Bug#696110: X segfaults in lnxACPIGetEventFromOs when ACPI event has less than 4 parameters
Control: forwarded -1 https://bugs.freedesktop.org/show_bug.cgi?id=55329 On Mon, Dec 17, 2012 at 07:47:19PM +0100, Julien Cristau wrote: > On Sun, Dec 16, 2012 at 21:44:17 +0100, Evgeni Golov wrote: > > > after the "recent" upgrade of acpid [1], the Xorg on my ThinkPad X61 Tablet > > started to segfault while moving the screen from Laptop to Tablet and back. > > A short backtrace (attached) and chat with ron in #debian-x revieled the > > adding of two new ACPI messages to acpid [2] as the evil. > > > > acpid now sends "video/tabletmode TBLT off" and "video/tabletmode TBLT on" > > when I move my screen. This is fine, until you read the source of > > lnxACPIGetEventFromOs in hw/xfree86/os-support/linux/lnx_acpi.c. The code > > just expects the video/* message to have four parameters, but the new ones > > acpid send have only three, resulting in a strtoul on NULL and a segfault. > > > Can't this be fixed in acpid? Are there other consumers for these > events that expect a different format? Hi Julien, well, yeah, acpid could be patched back to drop these two events and yes this will stop the Xorg on my tablet from crashing. But this would just hide the real problem again as Xorg should really try to check whether the acpi event is in a sane format before trying to parse it. The code is just waiting for explode on another occurence otherwise. Short Google search gives at least: - the upstream bug for that issue [1] - an entry in the ArchLinux forums about the same issue I cannot find any current consumers of the feature in acpid, but I do think it is useful as it gives a generic interface to catch tablet rotation (there was an event for that on IBM hardware before, but that was IBM specific). Greets Evgeni [1] https://bugs.freedesktop.org/show_bug.cgi?id=55329 [2] https://bbs.archlinux.org/viewtopic.php?id=150290 -- Bruce Schneier can read and understand Perl programs. -- To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20121219071838.gc3...@dorei.kerker.die-welt.net
Bug#649002: iBook touchpad stopped working
On 11/22/2011 10:53 PM, David A. Greene wrote: > I am also seeing this on a PowerBook G4 (PowerPC). It worked when I > installed squeeze but stopped working on upgrade to wheezy. Wheezy still has 1.4.1-1, which works just fine for me. Are you sure you aren't running sid? If you are, and you have 1.4.1-1 installed, it may be a different bug. -- To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4eccaf6a.1010...@debian.org
Bug#649002: iBook touchpad stopped working
This is a "me too". Works fine in 1.4.1-1, broken in 1.5.0-1. Lemme know if I can supply any usefull information. Btw, 12" iBook 2005 (1.33GHz) Regards Evgeni -- Bruce Schneier can read and understand Perl programs. -- To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/2019202305.gs17...@dorei.kerker.die-welt.net
Bug#536586: drm-modules-source: drm-modules fails to uninstall
force-merge 536586 539457 tags 536586 + patch thanks Had the same issue here, the problem is, that the preinst script add the diversion with drm-modules-KVERS-arch-amd64 (drm-modules-2.6.30-1-amd64-amd64 for me). The relevant line in modules.preinst.in is dpkg-divert --package drm-modul...@kernel@-amd64 --add --rename and should be changed to dpkg-divert --package drm-modul...@kernel@ --add --rename :) Regards Evgeni -- Bruce Schneier can read and understand Perl programs. -- To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: lrmi vs new kernels vs libx86
Hi David, On Sun, 8 Mar 2009 21:07:29 +0100 David Paleino wrote: > I prepared a package with an updated LRMI: > > http://alioth.debian.org/~hanska-guest/apt/unstable/libx86_1.1+ds1-3.dsc > > All the people involved: would you please testbuild your packages against it? > Thank you. Sorry that I didn't write yet, was somehow busy and forgot about this issue. Now I have tested -3 from your home and it works fine for me with v86d. Still no luck with including it into the initrd, but that is not your fault (it also happens when I build with the internal copy of lrmi/x86emu). Matthew, any plans on releasing a new version with David's patches upstream? Regards Evgeni pgpHvibJMkay0.pgp Description: PGP signature
lrmi vs new kernels vs libx86
Dear debian-devel,
dear maintainers of packages that contain lrmi.{c,h},
today Lucas has reported #518725 - atitvout FTBFS because of missing
*_MASK defines.
