Ubuntu on MacBook Pro

[phico]

Originally Posted by Gendo I was wondering if anyone resolved rapido's issue. I am having the same issue with my mbp, and I am running the kernel phico mentions, and I have even tried the latest alsa drivers which should be patched to fix this, but I can still not get the speakers to output any audio. I am wondering if anyone who has this working is using a specific .asoundrc or asound.conf that might make a difference. Any information would be appreciated.

I did not had to install anything special ( or I did not notice it or I was lucky ...)
Anyway If I can help you sharing config of so, just ask.

I have a MBP 17 with Sigmatel STAC9221 A1 chipset
and I choosed HDA Intel (Alsa Mixer) device in my sound config

[Gendo]

Sure if you are using an asound.conf or .asoundrc that information might be helpful as well as what is in /var/lib/alsa/asound.state . I did not realize you had a 17 inch macbook pro, I am using a 15 inch that was released before the glossy and 17 inch models, so I am wondering if maybe something is different with the 17 inch models. From the info in patch_sigmatel.c from the alsa drivers the note on the macs mentions
/*
* Early 2006 Intel Macintoshes with STAC9220X5 codecs seem to have a
* funky external mute control using GPIO pins.
*/
So I guess it is possible they changed the hardware a bit on the 17 inch models.

[phico]

Originally Posted by Gendo Sure if you are using an asound.conf or .asoundrc that information might be helpful as well as what is in /var/lib/alsa/asound.state .

I have no asound.conf neither .asoudrc on my system.
asound.state is in attachment

Originally Posted by Gendo From the info in patch_sigmatel.c from the alsa drivers the note on the macs mentions /* * Early 2006 Intel Macintoshes with STAC9220X5 codecs seem to have a * funky external mute control using GPIO pins. */ So I guess it is possible they changed the hardware a bit on the 17 inch models.

Yes it might be. Did you check which chipset you have ?
I though that patch could help me also because, as said above, my
microphone is not yet working.

[sam666]

Guys,

I installed Ubuntu on My 15" 2.1Ghz macbook pro. I follwed instructions but I noticed the following:


- ATI Graphics is not fully accelerated.. some 2D displays are fast (moving windows, VMware display etc...), but resizing a terminal window or switching from one desktop to another lags considerably.

Did I miss something in the installation? Any ideas?

Should I use 8.26.18 Ati drivers or 8.25.18 ( I used the ones that matched xorg-ati-drivers..: 8.25.18... but tested 8.26.18 as well...)?

I read another post mentionning that to get full acceleration the kernel had to be recompiled without imacfb or something else like Frame buffer... Haven't tried yet. Does anyone know about this?

- Sound does not work, even with 2.6.16-25 686 (didn't investigate a lot yet...)

- Also, the CPU being multi-core wouldn't it be better to use a SMP kernel? are both core used with an non SMP kernel?

- hdparm -t gives me very bad statistics for my internal (and firewire) drive: something between 20-30 MB/s reading. I had much more under debian on a less powerfull Dell laptop.

- most hdparm parameters (DMA etc..) return an error...

Any suggestion?

Thanks,
Sam

[phico]

Originally Posted by sam666 - ATI Graphics is not fully accelerated.. the kernel had to be recompiled without imacfb or something else like Frame buffer...

I did not check that myself. It is true that some screensavers do not seem to rue very smootly.


imacfb is closely linked to efi. I recompiled the kernel 2.6.17 with Macbook Pro options with this config :
http://modular.math.washington.edu/m...nfig-2.6.16.16
and add imacfb but it did not changed anything excepted I lost sound ! A patch is needed
with 2.6.17 It has been posted on mactel-linux-user mailing list


If you do it please feedback.

For sound, It seems some MBP does not have the same chipset as
mine. That could explain. It works with 2.6.15-25-686

Originally Posted by sam666 both core used with an non SMP kernel?

My kernel is 2.6.15-25-686 #1 SMP PREEMPT Wed Jun 14 11:34:19 UTC 2006 i686 GNU/Linux
Both core are used

Originally Posted by sam666 - hdparm -t gives me very bad statistics for my internal (and firewire) drive: something between 20-30 MB/s reading. I had much more under debian on a less powerfull Dell laptop.

