FS#16435 - [kernel26] Revert bad commit to joydev

Attached to Project: Arch Linux
Opened by (beomuex) - Friday, 02 October 2009, 16:42 GMT
Last edited by Paul Mattal (paul) - Saturday, 06 February 2010, 13:58 GMT
Task Type Feature Request
Category Upstream Bugs
Status Closed
Assigned To Tobias Powalowski (tpowa)
Thomas Bächler (brain0)
Architecture All
Severity Medium
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 3
Private No

Details

A commit to the joydev module in 2.6.30's upstream broke the support for joysticks with many buttons (like the Saitek X52 series).
There's a patch available to revert this bad commit.

http://bugzilla.kernel.org/show_bug.cgi?id=14049

Since many other distributions have already reverted this commit, it shouldn't affect any other functionalities of joydev.
This task depends upon

Closed by  Paul Mattal (paul)
Saturday, 06 February 2010, 13:58 GMT
Reason for closing:  Won't fix
Additional comments about closing:  It appears kernel devs have decided not to fix this. Reopen against 2.6.33 if it's not fixed in that line.
Comment by Roman Kyrylych (Romashka) - Sunday, 04 October 2009, 18:58 GMT
2.6.31.1 is going to be moved to Core soon.
I suppose this is fixed in .31 already.
Comment by Roman Kyrylych (Romashka) - Sunday, 04 October 2009, 19:05 GMT
Hm, looks like it's not fixed in .31, but there are works for fixing it for .32.
I'm not sure we should patch stuff ourselves. The commit fixes one bug and introduces another - which one is more important and for how many users?

For referrence: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/300143
Comment by (beomuex) - Sunday, 04 October 2009, 19:06 GMT
unfortunately it isn't:

http://tinyurl.com/ya77lfj

The latest patch-2.6.31.1-1-ARCH.bz2 doesn't include the fix either.
Comment by Laszlo Papp (djszapi) - Saturday, 07 November 2009, 07:41 GMT
I agree with Roman here, if I need for some patch, I rebuild the kernel myself, but if this patch is required by so much users, then it's worth to thinking of it.
Comment by Paul Mattal (paul) - Sunday, 06 December 2009, 19:38 GMT
Let's revisit this in .32. If it's fixed, we can avoid the hard decision. Otherwise, let's try to make that decision in January.

Having been in kernel patching quandaries myself, my feeling is it's okay to patch a small bug if it helps out a lot of people, but NOT if it clearly introduces some other issue. So based on my limited knowledge of this situation, I'd be in favor of not patching here if it introduces another issue.
Comment by (beomuex) - Wednesday, 09 December 2009, 15:43 GMT
Not fixed in 2.6.32 (vanilla from kernel.org).
Comment by (beomuex) - Wednesday, 09 December 2009, 16:11 GMT
---
Comment by Adam Lintz (lbcongo) - Saturday, 26 December 2009, 04:48 GMT
I'm almost certain that this regression is effecting my system. I've not ever tried my Saitek X52 on Arch, but am experiencing the symptoms of the device appearing as event in /dev/input but not as /dev/input/jsX. Is there info I can provide for a fix? Is there a known workaround?
Comment by Adam Lintz (lbcongo) - Sunday, 27 December 2009, 05:44 GMT
Hi all,

Not sure if this info is valuable, but I applied diff (from http://bugzilla.kernel.org/show_bug.cgi?id=14049), compiled kernel, and now my Saitek X52 is functioning properly.

-Adam
Comment by Paul Mattal (paul) - Monday, 04 January 2010, 02:50 GMT
I'll leave it to tpowa to decide if he wants to apply this patch. We'll close this in February, as a tacit "won't fix" if it's not done by then (since if it's not done by then, it will probably be fixed in 2.6.33).

Loading...