Macbook Pro

Message

je_fro · Post by **je_fro** » Sat May 26, 2007 7:55 pm

I'm using 2.6.21-r3 from Pic0's overlay and everything is fine...

Pic0 · Post by **Pic0** » Mon May 28, 2007 4:59 pm

FYI, I have totally crashed my MacBook Pro (I had to format the HDD). Since I absolutly need a working laptop tomorrow, I took the opportunity to try Kubuntu.
So I will not update the overlay for a little while (at least a week, maybe more if I stick to Kubuntu). In the meantime (and in the future), if someone wants to update things, I can create one (or more) users on the SVN repository with write access. If you are interested, feel free to contact me by email <guill.p (@) gmail.com>.

BTW, madwifi-ng r2381 and above is broken (kernel panic when modprobing ath_pci). You should stay at r2380 if you want a working madwifi.

Regards,

synss · Post by **synss** » Fri Jun 01, 2007 3:29 am

Hello everybody, it seems you have made a nice job at getting everything to work on the MBP. But this has needed a 24+pages thread I am not sure I (and others) are willing to read. Could somebody be kind enough to sum up what actually works (64bit + wifi?, Ati card? Hibernation? etc.) and update the how to in the wiki?

http://gentoo-wiki.com/HARDWARE_Apple_MacBook

Thanks

stardotstar · Post by **stardotstar** » Mon Jun 04, 2007 5:47 am

Guys, I need to get what our company call an "air card" working somehow...

Since the standard offering here is ThinkPads with WinXP I do all my windows stuff in a vmware workstation guest. Works really well.

I believe I have the option of getting an ExpressCard/34 wireless broadband (aircard) for the NextG Telstra Network but doubt it would be supported in Linux or detected in Windows guest virtual machine. Should I consider the USB version then?

Anyone have any experience with the Macbook Pro ExpressCard slots and how they interact with gentoo?

Not knowing exactly how these little wireless beasts work I don't know how likely it is that it could be configured in linux... If it could I could NAT to the VM so that would be fine.

Other option I suppose is to use it in OSX which I think would be fine at a pinch.

genmich · Post by **genmich** » Tue Jun 05, 2007 12:59 pm

The new MBP are out: http://www.apple.com/macbookpro/

Now with NVIDIA graphic cards and "power-efficient" LED-backlit. Sounds like an easier to go linux MBP

stardotstar · Post by **stardotstar** » Tue Jun 05, 2007 1:08 pm

NVIDIA!
1920-by-1200

someone's listening eh?

in every other respect I am still content with mine

stardotstar · Post by **stardotstar** » Wed Jun 06, 2007 12:01 pm

I have just dived in and done a kernel.org 2.6.21.3 install and various module and dependent package rebuilds (used the necessary patches from mactellinux) and although the system is running various things are slow and some hang - beryl won't work anymore (may be something to do with the previous emerging though what with the latest ati drivers and what not)

after finally getting vmware modules to config and load safely though I find that my cdrom has gone awol...

Can't make sense of the comparisons between the configs of 2.6.17.11 and 2.6.21.3 and simply can't grok why /dev/hda has disappeared...

I have done some searching but it is too late and after too much compiling and rebooting to press on........ if anyone can hint me I would most appreciate it... perhaps 2.6.21.3 is a BAD THING ... but then again I had to move ahead sooner or later

tia

Will

je_fro · Post by **je_fro** » Wed Jun 06, 2007 5:12 pm

I'm slowly working on 2.6.21.3...
For now I'm using 2.6.21.1 and the mactel patchset -r 124. It's working fine.

stardotstar · Post by **stardotstar** » Thu Jun 07, 2007 10:35 am

je-fro, may I compare my current 2.6.21.3 .config with your work in progress?

I have most things working - learned some hard lessons along the way with this build - but on the whole I think the basics are there.

My biggest prolem is that my CD Rom is AWOL and this is weird because the sections that pertain to it (or at least those I readily identify with) are the same as 2.6.17.11 - but then again this is a bit of a jump as I have found when diffing the two configs.

From what I can see the cd rom is found but won't assign as a drive in /dev/hda

the message I get is:

Code: Select all

 unable to mount the selected volume; mount:  special device /dev/hda/ does not exist

I have been toying with the kernel configs to try and get all the required support sussed but I am not getting anywhere...

