Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 87650 - 2005.0 hangs after Penguin with PCI resource allocation error for ATI Radeon 9200
Summary: 2005.0 hangs after Penguin with PCI resource allocation error for ATI Radeon ...
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Release Media
Classification: Unclassified
Component: Everything (show other bugs)
Hardware: PPC Linux
: High normal (vote)
Assignee: Gentoo Release Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-04-02 03:49 UTC by Mike Hammill
Modified: 2005-09-01 16:11 UTC (History)
1 user (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Mike Hammill 2005-04-02 03:49:15 UTC
When I boot from the 2005.0 minimal LiveCD, dated 2005-03-20, it gets no further than putting up the Penguin image and then printing the single line message:

PCI: Cannot allocate resource region 1 of device 0001:11:02.0

The same systems boots from the LiveCD 2004.3 minimal disk (2004-10-30) without problems. In that environment, an lspci shows that the device is 
0001:11:02.0 VGA compatible controller ATI Technologies Inc RV280 [Radeon 9200] (rev 01) 

Indeed, this is the only PCI or AGP video card in the machine.  An lspci listing can be provided if you like, since I can boot the machine using a gentoo-source kernel I configured manually using the minimal LiveCD 2004.3.

Reproducible: Sometimes
Steps to Reproduce:
1. Insert LiveCD 2005.0 disk into disk drive
2. Boot Mac with "c" key held down (which tells open firmware to boot from cdrom)
3. At kernel selection screen, choose G4
    Alternatively, choose G4 nodetect
    Alternatively, choose G4 nodetect video=radeon:1028x1024
    Alternatively, choose G4 video=radeon:1028x1024
Actual Results:  
The Penguin appear then directly below it, with nothing else on the screen, the one line error message 
given above. This happened the approx. 6-8 times I tried it with 2 exceptions.  In both of the exception 
situation, I had entered the nodetect option, but oddly, still received the PCI error message above.  
However, it then went on to at least try to boot up.  However, I then got another error message (which I 
believe to be bogus as well) that my root partition is problematic:

"EXT3-fs warning: maximal mount count reached, running e2fsck is recommended". 

I have double checked my root ext3 partition using e2fsck and no errors were found.

Expected Results:  
Booted into the PPC Gentoo LiveCD (minimal) 2005.0 environment.

Machine:
PowerMac G4 400MHz, model: "Digial Audio"
The "top" command gives: 255360k memory
New disk drive: Seagate 200GB
No card in the AGP slot, only a new ATI Radeon 9200 in the first PCI slot.  
No accelerators, overclocking, scsi, or other paraphenalia.  One of your more basic PPC systems.  
Basically, an old machine the university was throwing away that I wanted to use Gentoo on.
http://forums.gentoo.org/viewtopic-t-317356-highlight-.html
Problem has also been posted to the forums, with no response so far.
Comment 1 Lars Weiler (RETIRED) gentoo-dev 2005-05-07 16:33:17 UTC
Sorry for the long delay.

I tested the 2005.0-CD with a PowerMac 533.  The boot up takes quite long and there is a phase (when only the penguin is shown) where the CDs seems to hang for about a minute or two.  Well, this machine has an AGP and a PCI-card built in.  I can also test without the AGP-card.

Furthermore, you should use video=radeonfb: instead of video=radeon:
Comment 2 Mike Hammill 2005-05-11 14:06:17 UTC
I took the time to test the same 2005.0 CD 10 times on the same machine.  1 out of 10 times, it made it beyond the error message described before.  All the other times (9 of 10), it hangs, just as I stated.  I let the machine sit over 30 minutes after the error message hoping it was just an issue with slowness, but nothing happened so slowness is not the issue.   The machine is running Gentoo with a 2.6.10-gentoo-r6 kernel, so I think the problem is with the CD, not the machine.  I do have one other issue (which I'll submit as a bug now).  It is that "emerge" (metadata stage) of all things, hangs this same machine.  Nothing else does.

