Toshiba Satellite L10-101 Smart Battery System works with patched DSDT
Hi,
I just want to add this one to the list of working Smart Battery Systems with the DSDT patch from https://sourceforge.net/projects/sbs-linux/.
Are there still efforts for integrating something into the kernel to let those systems run out of box?
Thomas
-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl
Quoted 7 lines · Thomas Renninger
Hi,
I just want to add this one to the list of working Smart Battery Systems with the DSDT patch from https://sourceforge.net/projects/sbs-linux/.
Are there still efforts for integrating something into the kernel to let those systems run out of box?
That would be really interesting. The problem is that the ec-nospinlock patch in that project just works for kernels 2.6.10/11/12 but not for the current 2.6.13 kernel. I already tried to adapt it but i need a lot more knowledge of acpi/kernel hacking to achieve this.
To summarize it: The above patch lets you see your smart battery systems but depending on what kernel you are using you will lose interrupts which results in e.g. lost keypresses if the battery is read. If any of the kernel hackers could drop a note what is planned to solve these sbs-related issues it would make some people happy. The patch above is written by Rich Townsend who seem not to continue his project.
Signature · 3 lines
Best regards, Fred Thiele GnuPG-Key: 0x4F6BE238 <[email protected]>
Hi Fred,
Quoted 4 lines
That would be really interesting. The problem is that the ec-nospinlock patch in that project just works for kernels 2.6.10/11/12 but not for the current 2.6.13 kernel. I already tried to adapt it but i need a lot more knowledge of acpi/kernel hacking to achieve this.
have you tried to boot you machine with the kernel parameter ec_burst=1. This works on my machine. Up to kernel 2.6.11 I've used Rich Townsends spinlock patch. But afterwards i found a posting from Luming Yu in which he asked to try his patche which should solve this interrupt issue. And fortunately his patches found the way to the kernel-sources :-)
Perhaps you should have a look at /drivers/acpi/ec.c. If there is something like setup("ec_burst= ... you may give this a try.
regards and good luck
Olaf
-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl
Quoted 7 lines · Olaf Jansen-Olliges
Hi Fred,
Quoted 4 lines
That would be really interesting. The problem is that the ec-nospinlock patch in that project just works for kernels 2.6.10/11/12 but not for the current 2.6.13 kernel. I already tried to adapt it but i need a lot more knowledge of acpi/kernel hacking to achieve this.
Quoted 7 lines
have you tried to boot you machine with the kernel parameter ec_burst=1. This works on my machine. Up to kernel 2.6.11 I've used Rich Townsends spinlock patch. But afterwards i found a posting from Luming Yu in which he asked to try his patche which should solve this interrupt issue. And fortunately his patches found the way to the kernel-sources :-)
Perhaps you should have a look at /drivers/acpi/ec.c. If there is something like setup("ec_burst= ... you may give this a try.
Ahh great. This is my log output:
Oct 19 20:39:40 sleipnir Kernel command line: root=/dev/hda7 resume=/dev/hda5 acpi_os_name="Microsoft Windows XP" ec_burst=1 lapic Oct 19 20:39:40 sleipnir ACPI: EC burst mode.
Can someone tell me what is the aim of the EC burst mode, what does it do?
Signature · 3 lines
Best regards, Fred Thiele GnuPG-Key: 0x4F6BE238 <[email protected]>