Summary: | wacom: horrifically out of sync from upstream | ||||||
---|---|---|---|---|---|---|---|
Product: | xorg | Reporter: | Daniel Stone <daniel> | ||||
Component: | Input/other | Assignee: | Xorg Project Team <xorg-team> | ||||
Status: | RESOLVED WONTFIX | QA Contact: | |||||
Severity: | minor | ||||||
Priority: | lowest | CC: | ajax, boris, dberkholz, pingc, researchlab | ||||
Version: | git | ||||||
Hardware: | All | ||||||
OS: | All | ||||||
Whiteboard: | |||||||
i915 platform: | i915 features: | ||||||
Attachments: |
|
Description
Daniel Stone
2004-11-23 03:04:38 UTC
Created attachment 1358 [details] [review] huge, but very well-tested, patch Comment on attachment 1358 [details] [review] huge, but very well-tested, patch make roland less angry (In reply to comment #0) > Since there is no parallel development going on and it has been well-tested by > most distributions (I believe Red Hat incorporates it also), please, please, > incorporate it for the next release, or drop the Wacom driver from the tree. We actually package the linuxwacom.sf.net drivers seperately in the linuxwacom rpm and disable the in-tree wacom drivers. I think this is the better way to go, with the modularization on the horizon. I would expect that the in-tree driver just goes away when we modularize the distribution, and linuxwacom.sf.net becomes the oficially sanctioned wacom driver. While I think the code drop is a bit much for a point-release, I don't see it affecting the overall server stability, and I agree that you probably can't break the in-tree driver more than it already is. It's worth checking out the discussion in bug #366 when considering this. > We actually package the linuxwacom.sf.net drivers seperately in the linuxwacom
> rpm and disable the in-tree wacom drivers. I think this is the better way to
> go, with the modularization on the horizon. I would expect that the in-tree
> driver just goes away when we modularize the distribution, and linuxwacom.sf.net
> becomes the oficially sanctioned wacom driver.
We just have it in our #000_* patches, which we don't really count in our 'how
many patches do we have' count, as they're typically taken from upstream anyway.
While I agree that pointing people to linuxwacom is the way to go in terms of
modularisation, the fact that the Wacom there is there by default indicates that
it's there to be used and it works, and I believe it should either be dropped,
or made vaguely useful.
Are there any objections? *** Bug 1917 has been marked as a duplicate of this bug. *** i'm going to merge this at some point i'm going to merge this at some point *** Bug 366 has been marked as a duplicate of this bug. *** while this will be addressed in future modular releases by just making linuxwacom.sf.net the official driver, we still need to resolve this for 7.0, either by merging daniel's patch or by dropping our copy of the wacom driver. just a friendly reminder. If no-one has any objections, I think we should drop our version. marked as deprecated on the wiki: http://xorg.freedesktop.org/wiki/DeprecatedInX11R7 leaving open until 7.0 release on the off chance that someone has a convincing argument for updating our version. Hi, I just saw this bug whilst investigating possible reasons for relative mode not working with the 2.6.13_rc1,2,3 kernels. Anyway, during my investigations I found that linuxwacom apparently still doesn't support dlloader. Not only is dlloader slated to be the default module loading method in new Xorgs, but also it's the only way to load modules on a pic/pie hardened system. Therefore, until this is implemented in linuxwacom I'd recommend not deprecating the in-built version which does appear to build and work with dlloader. Thanks, Mike 5:) dlloader support will take care of itself, in my opinion. someone will try to use linuxwacom with 6.9/7.0, notice that it fails, and run the ld magic to convert it to dlloader format. it's really quite trivial. with that said: the wacom driver is autotooled for 7.0, but is still being marked deprecated, and will not be updated to match linuxwacom. closing. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.