I am using Mint X [with the Ubuntu kernel] on my R705. I just want to add here that I initially encountered no problems with any hardware, aside from the inaccessibility of Toshiba-specific keys. I don't use Bluetooth so I can't vouch for that, but i have yet to engage in any real troubleshooting on this machine.
I finally got to checking the logs and notice i'm getting an MCP error message that has been previously posted on, in addition to acpi errors at boot:
intel ips 0000:00:1f.6: MCP power or thermal limit exceeded
the error is regular but less common than it seems to have been for other users, but i'm still noticing a high idling frequency even when the error isn't being reported, with the chassis hot to the touch. The temp hasn't execeed 60C, but apparently this is still too hot. i encounter the error booting from both battery and AC, and it seems to occur mostly when the machine is idle. to summarize what i've read on this thread and elsewhere, this problem appears to be the result of the current 2.6.35 kernel being unable to run the i3/i5/i7 series processors in low power mode. the forthcoming .36 kernel does not appear to address this issue either. i've had no issues with crashing, freezing or fan operation, but i'm definitely concerned since the hardware wasn't designed to run with an inefficient CPU, and running the CPU beyond its thermal limit can reduce its lifespan or kill it.
Someone said they hadn't seen this MCP error on a 705, but I can assure you I'm seeing it. I have an i3, so i'm not sure if this makes a difference.
Just curious to know if anyone else on this thread is still having this problem, or if i possibly overlooked a solution. The only solutions I am seeing are:
1) disabling ips logs
2) rolling back the kernel
While the following solutions i have personally ruled out:
-bios update
-intel microcode update
From what i have read, ips alerts only disappear with a rollback to 2.6.32 because such errors weren't logged in the previous kernel, so basically both are only ignoring the issue, thus none of these offers any solution.
Looking into running toshset, but I'm not seeing any indication it will be effective at addressing this issue. Also wondering if manually setting the thermal limit might be an option. Either way, without a solution this issue is a border-line deal breaker for me, so perhaps it could be noted on the first post under a Power Management or CPU header. Either way, its too late for me, so I'll be searching tirelessly for a solution...
Last edited by sumwon; December 24th, 2010 at 09:58 PM. Reason: updating with info from further research
I tried patching my kernel (2.6.35-24-generic i686) with a toshset patch, as recommended by this bug post on launchpad.
Build fine, installation fine, but I still get:
Code:anydroid@tosh:~$ toshset required kernel toshiba support not enabled.
I just looked at the gnome-power-monitor while recharging my r630. It seems there is something wrong with the acpi-module, since the rate you get while recharging isn't the energy which goes into the battery, but the energy the system consumes. Since gnome-power-manager estimates the time to recharge by this rate, this has to be an error.
For example:
system idles with backlight low: rate = ~9W; time to recharge = ~7.5h
system idles with backlight full: rate = ~11W; time to recharge = ~5.5h
When I calculate the rate by looking at the battery charge i get in one hour, I get an average rate of 26W/h which looks quite correct.
You should check to see if you have another toshiba_acpi.ko somewhere, superseding the toshiba_acpi.ko you compiled. On my system, I had a /lib/modules/2.6.35-24-generic/kernel/drivers/platform/x86/toshiba_acpi.ko that was preventing the compiled one, which was placed in /lib/modules/2.6.35-24-generic/extra/toshiba_acpi.ko, from being loaded.
Bookmarks