Re: PCI PC/Hardware in Yosemite Powermac

2005-12-04 Thread Christian Walther
Hi Robin,

I'm using standard PC cards (USB, Firewire, Network Interface) with my
Umax Pulsar without any problem - and this includes OS X. As a
guideline, you should make sure that the card (or chipset) is
supported by Linux in general, and that the code required is part of
the mainstream kernel. Any patches required might hack the kernel in a
x86 specific way.

Of course the card won't be recognised by the systems OpenFirmware, so
you can't boot from any device connected to it. What definetely won't
work are PCI cards that contain a BIOS and do specific
initializations. Cards with their own BIOSes are IDE RAID adaptors,
for example. So I'm not sure if that applies to your Serial ATA card.

If you're unsure, it would be the best to choose a dealer who allows
you to return a card that doesn't work in your setup. Even if you pay
more for a card, you're saving yourself the pain of having an unusable
card at home.
Additionally, it's not a bad idea to choose PCI cards that are older,
and thus maybe better supported. I recently had a USB2 card that was
supposed to work even with OS X (driver was included), but the card
failed every time a USB2 device was connected to it. I'm still waiting
for a reply of the dealer where I purchased the card...

HTH,
Christian



Re: yaboot and console font

2005-12-04 Thread Michel Dänzer
On Fri, 2005-12-02 at 10:56 +0100, Bin Zhang wrote:
 
 I use only CONFIG_FONT_6x11=y for setting console font in 2.6.14, It works.
 
 But It doesn't work with 2.6.15-rc2, so I add two console fonts for
 kernel 2.6.15-rc4
 CONFIG_FONT_8x16=y
 CONFIG_FONT_6x11=y
 CONFIG_FONT_7x14=y
 The kernel boots with the vga 8x16 font.
 Is there a option in yaboot.conf which let me try the Mac 6x11 font ?

Something like

fbcon=font:Mac6x11


-- 
Earthling Michel Dänzer  | Debian (powerpc), X and DRI developer
Libre software enthusiast|   http://svcs.affero.net/rm.php?r=daenzer



Re: yaboot and console font

2005-12-04 Thread Bin Zhang
On 12/4/05, Michel Dänzer [EMAIL PROTECTED] wrote:
 On Fri, 2005-12-02 at 10:56 +0100, Bin Zhang wrote:
 
  I use only CONFIG_FONT_6x11=y for setting console font in 2.6.14, It works.
 
  But It doesn't work with 2.6.15-rc2, so I add two console fonts for
  kernel 2.6.15-rc4
  CONFIG_FONT_8x16=y
  CONFIG_FONT_6x11=y
  CONFIG_FONT_7x14=y
  The kernel boots with the vga 8x16 font.
  Is there a option in yaboot.conf which let me try the Mac 6x11 font ?

 Something like

 fbcon=font:Mac6x11

Thanks for your help.
I tested
fbcon=font:Mac6x11
fbcon=font:6x11
append=font:6x11
append=font:Mac6x11
They didn't change the vga 8x16 console font.

Best regards,
Bin

 --
 Earthling Michel Dänzer  | Debian (powerpc), X and DRI developer
 Libre software enthusiast|   http://svcs.affero.net/rm.php?r=daenzer




Re: sudden power loss (WAS: sudden power loss on tibook)

2005-12-04 Thread Jesus Climent
On Mon, Nov 28, 2005 at 01:54:51PM +1100, Benjamin Herrenschmidt wrote:
 On Sun, 2005-11-27 at 20:52 -0600, Tommy Trussell wrote:
 
 At this point, I would really need you guys to do some dichotomy to try
 to isolate more precisely what kernel change introduced the problem.
 That is if you confirm it's between 2.6.11 and 2.6.12, then test the
 various -rcX and possible git/bk snapshots in between those 2 kernels or
 do a git bisect. Will take a while, but that's all I can find for now.

I have started to get the power problems after i switched to 2.6.14. Since i
purchased a zd1211 wireless card, I started from the oh dear, the driver for
this card is totally fscked but then i remembered this thread and decided to
try 2.6.12.

So far, I have been using debian kernels, and have had not a single problem
with 2.6.12 in a very long time, so i have a strong believe it has been
introduced with the 2.6.12 - 2.6.14, but that is not saying much, I believe.

So far it has started happening mostly when I do very intensive tasks, like
(mostly) transfering files, at 2.2Mps, so that is why i thought it was related
to the zd1211 driver, but tonight it also happened when transfering files to
my mp3 player, although i got an Oops, not a shutdown, which of course I dod
not copy :(

I will try 2.6.12 with the zd1211 and see if it still happens.