...some logs first:

Code: Select all

lshal yields
udi = '/org/freedesktop/Hal/devices/pci_8086_27df_scsi_host_scsi_device_lun0'
  info.bus = 'scsi'  (string)
  info.parent = '/org/freedesktop/Hal/devices/pci_8086_27df_scsi_host'  (string)
  info.product = 'SCSI Device'  (string)
  info.subsystem = 'scsi'  (string)
  info.udi = '/org/freedesktop/Hal/devices/pci_8086_27df_scsi_host_scsi_device_lun0'  (string)
  linux.hotplug_type = 2  (0x2)  (int)
  linux.subsystem = 'scsi'  (string)
  linux.sysfs_path = '/sys/devices/pci0000:00/0000:00:1f.1/host0/target0:0:0/0:0:0:0'  (string)
  scsi.bus = 0  (0x0)  (int)
  scsi.host = 0  (0x0)  (int)
  scsi.lun = 0  (0x0)  (int)
  scsi.model = 'DVD-R   UJ-846'  (string)
  scsi.target = 0  (0x0)  (int)
  scsi.type = 'cdrom'  (string)
  scsi.vendor = 'MATSHITA'  (string)

and in messages from boot

Code: Select all

Jun  7 20:20:29 spitfire ata_piix 0000:00:1f.1: version 2.10ac1
Jun  7 20:20:29 spitfire ACPI: PCI Interrupt 0000:00:1f.1[A] -> GSI 18 (level, low) -> IRQ 18
Jun  7 20:20:29 spitfire PCI: Setting latency timer of device 0000:00:1f.1 to 64
Jun  7 20:20:29 spitfire ata1: PATA max UDMA/133 cmd 0x000101f0 ctl 0x000103f6 bmdma 0x000140c0 irq 14
Jun  7 20:20:29 spitfire ata2: PATA max UDMA/133 cmd 0x00010170 ctl 0x00010376 bmdma 0x000140c8 irq 15
Jun  7 20:20:29 spitfire scsi0 : ata_piix
Jun  7 20:20:29 spitfire ata1.00: ATAPI, max UDMA/66
Jun  7 20:20:29 spitfire Clocksource tsc unstable (delta = -213926640 ns)
Jun  7 20:20:29 spitfire ata1.00: configured for UDMA/66
Jun  7 20:20:29 spitfire scsi1 : ata_piix
Jun  7 20:20:29 spitfire ATA: abnormal status 0x7F on port 0x00010177
Jun  7 20:20:29 spitfire scsi 0:0:0:0: CD-ROM            MATSHITA DVD-R   UJ-846   FQ3T PQ: 0 ANSI: 5
Jun  7 20:20:29 spitfire scsi 0:0:0:0: Attached scsi generic sg0 type 5
Jun  7 20:20:29 spitfire ata_piix 0000:00:1f.2: MAP [ P0 P2 XX XX ]
Jun  7 20:20:29 spitfire ata_piix 0000:00:1f.2: invalid MAP value 0
Jun  7 20:20:29 spitfire PCI: Enabling device 0000:00:1f.2 (0005 -> 0007)
Jun  7 20:20:29 spitfire ACPI: PCI Interrupt 0000:00:1f.2[B] -> GSI 19 (level, low) -> IRQ 19
Jun  7 20:20:29 spitfire PCI: Setting latency timer of device 0000:00:1f.2 to 64
Jun  7 20:20:29 spitfire ata3: SATA max UDMA/133 cmd 0x000140d8 ctl 0x000140f6 bmdma 0x00014020 irq 19
Jun  7 20:20:29 spitfire ata4: SATA max UDMA/133 cmd 0x000140d0 ctl 0x000140f2 bmdma 0x00014028 irq 19
Jun  7 20:20:29 spitfire scsi2 : ata_piix
Jun  7 20:20:29 spitfire ATA: abnormal status 0x7F on port 0x000140df
Jun  7 20:20:29 spitfire ATA: abnormal status 0x7F on port 0x000140df
Jun  7 20:20:29 spitfire ata3.01: ATA-7: FUJITSU MHV2120BHPL, 0081702E, max UDMA/100
Jun  7 20:20:29 spitfire ata3.01: 234441648 sectors, multi 16: LBA48 NCQ (depth 0/32)
Jun  7 20:20:29 spitfire ata3.01: configured for UDMA/100
Jun  7 20:20:29 spitfire scsi3 : ata_piix
Jun  7 20:20:29 spitfire ATA: abnormal status 0x7F on port 0x000140d7
Jun  7 20:20:29 spitfire scsi 2:0:1:0: Direct-Access     ATA      FUJITSU MHV2120B 0081 PQ: 0 ANSI: 5
Jun  7 20:20:29 spitfire SCSI device sda: 234441648 512-byte hdwr sectors (120034 MB)
Jun  7 20:20:29 spitfire sda: Write Protect is off
Jun  7 20:20:29 spitfire sda: Mode Sense: 00 3a 00 00
Jun  7 20:20:29 spitfire SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Jun  7 20:20:29 spitfire SCSI device sda: 234441648 512-byte hdwr sectors (120034 MB)
Jun  7 20:20:29 spitfire sda: Write Protect is off
Jun  7 20:20:29 spitfire sda: Mode Sense: 00 3a 00 00
Jun  7 20:20:29 spitfire SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Jun  7 20:20:29 spitfire sda: sda1 sda2 sda3 sda4
Jun  7 20:20:29 spitfire sd 2:0:1:0: Attached scsi disk sda
Jun  7 20:20:29 spitfire sd 2:0:1:0: Attached scsi generic sg1 type 0
Jun  7 20:20:29 spitfire ACPI: PCI Interrupt 0000:00:1d.7[A] -> GSI 23 (level, low) -> IRQ 20
Jun  7 20:20:29 spitfire PCI: Setting latency timer of device 0000:00:1d.7 to 64

