Re: powerpc d-i daily builds reactivated, use 2.6.12 kernels, including 64bit kernels, miboot floppies dropped for now.
On Sun, Aug 21, 2005 at 08:47:16PM -0300, Rogério Brito wrote: > There is current discussion on the lkml discussing which kernels one > should use with small systems (like old 486s and such) and the consensus > seems to be that one should use either a 2.6 kernel (perhaps using some > of the linux-tiny patches) or use a really ancient kernel. Do you have a link to this discussion? Can't see anything abvious in the archives. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Booting with yaboot and a new kernel
On Mon, Aug 22, 2005 at 05:40:47PM +0200, Pablo Gil wrote: > As far as I know this it is right. When I start the computer the yaboot > propmt appears, then I press tab to be able to see the list. Then I > choose "LinuxNuevo" to boot the new kernel. Yaboot throws some lines as > usual but the linux never starts loading (usualy I get tux in one corner > and that system information). My computer gets stuck and I only can hear > the hard drisk spinnig as if something was trying to read it. Did you rerun ybin after canging yaboot.conf? What are the exact last lines you see? It could you have not selected the right video driver and the machine is really booting up nicely, since you hear the disk spinning. But you would see that in /var/log/messages after booting your old kernel again. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: SysRq on titanium powerbook
On Sat, Aug 27, 2005 at 10:46:21PM +0300, Jesus Climent wrote: > On Wed, Aug 17, 2005 at 02:10:47PM +0200, Michael Schmitz wrote: > > > > > > Code below uses Alt-F11, but I'm not sure if that is specific enough > > > as it could be used by normal applications. Any sugestion for a better > > > key combination to use? > > > > F11 is also mapped as a button by some people. > > > > Can Alt-Power be used for this? > > Could this patch be passed around? I would be really happy to have it... With what key combination? I don't have a patch that uses Alt-Power, because pressing Power too long turns off the computer. The best thing I have come up with is Alt-F11 but as Michael noted it means this key-combination does not work any more in normal applications. I've included that patch earlier in this thread. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [debian-powerpc] New Ibook no sound , Works only for the headphones
Sven, Saladino, 1. Was your model detected as a snapper before your fix? You can also check the "compatible" file in the open firmware 'sound' directory. 2. Can you send a listing of the open firmware 'sound' directory? It would be interesting to see if there are any new files here. Output of 'od -x sample-rates' would also be usefull. Maybe others as well, not sure now. 3. To get this into the kernel please post the fix to alsa-devel@lists.sourceforge.net with a signed-off line and explanation what model chip this is, or at least the /proc/cpuinfo output. Read Documentation/SubmittingPatches for more info. I can forward it as well, if you get the info and sign it off. Best regards, -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets --- On Thu, Sep 15, 2005 at 08:13:47PM +0200, Sven Henkel wrote: > Hi! > > > Misteriously ive found that when i use mpg321 it works with the > > headphones!!! But none of the controls in alsamixer worked with this > > sound. > > ... > > I had the same problems on my brand new 12" iBook, i.e.: > - modprobing dmasound_pmac fails with "No such device" > - snd_powermac results in sound on headphones, no sound in speakers and an > ineffective mixer > > Those results were seen on a vanilla 2.6.13 kernel, so my scenario looks > pretty similar to yours. ;) > > Reading through latest kernel-changes, I got inspired by: > > http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=821690cdc82e4090ef6f91947f76a231fad5cbb1 > > Unfortunately that patch didn't help, but I modified it to exactly match my > soundcard-model. So, here it is: > > --- linux-2.6.13/sound/ppc/pmac.c 2005-08-29 14:11:45.0 +0200 > +++ linux-2.6.13.patched/sound/ppc/pmac.c 2005-09-15 19:57:16.0 > +0200 > @@ -988,6 +988,7 @@ static int __init snd_pmac_detect(pmac_t > case 0x33: > case 0x29: > case 0x24: > + case 0x50: > chip->num_freqs = ARRAY_SIZE(tumbler_freqs); > chip->model = PMAC_SNAPPER; > chip->can_byte_swap = 0; /* FIXME: check this */ > > This patch gives me sound on speakers and headphones and a lot of working > mixer-controls, so it looks good... > > I hope it helps, > > cheers, > Sven -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [debian-powerpc] New Ibook no sound , Works only for the headphones
Sven, Thanks for the additional info. Some notes below. On Fri, Sep 16, 2005 at 04:18:02PM +0200, Sven Henkel wrote: > Hi! > > Being a mac-owner for only 48 hours now, I don't really understand too much > of > the hardware-side of my iBook. So, I might be misunderstanding some of your > questions, but I'll try and give my best in answering them. :-) > > > 1. Was your model detected as a snapper before your fix? > > No, it was detected as an awacs-card. That's also what hwinfo tells me. > > > You can also > > check the "compatible" file in the open firmware 'sound' directory. > > /proc/device-tree/[EMAIL PROTECTED]/[EMAIL PROTECTED]/[EMAIL > PROTECTED]/[EMAIL PROTECTED]/sound# cat > compatible > AOAbase > > This doesn't really mean anything to me. > > > 2. Can you send a listing of the open firmware 'sound' directory? It would > > be interesting to see if there are any new files here. > > Sure, here it is: > > /proc/device-tree/[EMAIL PROTECTED]/[EMAIL PROTECTED]/[EMAIL > PROTECTED]/[EMAIL PROTECTED]/sound# ls -la > total 8 > dr-xr-xr-x 2 root root 0 Sep 16 15:47 . > dr-xr-xr-x 3 root root 0 Sep 16 15:47 .. > -r--r--r-- 1 root root 0 Sep 16 15:57 built-in > -r--r--r-- 1 root root 8 Sep 16 15:57 compatible > -r--r--r-- 1 root root 10 Sep 16 15:57 device_type > -r--r--r-- 1 root root 4 Sep 16 15:57 layout-id > -r--r--r-- 1 root root 4 Sep 16 15:57 linux,phandle > -r--r--r-- 1 root root 6 Sep 16 15:57 name > -r--r--r-- 1 root root 4 Sep 16 15:57 object-model-version > -r--r--r-- 1 root root 4 Sep 16 15:57 platform-tas-codec-ref > -r--r--r-- 1 root root 4 Sep 16 15:57 vendor-id > > > Output of 'od -x sample-rates' would also be usefull. > > Sorry, I don't have that file. Thats is usefull to know as well. It does mean you're restricted to 44.1KHz now. > > Maybe others as well, > > not sure now. > > Okay, here we go: > > od -x layout-id > 000 0050 > 004 > > od -x linux,phandle > 000 ff98 5200 > 004 > > od -x platform-tas-codec-ref > 000 ff98 cdd8 > 004 This tells me is that you definitely have a TAS chipset in there. Don't know what extact version it is though. > od -x vendor-id > 000 106b > 004 Apple > cat device_type > soundchip > > cat name > sound > > > 3. To get this into the kernel please post the fix to > > alsa-devel@lists.sourceforge.net > > with a signed-off line and explanation what model chip this is, or at > > least the /proc/cpuinfo output. Read Documentation/SubmittingPatches for > > more info. I can forward it as well, if you get the info and sign it off. > > Well, I don't really have a clue why that patch works, so I don't like to > submit it to mainline before I have some deeper understanding of what I > actually did... ;-) So, I will submit it as soon as someone with more > knowledge of apple's sound-hardware confirms that the patch looks reasonable. Basically your patch identifies your sound chip as the TAS3004, usually called Snapper by Apple. I can't say if this is the precise chip you have. You could also be having a later model which behaves compatible (so far) with the TAS3004. Either way, the TAS3004 code is the most advanced chip in the powermac code. I would not have recommended submitting the patch if I did not think it looked okay... -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [debian-powerpc] New Ibook no sound , Works only for the headphones
This issue has been reported as alsa bug 1430: https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1430 (use guest login). -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How it to understand? (RTC trouble)
Your machine probably does not use the 'rtc' code, but the 'genrtc' code. genrtc does not support periodic timers, though there is a patch somewhere for that. You do not have the file because it is not supported. I don't know if there are any powerpc machines that use rtc. Martin On Tue, Sep 20, 2005 at 09:36:36AM +0400, aleuni wrote: > Kind localtime! > > I shall not understand, where has got to/proc/sys/dev/rtc/, and in > particular:/proc/sys/dev/rtc/max-user-freq? Has looked configs of a > kernel - rtc it is compiled in it. > Naturally, has found out it in/proc/drivers. > A question: where mine/proc/sys/dev/rtc/? > System: Debian sarge 3.1 r0a > > I hope for understanding. :) > > -- > bw, > aleuni -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: summary of ALSA status on PPC as of kernel 2.6.13 (ALSA 1.0.9)