Cheers,

-- 
Jesus Climent  info:www.pumuki.org
Unix SysAdm|Linux User #66350|Debian Developer|2.6.14|Helsinki Finland
GPG: 1024D/86946D69 BB64 2339 1CAA 7064 E429  7E18 66FC 1D7F 8694 6D69

We seem to be made to suffer. It's our lot in life.
--C-3PO (Star Wars)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: etch graphical installer does not work on G5 PowerMac?

2005-12-04 Thread Benjamin Herrenschmidt
On Sat, 2005-12-03 at 08:05 -0800, Shyamal Prasad wrote:

 Yes, it is an Nvidia GeForce card, unfortunately that is what Apple
 sells with these machines. I'll swap it when I can but for now I'm
 stuck with it.
 
 So, I tried the daily build. It boots fine, but the installer (more
 likely the fb itself) does not do anything useful with the nvidia
 card. It just flashes (white) the screen (black background) every few
 seconds. I can switch to a vt and use the shell, but the screen
 continues the flash at the same rate.

nvidiafb  on recent kernels should work fine, as should the X nv
driver...

Ben.



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: etch graphical installer does not work on G5 PowerMac?

2005-12-04 Thread Sven Luther
On Mon, Dec 05, 2005 at 09:12:21AM +1100, Benjamin Herrenschmidt wrote:
 On Sat, 2005-12-03 at 08:05 -0800, Shyamal Prasad wrote:
 
  Yes, it is an Nvidia GeForce card, unfortunately that is what Apple
  sells with these machines. I'll swap it when I can but for now I'm
  stuck with it.
  
  So, I tried the daily build. It boots fine, but the installer (more
  likely the fb itself) does not do anything useful with the nvidia
  card. It just flashes (white) the screen (black background) every few
  seconds. I can switch to a vt and use the shell, but the screen
  continues the flash at the same rate.
 
 nvidiafb  on recent kernels should work fine, as should the X nv
 driver...

Well, this is 2.6.14, is this recent enough ? And the problem is not so much
plain fbdev, but the directfb and gtk-directfb layers over it. 

Could you try passing the directfb noaccel option ? not sure exactly how, but
Attilio could help you there. This is similar of what happens with atyfb and
rage 128 chips. Right now we only had success on the x86 vesafb and radeonfb.

Friendly,

Sven Luther


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: etch graphical installer does not work on G5 PowerMac?

2005-12-04 Thread Benjamin Herrenschmidt
On Sun, 2005-12-04 at 23:34 +0100, Sven Luther wrote:
 On Mon, Dec 05, 2005 at 09:12:21AM +1100, Benjamin Herrenschmidt wrote:
  On Sat, 2005-12-03 at 08:05 -0800, Shyamal Prasad wrote:
  
   Yes, it is an Nvidia GeForce card, unfortunately that is what Apple
   sells with these machines. I'll swap it when I can but for now I'm
   stuck with it.
   
   So, I tried the daily build. It boots fine, but the installer (more
   likely the fb itself) does not do anything useful with the nvidia
   card. It just flashes (white) the screen (black background) every few
   seconds. I can switch to a vt and use the shell, but the screen
   continues the flash at the same rate.
  
  nvidiafb  on recent kernels should work fine, as should the X nv
  driver...
 
 Well, this is 2.6.14, is this recent enough ? And the problem is not so much
 plain fbdev, but the directfb and gtk-directfb layers over it. 

I fixed various issues with nvidiafb in 2.6.15, the patches didn't make
it in time for 2.6.14.

 Could you try passing the directfb noaccel option ? not sure exactly how, but
 Attilio could help you there. This is similar of what happens with atyfb and
 rage 128 chips. Right now we only had success on the x86 vesafb and radeonfb.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: PowerBook5,8 - TrackPad update

2005-12-04 Thread Michael Hanselmann
On Fri, Dec 02, 2005 at 03:28:31PM +0100, Stelian Pop wrote:
 Is this version really working well on the new Powerbooks ? From what
 I've seen in this thread there are still issues and it's still a work
 in progress, so it may be too early to integrate the changes in the
 kernel.

It works fine for the 15 PowerBooks (Oct 2005) and has been tested by
at least three people. 17 should work too, but I can't test that until
later this week.

 +#if defined(CONFIG_RELAYFS_FS) || defined(CONFIG_RELAYFS_FS_MODULE)
 +#include linux/relayfs_fs.h
 +#endif

 While the relayfs code is ok for debugging, I'm wondering if it should
 be left in the final version at all.

