[2.6.32] Please cherry-pick Toshiba Satellite DSDT corruption workaround
A message that is not in the archive (sent off-list or not yet received): 20111118200908.3274.86645.reportbug@pingwin
Hi Greg,
Mirosław Zalewski wrote[1]:
Quoted 2 lines
When kernel is invoked with ACPI support, it has various problems with USB detection
[...]
Quoted 4 lines
When finally booted, at least wireless driver (ath9k)
does not work. I heard that sound does not work as well, but haven't tested it myself. With ACPI enabled, kernel is unable to shutdown computer properly.
[...]
Quoted 8 lines
This is known flaw of Toshiba Satellite C650/kernel. You can find various posts around the net about it.
Toshiba Satellite C650 comes in four variants:
PSC08E
PSC10E
PSC12E
PSC14E
[...]
Quoted 3 lines
In July there was a BIOS update for 12E and 14E. The new BIOS is version 1.5 and it is said to fix ACPI issues on Linux [3]. Unfortunetly there is no new BIOS for 08E and 10E machines.
Mirosław confirmed that cherry-picking the following six patches on top of your linux-2.6.32.y tree brings his machine back to sanity:
729df0f848da (ACPICA: Add detection of corrupted/replaced DSDT, 2010-04-01) 69ec87efa815 (ACPICA: Add subsystem option to force copy of DSDT to local
memory, 2010-04-01)
43323cb4c4b6 (ACPICA: Update DSDT copy/detection, 2010-04-07) aa2110cb1a75 (ACPI: add boot option acpi=copy_dsdt to fix corrupt DSDT,
2010-04-08)
cce4f632db20 (ACPI: fix early DSDT dmi check warnings on ia64,
2010-05-12)
100cf87788c0 (ACPI: invoke DSDT corruption workaround on all Toshiba
Satellite, 2010-09-28)
Patches 1-5/6 have been in Debian squeeze since last September and seem to work well. Patch 6/6 turns on acpi=copy_dsdt by default for additional machines that need it, including Mirosław's.
The DMI pattern in 6/6 overreaches a little in my opinion --- it would be better to look at a BIOS version number or something instead. But it is what was done in mainline, and false positives should be harmless. I have tested on my PSC16U (which has BIOS 1.60 and doesn't need the workaround) that I get
TOSHIBA Satellite detected - force copy of DSDT to local memory
and no other noticeable side-effect.
The patches apply cleanly to linux-2.6.32.y except for patch 2/6, which has a boring textual conflict with the inverse of 4cdf1a5 (and the obvious conflict resolution works).
Patches 1-5/6 hit mainline in v2.6.35-rc1~477. Patch 6/6 is v2.6.36-rc7~25^2^5.
Anyway, I think these make good candidates for application to the 2.6.32.y longterm tree. What do you think?
Regards,
Jonathan
[1] http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=55;bug=598104
Quoted 1 line
[3] - http://ubuntuforums.org/showthread.php?p=9622930#post9622930
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 39 lines · Jonathan Nieder
Hi Greg,
Mirosław Zalewski wrote[1]:
Quoted 2 lines
When kernel is invoked with ACPI support, it has various problems with USB detection
[...]
Quoted 4 lines
When finally booted, at least wireless driver (ath9k)
does not work. I heard that sound does not work as well, but haven't tested it myself. With ACPI enabled, kernel is unable to shutdown computer properly.[...]
Quoted 8 lines
This is known flaw of Toshiba Satellite C650/kernel. You can find various posts around the net about it.
Toshiba Satellite C650 comes in four variants:
PSC08E
PSC10E
PSC12E
PSC14E[...]
Quoted 3 lines
In July there was a BIOS update for 12E and 14E. The new BIOS is version 1.5 and it is said to fix ACPI issues on Linux [3]. Unfortunetly there is no new BIOS for 08E and 10E machines.
Mirosław confirmed that cherry-picking the following six patches on top of your linux-2.6.32.y tree brings his machine back to sanity:
729df0f848da(ACPICA: Add detection of corrupted/replaced DSDT, 2010-04-01)69ec87efa815(ACPICA: Add subsystem option to force copy of DSDT to local
memory, 2010-04-01)
43323cb4c4b6(ACPICA: Update DSDT copy/detection, 2010-04-07)aa2110cb1a75(ACPI: add boot option acpi=copy_dsdt to fix corrupt DSDT,
2010-04-08)
cce4f632db20(ACPI: fix early DSDT dmi check warnings on ia64,
2010-05-12)
100cf87788c0(ACPI: invoke DSDT corruption workaround on all Toshiba
Satellite, 2010-09-28)
Quoted 3 lines
Patches 1-5/6 have been in Debian squeeze since last September and seem to work well. Patch 6/6 turns on acpi=copy_dsdt by default for additional machines that need it, including Mirosław's.
These all do not look like the match up with the stable_kernel_rules.txt file, do they?
Because of that, I can't take these patches, sorry.
greg k-h
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 2 lines · Greg KH
These all do not look like the match up with the stable_kernel_rules.txt file, do they?
I assume you mean because they are too big / nontrivial? If so, then that makes sense.
Thanks for explaining.
Jonathan