Martin, Am very glad to see this. Good effort. Some more details on the G4 PowerBook (PowerBook 3,4) with the Snapper chipset I reported originally: I'd put a yes in all collumns. The only glitch with line input is that channels switch sometimes with mono input. Never during one recording, but sometimes between takes (i.e. when the device gets closed/opened). I'm keeping an eye on this, but have not found a pattern yet. Martin On Sat, Sep 17, 2005 at 10:02:35PM +0300, Martin-Éric Racine wrote: > The long promised table of ALSA PowerMac support is finally available. > > The list is in chronological order and it cross-references Apple models > with audio chipsets. A lot of data is still missing. > > Point your browsers to http://wiki.debian.net/?PowerpcSoundcards and > feel free to contribute additional data. > > If anyone wants to discuss and compare their findings before committing > changes to the document, please drop by #DebianPPC on Freenode. > > -- > Martin-Éric Racine > http://q-funk.iki.fi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: sound problem with g4 pb and selfmade 2.6.12
On Sun, Sep 18, 2005 at 10:33:00PM +0200, Michael Lampard wrote: > - in the kernel config i have activated CONFIG_SOUND=m, > CONFIG_DMASOUND_PMAC=m and CONFIG_DMASOUND=m as well as > CONFIG_SND_POWERMAC=m The DMASOUND settings are for the deprecated OSS system. The last one is the newer ALSA system. > has anybody a hint on how to track this problem down and what might be > the reason of it? The first question is, which of the above modules are loaded? You should either have dmasound_pmac loaded or snd_powermac, but not both. If you have dmasound_pmac loaded try switching to alsa. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How it to understand? (RTC trouble)
On Tue, Sep 20, 2005 at 03:51:45PM +0400, aleuni wrote: > What to me to do? Any video player does not work for me. xine - hangs. > vlc - falls in segfault. Not sure. xine works fine for me. I don't use vlc. Maybe best to report it to your distribution. I'd expect some kind of error messages on failing I/O controls if this was a rtc related problem. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: mounting a HFS drive on a Linux machine clobbers it?
On Mon, Sep 26, 2005 at 11:13:06AM -0700, Kim Cascone wrote: > no one seemed to have an answer for this one so I am posting it again: > When I booted into Linux with my 300G HFS+ formatted backup drive > hanging off the iMac the drive came up and I could see all the .dmg > backups on it... > but when I rebooted into OS X the HFS drive came up as > unrecognized...WTF?! > my question is: > - how do I fix this? > - is there a Unix util that can fix this? I have Fink and Darwinports > so I can DL it from them... So what is it? HFS or HFS+? Big difference for Linux cause different drivers are used (hfs and I think hfsplus). Not sure if OS X still recognizes a HFS drive. Maybe the partition table on the drive is no longer recognized. Wat do fdisk (in Linux and OS X) make of it? How did you partition and format the drive initially? -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: serial console
On Tue, Oct 04, 2005 at 02:56:54PM +0800, KokHow Teh wrote: > When I pass "console=tts/0,115200n8" to the kernel, I don't get any > console output until the login prompt which I set to 115200 in > /etc/inittab. Make sure you have the BOOTX_TEXT config option set. > This is what I end up without "console=" option passed to the kernel: > > > Mounted devfs on /dev > Freeing unused kernel memory: 232k init 4k prep > ..���.�...��������.��.�...�...���..�..�...���..�.�.��.��.��[42;124H. If the init scripts don't set the tty speed it will probably default to 9600n8. Maybe adding an stty early on will help. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: serial console
On Wed, Oct 05, 2005 at 11:06:16AM +0800, KokHow Teh wrote: > If the init scripts don't set the tty speed it will probably default to > 9600n8. Maybe adding an stty early on will help. > > I need more help on this. Do you refer to sbin/init or some scripts > in /etc/ folder? Where should I set it? I was thinking of a script in /etc/init.d that is run from rcS.d. On my machine I have /etc/init.d/etc-setserial (linked from /etc/rcS.d/S30etc-setserial), which where I would add something like: stty /dev/tts/0 115200 -evenp -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: serial console
Sorry, no clue. Unless there is a script earlier than /etc/rcS.d/S00e* Martin On Thu, Oct 06, 2005 at 12:28:40PM +0800, KokHow Teh wrote: > This works. Notice that the correct syntax is "stty -F /dev/tts/0 115200 > -evenp". > > I put it as /etc/rcS.d/S00etc-setserial. However, there is still little > garbled message printed out. I am not a perfectionist but I am just curious > what that little garbled message is printed that it is not affected by the > S00etc-setserial script. > > > VFS: Mounted root (nfs filesystem). > Mounted devfs on /dev > Freeing unused kernel memory: 248k init 4k prep 4k openfirmware > ..���.�...��������.��.�ctivating swap. > System time was Thu Jan 1 00:10:12 UTC 1970. > > > > VFS: Mounted root (ext2 filesystem). > Mounted devfs on /dev > Freeing unused kernel memory: 248k init 4k prep 4k openfirmware > ..���.�...��������.��.�ctivating swap. > Checking root file system... > > > > Any idea..? Thanks. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Boot message : Unknown symbol ...
Try running 'depmod -e'. Maybe your modules.dep is wrong/old. > This happens after udev upgrade (hotplug deleted). I don't know if > this is a udev problem or a kernel problem. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: yaboot and runlevel
Just "Linux 1" should do it. Martin On Wed, Nov 02, 2005 at 10:24:56AM -0700, Spuhler, Peter wrote: > Is it possible with yaboot to specify a runlevel to boot into? > I've tried "Linux init 1" at the yaboot prompt, but I keep ending up in > runlevel 5 at which point my keyboard stops working. I need to get into > my system to fix networking and the X keyboard issue. My networking is > also misconfigured so a remote terminal session is not an option right > now. > Thanks for any help > Peter -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: No sound: Deb Sarge on G4 iMac
On Sun, Jan 09, 2005 at 08:02:36PM -0500, Harvey Ussery wrote: [snip] > Progress! Aplay at least generates sound--the first peep I've heard from > the system since I installed. Here's a sample output: OK, so it seems alsa is working at least. [snap] > [EMAIL PROTECTED]:~$ uname -r > 2.6.8-powerpc > > Thanks for your help--at least now I know the system is capable of > sound. Any idea why the cd player, though clearly playing a music cd, is > not generating any audio? (And I still can't ring the system bell?) What is the name of your cd player application? (it's more than likely that I'm not familiar with it). Maybe it uses the OSS emulation, so check it's setup if it has any. If you use KDE or Gnome you may have a sound deamon running that needs to be reconfigured. I'm not familiar with these, but you can find out who is using/blocking the sound device by doing: lsof | egrep "/dev/snd|/dev/audio" You may want to run this while playing a CD. I've never been able to ring the system bell in 2.6, but haven't bothered to figure out why cause I hate the thing. Worked for me with OSS in 2.4. So the short answer is that your kernel, modules and alsa are okay. Must be application problems. Good luck, -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: gramofile wav format not accepted by other sound programs
On Mon, Jan 24, 2005 at 12:43:43PM +0100, Daniel Kobras wrote: > On Mon, Jan 24, 2005 at 10:27:10PM +1100, Erik de Castro Lopo wrote: > > My guess is that gramofile uses its own WAV file writing code instead of > > using something proven like libsndfile: > > > > http://www.mega-nerd.com/libsndfile/ > > > > Try grabbing the the sndfile-programs package and use sndfile-info on > > the file. You can post the output here and I'll have a look. > > Gramofile uses a forked version of bplay for all audio i/o, and a quick > look confirms that it indeed suffers from the usual endianness issues. > Nick, feel free to file me a bug on gramofile about this. It should be > fairly easy to fix, but I'm a bit busy right now and might need a few > days to provide a patch. You could try the -x switch on sox. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [PATCH] Set accel in radeon driver
Just wanted to verify if my fix has been included. Looks like it might
have fallen between the cracks. If needed I can repost it.
Thanks,
Martin
On Thu, Jan 20, 2005 at 03:18:45PM +1100, Benjamin Herrenschmidt wrote:
> On Sat, 2005-01-08 at 00:00 +0100, Sven Luther wrote:
> > On Fri, Jan 07, 2005 at 05:34:18PM +0000, Martin Habets wrote:
> > > The patch below enhances the FBIOGET_FSCREENINFO ioctl. It is used
> > > by the lcd_hack tool that Ben H wrote. Before this fix accel is set
> > > to 0 and lcd_hack aborts. With it, the tool works again.
> > >
> > > Please apply. Thanks,
> > > --
> > > Martin
> > >
> > > --- 2.6.8/drivers/video/aty/radeon_base.c.orig2005-01-07
> > > 16:55:49.0 +
> > > +++ 2.6.8/drivers/video/aty/radeon_base.c 2005-01-07 16:57:22.0
> > > +
> > > @@ -1812,9 +1812,12 @@ static int __devinit radeon_set_fbinfo (
> > >
> > > fb_alloc_cmap(&info->cmap, 256, 0);
> > >
> > > - if (noaccel)
> > > + if (noaccel) {
> > > info->flags |= FBINFO_HWACCEL_DISABLED;
> > > + info->fix.accel = FB_ACCEL_NONE;
> > > + } else
> > > + info->fix.accel = FB_ACCEL_ATI_RADEON;
> > >
> > > return 0;
> > > }
> >
> > benh, could you comment on this ? Does it make sense to include this in the
> > debian kernel ?
>
> Yes.
>
> Ben.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: bounty for fixing ALSA's powermac driver
Rather than complainting, maybe someone can offer access to their Mac. Either physical or via ssh. Usually the real guru's don't need much time to put things right. (having said that, the cleanup Ben suggested would be a big effort) I use the powermac driver on 2.6.8 with ALSA 1.0.4 and it works fine for me. But my mac has the snapper chip, not the burgundy one. I am able to help debugging, though I'm not a real expert on this driver. Martin On Fri, Feb 04, 2005 at 02:21:33PM +0200, Martin-?ric Racine wrote: > According to the following BTS entry, the ALSA maintainer assigned to the > snd-powermac driver doesn't even have any Mac to play with: > > https://bugtrack.alsa-project.org/alsa-bug/view.php?id=306 > > This prooves that all attempts at fixing the driver have been shooting in > the dark. No f* wonder the snd-powermac driver is such a mess! > > Wouldn't it be a good idea to place a bounty to encourage Ubuntu PPC > developers to get around fixing snd-powermac for all Mac PPC chips? > > -- > Martin-?ric Racine > http://www.iki.fi/q-funk/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: eject CD
I changed the "kb:" entry in /etc/inittab to execute eject. ALT-UpArrow will now eject the CD. Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: record mic and linein on powerbook G3
On Fri, Mar 04, 2005 at 08:49:13AM -0800, gm c wrote: > > > I was wondering if it is now possible to record from the mic or linein on the > powerbook G3 pismo? If so , how? Don't know about OSS. With current ALSA you have an "Input Source" selector in amixer/alsamixer. It was a later addition, though I'm pretty sure it is in ALSA 1.0.6. On a 2.6 kernel you may have to use OS X to set the input gain. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: sound in new powerbooks
If you have MAC OS installed, maybe you can unmute it there? (and hopefully preserve that over a reboot...) Just a thought. On Tue, Mar 15, 2005 at 01:05:01PM +, Jochen Voss wrote: > Hello Jonathan, > > On Tue, Mar 15, 2005 at 01:52:36PM +0100, Chocron J. wrote: > > Nope, however, the card seems to be detected correctly, and the > > /dev/ entries get created, my guess is this is an API thing, I think > > we can't unmute it... > yes, I suspect something similar. If I try to play a sound it even > takes the appropriate time to do so, so maybe the data is actually > moved to the hardware and only inaudible due to some hidden mute button. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Complie error while build cross compile toolchain for i386-linux
The assembler used does not support x86_64. Check the binutils build, and run the cross-assembler with --version. If the cross-assembler is okay, maybe the wrong one is used by xgcc. There's an option to gcc I think to print what commands it uses. Martin On Thu, Mar 17, 2005 at 12:58:52AM +0800, GONG Jie wrote: > Assembler messages: > Fatal error: No compiled in support for x86_64 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: list of PowerMac sound hardware TO-DO (help welcome)
On Tue, Mar 22, 2005 at 04:13:15PM +0200, Martin-?ric Racine wrote: > On Tue, 22 Mar 2005, Benjamin Herrenschmidt wrote: > > > On Mon, 2005-03-21 at 15:19 +0200, Martin-?ric Racine wrote: > > > Regarding overall pmac ALSA support, is there a table that shows what > > > works and what is broken for each supported sound hardware? Such a table > > > could go a long way towards helping whoever will pocket the $500 Ubuntu > > > bounty for fixing ALSA for all pmac sound hardware variants. > > > > Not exactly, you are welcome to start such a table though :) > > This is what I have so far: > > --works y/n-- > CHIPSET MODELS BEEPALSA > awacs > burgundyG3 iMac no no > daca > keywest > tumbler > > Other people are quite welcome to contribute additional data, especially > which PowerMac models use which chipset. Tools that can be used to find > the info include 'lspci' 'dmesg' and 'hwinfo' - just to name a few. > > -- > Martin-?ric Racine > http://www.iki.fi/q-funk/ > snapper PowerBook3,4yes (2.4.25-benh) yes (1.0.3) no (2.6.8.1)yes (1.0.4) Hope this is clear, but this is an older powerbook (around 2002). Beep doesn't work in 2.6 because it is disabled in the code (don't know why). -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: list of PowerMac sound hardware TO-DO (help welcome)
On Tue, Mar 22, 2005 at 02:09:57PM -0500, Michel D?nzer wrote: > On Tue, 2005-03-22 at 16:59 +0000, Martin Habets wrote: > > > > snapper PowerBook3,4yes (2.4.25-benh) yes (1.0.3) > > no (2.6.8.1)yes (1.0.4) > > > > Hope this is clear, but this is an older powerbook (around 2002). > > Beep doesn't work in 2.6 because it is disabled in the code (don't know > > why). > > The beep works fine here with recent 2.6 kernels on a PowerBook3,5 with > a snapper. Make sure the 'Beep' mixer has a non-zero value. I just compiled 2.6.11 and an ALSA 1.0.9rc1 snapshot, and beep works there. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
RTC_PIE possible using via-pmu?
Hi, One of my major struggles in doing audio with my powerbook is the lack of a programmable RTC. I have a via-pmu based pmac and use the genrtc module. More presicely I need these iotctls: RTC_IRQP_SET, RTC_PIE_ON, RTC_PIE_OFF. Would it be possible/wise to add these using via-pmu timer 1? Is there a PMU command for this, or would it require direct IO like via_calibrate_decr()? Thanks, -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RTC_PIE possible using via-pmu?
RTC_IRQP_SET sets the periodic interrupt rate. RTC_PIE_ON turns on periodic interrupts. RTC_PIE_OFF turns off periodic interrupts. There are similar ioctls for setting alarms, but I don't see those used very often. I've pasted the relevant part of my dmesg below. It seems the OpenPIC timer frequency will limit the usability of above iotctls to 4.16MHz. Is that right? Thanks, Martin PowerMac using OpenPIC irq controller at 0x8004 OpenPIC Version 1.2 (4 CPUs and 64 IRQ sources) at fc5e4000 OpenPIC timer frequency is 4.16 MHz PID hash table entries: 4096 (order: 12, 65536 bytes) GMT Delta read from XPRAM: 0 minutes, DST: on time_init: decrementer frequency = 33.331553 MHz On Tue, Mar 29, 2005 at 08:38:48AM +1000, Benjamin Herrenschmidt wrote: > On Mon, 2005-03-28 at 18:45 +0100, Martin Habets wrote: > > Hi, > > > > One of my major struggles in doing audio with my powerbook is the > > lack of a programmable RTC. I have a via-pmu based pmac and use > > the genrtc module. > > > > More presicely I need these iotctls: RTC_IRQP_SET, RTC_PIE_ON, > > RTC_PIE_OFF. Would it be possible/wise to add these using > > via-pmu timer 1? > > Is there a PMU command for this, or would it require direct IO like > > via_calibrate_decr()? > > What are those things supposed to be ? > > Ben. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RTC_PIE possible using via-pmu?
On Wed, Mar 30, 2005 at 10:36:15AM +1000, Benjamin Herrenschmidt wrote: > On Tue, 2005-03-29 at 13:17 +0100, Martin Habets wrote: > > RTC_IRQP_SET sets the periodic interrupt rate. > > RTC_PIE_ON turns on periodic interrupts. > > RTC_PIE_OFF turns off periodic interrupts. > > > > There are similar ioctls for setting alarms, but I don't see those > > used very often. > > > > I've pasted the relevant part of my dmesg below. It seems the OpenPIC > > timer frequency will limit the usability of above iotctls to 4.16MHz. > > Is that right? > > Oh, those periodic interrupt things are features of x86 RTC chips, they > aren't available on other architectures, at least not all of them, > definitely not on pmac. Well I was wondering if I could use the via-pmu timer to implement such a periodic interrupt. > Why not just normal timers ? Not sure what exactly you mean by "normal". I created a genrtc patch [1] for the ioctls above using add_timer(). But the resulting timer responses using a simple test program [2] are not that good (the last percentage is the accuracy), and above 1024Hz completely unusable: $ ./rtctest2 20 loops at 2Hz = 9997669 usec, 499883 usec/loop. deviation is 117 usec (0%). 20 loops at 4Hz = 4999186 usec, 249959 usec/loop. deviation is 41 usec (0%). 20 loops at 8Hz = 2499566 usec, 124978 usec/loop. deviation is 22 usec (0%). 20 loops at 16Hz = 1239760 usec, 61988 usec/loop. deviation is 512 usec (0%). 20 loops at 32Hz = 619856 usec, 30992 usec/loop. deviation is 258 usec (0%). 20 loops at 64Hz = 299902 usec, 14995 usec/loop. deviation is 630 usec (4%). 20 loops at 128Hz = 139931 usec, 6996 usec/loop. deviation is 816 usec (10%). 20 loops at 256Hz = 59947 usec, 2997 usec/loop. deviation is 909 usec (23%). 20 loops at 512Hz = 21003 usec, 1050 usec/loop. deviation is 903 usec (46%). 20 loops at 1024Hz = 19906 usec, 995 usec/loop. deviation is 19 usec (1%). 20 loops at 2048Hz = 19954 usec, 997 usec/loop. deviation is 509 usec (104%). 20 loops at 4096Hz = 19955 usec, 997 usec/loop. deviation is 753 usec (308%). 20 loops at 8192Hz = 19956 usec, 997 usec/loop. deviation is 875 usec (717%). Oddly enough (well, odd to me :), 1024Hz is quite good. I have been able to do some latency tests with that setting [3] which point to latency issues with the radeon and/or drm modules. If I get those figured out (video is NOT my area) I'll be kinda stuck. Hints/tips/suggestions welcome... Must say 2.6.11 looks better than 2.6.8. Thanks, -- Martin Signed-off-by: Martin Habets <[EMAIL PROTECTED]> [1] http://www.mph.eclipse.co.uk/pub/linux/patches/2.6.11/genrtc.c.patch [2] http://www.mph.eclipse.co.uk/pub/linux/patches/rtctest2.c [3] http://www.mph.eclipse.co.uk/pub/linux/latency/test3/index.html -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: eject debian 3.0 inittab
William, In /etc/inittab you probably have a line like this: #kb::kbrequest:/bin/echo "Keyboard Request--edit /etc/inittab to let this work." If you have a line that starts with "kb:" (without the #) there is already some action assigned, which maybe you want to keep. On my machine I changed the above line to: kb::kbrequest:/usr/bin/eject After changing the file run telinit q to activate the change. You can also reboot if you like. Notes: - You need to be root to make this change and run telinit. - Use your favorite editor. If you don't have one try pico or nano. - eject could be in a different place on your machine (unlikely though). To find out where it is run type eject If you get something saying 'type: eject: not found' you'll need to install eject. On debian that is done using apt-get install eject That's all I can think off. Let me know if something is not clear or if you need more info. Martin On Fri, Apr 01, 2005 at 11:31:16AM +0100, William Besson wrote: > Hi Martin, > > I came accross your answer on lists.debian.org. about eject on Debian > 3.0. > I am having the same problem. I have honestly tried everything written > on this forum but nothing will work. I am sure your "kb:" change can be > a good solution but I am quite new to Linux. Would you be kind enough to > tell me exactly what I should write down in the inittab file. > > Thanks for your time > > William -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: halt vs. poweroff; server_mode variable (kernel 2.6.10)
On Fri, Apr 01, 2005 at 06:27:22PM -0500, Charles Lepple wrote: > It looked like that's what would happen if I set server_mode=1 in > /proc/pmu/options and executed halt. But when I ran 'halt', it turned > off the power, just as though I had run 'poweroff'. Have a look at /etc/default/halt. You'll want it set to halt in there. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets --- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: pre-release 2.6.11-0.1 debian/powerpc kernel.
On Wed, Apr 06, 2005 at 11:24:43PM +0200, Paul van Tilburg wrote: > OK, I want to come back to this... I do seem to have a strange problem > with my sound (running kernel-image-2.6.11-powerpc 2.6.11-1) on my > TiBook IV and using ALSA. What chipset does it say in /proc/asound/cards? > When I listen music with mplayer/rhythmbox/ogg123/... after a random > period (can vary from 5secs to 20mins) the sound device just blocks and > the sound stops, I have to quit the app and restart it to have sound > again (for a short while)... Are you using ALSA or OSS drivers or ALSA OSS emulation? I have not been able to reproduce this sofar, using Snapper chipset and ALSA drivers. But I am on ALSA 1.0.9rc1. I think the ALSA version has changed between 2.6.10 and 2.6.11. Could you compare /proc/asound/version for both kernels? Could be your alsa-libs need to be upgraded... -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: modem on ibook g4
It is probably a conexant modem. Have a look at http://www.linuxant.com. I had the 0.99 version of their driver working in 2.4, but have not bothered since. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets --- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: alsa and jack latency
In order to lower the latency you first need to be able to measure it... I have been able to do some latency testing using the ALSA latencytest tools on my powerbook using 2.6.11 and some patches: [1] A patch for the genrtc kernel driver http://www.mph.eclipse.co.uk/pub/linux/patches/2.6.11/genrtc.c.patch The ioctls added can be tested using http://www.mph.eclipse.co.uk/pub/linux/patches/rtctest2.c It is important to run to this to find out the accuracy of the clock on your platform. On PPC 1024Hz is probably the highest accurate frequency. Any machine using genrtc should be able to use these ioctls, not just PPC. [2] A patch against latencytest.0.5.5 http://www.mph.eclipse.co.uk/pub/linux/patches/latencytest.patch With USE_GENERIC_TIMER selected in the makefile, this should work on any platform. For PPC I'd recommend setting USE_PPC_VEA_TIMEBASE in the Makefile. Results: Initial testing showed serious latency from the X11 tests, but the rest looks okay. [3] http://www.mph.eclipse.co.uk/pub/linux/latency/test3/index.html With a (questionable) DRM patch for the Radeon driver the results become: [4] http://www.mph.eclipse.co.uk/pub/linux/latency/test7/index.html Jack shows 11.6msec latency with -r44100 -p128 -n4. I'm working with Takashi to get [2] included. I'm not convinced [1] is needed/usefull for inclusion in the default kernel, unless I hear otherwise. Martin On Sat, May 14, 2005 at 09:56:40PM +0100, Paul Brossier wrote: > Hi all, > > Trying to reduce the latency of jack on my ibook G4, i realised i can go > down to 8.7ms using the oss mode, which is really good. But alsa won't > accept running with less than 22ms and click on most X request: > > $ jackd -R -doss -r44100 -p128 -n3 -w16 > works perfectly, 128*3/44100=8.7ms latency, hardly any xrun on heavy load > > $ jackd -R -doss -r44100 -p256 -n2 -w16 > works fine, 11.6ms latency > > $ jackd -R -dalsa -r44100 -p128 -n3 -S > runs but no clients can connect > > $ jackd -R -dalsa -r44100 -p256 -n2 -S > fails saying ... > ALSA: cannot set number of periods to 2 for capture > ALSA: cannot configure capture channel > > $ jackd -R -dalsa -r44100 -p256 -n4 -S > would run with 23.2ms latency, but xruns on most X queries > > It looks to me that this is related to the pmac driver. I have similar > problems with snd-usb-audio and an edirol usb, which will only let jack > run in oss mode. > > I wonder what prevents the pmac module to do so while the oss emulation > layer does it very well. I am running 2.6.12-rc4, but had this problem > with earlier 2.6 kernels. Any thoughts? > > Many thanks, Paul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: munged yaboot.conf
On Tue, May 31, 2005 at 09:00:56PM -0700, james houston wrote: > I have OS X (10.4.1) and Sarge on my iBook (dual USB). > > I tried to customize my yaboot.conf file and messed up > the syntax so now when I choose "l" in yaboot, I get > this error: > > Config file error: newline is not allowed in quoted > strings near line 7 in file \\yaboot.conf > Syntax error or read error config > > How can I edit the yaboot.conf file safely? I read in > the archives about editing it from OS X, but that was > discouraged. What is the best direction to go then? Since you get the yaboot prompt, you could directly type the path to your kernel, like: hd:3,/boot/vmlinux if your root is on partition 3. The first partition is number 1 in yaboot. > I tried booting from my Sarge installation CD and > choosing "Execute a shell" but my hard drive was not > mounted on "/target" as it said it would be. The hard driver partition gets mounted at a later time in the install process. You could step through more steps of the install process if you don't know where your root partition is, or mount it manually from your shell using: mount /dev/hda3 /target if your root is on partition 3. After editing yaboot.conf run ybin like ybin -C /target/etc/yaboot.conf because of the non-standard path. I hope you actually don't need this tip anymore, but: always keep at least one safe backup choice in your yaboot.conf. Something like: image=/vmlinux label=Rescue read-only append="init=/bin/sh" -- Martin ------- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets --- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Partition check
On Tue, May 31, 2005 at 09:12:47PM -0300, Gunther Furtado wrote: > Hello, > > This is the reading I get whe I boot the 2.4.25-behn kernel. > > Parition check: > hda: [mac] hda1 hda2 hda3 hda4 hda5 hda6 hda7 hda8 hda9 hda10 hda11 hda12 > > And this is when i boot a kernel I compiled from linux-2.4.26 sources. > > Partition check: > /dev/ide/host0/bus0/target0/lun0: [mac] p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 p11 p12 > > Can anyone tell what did I switched off from Behn's kernel configuration? The benh kernel accesses the hard disk as an IDE device. Your later kernel sees it as a SCSI device. First check if you have a boot parameter like "hda=scsi" and if so, remove it. If that isn't it, check the SCSI and IDE settings in your config. Sorry, I don't know exactly which ones. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets --- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sound in g4
With ALSA/Debian I have the issue that I have to add snd to /etc/modules manually. /etc/init.d/alsa won't do anything otherwise, as /proc/asound does not exist. Not sure if the init script is from debian or alsa. This happens on all hardware platforms. In your case, a modprobe snd-powermac should load the sound driver for you. Run alsamixer to unmute it. Martin On Tue, Jun 07, 2005 at 02:30:51AM +, Mike S wrote: > this is probably a totally newbie question, but I have just switched to > debian from ubuntu, and am working on a self compiled 2.6.11 upstream > kernel. I compiled all the sound stuff as modules, but I do not know > what file to edit to load them, or even how to start. KDE reports no > /dev/dsp on startup, and when I run lsmod it shows no snd or anything > like that, so could someone point me in the right direction to start > setting up sound on this thing? > > --Mike S > if specs are needed this is a G4 Dual 450 (Gigabit Ethernet model) and I > am not sure what the sound card is. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [ALSA] Problem on G3-266
On Thu, Nov 04, 2004 at 01:46:26PM +0100, Matthieu Oppliger wrote: > Hello ! > I try to use ALSA on my G3-266. > > After recompilation of my kernel (2.6.8),with ALSA drivers as modules, > i did apt-get install alsa. > > Then alsaconf - finds powermac - PowerMac sound card. It seems that > alsaconf did'nt put snd and snd-powermac in my /etc/modules, so I did it > myself. I think you'll find it has updated /etc/modules.d for you. > After alsaconf, it tells me : > _ > Running update-modules... > Loading driver... > Starting ALSA.../etc/init.d/alsa: Error: alsactl restore failed with > message 'ALSA lib control.c:654:(snd_ctl_open_noupdate) Invalid CTL hw:0 > alsactl: set_controls:986: snd_ctl_open error: No such file or > directory'. > done. > Setting default volumes... > > === ... > Any idea of something missing ? I think you are missing the device files in /dev/snd/. If so, you need to run the snddevices script found in the alsa-driver sources. If that does not help, send the output of lsmod and relevant dmesg output. -- Martin
Re: Music On Console Player
On Sun, Nov 14, 2004 at 03:21:43PM +0100, Elimar Riesebieter wrote: > It is a ncurses based music player. It supports mp3, ogg on ppc. > On i386 wav ist tested as well. > > PPC-Coders: Could you please help us to support wav on ppc? > > Just a comment from Damian Pietras, the creator of MOC: > "Endianes and byte order conversions must be done when reading a > wave file, but I will not write the code, because I can't test it, > sorry" > I'd suggest linking with libsndfile (http://www.mega-nerd.com/libsndfile/) It supports oodles of formats and works on all endianes. -- Martin
Re: 2.4 -> 2.6 kernel upgrade problem
I read a while ago that the debian package somehow recreates /dev/null as a regular file. Looks like your /dev/console might be hosed as well. Here's what they should be: crw-rw-rw- 1 root root 1, 3 Sep 12 18:58 /dev/null crw--- 1 root tty 5, 1 Sep 12 18:59 /dev/console You probably have to boot from CD (or net) to get back into your system, but you could try init level 1. When you get in, use mknod to recreate these files if needed. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets ---
Re: Sound output problem
On Fri, Dec 31, 2004 at 03:24:51PM +0100, Pablo Gil wrote: > I've typed killall esd as root and there was no process called that way. > Does the daemon have other name? > I've also switched to the ALSA plugin and still doesn't work... > > The possible errors are: > 1-Wrong plugin > 2- Sound card blocked > 3- Sound card wrongly configured > > I think it only can be 2 but if there is no esd, who is using it? To figure out who is using it, try this: lsof | egrep "snd|dsp" Also, have a look at the files in /proc/asound. The cards and devices files should be interesting. devices should have entries for ctl, playback, capture and timer. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets ---
Re: mixer doesn't work
On Sat, Jan 01, 2005 at 08:43:25PM +0800, William Xuuu wrote: > Comparing my config file and yours, the main difference is that most of > the items, you've made them built as modules, i, instead, have made > them built in the kernel. Can this be the reason? So odd if it's . And, > What's your kernel version? Mine is 2.6.9. CONFIG_DMASOUND_PMAC is for the OSS driver. CONFIG_SND_POWERMAC is the ALSA driver. You should not use both, but choose one. Use aumix with the OSS driver, and alsamixer with the ALSA driver. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets ---
[PATCH] Set accel in radeon driver
The patch below enhances the FBIOGET_FSCREENINFO ioctl. It is used
by the lcd_hack tool that Ben H wrote. Before this fix accel is set
to 0 and lcd_hack aborts. With it, the tool works again.
