[Bug 34508] 2.6.15-19 kernel fails to boot on ppc machine
Ben Stanley
Ben.Stanley at exemail.com.au
Thu Apr 6 04:07:39 UTC 2006
I updated from a perfectly working Breezy to Dapper by apt-get dist-upgrade on 5/4/06, and now I have a wonderful paperweight that won't boot. The machine is a 2000 PowerBook "Pismo" model. (/proc/cpuinfo says PowerBook3). The installed kernel version is 2.6.15-19-powerpc.
When the machine tries to boot using the default (2.6.15-19) kernel, it displays only the following message:
[ 24.007715] no framebuffer address found for /pci at f0000000/ATY,RageM3pParent at 10/ATY,RageM3pB
It then hangs for a few minutes and then displays the following message:
ALERT! /dev/hda13 does not exist. Dropping to a shell!
then drops into BusyBox.
If I try to boot using the previously working Breezy kernel (2.6.12-10), I get the following behaviour:
[ ??.??????] aty128fb: Invalid ROM signature 0 should be 0xaa55
[ ??.??????] no framebuffer address found for /pci at f0000000/ATY,RageM3pParent at 10/ATY,RageM3pB
Segmentation fault
<repeats for many screenfulls>
ALERT! /dev/hda13 does not exist. Dropping to a shell!
[Note that the messages about aty128fb and RageM3p were also displayed during normal operation of Breezy.]
If I try to boot using Linux debug at the yaboot prompt, it probes the ide devices (not the partition tables), starts up the TCP stack, and the last few messags are:
[ 28.852783] Freeing unused kernel memory: 180k init
[ 29.013427] Capability LSM initialized
[ 29.121935] Found KeyWest i2c on "uni-n", 2 channels, stepping: 4 bits
[ 29.128864] Found KeyWest i2c on "mac-io", 1 channel, stepping: 4 bits
Then it hangs for a little while and drops into BusyBox.
These tests were conducted with the laptop connected only to the power supply, an ethernet and a Logitech 3 button USB wheelmouse. Removing the mouse did not seem to have any effect.
The issue I am describing here seems to be a little different from that described by others.
--
2.6.15 kernel fails to boot on ppc machine
https://launchpad.net/malone/bugs/34508
More information about the kernel-bugs
mailing list