this seems to me to show that the sda drive is configuring correctly but the cd rom is not...

maybe some scsi setting I have missed.

BERYL working, Celestia and googleearth working with the latest ati drivers - so given sound, networking and graphics are cool it is looking promising...

I think I might open a new thread dedicated to macbook pro kernel configs to assist in not avalanching this venerable one...

Regards,
Will

stardotstar · Post by **stardotstar** » Thu Jun 07, 2007 11:40 am

Found it - new libata driver assigns drive as:

/dev/sr0

now to work out how to get all the media players to go there

btw totem sees it when the fstab is changed to reflect the mount - but xine does not...

je_fro · Post by **je_fro** » Thu Jun 07, 2007 3:37 pm

Looks like I'm a bit late to the party, but here's a present anyway:
http://je-fro.net/linux-2.6.21-macbook-pro-core2.config
Like I said, it's patched with mactel-linux -r124

je_fro · Post by **je_fro** » Thu Jun 07, 2007 5:51 pm

For those of you using Pic0's overlay I've bumped the kernel to 2.6.21.3 with mactel-patches revision 126.
Also, madwifi-ng was bumped to subversion revision 2431.
Everything is working fine here.

ath0 used to drop out occasionally with the older madwifi revision, but it hasn't done it yet with -r 2431.

Let me know if anything breaks

stardotstar · Post by **stardotstar** » Fri Jun 08, 2007 3:19 am

thanks for the config - I will have a play.

I have reverted to my last known good 2.6.17.11 in the mean time because even though I got everything working seemingly I ended up with hard x locks and mysterious oops on boot.

I'll post back when I have anything useful

Will

stardotstar · Post by **stardotstar** » Tue Jun 12, 2007 9:30 am

No luck with 2.6.21 yet - gentoo r3 or .3 from kernel.org. I get mysterious X Locks either during a session or on starting metacity/gnome or beryl. Might be ATIDRIVERS with that kernel. I have been recompiling madwifi-ng, vmware modules and ati drivers on each kernel attempt but the locks continue. network and vmware all seem fine - just X...

stardotstar · Post by **stardotstar** » Sat Jun 16, 2007 12:24 am

And suddenly yesterday the mouse button began to feel less "clicky" but still worked and I thought Wha?? and then the battery appeared to die suddenly and on investigation I have a swollen battery which was pushing up on the mouse button and causing the mechanism to not feel right - so It looks like I have "one of those ones" afterall. Shop tells me that I must bring the mac and battery in and they will send away for a new battery - I told them that I need the machine and they said that's fine but I have to bring it in to get the serial number off it - yeah right - they want to see what I am running I bet. But it shouldn't be a problem - they better honor the warranty - this is a hardware issue - I have all recent required firmware and still run OSX from time to time to check for updates...

