New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Lenovo Y570 missing dsm call #2
Comments
This is a known issue, if you're using nouveau, please try Bumblebee-Project/Bumblebee-old#149 (comment) and report back.
|
@SimonSheehan could you try the latest version? |
@Lekensteyn here is the info you wanted from Reporting bugs: http://pastebin.com/jb80Fyc1 (Lenovo Ideapad Y570) |
I have the same issue on IdeaPad Y570 with Ubuntu 11.10. I tried the 'known issue' link above but it is giving me a 404. Could someone please respond with the correct link? |
Commit 11ddd34 should have added support for the Lenovo Ideapad Y570. Could you try the module from develop? |
I tried the version from the test PPA, is there a newer one? |
Nothing here, Y570: http://pastebin.com/HnPMtyY9 |
I can enable and disable Nvidia card without any problem using acpi_call. Is it helpful for you? |
I found the issue, the commit checked for the NVIDIA DSM UUID where the Lenovo Y570 uses the new Optimus UUID. I'll fix it tomorrow. |
This should finally support machines like the Lenovo Y570. WMMX methods should be more reliable than direct DSM probing so let's try that before calling DSM directly.
Please test the new code from develop! On my machine (BTO P-BOOK 17CL34 a.k.a. Clevo B7130), the WMMX method works. |
Module loaded successfully :-), but seems PM doesn't work: http://pastebin.com/emrSCDP6 . |
@brutall Well, we've progressed then, at least the method is detected well. Could you join #bumblebee on Freenode IRC for faster feedback? |
Ugh... The method |
Is this fixed now? I'm trying to get the code from https://github.com/Bumblebee-Project/bbswitch. The latest I can find is 0.4.6 and I can't even load that version on my Y570. |
It's not fixed yet. I'll have to make a very hacky, specific workaround for this bug and am looking for a clean and compatible way to do this. |
On http://forums.lenovo.com/t5/IdeaPad-Y-and-U-series-Laptops/Y570-How-to-enable-Nvidia-Optimus-Technology/ta-p/479197 I see a switch. If this exists on the Y570 or Y470, please generate a new acpi_dump_info report as described in the README. (and mention whether it's a Y470 or Y570) |
This switch does exist on the Y470; however it is a spring loaded switch that always returns to the same position and toggles whether the LED next to it is on (either switchable or integrated only) but it only has any effect on the LED in Windows 7, so I think it might be a software switch. Unfortunately I cannot test right now as the LCD on mine just died and I have to send it in for repairs. |
Yes, it exists, but it's a software switch, I think. I mean it isn't a two-state switch: you can push/press it (actually you drag it to right and then it returns to left), so discrete card is enabled/disabled. There is also a LED indicating current state of Optimus. Switch/LED doesn't work outside of windows and this is the reason, why I think it's controlled by the software. In Ubuntu it lights all the time - no matter if you push the switch or disable discrete card using ascpi_call. |
I wasn't clear on whether this was confirmed on the Y570 or not, but yes it is on the Y570 as well. Behavior is the same as above. Would be great if it could be made to work in Linux! |
Yes its a 1 state switch. (Could be a button as well) It only works for sofware, The white led is always on in Linux *shows that GPU acceleration is enabled). No function for button yet In Linux Mint Lisa. |
I'm about to report this issue on the kernel bugtracker. The wrong handle is being detected (that is arguable btw...). |
@Lekensteyn |
Well, the solution to Lenovo's buggy BIOS likely fixes the VBIOS/ROM issues of at least the nouveau driver. I'm already looking for possible ways to get the code patched for these laptops. I'm available on #bumblebee |
I've created a bug report for this: https://bugzilla.kernel.org/show_bug.cgi?id=42696 |
If anyone finds out how to enable nvidia gpu (manually or with optimus) on y570, please let me know. I'm newb in linux, but thi but thi is the only thing that keeps me from porting over. |
for my y480 and @hadrons123 Y580 at least; yes |
My Y570 is working without any patches / hacks too :) |
fixed in 3.8.5 upstream kernel too. |
Confirm that 3.8.5 - stable is working without any patches / hacks and etc. on Lenovo IdeaPad Y570 |
Can we consider closing this issue now that it is fixed upstream? |
Sure, I have updated the nobody-READs-ME of the lenovo-hack branch already, so that should be informative enough. |
Guys, this patch caused an regression on #65. Can users of 3.8.5+, 3.9-rc1+ (and thus 3.10/3.11 are included) confirm that applying the following commits still work for you:
Without the first patch, I expect bbswitch to get broken again. However, with the second patch applied on top of the first, things should not break. Please confirm if it still works for you, otherwise v3.10/v3.11 will be a regression for you. |
both patches on top of my 3.10 and everything still works |
@TheSiege Thank you for confirming! |
@TheSiege, Could you test the following patch with commit c7d9ca9 which has been merged into v3.11-rc1? Check whether bbswitch works on your machine with them. commit c7d9ca9 link. |
can't test rc1 doesn't boot for me |
You also can test both patches on v3.10. Commit c7d9ca9 should be applied firstly. |
reverting the previous patches and applying the newly suggested ones; |
@TheSiege Great, Thanks for test. |
If you have time, please test https://bugzilla.kernel.org/attachment.cgi?id=107027&action=diff&format=raw and report any regressions (or none). I do not expect regressions though. |
everything keeps working (tested on 3.11-rc2) |
@TheSiege Thank you for reporting back! |
Affected user of this bug, can you please test the patch in https://bugzilla.kernel.org/show_bug.cgi?id=60829#c12? It can be applied on top of 3.10.10+ or 3.11-rc1+. It should not introduce a regression for your laptop, but we would like to have a confirmation. |
all is well
|
@Lekensteyn yes I indeed applied the patch sorry for the confusion; wasn't used to geting these warnings; but they happen on unpatched as well |
Thank you for confirming. |
Update 2013-03-16: Linux 3.9-rc1 finally fixed the bug, everyone using 3.9-rc1 or newer can drop the acpi-handle-hack. ~L
Update 2013-03-29: Linux 3.8.5 also got the fix, you do not need any patching yourself for that kernel and newer!
Those who are looking for a solution, look at #2 (comment)
Running Fedora 16
The text was updated successfully, but these errors were encountered: