Bugzilla Bug 40314

Book xfree-drm # emerge ./xfree-drm-4.3.0-r7.ebuild 
Calculating dependencies ...done! 
>>> emerge (1 of 1) x11-base/xfree-drm-4.3.0-r7 to / 
>>> md5 src_uri  xfree-drm-4.3.0-r7-gentoo-0.3.tar.bz2 
>>> md5 src_uri  linux-drm-4.3.0-kernelsource-20031202.tar.bz2 
>>> md5 src_uri  linux-drm-mach64-4.3.0-kernelsource-20031202.tar.bz2 
video_cards_mach64 
video_cards_mach64 
video_cards_mach64 
>>> Unpacking source... 
video_cards_mach64 
>>> Unpacking linux-drm-mach64-4.3.0-kernelsource-20031202.tar.bz2 to /var/tmp/portage/xfree-drm-4.3.0-r7/work 
>>> Unpacking xfree-drm-4.3.0-r7-gentoo-0.3.tar.bz2 to /var/tmp/portage/xfree-drm-4.3.0-r7/work 
video_cards_mach64 
* Updating for mach64 build... 
* Applying various patches (bugfixes/updates)... 
* 001_ppc_drm-ioremap-agp.patch... 

* Failed Patch: 001_ppc_drm-ioremap-agp.patch! 
* 
* Include in your bugreport the contents of: 
* 
* /var/tmp/portage/xfree-drm-4.3.0-r7/temp/001_ppc_drm-ioremap-agp.patch-17214.out 


!!! ERROR: x11-base/xfree-drm-4.3.0-r7 failed. 
!!! Function epatch, Line 322, Exitcode 0 
!!! Failed Patch: 001_ppc_drm-ioremap-agp.patch! 


next ... /var/tmp/portage/xfree-drm-4.3.0-r7/temp/001_ppc_drm-ioremap-agp.patch-17214.out 

***** 001_ppc_drm-ioremap-agp.patch ***** 

========================================= 

PATCH COMMAND: patch -g0 -p0 < /var/tmp/portage/xfree-drm-4.3.0-r7/work/patch/001_ppc_drm-ioremap-agp.patch 

========================================= 
can't find file to patch at input line 4 
Perhaps you used the wrong -p or --strip option? 
The text leading up to this was: 
-------------------------- 
|diff -urN drm-old/drmP.h drm/drmP.h 
|--- drm-old/drmP.h 2003-10-14 19:20:55.000000000 -0400 
|+++ drm/drmP.h 2003-10-14 19:22:18.000000000 -0400 
-------------------------- 
No file to patch. Skipping patch. 
2 out of 2 hunks ignored 
patching file drm/drmP.h.orig 
can't find file to patch at input line 1062 
Perhaps you used the wrong -p or --strip option? 
The text leading up to this was: 

and a lot of more... same that ^^

When I try to emerge it the portage are updated, but the rest of the "world"
not. (Xfree-4.3.0-r3) now is updating...

I try to emerge 20 times, with the same result :(

How I can do??

Created an attachment (id=24903) [details]
xfree-drm-4.3.0-r7.ebuild.patch

cd /usr/portage/x11-base/xfree-drm
patch -p0 /path/to/xfree-drm-4.3.0-r7.ebuild.patch

Note: while this should enable you to compile it, it may not work because the
I/O-remapping will be gone. I've heard a rumor that it may work anyway, but I'd
like to some feedback.

I have the same problem with radeon on ppc. I'm trying it without the remap
patch now.

It works for me without the remap patch on my Radeon 9100. BUT, I do not use
AGP-mode as there is no working AGP driver available for my platform yet
(Pegasos II).

I'm gonna give it a shot on the powerbook (AGP + Rage128)... gotta build a 2.4
kernel first though.

Ok, I gave it a shot on the PowerBook. I couldn't get either -r6 or -r7 to work
in AGP mode though :(. Both, however, worked just fine if I forced PCI mode.
That machine is rather old though so it would be great if someone on a newer
Mac could test it as well.

Committed my patch from comment #2 to fix mach64 ppc unpack. It's still not
apparent, however, whether anything will work on mach64.

David, did you mean to post those to the 4.3.0-r7 testing bug? Also, things are
a little unclear: are you saying the ioremap patch won't apply for radeon/r128
etc as well as mach64?

That is correct. The patch failed on both r128 and radeon on my PowerPCs. I'll
check why and get back on the matter.

The io-remap patch seems to contain some garbage. It seems the patch was
created without first removing the *.orig files. Therefore it passes a dry run
but when it is applied for real it will first create drmP.h.orig during the
first subpatch, then the next step it tries to create drmP.h.orig from the
original patch-dir (when the patch was created) which will of course fail.

The *.orig entries should be removed from the patch. Then it'll work.

K, I'll throw up a new patchset later today or so.

I read all about it, I'm the first writer of this bug :P

Now I d a grep -e WW -e EE and find this

         (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(II) Loading extension MIT-SCREEN-SAVER
(WW) ATI(0): Cannot shadow an accelerated frame buffer.
(WW) ATI(0): Extraneous XF86Config HorizSync specification(s) ignored.
(WW) ATI(0): Extraneous XF86Config VertRefresh specification(s) ignored.
(WW) ATI(0): [agp] AGP not available
(WW) ATI(0): [agp] AGP failed to initialize -- falling back to PCI mode.
(WW) ATI(0): [agp] Make sure you have the agpgart kernel module loaded.

AGP not really works, but DRM works as PCI... It's ioremap problem? ;)

I'm very happy with my gentoo Linux on my little iBook (1st generation) with DRM working :D

I fixed the broken patch and uploaded a new patchset to distfiles-local. Please
test it as soon as it's been seeded to the servers. (I placed
xfree-drm-4.3.0-r7 in ~ppc as well)

David when you change what's inside the patchset you ALSO need to bump the
PATCH_VER, otherwise it breaks everything for people who already download the
old patchset because it's the same file name with a different digest.

BTW, I'll fix this.

David, maybe I'm missing something but it still doesn't apply cleanly to
mach64. Help me out here? `VIDEO_CARDS=mach64 emerge xfree-drm`

Created an attachment (id=25682) [details]
distfiles/xfree-drm-4.3.0-r7-gentoo-0.4.tar.bz2

Try this new patchset please

Created an attachment (id=25683) [details]
xfree-drm-4.3.0-r7.ebuild

Still b0rked on mach64, hopefully fixed by this new patchset. I haven't added
it to portage though.

I'm gonna change the patch number of the 000 patch, make one additional change
to the ebuild (excluding 001 for USE=gatos because it fails, and I have no clue
whether gatos stuff works for ppc), and commit this stuff.

OK finally got this fixed.