Seeing that bug and remembering fun with lrmi myself, I thought I can
have a look how many other packages will FTBFS.
The following packages contain lrmi.{c,h}, and the ones with the *
FTBFS:
read-edid*
atitvout*
s3switch*
libx86
v86d
xresprobe*
zhcon*
lphdisk*
svgalib*
usplash
makes 7 of 10 - nice quota. btw v86d and usplash do not build the
shipped lrmi.c but link against libx86 which has a patched one.
The patch is trivial an can be found in the current libx86 package.
But actually we should stop duplicating code (esp. OLD code - some
packages have lrmi.c from lrmi 0.6, 0.10 is latest) and using libx86
only.
If only libx86 would have latest lrmi code...
David, is there any chance that libx86 will be updated someday? Esp
because upstream of v86d has an updated 0.10 in his git at
http://repo.or.cz/w/v86d.git and Debian's v86d is not using it in
favour of not build duplicate code.
All other (incl David), is there any interest in forking libx86 and
using it globally instead of fixing that ftbfs 7 times?
Attached is a dd-list of the failing packages, just for the case :)
Regards
Evgeni
--
Bruce Schneier Fact Number 127:
Bruce Schneier's DNA is a secure platform and cannot be cloned.
Philippe Coval
atitvout
Debian X Strike Force
xresprobe
Yu Guanghui
zhcon
Guido Guenther
s3switch
Guillem Jover
svgalib
Roberto Lumbreras
lphdisk
David Nusinow
xresprobe (U)
Sam Hocevar (Debian packages)
svgalib (U)
Branden Robinson
read-edid
Otavio Salvador
xresprobe (U)
pgpjCgrm2a5xV.pgp
Description: PGP signature
Re: RFA: atitvout -- ATI TV Out Support Program
On Mon, 28 Apr 2008 01:17:48 +0200 Julien Cristau wrote: > > I'm just wondering, shouldn't we remove atitvout at all? AFAIK the > > TVout on Radeons is supported by xserver-xorg-video-ati through the > > XRandR1.2 interface. > > My Radeon does not have a TVout, so I cannot test it, but I'm CC'ing > > [EMAIL PROTECTED], they should be able to give a short comment on that :) > > > the atitvout package description talks about ati rage mobility, not > radeon, though? Uhm, /me has his ATI-filter on :) Sorry for that. But isn't the rage series supported by -ati (<= 6.8.0) and -r128? Or is rage mobility something different? (If so, please ignore my mails). Regards Evgeni pgpCxWUuLCDj3.pgp Description: PGP signature
Re: RFA: atitvout -- ATI TV Out Support Program
Hi, I'm just wondering, shouldn't we remove atitvout at all? AFAIK the TVout on Radeons is supported by xserver-xorg-video-ati through the XRandR1.2 interface. My Radeon does not have a TVout, so I cannot test it, but I'm CC'ing [EMAIL PROTECTED], they should be able to give a short comment on that :) Regards Evgeni pgpQxMMpIzUYW.pgp Description: PGP signature
Bug#474395: iceweasel: some tabs have spurious lines, making display unreadable
reassign 474395 libcairo2 found 474395 1.5.8-1 kthxbye On Mon, 7 Apr 2008 12:23:28 +0200 Evgeni Golov wrote: > On Mon, 7 Apr 2008 10:51:56 +0200 Mike Hommey wrote: > > > > > If that's helpful in any way, I'd like to mention that I see exactly the > > > > same effect with fglrx (non-free cr**, I know). > > > > > > And me sees it with xserver-xorg-video-ati (on a Mob Radeon M6 LY). So > > > IMHO it's not a driver issue, but firefox^Wiceweasel one. > > > > Then it could also be a cairo issue, or a Xrender issue (or whatever cairo > > uses to do rendering). It looks unlikely to be an iceweasel specific issue, > > but you can still prove me wrong. > > Yepp, it *could* be also in cairo or xrender or ..., however > iceweasel3 is the only one who shows that behaviour at the moment. I'll > try cairo 1.5.18 today, maybe that will help, if not, any debugging > hints are welcome :) It IS cairo (at least for me). For testing, I used the Iceweasel BTS page [1] which renders realy badly with libcairo2 1.5.8-1. That does NOT happen with binary firefox from [2] and libcairo2 1.5.8-1. And it does NOT happen when I use a localy built libcairo2 1.5.18. Can someone confirm that with my libcairo2_1.5.18-1_i386.deb [3]? Regards Evgeni Golov [1] http://bugs.debian.org/iceweasel [2] http://en-us.www.mozilla.com/en-US/firefox/all-beta.html [3] http://debian.die-welt.net/cairo/libcairo2_1.5.18-1_i386.deb -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#474395: iceweasel: some tabs have spurious lines, making display unreadable
On Mon, 7 Apr 2008 10:51:56 +0200 Mike Hommey wrote: > > > If that's helpful in any way, I'd like to mention that I see exactly the > > > same effect with fglrx (non-free cr**, I know). > > > > And me sees it with xserver-xorg-video-ati (on a Mob Radeon M6 LY). So > > IMHO it's not a driver issue, but firefox^Wiceweasel one. > > Then it could also be a cairo issue, or a Xrender issue (or whatever cairo > uses to do rendering). It looks unlikely to be an iceweasel specific issue, > but you can still prove me wrong. Yepp, it *could* be also in cairo or xrender or ..., however iceweasel3 is the only one who shows that behaviour at the moment. I'll try cairo 1.5.18 today, maybe that will help, if not, any debugging hints are welcome :) Regards Evgeni -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#474395: iceweasel: some tabs have spurious lines, making display unreadable
On Sat, 05 Apr 2008 17:16:29 +0200 Krzysztof Sobolewski wrote: > Mike Hommey pisze: > > >> This occurs in xserver-xorg-video-intel from both sid and experimental, > >> but this is the only program that has this problem, and only in 3.0b5. > >> Iceweasel 2 is unaffected by this bug. Since the only thing that has > >> changed is Iceweasel, I presume that's where the problem is. > > > > There is a big difference in rendering between iceweasel 2 and iceweasel > > 3, one of which is that it now uses cairo. There are plenty of subtle > > ways of triggering video driver bugs, which is why they sometimes only > > occur in one application. > > If that's helpful in any way, I'd like to mention that I see exactly the > same effect with fglrx (non-free cr**, I know). And me sees it with xserver-xorg-video-ati (on a Mob Radeon M6 LY). So IMHO it's not a driver issue, but firefox^Wiceweasel one. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#455065: libgl1-mesa-glx and fglrx-driver (proprietary) share /usr/lib/libGL.so.1.2
reassign 455065 fglrx-driver retitle 455065 performace drop and broken fgl_glxgears when using debian-packages kthxbye Again: *PLEASE* keep the BTS cc:ed, even Mutt knows "reply-all" On Sun, 9 Dec 2007 01:36:38 +0100 Gerard Robin wrote: > >So 8.43.2-2 is working fine together with libgl1-mesa-glx? > yes. Fine, so at least our packages are not completelly broken. > >What do you mean by "fgl_glxgears didn't work"? Does the normal > >glxgears work? Or other 3D-apps? > > fgl_glxgears gives: > Using GLX_SGIX_pbuffer > X Error of failed request: GLXUnsupportedPrivateRequest >Major opcode of failed request: 143 (GLX) >Minor opcode of failed request: 16 (X_GLXVendorPrivate) >Serial number of failed request: 28 >Current serial number in output stream: 30 I wonder if fgl_glxgears is still usefull, as we don't have an own libGL anymore, however, could you show glxinfo? > glxgears gives: > 2668 frames in 5.2 seconds = 513.488 FPS > 2760 frames in 5.1 seconds = 540.326 FPS > 2760 frames in 5.1 seconds = 536.650 FPS > 2760 frames in 5.0 seconds = 547.812 FPS > .. > > with the ati drivers 8.433 (proprio): > glxgears: > 8883 frames in 5.2 seconds = 1776 FPS glxgears is not a benchmark, but it should give similiar results when run with the same driver on the same hardware, so we fail here? Could you try some "real" benchmark? Like playing Quake and look there at the FPS counter? > . > fgl_glxgears: > 1767 frames in 5.2 seconds = 353 FPS I really would love to know what difference is in that binary. However, it seems we do something wrong, maybe Romain has time to look at that. pgpM0FUrKp7Pw.pgp Description: PGP signature
