc: lost 787 interrupts" before freezing.
The workaround "hpet=disable" works mostly.
From the ChangeLog-2.6.27-rc5:
>> commit 38c052f8cff1bd323ccfa968136a9556652ee420
>> Author: Ingo Molnar <mingo@elte.hu>
>> Date: Sat Aug 23 17:59:07 2008 +0200
>> rtc: fix deadlock
>> if get_rtc_time() is _ever_ called with IRQs off, we deadlock badly
>> in it, waiting for jiffies to increment.
>>
>> So make the code more robust by doing an explicit mdelay(20).
>> This solves a very hard to reproduce/debug hard lockup reported
>> by Mikael Pettersson.
Kernel 2.6.27-rc5 solved the freeze problem on my Thinkpad T60!
If others can confirm this solution, maybe the developers can backport this fix to 2.6.26.
Thanks a lot especially to Ingo Molnar
Helle
Note also the patch in bug #11153 comment #18 ... and that whole bug; the whole "HPET emulation" scheme for RTC interrupts is problematic. Of course, expecting ACPI interrupts to work is turning up problems too ... we might need to just hang the RTC irq handler directly off SCI (which is where the HPET "legacy replacement mode" re-routes it) rather than using ACPI mechanisms. (In reply to comment #1) > From the ChangeLog-2.6.27-rc5: > >> commit 38c052f8cff1bd323ccfa968136a9556652ee420 > >> Author: Ingo Molnar <mingo@elte.hu> > >> Date: Sat Aug 23 17:59:07 2008 +0200 > > Kernel 2.6.27-rc5 solved the freeze problem on my Thinkpad T60! > If others can confirm this solution, maybe the developers can backport this > fix to 2.6.26. I've applied the above patch over a stock 2.6.26, I have now +15h uptime without encountering any freeze on my T60p. Thanks for the hint! So it looks like this issue was resolved with 2.6.27, and this bug can be closed. |