On my system I get this result on internal 7200rpm drive :
Timing buffered disk reads: 134 MB in 3.01 seconds = 44.55 MB/sec

[Gendo]

Here is what I did for ATI accelerration

http://wiki.cchtml.com/index.php/Ubu...allation_Guide

Here is for compiz and XGL

http://wiki.cchtml.com/index.php/Xgl-Compiz-Dapper

Good luck!

[phico]

Originally Posted by Gendo Here is what I did for ATI accelerration http://wiki.cchtml.com/index.php/Ubu...allation_Guide Here is for compiz and XGL http://wiki.cchtml.com/index.php/Xgl-Compiz-Dapper Good luck!

Thanks for those links, I added your contribution to the beginning of the thread.

Also I tried to configure the right click on the touchpad.
I managed to get is working but it freeze the touchpad after each click ..
If anyone could find out why .. that could be useful !

Install mouseemu (with synaptic or apt-get --install mouseemu)
Edit config file:

sudo vi /etc/default/mouseemu

Add that content :

MID_CLICK="-middle 56 272" # Alt + mouse click = middle click RIGHT_CLICK="-right 464 272" # Fn + mouse click = right click SCROLL="-scroll 87" # F11 + mouse movement TYPING_BLOCK="-typing-block 300" # block mouse for 300ms after a keypress

Restart mouseemu :

sudo /etc/init.d/mouseemu restart

[Gendo]

I have been do some research on the sound issue and I have found there does seem to be a difference between the 17" mbp and the 15" mbp. The 17" mbp seems to be more like the mini in the way it behaves which is why I think the speakers are working on the 17".

For example in windows to disable the internal speaker and have the headphone jack only on the 17" and mini you need to change the 00 REG_BINARY at HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\
Control\Class\{4D36E96C-E325-11CE-BFC1-08002BE10318} from 85 to 05, but this does not work on the 15" mbp and this needs to be changed from 85 to 80.

I have been poking around in the alsa code (patch_sigmatel.c) trying to add a switch that might toggle this, I have manged to get a switch to display but it really does not do anything as I am really not well versed in alsa's workings. I do think it may have to do with the two calls to stac922x_gpio_mute in stac92xx_init. That or for whatever reason the AUTO_PIN_XXXXs are not getting enumerated properly. However like I said I am not well versed in ALSA and how it works so I am pretty much just throwing darts.

I also found this table, I am not sure if it is pertinent or not, but just throwing it out there.

00,01
85 88 both
85 85 both
88 88 head phone
88 85 HEADphone
85 both
88 head

[rapido]

Gendo,

I have a little more information to possibly help with the research. Mostly, on which machines seem to work and which fail ( sound-wise ).

I have both a 15" mbp, and a 20" Intel Imac. I have installed Ubuntu on both, and have the same end result on both as far as sound is concerned. They will each play over the headphone jack only. I believe that both of these are considered "Early 2006" as far as firmware, etc. My thought is that if we find a solution for one of these machines, that both will work.

I have not yet confirmed any fully working audio examples of these machines.

But, I have heard of successes with the new Macbook ( http://desrt.mcmaster.ca/macbook.xhtml ) and the 17" mbp ( this thread ).

[Gendo]

rapido, I had seen that link earlier, and the patch referenced is what is merged into the latest dev alsa drivers at ftp://ftp.alsa-project.org/pub/drive....12rc1.tar.bz2 and similarly the ubuntu kernel has been patched with similar code.

The way I figure it the pin settings for the iMac and 15" mbp pro are the same (not working) , and the 17" mbp, mac mini, and regular macbook are the same (working).

I am trying to see if I can reverse the behavior in the alsa code with a toggle, so maybe it could be merged in. But so far I have not had much success other than making the line out and the headphones work for output without being toggled to and unable to be turned off.

I am going to post my results to the alsa devs (see bugtrack that had been started by someone else https://bugtrack.alsa-project.org/al...ew.php?id=2198) in a day or so since there are a few more things I am going to try before giving up.