Bug#455065: libgl1-mesa-glx and fglrx-driver (proprietary) share /usr/lib/libGL.so.1.2
*PLEASE* keep the BTS CC:ed, so others can follow the discussion. On Sat, 8 Dec 2007 23:53:23 +0100 Gerard Robin wrote: > >> fglrx-driver_8.433 > g > >That version is not in Debian? > >I see only 8.43.2-2 here, which *should* use libGL.so from mesa. > > I submitted this request because the proprietary drivers 8.39 worked > fine but after that I upgraded with the packages 8.43.2-2, fgl_glxgears > didn't work. But should have worked. > Anyway, I have reinstalled the 8.43.2-2 version and now hibernate is ok > it is more important that fgl_glxgears with a laptop. So 8.43.2-2 is working fine together with libgl1-mesa-glx? What do you mean by "fgl_glxgears didn't work"? Does the normal glxgears work? Or other 3D-apps? Regards Evgeni pgpWKz9icUVGG.pgp Description: PGP signature
Bug#455065: libgl1-mesa-glx and fglrx-driver (proprietary) share /usr/lib/libGL.so.1.2
On Sat, 08 Dec 2007 20:19:13 +0100 Gerard Robin wrote: > fglrx-driver_8.433 That version is not in Debian? I see only 8.43.2-2 here, which *should* use libGL.so from mesa. Regards Evgeni pgp4jCqLbAhMK.pgp Description: PGP signature
Bug#446901: compiz ignores gconf modifications and number_of_desktops value
On Wed, 05 Dec 2007 11:33:37 +0100 Giacomo Montagner wrote:
> - run command:
> ps -ef | grep compiz.real | grep -v grep \
> | awk '{print "kill -HUP "$2}' | bash
Why not just
kill -HUP `pidof compiz.real`
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#452959: xserver-xorg-core 1.4.1 introduces two new displaying errors
Package: xserver-xorg-core Version: 2:1.4.1~git20071119-1 Hi, since the upgrade from 2:1.4-3 my iceweasel has two nice displaying errors: - the tab-close button of the active tab is one pixel too high after moving the mouse over it (http://files.die-welt.net/2007-11-25-223855_1024x768_scrot.png) - the upper half of the fonts in the content is shifted one pixel to the right (or left) The first one occurs always, the second one only randomly (this is why I do not have a screenshot at the moment). Regards Evgeni PS: I use the latest xserver-xorg-video-ati from experimental, but the problem also occurs with the version in unstable and the latest git. pgpxdZ5osWTb9.pgp Description: PGP signature
Bug#445808: File libtype1.so is NOT included in xserver-xorg-core package
Privet Anton! On Mon, 08 Oct 2007 17:46:35 +0700 AntonSerov wrote: > Justification: renders package unusable Why? Here it works fine without type1. Regards Evgeni -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#445746: screen looks very strange after the update to 6.7.195
Package: xserver-xorg-video-ati Version: 1:6.7.195-1 after the update, my screen terribly misses colors and is a bit garbled. This is known upstream (https://bugs.freedesktop.org/show_bug.cgi?id=12720) and has been fixed in commit 051435610a66735fd455bbb526fa294fcfe8c0b6. Could you please fetch the changes and put them into the package? Regards Evgeni -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#437023: xserver-xorg-input-kbd: X does not start if I upgrade the package
On Thu, 09 Aug 2007 08:51:01 +0300 Mikko Heimola wrote: > If I upgrade this package to the testing/unstable version, X refuses to > start. > Downgrading back to 1:1.1.0-4 helps. You're running testing/unstable? So you did read /usr/share/doc/xserver-xorg-input-kbd/NEWS.Debian.gz? ;) keyboard is obsolete, one should use kbd instead. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#391228: compiz: does not work correctly with twinview