It;s quite spectacular - the battery case is splitting away top and bottom and although the fuel guage says its full it won't work in the machine for more than a minute or two then OFF.

UPDATE:

they are sending the battery off - there was no need to turn the machine on - simply reported as swollen battery and not even a warranty claim - providing apple have them in stock two to three days replacement.

echo6 · Post by **echo6** » Mon Jun 18, 2007 4:58 pm

I just want to pass on my thanks to je_fro and Pic0 for their efforts here.

I finally had some downtime to get my MacBook Pro sorted, and using the Pic0's overlay with je_fro kernel config saved me lots of agro

e.g. got ati drivers, sound and wifi working (well wifi is almost working )

p.s. I cheated a little and installed Sabayon, but I now have a MacBook Pro with OSX, XP and Linux working with rEFIt.

Sixtiz · Post by **Sixtiz** » Sun Jun 24, 2007 9:39 am

Has anyone else noticed that the brightness keys started working as I mention in http://forums.gentoo.org/viewtopic-t-56 ... cbook.html ?

stardotstar · Post by **stardotstar** » Mon Jun 25, 2007 12:03 am

yes. not sure if they are properly activated by the ambient light sensor though (they certainly come on in the dark - or when the sensor is covered but don't return). Mine seem to come on at different levels and Fn F10 is up, Fn F8 off but Fn F9 does not work anymore. Previously I had set them up manually. must look into this when I get a chance. The current mapping of the Brightness is fine but my volume Fn F3 works for Mute on off but F4 and F5 do not respond anymore - once again they had been set manually but are now not working.

stardotstar · Post by **stardotstar** » Mon Jun 25, 2007 12:19 am

Battery was promptly replaced and sent out to me - very efficient - very good.

On another note does anyone find that the touchpad will suddenly become constrained to about a band of 1/3rd of the screen in the middle - like a belt above and below which the pointer won't go?

This is very annoying cause it often requires me to ctrl-alt-delete my x session... One the whole the touchpad is very good - single, two finger and three finger taps for three buttons, accel and sensitivity all good.

Here is my config if it could prompt any thoughts - how much has changed in the touchpad code on kernels more recent than 2.6.17.11 (still having no luck with 2.6.21+)

Code: Select all

Section "InputDevice"
        Identifier  "AppleTouchPad"
        Driver      "synaptics"
        Option      "SHMConfig" "on"
#       Option          "CorePointer"
        Option      "Device" "/dev/input/appletouchpad"
        Option      "Protocol" "event"
        Option      "LeftEdge" "100"
        Option      "RightEdge" "1120"
        Option      "TopEdge" "50"
        Option      "BottomEdge" "310"
        Option      "FingerLow" "25"
        Option      "FingerHigh" "30"
        Option      "MaxTapTime" "180"
        Option      "MaxTapMove" "220"
        Option      "MaxDoubleTapTime" "180"
        Option      "VertScrollDelta" "20"
#       Option  "HorizScrollDelta"      "50"
        Option      "MinSpeed" "0.89"
        Option      "MaxSpeed" "0.98"
        Option      "AccelFactor" "0.0095"
EndSection

Will

echo6 · Post by **echo6** » Wed Jul 04, 2007 8:04 pm

Annoying issue, any ideas how where to look to trouble shoot the following issue.

Flashing cursor and
Drawing anything in a term windows especially an ncruses window e.g. make menuconfig is extremely slow

dmvianna · Post by **dmvianna** » Wed Jul 04, 2007 9:06 pm

stardotstar wrote:On another note does anyone find that the touchpad will suddenly become constrained to about a band of 1/3rd of the screen in the middle - like a belt above and below which the pointer won't go?

