Bug 133249 - PCMCIA - Acer TravelMate 8100 - SUSE LINUX Boot DVD 10.0 oopses in yenta PCMCIA driver
Summary: PCMCIA - Acer TravelMate 8100 - SUSE LINUX Boot DVD 10.0 oopses in yenta PCMC...
Status: RESOLVED FIXED
Alias: None
Product: SUSE LINUX 10.0
Classification: openSUSE
Component: Kernel (show other bugs)
Version: Final
Hardware: i686 Other
: P5 - None : Normal
Target Milestone: ---
Assignee: Jaroslav Kysela
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-11-10 13:41 UTC by Jaroslav Kysela
Modified: 2006-01-03 10:28 UTC (History)
0 users

See Also:
Found By: Development
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---

Attachments
PCI subordinate initialization fix from the 2.6 mainstream (1.24 KB, text/x-diff)
2005-11-11 10:38 UTC, Jaroslav Kysela 		Details
Prevent Yenta pcmcia driver to hang machine when subordinate for the PCI device is NULL (617 bytes, text/x-diff)
2005-11-11 10:41 UTC, Jaroslav Kysela 		Details
57055: Prevent Yenta pcmcia driver to hang machine when subordinate for the PCI device is NULL v2 (615 bytes, text/x-diff)
2005-11-11 10:50 UTC, Jaroslav Kysela 		Details
Show Obsolete (1) View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Jaroslav Kysela 2005-11-10 13:41:36 UTC 
SUSE LINUX Boot DVD 10.0 oopses in yenta PCMCIA driver on Acer TravelMate 8100 notebook. I will try to collect more information - the crash is caused in yenta_config_init+0x87/0x100 code. 10.0 cannot be installed, becuase the kernel hangs when next module (ahci) is loaded, of course.
Comment 1 Jaroslav Kysela 2005-11-10 19:57:13 UTC 
The oops is from yenta_config_init() function drivers/pcmcia/yenta_socket.c -
the PCI dev->subordinate member is NULL.

It seems that the 'pci=assign-busses' should be passed as the boot option, but for Acer TM81xx notebooks should be applied also this patch (I will try it) from 2.6 mainstream:

commit 12f44f46bc9c6dc79255e661b085797af395d8da
tree cec3ab7b0bfd7f90fba01b6fb6ecb896eed00079
parent bfab08c097df5f6a60dd07f0f8c78c54e725780b
author Ivan Kokshaysky <ink@jurassic.park.msu.ru> Thu, 22 Sep 2005 21:06:31 -0700
committer Linus Torvalds <torvalds@g5.osdl.org> Fri, 23 Sep 2005 08:05:16 -0700

    [PATCH] pci: fixup parent subordinate busnr

    I believe the change that broke things is introduction of
    pci_fixup_parent_subordinate_busnr().

    The patch here does two things:
    - hunk #1 should fix the problems you've seen when you boot without
    additional "pci" kernel options;
    - hunk #2 supposedly fixes boot with "pci=assign-busses" option which
    otherwise hangs Acer TM81xx machines as reported.

    Please try this with and without "pci=assign-busses". If it boots,
    I'd like to see 'lspci -vvx' for both cases.

    Signed-off-by: Andrew Morton <akpm@osdl.org>
    Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Comment 2 Jaroslav Kysela 2005-11-11 10:38:34 UTC 
Created attachment 57053 [details]
PCI subordinate initialization fix from the 2.6 mainstream
Comment 3 Jaroslav Kysela 2005-11-11 10:41:08 UTC 
Created attachment 57055 [details]
Prevent Yenta pcmcia driver to hang machine when subordinate for the PCI device is NULL
Comment 4 Jaroslav Kysela 2005-11-11 10:50:13 UTC 
Created attachment 57059 [details]
 57055: Prevent Yenta pcmcia driver to hang machine when subordinate for the PCI device is NULL v2
Comment 5 Jaroslav Kysela 2005-11-21 13:37:48 UTC 
Seems like a dup of bug#113778, but the used DVD was 10.0 final from ftp.suse.com,  so probably the fixed kernel is not there.
Comment 6 Jaroslav Kysela 2006-01-03 10:28:47 UTC 
This problem is fixed in upcoming 10.1 kernel.