It doesn't hurt as it's not much code, it makes updating the driver for
newer devices easier and it's only enabled if requested. Because of
that, I would leave it in. However, I can split it into a separate patch
if wanted.

 +   int overflowwarn;   /* overflow warning printed? 
 */

 I would use a static variable in the case -OVERFLOW: block here.

As there could be, due to whatever reason, multiple devices using the
same driver, one could overflow while another doesn't. Therefore, it
should be device specific.

 +   dev-xy_cur[i++] = dev-data[19];
 +   dev-xy_cur[i++] = dev-data[20];
 +   dev-xy_cur[i++] = dev-data[22];
 +   dev-xy_cur[i++] = dev-data[23];

 There is obviously a pattern here:

   for (i = 0; i  15; i++)
   dev-xy_cur[i] = dev-data[ 19 + (i * 3) / 2 ]

It's not that easy. This code wouldn't work as it should.

I wrote a working loop into the patch below, but as the code's called
about 1'000 times a second I unrolled the loop myself. A friend of mine
calculated that the code would take at least twice the time to run when
written using a loop.

 I'm wondering if the same formula doesn't apply for more X and Y
 sensors (like 16 X and 16 Y sensors on the old Powerbooks, 26 for the
 17 models)

It does and is addressed in the new patch below.

 What is the point in doing this since the dbg_dump is called a few lines
 later ?

Those were some leftovers from my debugging and would have been deleted
before submitting. I just didn't want to take everything out if someone
else wants to do some extensive debugging or so.

Here's an updated patch including support for the 17 PowerBooks (Oct
2005). The informations about the 17 one are from Alex Harper.

---
--- linux-2.6.15-rc5/drivers/usb/input/appletouch.c.orig2005-12-04 
20:25:21.0 +0100
+++ linux-2.6.15-rc5/drivers/usb/input/appletouch.c 2005-12-04 
23:37:29.0 +0100
@@ -6,9 +6,19 @@
  * Copyright (C) 2005  Stelian Pop ([EMAIL PROTECTED])
  * Copyright (C) 2005  Frank Arnold ([EMAIL PROTECTED])
  * Copyright (C) 2005  Peter Osterlund ([EMAIL PROTECTED])
+ * Copyright (C) 2005  Parag Warudkar ([EMAIL PROTECTED])
+ * Copyright (C) 2005  Michael Hanselmann ([EMAIL PROTECTED])
  *
  * Thanks to Alex Harper [EMAIL PROTECTED] for his inputs.
  *
+ * Nov 2005 - Parag Warudkar 
+ *  o Added ability to export data via relayfs
+ *
+ * Nov 2005 - Michael Hanselmann
+ *  o Compile relayfs support only if enabled in the kernel
+ *  o Enable relayfs only if requested by the user
+ *  o Added support for new October 2005 PowerBooks (Geyser 2)
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
@@ -35,8 +45,13 @@
 #include linux/input.h
 #include linux/usb_input.h
 
+#if defined(CONFIG_RELAYFS_FS) || defined(CONFIG_RELAYFS_FS_MODULE)
+#include linux/relayfs_fs.h
+#endif
+
 /* Apple has powerbooks which have the keyboard with different Product IDs */
 #define APPLE_VENDOR_ID0x05AC
+#define GEYSER_2_PRODUCT_ID0x0215
 
 #define ATP_DEVICE(prod)   \
.match_flags = USB_DEVICE_ID_MATCH_DEVICE | \
@@ -51,14 +66,17 @@
 static struct usb_device_id atp_table [] = {
{ ATP_DEVICE(0x020E) },
{ ATP_DEVICE(0x020F) },
+   { ATP_DEVICE(GEYSER_2_PRODUCT_ID) },/* PowerBooks Oct 2005 */
{ ATP_DEVICE(0x030A) },
{ ATP_DEVICE(0x030B) },
{ } /* Terminating entry */
 };
 MODULE_DEVICE_TABLE (usb, atp_table);
 
-/* size of a USB urb transfer */
-#define ATP_DATASIZE   81
+#if defined(CONFIG_RELAYFS_FS) || defined(CONFIG_RELAYFS_FS_MODULE)
+struct rchan* rch = NULL;
+struct rchan_callbacks* rcb = NULL;
+#endif
 
 /*
  * number of sensors. Note that only 16 instead of 26 X (horizontal)
@@ -73,6 +91,7 @@
 
 /* maximum pressure this driver will report */
 #define ATP_PRESSURE   300
+
 /*
  * multiplication factor for the X and Y coordinates.
  * We try to keep the touchpad aspect ratio while still doing only simple
@@ -108,6 +127,8 @@
signed char xy_old[ATP_XSENSORS + 

Proprietary video formats on PPC.