Hi Brice, On Wed, 06 Jun 2007 21:45:06 +0200 Brice Goglin wrote: > Does this problem with compiz in twinview still occurs with latest > compiz 0.5, xserver-xorg-core 1.3 and nvidia drivers? Can you reproduce > with another xinerama-like extension that does not need on the nvidia > binary? Nobody else reported such a problem with a free driver, so it > might be a nvidia-specific bug. Well, I do not have a NVIDIA card here anymore, so I cannot test it. Sorry. But if you cannot find anyone, I'll maybe ask some friend of mine, he still should have a twinview nvidia setup at home. Regards Evgeni -- ^^^| Evgeni -SargentD- Golov ([EMAIL PROTECTED]) d(O_o)b | GPG/PGP-Key-ID: 0xAC15B50C >-|-< | 0C04 F872 0963 ADC9 AA83 882B 24A0 1418 AC15 B50C / \| http://www.die-welt.net - [EMAIL PROTECTED] Spam botnets are a real boon when setting up new MTAs -- you get instant testing. (Lars Magne Ingebrigtsen - gmane.discuss) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#426947: xserver-xorg-core should conflict only with fglrx-driver << 8.37.6
Package: xserver-xorg-core
Version: 2:1.3.0.0.dfsg-5
As of fglrx 8.37.6, Xorg 1.3 is supported (I've successfully installed
the debs with --force-conflicts and it's running).
So could you please change the conflict to a versioned one? <<8.37.6-1
should be fine I think (no, it is not yet in Debian, but as soon it
will be, xorg should be installable again ;]).
Regards
Evgeni
--
^^^| Evgeni -SargentD- Golov ([EMAIL PROTECTED])
d(O_o)b | GPG/PGP-Key-ID: 0xAC15B50C
>-|-< | 0C04 F872 0963 ADC9 AA83 882B 24A0 1418 AC15 B50C
/ \| http://www.die-welt.net - [EMAIL PROTECTED]
foreach gf array_get(gf) { alloc(time+money) }
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#391228: compiz: does not work correctly with twinview
Package: compiz Version: 0.0.13+git20060928-2 Severity: normal Hi, I've just tried out the new compiz packages and they seem to work ok with a single-screen setup. But I have a dual one (1024x768+1280x1024) and compiz seems not to recognize that. When I maximize a window, it gets as big as my whole virtual screen (2304x1024) so I'm not able to see the parts of the window which are on the smaller screen above the 768 pixel. You can image it like this: ___ X | || 1024 768 || 1024 + 1280 I can see the white space, but not the X'es :( xfwm4 (my original window manager from Xfce4.4) does handle this correctly, so it seems to be a compiz issue. Kind regards Evgeni Golov -- ^^^| Evgeni -SargentD- Golov ([EMAIL PROTECTED]) d(O_o)b | PGP-Key-ID: 0xAC15B50C >-|-< | WWW: http://www.die-welt.net ICQ: 54116744 / \| IRC: #sod @ irc.german-freakz.net Fuer windows hab keine Zeit weil wenn ich nachhause komm will ich lieber poppen und nich noch mein OS reparieren (Therion - german-freakz.net) pgpDA3GaGfbN5.pgp Description: PGP signature
Bug#390503: killing gconf helps, unseting not
Hi, deleting the gconf* and gnome2* folders helps - after that I had window-borders with compiz. Is this a general compiz/gconf bug, or did it happen because of the previos installation of compiz from the ubunti package from compiz.net? Unsetting XLIB_SKIP_ARGB_VISUALS did not help (the variable wasn't set before, as I saw). I still have some issues with TwinView, but thats nothing for THIS bug ;-) Regards Evgeni Golov -- ^^^| Evgeni -SargentD- Golov ([EMAIL PROTECTED]) d(O_o)b | PGP-Key-ID: 0xAC15B50C >-|-< | WWW: http://www.die-welt.net ICQ: 54116744 / \| IRC: #sod @ irc.german-freakz.net Viele geniale Leute sind schwierig und kommen mit Gott und der Welt nicht klar. Leider folgern daraus viele Leute, die mit Gott und der Welt nicht klar kommen, dass sie genial sind. (CodeWicht @ heise.de) pgp0b0jiDcCOi.pgp Description: PGP signature