By the way, this machine is a G4.  /proc/cpuinfo gives:

audi root # cat /proc/cpuinfo 
processor       : 0
cpu             : 7400, altivec supported
temperature     : 23-25 C (uncalibrated)
clock           : 400MHz
revision        : 2.9 (pvr 000c 0209)
bogomips        : 794.62

total bogomips  : 794.62
machine         : PowerMac3,3
motherboard     : PowerMac3,3 MacRISC2 MacRISC Power Macintosh
detected as     : 65 (PowerMac G4 AGP Graphics)
pmac flags      : 00000004
L2 cache        : 1024K unified
memory          : 256MB
pmac-generation : NewWorld
Comment 3 Mike Hammill 2005-05-11 14:10:56 UTC
I forgot to add that I tried video=radeonfb on a few of the 10 trial (just as I had before submitting the bug report).  I have NEVER made it beyond the hanging state with that setting.  I have only made it beyond it without using it (i.e., no kernel options, just "G4").  However, 10+ trial is not extensive testing.
Comment 4 Chris Gianelloni (RETIRED) gentoo-dev 2005-05-12 06:08:37 UTC
This is on an IDE CDROM, correct?

Have you tried booting with ide=nodma just to be sure that it isn't a buggy MDA implimentation for your chipset?
Comment 5 Mike Hammill 2005-05-21 13:30:29 UTC
(In reply to comment #4)
> This is on an IDE CDROM, correct?
> 
> Have you tried booting with ide=nodma just to be sure that it isn't a buggy MDA implimentation for 
your chipset?

Yes, IDE CD-ROM.  According to dmesg it's a:
hdc: CD-ROM CDU311, ATAPI CD/DVD-ROM drive
hdc: ATAPI 8x CD-ROM drive, 256kB Cache, DMA
Uniform CD-ROM driver Revision 3.20

I tried ide=nodma by itself.  (i.e., G4 ide=nodam).  I got the same message.
I also tried it in cobination with video=radeonfb and even with nodetect.  All got the same response.

By the way, in the forum thread (mentioned earlier), I see someone else has now posted they get the 
same error message.

I think I already noted that I get this error message with my currently running system (originally 
installed with the 2004.3 CD.  I tried the card in a similar machine with ubuntu's install disk and the 
same error pops up, but the installer does not hang on it.  I also tried the FC3 boot disk and it did not 
show the message or hang.  In short, my current gentoo system does not hang on it.  Ubuntu install 
disk 1 and FC3 do not hang on it (in another machine), and, others are getting the same message.  I 
think it might be time for a serious look into the matter.
Comment 6 Mike Hammill 2005-07-04 14:30:22 UTC
A little good news on this problem.  I found that if I specify certain kernel options, I can move beyond 
the hanging point.  Specifically, if I use:

video=ofonly  (it often continues beyond the PCI error)
debug (it always continues beyond the PCI error)

More info:  I have updated by ATI Radeon 9200 (a PCI card) with the newest ROM update from ATI:
ATI 4.5.1
Card model: ATY, RV280
ROM: ATY, Bugsy_A
FCode version: 1.94
NDRV version: 1.0.1f59
VRAM: 128 MB

Other info:

The PPC Rawhide boot disk from the Fedora PPC project (June 18, 2005) also hangs with the same 
message.  Both Fedora kernels 2.6.11-1.1381_FC5 and 2.6.12-1.1411_F are effected by exactly the 
same bug.  The Ubuntu 5.04 LiveCD (April 6, 2005), however, does not have this problem and X even 
works so there must be a solution.

I'd still love to have Gentoo working on a PPC machine.  Hope the above information helps.  If I can 
provide anything else, let me know.
Comment 7 Joe Jezak (RETIRED) gentoo-dev 2005-07-11 21:50:52 UTC
What kernel was used on the Ubuntu CD that worked?  Since you were able to boot
with other CDs did you succeed in installing Gentoo or another version of Linux?
 If so, which kernel version and what settings are you using so you do not have
this problem?
Comment 8 Joe Jezak (RETIRED) gentoo-dev 2005-09-01 16:11:58 UTC
Please reopen if it's still a problem with 2005.1.