Please apply. Thanks,
--
Martin
--- 2.6.8/drivers/video/aty/radeon_base.c.orig 2005-01-07 16:55:49.0
+
+++ 2.6.8/drivers/video/aty/radeon_base.c 2005-01-07 16:57:22.0
+
@@ -1812,9 +1812,12 @@ static int __devinit radeon_set_fbinfo (
fb_alloc_cmap(&info->cmap, 256, 0);
- if (noaccel)
+ if (noaccel) {
info->flags |= FBINFO_HWACCEL_DISABLED;
+ info->fix.accel = FB_ACCEL_NONE;
+ } else
+ info->fix.accel = FB_ACCEL_ATI_RADEON;
return 0;
}
Re: No sound: Deb Sarge on G4 iMac (Addendum)
On Sat, Jan 08, 2005 at 11:50:33AM -0500, Harvey Ussery wrote: > >[EMAIL PROTECTED]:~$ cat /dev/sndstat > >Sound Driver:3.8.1a-980706 (ALSA v1.0.4 emulation code) > >Kernel: Linux doob 2.6.8-powerpc #1 Sun Oct 3 13:22:21 CEST 2004 ppc > >Config options: 0 > > > >Installed drivers: > >Type 10: ALSA emulation > > > >Card config: > >PowerMac Tumbler (Dev 29) Sub-frame 0 > > > >Audio devices: > >0: PowerMac Tumbler > > > >Synth devices: NOT ENABLED IN CONFIG > > > >Midi devices: NOT ENABLED IN CONFIG > > > >Timers: > >7: system timer > > > >Mixers: > >0: PowerMac Tumbler > >[EMAIL PROTECTED]:~$ > > "Audio devices" apparently config'd. Since I'm only interested in basic > audio, I shouldn't be concerned that Synth & Midi are not config'd, right? This looks okay. Synth and Midi show the same for me too. I think this is normal as this is the OSS emulation info. > What about sound drivers? You already have that loaded. It's the snd_powermac module. > Again, many thanks to anyone who can get me started on the correct > diagnostic path why I don't have sound enabled.--Harvey The first think I'd try next is to run alsamixer and make sure the volume is turned up. Next, try to play a sample sound using aplay. If you get any errors, post them. If you hear no sound with aplay, try the '-v' and '-vv' switches. The last one should show continuous output for a while. The only other thing I can think of is that the OSS driver might be built into your kernel. I don't know if this is a Debian practice or not. What kernel are you on? (uname -r) -- Martin
Re: No sound: Deb Sarge on G4 iMac
On Sun, Jan 09, 2005 at 08:02:36PM -0500, Harvey Ussery wrote: [snip] > Progress! Aplay at least generates sound--the first peep I've heard from > the system since I installed. Here's a sample output: OK, so it seems alsa is working at least. [snap] > [EMAIL PROTECTED]:~$ uname -r > 2.6.8-powerpc > > Thanks for your help--at least now I know the system is capable of > sound. Any idea why the cd player, though clearly playing a music cd, is > not generating any audio? (And I still can't ring the system bell?) What is the name of your cd player application? (it's more than likely that I'm not familiar with it). Maybe it uses the OSS emulation, so check it's setup if it has any. If you use KDE or Gnome you may have a sound deamon running that needs to be reconfigured. I'm not familiar with these, but you can find out who is using/blocking the sound device by doing: lsof | egrep "/dev/snd|/dev/audio" You may want to run this while playing a CD. I've never been able to ring the system bell in 2.6, but haven't bothered to figure out why cause I hate the thing. Worked for me with OSS in 2.4. So the short answer is that your kernel, modules and alsa are okay. Must be application problems. Good luck, -- Martin
Re: Usb harddisk trouble...
On Sun, Jun 12, 2005 at 06:07:24PM +0200, Peter Plessas wrote: ... > At power on, the HD gets listed in /proc/partitions: > >8 0 195360984 sda >8 1 32764536 sda1-> FAT32 >8 2 32764567 sda2-> FAT32 >8 3 32764567 sda3-> FAT32 >8 4 97064730 sda4-> ntfs > > The partitions were made on a win2k box. In you kernel config, make sure you have CONFIG_MSDOS_PARTITION set (file systems->partition types). But I suspect you already have this as the partitions get recognized. ... > As i try to mount the first partition: > gonzo:~# mount -v -t msdos /dev/sda1 /fw/ > mount: /dev/sda1 is not a valid block device Try if any of these work: mount -v -t vfat /dev/sda1 /fw/ mount -v -t fat /dev/sda1 /fw/ mount -v /dev/sda1 /fw/ If that does not work, run fdisk /dev/sda and see if that works okay. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Newbie: Firewire Booting
It is not necessary to have linux on the internal disk. You can boot it from any place the open firmware recognizes. That should include external disks. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets --- On Thu, Jun 16, 2005 at 11:20:14AM +0300, Panagiotis Papadopoulos wrote: > Hello, > I posted this before but got no answer. Is it possible to boot Debian from > an external drive or is it necessary to install it in the internal hard > disk? > If this is too newbie please forgive me > Panagiotis -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Debian as only OS on a PowerBook G3
On Tue, Jul 12, 2005 at 07:54:54PM +0200, Klaus Ethgen wrote: > Hello, > > Now I'm completely stuck. I tryed to install debian woody on a PowerBook > G3. The think worked well until I tryed to boot the new OS. All I tryed > did not help booting from the HD. > > The Technical: > - PowerBook G3 (black case) > - OldWorld MAC > - 20 GB HD > - Partitionated with first partition / (including /boot) > - Bootloader quik > > The first strange was that ofpath shows /pci/mac-io/ata0/ata-disk where > the system only has a /pci/mac-io/ata0/[EMAIL PROTECTED],0 at openfirmware. > Next > funny was that I could only access the firmware AFTER installing linux > not bevore. > > Now the important: If I do a "dev /pci/mac-io/ata0/[EMAIL PROTECTED],0" > followed by > an "words" I get NO words. So it do not wonder why boot ata0/[EMAIL > PROTECTED],0:2 > give an "can't OPEN". Yes, I think this means there's no disk there. Are you sure it's an ATA disk? I have a SCSI one and boot from scsi/sd. During installation did it use hda or sda? Try "devalias" and check for other disk with "words" like you did above. > Is there any I can do to let this mac boot into linux? Maybe there is a > way to create a bootdisk? The installation menu says that is not > supportet. But the installation starts from the floppy so it must be > possible to boot the normal system from floppy too. You can boot from the installation disk, go through the first couple of screens, and in the menu scroll down to "execute a shell". From there maybe you can identify your disk. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets --- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Few problems with PowerBook(5,7)
As for the sound, install alsa-base. It should load snd_powermac for you. Don't forget to unmute and turn up the volume with amixer or alsamixer. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
SysRq on titanium powerbook
I could not find any way to get keycode 0x69 to come out of the
adb keyboard on my powerbook. The patch below (against 2.6.11)
changes Fn+F12 to provide this.