2005-12-04 Thread Federico Pistono
Hi all,
I'm sure everyone during his experience with Linux on a PPC had to
deal with the problem of proprietary video formats. At first I tried
all the possible hacks I found googling, like the experimental use of
ffmpeg on wmv3

http://multimedia.cx/eggs/?p=129
http://ffmpeg.sourceforge.net/download.php

and other projects, but none of them seemed to work. I always wondered
why could we use the OS X libraries, but I guess there's a problem of
compatibility that I do not understand.

After the breaking news that Apple was switching to x86 I felt like I
had to accept the fact that I could never watch a wmv, asf, rm, flash,
on my notebook under Linux.

Don't get me wrong, I HATE these formats, they are lossy, they're
closed source and most of the footage you find is low level quality.
But there is tons of interesting stuff that for some stupid reason
(somebody said Windows?) is reachable only with these formats, so you
loose a lot of info, unless of course you:
1. Reboot, this doesn't sound like a decent solution to me
2. Run the Mac On Linux, a bit better, but I've got some problems with
the audio, though
3. Use a x86 machine that is handy at the time. Hey, if I wanted an
x86 I could have just bought an IBM notebook...

Then something happened. You all know about the XboX 360, right? The
console hardware is based on a custom IBM PowerPC-based Xenon
central processing unit (CPU), and it has functions of a media
center (we all know that's not true, but let it be. Vorbis, uh?).

So, if they made the 360 working (not too well
http://games.slashdot.org/article.pl?sid=05/11/23/138200tid=128tid=211tid=10),
they also made proprietary codecs and DLL for PPC right?

If this project grow as it predecessor:
http://www.xbox-linux.org/wiki/Main_Page
shouldn't we then be able to play the damn videos?


Just wondering..

--
  Federico Pistono
  Department of Computer Science - Verona, Italy
  My website : http://www.federicopistono.org
  My personal website  http://pain.altervista.org
  Ex-College Website : http://pain.altervista.org/flatnukeuwcad/
  Linux Registered User :: #340392
   Though I walk through the valley of the shadow of death,
I will fear no evil: for thou art with me;
thy rod and thy staff they comfort me.



Re: Proprietary video formats on PPC.

2005-12-04 Thread Hisham Muhammad
On 12/5/05, Federico Pistono [EMAIL PROTECTED] wrote:
 Hi all,
 I'm sure everyone during his experience with Linux on a PPC had to
 deal with the problem of proprietary video formats. At first I tried
 all the possible hacks I found googling, like the experimental use of
 ffmpeg on wmv3

You can play wmv3 on PPC using Jon Johansen's patches for VLC:

http://www.nanocrew.net/2005/09/01/compiling-vlc/

Hope that helps,

-- Hisham



Re: Proprietary video formats on PPC.

2005-12-04 Thread Dean Hamstead

After the breaking news that Apple was switching to x86 I felt like I
had to accept the fact that I could never watch a wmv, asf, rm, flash,
on my notebook under Linux.


most wmv and asf files will play in mplayer (using ffmpeg and friends)
some rm's will also. i can honestly say that mplayer beats windows
hands down.


Don't get me wrong, I HATE these formats, they are lossy, they're
closed source and most of the footage you find is low level quality.


lossey codecs arent bad. ie jpeg and mpegs are both lossey but
use tricks to lose stuff thats irrelevant (psychotropic compression
i believe its called)


Then something happened. You all know about the XboX 360, right? The
console hardware is based on a custom IBM PowerPC-based Xenon
central processing unit (CPU), and it has functions of a media
center (we all know that's not true, but let it be. Vorbis, uh?).


there will be similar problems as with using darwin binaries in
linux.

try mplayer, you should be able to watch most files.

Dean
--
WWW: http://deanpatrick.tk
LAN: http://www.bong.com.au
EMAIL: [EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Proprietary video formats on PPC.

2005-12-04 Thread Jack Malmostoso
On Mon, 05 Dec 2005 07:00:10 +0100, Federico Pistono wrote:

 I could never watch a wmv, asf, rm, flash, on my
 notebook under Linux.

You can watch Real Video: Real Player is distributed also for PPC linux.
Not a free solution, but still, most online content is offered for both
Real and WMV.

-- 
Best Regards, Jack
Linux user #264449
Powered by Debian PPC


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]