Bug 106591
Summary: | Toshiba L750D - system doesn't recognise the battery - AMD A6 | ||
---|---|---|---|
Product: | ACPI | Reporter: | Bartosz (oliwa.bartosz) |
Component: | Power-Battery | Assignee: | Zhang Rui (rui.zhang) |
Status: | CLOSED WILL_NOT_FIX | ||
Severity: | high | CC: | aaron.lu, lenb, lv.zheng, rui.zhang |
Priority: | P1 | ||
Hardware: | x86-64 | ||
OS: | Linux | ||
Kernel Version: | 3.16.7-24-desktop | Subsystem: | |
Regression: | No | Bisected commit-id: | |
Attachments: |
Acpidump
dmesg Debug-patch-to-force-disabling-ACPI-4.0-battery |
Description
Bartosz
2015-10-25 21:35:12 UTC
acpidump please: # acpidump > acpidump.txt Created attachment 191291 [details]
Acpidump
Also making lighter/brighter or darker screen doesn't work to, either with fn keys or in the system. I feel it is conencted with battery issue, because settings are in the power management options. I can't even disassemble the table: $ iasl -d dsdt.dat Intel ACPI Component Architecture ASL+ Optimizing Compiler version 20150515-64 Copyright (c) 2000 - 2015 Intel Corporation Reading ACPI table from file dsdt.dat - Length 00070373 (0x0112E5) ACPI: DSDT 0x0000000000000000 0112E5 (v01 TOSQCI TOSQCI00 F0000000 ACPI 00040000) Acpi table [DSDT] successfully installed and loaded Pass 1 parse of [DSDT] Pass 2 parse of [DSDT] Parsing Deferred Opcodes (Methods/Buffers/Packages/Regions) Parsing completed ACPI Error: External method arg count mismatch _SB_.ALIB: Current 5, attempted 2 (20150515/dmextern-819) ACPI Error: External method arg count mismatch _SB_.ALIB: Current 5, attempted 6 (20150515/dmextern-819) ACPI Error: External method arg count mismatch _SB_.ALIB: Current 5, attempted 3 (20150515/dmextern-819) ACPI Error: External method arg count mismatch _SB_.ALIB: Current 5, attempted 2 (20150515/dmextern-819) ACPI Error: External method arg count mismatch _SB_.ALIB: Current 5, attempted 2 (20150515/dmextern-819) ACPI Error: External method arg count mismatch _SB_.ALIB: Current 5, attempted 2 (20150515/dmextern-819) ACPI Error: External method arg count mismatch _SB_.ALIB: Current 5, attempted 2 (20150515/dmextern-819) ACPI Error: External method arg count mismatch _SB_.ALIB: Current 5, attempted 2 (20150515/dmextern-819) Found 2 external control methods, reparsing with new information Pass 1 parse of [DSDT] Pass 2 parse of [DSDT] Parsing Deferred Opcodes (Methods/Buffers/Packages/Regions) Parsing completed Segmentation fault (core dumped) Maybe I need to upgrade the iasl tool shipped in Fedora 22... Upgraded to acpica-tools-20150818-2.fc22.x86_64 works now. I'll need Lv's help to take a look at the _BIX control method, as for the backlight issue, do you have any directory under /sys/class/backlight? Please also attach full dmesg after boot. Created attachment 191411 [details]
dmesg
https://bugzilla.kernel.org/show_bug.cgi?id=106591 bartosz@OpenSuse:/sys/class/backlight> pwd /sys/class/backlight bartosz@OpenSuse:/sys/class/backlight> ls radeon_bl0 toshiba (In reply to Bartosz from comment #7) > https://bugzilla.kernel.org/show_bug.cgi?id=106591 > > bartosz@OpenSuse:/sys/class/backlight> pwd > /sys/class/backlight > bartosz@OpenSuse:/sys/class/backlight> ls > radeon_bl0 toshiba Does manually change backlight work there? e.g. to manually change backlight using the radeon_bl0 interface, you can: # cd radeon_bl0 # cat max_brightness XXX # echo a_value_smaller_than_XXX > brightness see if backlight changes. Hi, thank you for your response. Your method works! The monitor went darker when I typed 100. And what about the other interface, i.e. toshiba? OpenSuse:/sys/class/backlight/toshiba # ls actual_brightness bl_power brightness device max_brightness power subsystem type uevent changing brightness in toshiba doesn't work, it is the same effect when I hit Fn + F6 or F7 (the dialog shows that brightness is changing in KDE but in fact nothing happens) Then I suppose you can add: acpi_backlight=native to kernel cmdline and see if the radeon interface is the only one left in /sys/class/backlight directory, then the user space should pick the correct interface to use. where, how to do it? Please google how to add kernel cmdline option, you are probably using grub2, add that to search keyword too. An example: http://askubuntu.com/questions/19486/how-do-i-add-a-kernel-boot-parameter Ok I did: 1. sudo nano /etc/default/grub 2. in file added a cmdline: GRUB_CMDLINE_LINUX_DEFAULT="quiet splash acpi_backlight=native" 3. saved and then: 4. sudo update-grub 5. sudo reboot 6. bartosz@bartosz-L750D:/sys/class/backlight$ ls radeon_bl0 toshiba nothing happens, results the same, fn + f6/f7 don't make any effect. It seems you are using an old kernel, the said cmdline option works with a v4.2 or above kernel. I'm using 3.19, so I have to wait for a new version of linux (kubuntu)? You can build a new kernel. If that's not an option, maybe you can try: acpi_backlight=video. This will, theoretically, disable the vendor(toshiba here) interface. But I'm afraid it will make the acpi_videoX interface appear. Anyway, worth a try :-) Lv, Any hint if the problem of _BIX is a firmware issue or ACPICA? I've made a Lubuntu 15.04 USB Drive and tried it live and could manage the brightness via function keys. But there were still problem with battery. [ 11.403369] ACPI Exception: AE_AML_PACKAGE_LIMIT, Index (0x000000011) is beyond end of object (length 0xD) (20140424/exoparg2-420) [ 11.403382] ACPI Error: Method parse/execution failed [\_SB_.BAT1.UBIX] (Node ffff8801fe05afb8), AE_AML_PACKAGE_LIMIT (20140424/psparse-536) [ 11.403391] ACPI Error: Method parse/execution failed [\_SB_.BAT1._BIX] (Node ffff8801fe05a0b8), AE_AML_PACKAGE_LIMIT (20140424/psparse-536) [ 11.403398] ACPI Exception: AE_AML_PACKAGE_LIMIT, Evaluating _BIX (20140424/battery-461) We can see that ACPI battery driver fails to probe the device because of broken _BIX control method. There are two ways to fix the battery issue. 1. push Toshiba to fix the BIOS for this platform. 2. Force disabling _BIX on this platform so that the ACPI battery driver can fall back to use _BIF, which should also work for you. Debug patch attached, but unfortunately I don't think this can be included in upstream kernel... IMO, this bug can be closed as it's not a kernel issue. Please feel free to reopen it if you still have any questions. Created attachment 198891 [details]
Debug-patch-to-force-disabling-ACPI-4.0-battery
Sorry I'm total beginner with this kind of issue? Can you write step by step what to do? Thanks. Okay. are you able to build a custom kernel? Say, download the kernel source and build the kernel by yourself. If yes, I can show you how to patch your custom kernel with the patch file attached in comment #23. If no, then I'm afraid I cannot help on this... |