Now if I press Fn+Option+F12+command I get the SysReq stuff.
If anyone knows how to generate 0x69 let me know. Could not find
any info on the raw keycodes of the keyboard.
--
Martin
$ cat /proc/cpuinfo
processor : 0
cpu : 7455, altivec supported
clock : 800MHz
revision: 2.1 (pvr 8001 0201)
bogomips: 798.32
machine : PowerBook3,4
motherboard : PowerBook3,4 MacRISC2 MacRISC Power Macintosh
detected as : 73 (PowerBook Titanium III)
pmac flags : 001b
L2 cache: 256K unified
memory : 512MB
pmac-generation : NewWorld
---
Index: 2.6/drivers/macintosh/adbhid.c
===
--- 2.6.orig/drivers/macintosh/adbhid.c 2005-08-15 13:49:27.0 +0100
+++ 2.6/drivers/macintosh/adbhid.c 2005-08-15 15:05:47.0 +0100
@@ -70,6 +70,7 @@
#define ADB_KEY_CMD0x37
#define ADB_KEY_CAPSLOCK 0x39
#define ADB_KEY_FN 0x3f
+#define ADB_KEY_F120x6f
#define ADB_KEY_FWDEL 0x75
#define ADB_KEY_POWER_OLD 0x7e
#define ADB_KEY_POWER 0x7f
@@ -343,6 +344,14 @@
ahid->flags |= FLAG_EMU_FWDEL_DOWN;
}
break;
+#ifdef CONFIG_MAGIC_SYSRQ
+ case ADB_KEY_F12:
+ /* Map Fn+F12 to SysRq. Input driver checks that Alt is pressed
too. */
+ if (ahid->flags & FLAG_FN_KEY_PRESSED) {
+ keycode = 0x69;
+ }
+ break;
+#endif
#endif /* CONFIG_PPC_PMAC */
}
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Not able to control sound with gnome
Pablo, We would need some info on your setup. Things like: 1) What kernel are you using? (output of 'uname -r') 2) Are you using ALSA or OSS? Provide the output of 'cat /proc/asound/version'. Also, provide the output of 'cat /proc/asound/cards'. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: SysRq on titanium powerbook
On Tue, Aug 16, 2005 at 10:51:55AM +1000, Benjamin Herrenschmidt wrote:
> You need to keep track that it's pressed to send appropriate key up as
> well I suppose...
Found another snag with this. Using Fn activates the numeric keypad
emulation of the keyboard, effectively blocking commands p,m,o.
Unfortunately can't use F12 itself here as that is mapped as a button.
Code below uses Alt-F11, but I'm not sure if that is specific enough
as it could be used by normal applications. Any sugestion for a better
key combination to use?
--
Martin
Signed-off-by: Martin Habets <[EMAIL PROTECTED]>
---
Index: 2.6/drivers/macintosh/adbhid.c
===
--- 2.6.orig/drivers/macintosh/adbhid.c 2005-08-15 13:49:27.0 +0100
+++ 2.6/drivers/macintosh/adbhid.c 2005-08-16 22:48:15.0 +0100
@@ -69,7 +69,9 @@
#define ADB_KEY_DEL0x33
#define ADB_KEY_CMD0x37
#define ADB_KEY_CAPSLOCK 0x39
+#define ADB_KEY_LEFTALT0x3a
#define ADB_KEY_FN 0x3f
+#define ADB_KEY_F110x67
#define ADB_KEY_FWDEL 0x75
#define ADB_KEY_POWER_OLD 0x7e
#define ADB_KEY_POWER 0x7f
@@ -221,6 +223,8 @@
#define FLAG_FN_KEY_PRESSED0x0001
#define FLAG_POWER_FROM_FN 0x0002
#define FLAG_EMU_FWDEL_DOWN0x0004
+#define FLAG_LEFTALT_DOWN 0x0008
+#define FLAG_F11_DOWN 0x0010
static struct adbhid *adbhid[16];
@@ -343,6 +347,31 @@
ahid->flags |= FLAG_EMU_FWDEL_DOWN;
}
break;
+#ifdef CONFIG_MAGIC_SYSRQ
+ case ADB_KEY_LEFTALT:
+ /* Keep track of the Alt key state */
+ if (up_flag) {
+ ahid->flags &= ~FLAG_LEFTALT_DOWN;
+ /* Map Alt+F11 to SysRq. */
+ if (ahid->flags & FLAG_F11_DOWN) {
+ ahid->flags &= ~FLAG_F11_DOWN;
+ keycode = 0x69;
+ break;
+ }
+ } else
+ ahid->flags |= FLAG_LEFTALT_DOWN;
+ break;
+ case ADB_KEY_F11:
+ /* Map Alt+F11 to SysRq. */
+ if (ahid->flags & FLAG_LEFTALT_DOWN) {
+ keycode = 0x69;
+ if (up_flag)
+ ahid->flags &= ~FLAG_F11_DOWN;
+ else
+ ahid->flags |= FLAG_F11_DOWN;
+ }
+ break;
+#endif
#endif /* CONFIG_PPC_PMAC */
}
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: SysRq on titanium powerbook
On Wed, Aug 17, 2005 at 02:10:47PM +0200, Michael Schmitz wrote: > F11 is also mapped as a button by some people. Yes, I use it to emulate mouse button 2. But that does not interfere with this because I check for Alt-F11. > Can Alt-Power be used for this? Considered that, but if you press Power too long the hardware turns off the system. AFAIK the kernel can not interfere with this. Not sure if this would work for others, but personally I tend to hold the sysrq sequence for a while (seconds) while I go for that command button. Maybe I'm just slow :). Thanks for the thoughts. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[PATCH] Re: Weird display problems on wakeup from sleep (sometimes)
Looks like the same issue I sometimes have. You have an ATI graphics
card, right?
If so, get the lcd_hack too that Ben H. wrote. If you get an error
when running the tool, you may need the kernel patch from
http://lists.debian.org/debian-powerpc/2005/01/msg00219.html
Running the tool fixes the problem for me 90% of the time on 2.6.11,
and always fixed it on 2.6.8 and earlier.
Hmm, looks like it never got upstream. Can someone please push
the patch below? Please see the link above for explanation and approvals.
It applies clean to 2.6.11 as well.
Martin
Signed-off-by: Martin Habets <[EMAIL PROTECTED]>
---
--- 2.6.8/drivers/video/aty/radeon_base.c.orig 2005-01-07 16:55:49.0
+
+++ 2.6.8/drivers/video/aty/radeon_base.c 2005-01-07 16:57:22.0
+
@@ -1812,9 +1812,12 @@ static int __devinit radeon_set_fbinfo (
fb_alloc_cmap(&info->cmap, 256, 0);
- if (noaccel)
+ if (noaccel) {
info->flags |= FBINFO_HWACCEL_DISABLED;
+ info->fix.accel = FB_ACCEL_NONE;
+ } else
+ info->fix.accel = FB_ACCEL_ATI_RADEON;
return 0;
}
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Booting off usb.