Hi, Will. Yep, I have noticed something similar for a long time. Every time I unplug my USB mouse, udev (in my system) modprobes appletouch, and every time I plug it, udev rmmods appletouch. But for a long while now, after a rmmod/modprobe appletouch after booting, the touchpad will be unusable, with the cursor being unable to scroll vertically, blinking and looking very unstable. Hence, I have avoided as much as I could to unplug my USB mouse. But this week I found out that if I do it twice, the touchpad will work!

Hence, I would suggest you rmmod and modprobe appletouch every time it behaves strangely, and maybe you could try it twice in case it doesn't come back the first time. No idea how to fix it completely. I had the same issue with udev before, when plugging USB sticks and my iPod. I would have to try plugging them more than once, and at some point (usually after the second try) it would be correctly mounted. But it seems to have become more reliable in the last week, after an emerge -uDN world (I am running stable, with many packages unmasked in package.keywords).

Hope that helps.

stardotstar · Post by **stardotstar** » Fri Jul 06, 2007 11:28 pm

Pretty much invaluable info dmv - I actually have found myself that modprobe -r appletouch && modprobe appletouch works as a get out of jail free script - since the touchpad stops completely and then comes back alive immediately - but without any of the synaptics features acceleration, multi-finger taps etc...

Fortunately the fact that it can be used again at least means that I don't end up having to kill a whole x session and then reboot - which is very inconvenient esp when I have a live VM running which needs to be keyboard controlled.

I wonder if there is a way to reinvoke the synaptics module in X during the session so that after modprobing the appletouch module the touchpad can be used as normal.

I don't seem to get this from plugging and unplugging my usb mouse - seems to happen after long idle times and lid closed periods (my lid closed event is only designed to blank the screen so huh?>)

Thanks for that.

Will.

btw I am having complete success with 2.6.20 patched and built according to the wiki linked overlay

je_fro · Post by **je_fro** » Sat Jul 07, 2007 12:04 am

stardotstar wrote:without any of the synaptics features acceleration, multi-finger taps etc...

I'm getting the _exact_ same behavior, and none of the recent appletouch patches from mactel-linux seem to help either...

stardotstar · Post by **stardotstar** » Sat Jul 07, 2007 7:41 am

So can synaptics be reloaded during an X session?

What sort of symptoms do you get that require you to load and unload appletouch?

I have a pretty usual mix of:

1) pointer becomes constrained to a horizontal band on the screen - usually about 1/3 or 2/3 of the way down the screen and the pointer seems to bungy around when it is moved in this region.

2) it can become jittery and jumpy by itself

3) plugging in a usb mouse fixes 1 but not 2

Here is my synaptics:

Code: Select all

Section "InputDevice"
        Identifier  "AppleTouchPad"
        Driver      "synaptics"
        Option      "SHMConfig" "on"
#       Option          "CorePointer"
        Option      "Device" "/dev/input/appletouchpad"
        Option      "Protocol" "event"
        Option      "LeftEdge" "100"
        Option      "RightEdge" "1120"
        Option      "TopEdge" "50"
        Option      "BottomEdge" "310"
        Option      "FingerLow" "25"
        Option      "FingerHigh" "30"
        Option      "MaxTapTime" "180"
        Option      "MaxTapMove" "220"
        Option      "MaxDoubleTapTime" "180"
        Option      "VertScrollDelta" "20"
#       Option  "HorizScrollDelta"      "50"
        Option      "MinSpeed" "0.89"
        Option      "MaxSpeed" "0.98"
        Option      "AccelFactor" "0.0095"
EndSection

When it is working the mouse behavior is very good except for the occasional tendency to suddenly and quite randomly twitch to another place when typing - including a different field higher up the screen which is *really* annoying and I can't see what is causing it - not palms on the pad or anything.

Will

dmvianna · Post by **dmvianna** » Sat Jul 07, 2007 8:29 am

stardotstar wrote:When it is working the mouse behavior is very good except for the occasional tendency to suddenly and quite randomly twitch to another place when typing - including a different field higher up the screen which is *really* annoying and I can't see what is causing it - not palms on the pad or anything.

I could never make it behave quite as nicely as in OSX (without having palm mouseclicks

), and that's why I'm not using synaptics anymore (no touchpad mouseclick, hence no danger). But it jumps to the top of the screen even when using the USB mouse with me too, although it is kind of rare.

Macbook Pro

Mbp2