On Tue, Oct 19, 2004 at 11:20:24AM -0700, Mike Power wrote: > I am having trouble getting my iBook G4 USB 2.0 to boot off a SanDisk > Cruzer Micro 256 USB2.0 flash drive. I have the flash drive partitioned > with OSX's disk utility to have one partition on it with a HFS Standard > filesystem on it. On the filesystem I made a directory called boot. > Put in the directory a vmlinux, yaboot, yaboot.conf, initrd, ofboot.b. Make sure your kernel recognises. the HFS filesystem. Did you run ybin for this disk? > I use the following commands to display the location of the usb drive > > dev usb0 > ls > pwd > > The commands display the following output > ok > ff9c7700: /[EMAIL PROTECTED] > ok > /[EMAIL PROTECTED]/[EMAIL PROTECTED],1 ok > > I use the following command in open firmware to try and boot the device. > boot usb0/[EMAIL PROTECTED]:,\boot\yaboot > or > boot usb0/[EMAIL PROTECTED]:2,\boot\yaboot > or > boot /[EMAIL PROTECTED]/usb/@1b,1/[EMAIL PROTECTED]:,\boot\yaboot > or > boot /[EMAIL PROTECTED]/usb/@1b,1/[EMAIL PROTECTED]:2,\boot\yaboot > > After I execute the command I get a long pause, about 17 seconds, then I > get the following error > can't OPEN: usb0/[EMAIL PROTECTED]:,\boot\yaboot > > Does anyone know what I am missing when it comes to booting off usb > devices with an iBook? Maybe you need to do something like: boot usb0/[EMAIL PROTECTED]:,\\:tbxi or boot usb0/[EMAIL PROTECTED]:2,\\:tbxi Martin
Re: works, but stuck at 640 x 480
1) Make sure you use the framebuffer device. In you /etc/X11/XF86Config-4: Option "UseFBDev" 2) Look at /var/log/XFree86.0.log to see what X tries to do, and how it end up selecting 640x480. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets ---
Re: Configuring printing on Debian (Performa6230CD)
On Thu, Oct 28, 2004 at 06:20:53PM -0200, Gunther Furtado wrote: > Hi, > > Can anyone help me to set up a StyleWriter II and/or an ImageWriter II on a > Debian on a Performa6230CD. I was able to run http://localhost:631 but I > don't know exactly how to put down on some fields like location. I've tried > '/dev/printer' whit no success. Try /dev/lp0. -- Martin
Re: Alsa configuration problem
What do you have in /proc/asound/devices ? Do you have the alsa devices in /dev/snd/ ? Martin
Re: 2.6.4 kernel
> box: Tibook 1GHz G4, ATI radeon 9000M, 10/100/1000 > ethernet, orinico airport card(not extreme), > cd-rw/dvd-r, and a winmodem (nothing is perfect). > > i have a 2.4.21-ben2 kernel, and want to upgrade to > 2.6 series, so i have some questions before to make > the change: > > 2.6.4 has all the powerpc stuff?(no need of ben tree?) I still see Ben submitting some patches for some sleep stuff, so I'm going to stay on his tree for now. > I need to upgrade from modutils to module-init-tools? > (for module stuff) Yes, definitely. Before you boot your new 2.6 kernel. > ide-scsi emulation no longer needed? So they say, but I haven't tested it yet. > As far as i know hfsplus is not supported (yet) in the > 2.6 series, is this true? Yes, unfortunately. > i need to install alsa? If you go with that, which is recommended. You can still select the old OSS stuff in the kernel config. For alsa I think you need at least alsa-base and alsa-utils. > can i steel have 3d accel?(dri-trunk of daenzer? comes > included with the kernnel?) It works for my radeon 7500 using unstable. I need to use the lcd_hack program from Ben sometimes. There is a conflict between the radeon driver in DRI mode and some real-time programs like qjackctl and rosegarden4 (if you're into that). So I don't load DRI in that case. > I need kernel-headers-2.6.4-? (to recompile mol and > modem)Where can i get those? Have not seen a package for this (yet). > Can i have cpufreq? or still causes the box turn off > and clock reset to dec 31 1969? (by openfirmware >
Re: 2.6.4 kernel
To Linus ?? To this here list (the pmac serial sleep patch comes to mind)?? While Ben might still update linuxppc-2.5-benh occasionally, he said to use Linus' tree now. Yes, I know, but Martin seemed to say that Ben was still updating his own tree, which is why I was wondering. I was wrong. I've seen the patches submitted to this list and others, and I *assumed* they would be in his tree soon. After catching up on my reading I understand this won't happen any more (good thing). Sorry for the confusion. Which, of course, might be as broken as one may expect every now and then. I really enjoyed being one step removed from the lkm development so far :-) Until one of the ppc fixes breaks x86... That'll be something else for a change! :) Martin
Re: Problems with CD in 2.6.5
First of all, check your dmesg output for ide and hde (and maybe SCSI) related messages. Post that if it doesn't make complete sense. It seems strange to me that your cdrom drive would be hde. Do you have disks on hdb thru hdd? Check lsmod to see if any scsi modules are loaded (I don't think they should be). Regarding you .config, I have CONFIG_IDE_TASK_IOCTL turned on as well. Maybe most CD players use this raw access, I don't know. Martin
Re: problems with benh and suspend to disk
Your config looks ok to me. Do you have drivers/char/nvram.o? If so, remove it and rebuild. If that doesn't do it, try a make clean and rebuild. Martin
Re: problems with benh and suspend to disk
So my guess now is that you don't have drivers/char/nvram.o, but have generic_nvram.o in stead... Pleas confirm this. What kernel version are you on? I also have the function in arch/ppc/kernel/setup.c, along with an EXPORT_SYMBOL() for it. Do you have such code? Martin --- 30 years from now GNU/Linux will be as redundant a term as UNIX/MERT is today. - Martin Habets -- On Mon, Apr 12, 2004 at 03:59:23PM +0200, Maciej Jan Broniarz wrote: > On Mon, Apr 12, 2004 at 01:07:46PM +0100, Martin Habets wrote: > > > I did what you said - and heres what i get: > > drivers/built-in.o(.text+0x6b278): In function `platinum_init_fb': > : undefined reference to `nvram_read_byte' > drivers/built-in.o(.text+0x6b298): In function `platinum_init_fb': > : undefined reference to `nvram_read_byte' > drivers/built-in.o(.init.text+0x4f80): In function `init_control': > : undefined reference to `nvram_read_byte' > drivers/built-in.o(.init.text+0x5004): In function `init_control': > : undefined reference to `nvram_read_byte' > drivers/built-in.o(.init.text+0x5854): In function > `valkyrie_choose_mode': > : undefined reference to `nvram_read_byte' > drivers/built-in.o(.init.text+0x5874): more undefined references to > `nvram_read_byte' follow > make: *** [.tmp_vmlinux1] Error 1 > > Any idea what is wrong ? > > gausus >
Re: problems with benh and suspend to disk
On Wed, Apr 14, 2004 at 08:40:08AM +0200, Maciej Jan Broniarz wrote: > On Wed, Apr 14, 2004 at 02:48:42AM +0100, Martin Habets wrote: > > So my guess now is that you don't have drivers/char/nvram.o, > > but have generic_nvram.o in stead... Pleas confirm this. > > > wolverine:/usr/src/linux/drivers/char# ls generic_nvram* > generic_nvram.c > > i have no generic_nvram.o so the module doesnt build i guess I think this is causing the build errors. Do you have nvram.o? In the config you send earlier you had: CONFIG_GENERIC_NVRAM=y so generic_nvram.o should have been build. Is this still the ".config" file you have? Check your Makefile and Kconfig in drivers/char. Not sure what to tell you to look for... something wrong... > > What kernel version are you on? > > > v2.6.5-rc3-ben0 patched for suspend-2-disk Ok, have 2.6.5 here. Should be close enough. > > I also have the function in arch/ppc/kernel/setup.c, along with > > an EXPORT_SYMBOL() for it. Do you have such code? > wolverine:/usr/src/linux/arch/ppc/kernel# grep nvram setup.c > #include > nvram_sync(); > nvram_sync(); > nvram_sync(); > /* Generic nvram hooks used by drivers/char/gen_nvram.c */ > unsigned char nvram_read_byte(int addr) > if (ppc_md.nvram_read_val) > return ppc_md.nvram_read_val(addr); > EXPORT_SYMBOL(nvram_read_byte); > void nvram_write_byte(unsigned char val, int addr) > if (ppc_md.nvram_write_val) > ppc_md.nvram_write_val(addr, val); > EXPORT_SYMBOL(nvram_write_byte); > void nvram_sync(void) > if (ppc_md.nvram_sync) > ppc_md.nvram_sync(); > EXPORT_SYMBOL(nvram_sync); > > so it looks fine to me. Yes, looks good. But if CONFIG_GENERIC_NVRAM is not set (or it thinks so), the code won't be compiled. Martin
