Bug#650962: kdebase-workspace-bin: desktop is inverted
On Fri, Dec 9, 2011 at 22:39:40 +0100, Thomas Hahn wrote: If I configure desktop effects within KDE's system settings, there is a compositing combo box where you can switch between OpenGL and XRender. If I switch to OpenGL the desktop is being inverted when I press the Apply pushbutton. Choosing the XRender there is no such effect. For the time being I have switched of Desktop Effects in KDE. Hope that helps. Please attach your X log, and the output of glxinfo. Also make sure kwin uses direct rendering. Cheers, Julien -- 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/20111210132159.gc2...@radis.cristau.org
Bug#650962: kdebase-workspace-bin: desktop is inverted
On Sat, Dec 10, 2011 at 02:21:59PM +0100, Julien Cristau wrote: On Fri, Dec 9, 2011 at 22:39:40 +0100, Thomas Hahn wrote: If I configure desktop effects within KDE's system settings, there is a compositing combo box where you can switch between OpenGL and XRender. If I switch to OpenGL the desktop is being inverted when I press the Apply pushbutton. Choosing the XRender there is no such effect. For the time being I have switched of Desktop Effects in KDE. Hope that helps. Please attach your X log, and the output of glxinfo. Also make sure kwin uses direct rendering. Cheers, Julien Attached you will find the X log, glxinfo and the diff of .xsession-errors before and after switching on opengl. Please note that OpenGL options have Enable direct rendering clicked, but the info from .xsession-errors tells otherwise. Hope that helps. Tnx Thomas Xorg.0.log.kde.gz Description: Binary data glxinfo.lst.gz Description: Binary data xsession-errors.diff.gz Description: Binary data
Bug#650962: kdebase-workspace-bin: desktop is inverted
On Sat, Dec 10, 2011 at 15:49:03 +0100, Thomas Hahn wrote: On Sat, Dec 10, 2011 at 02:21:59PM +0100, Julien Cristau wrote: On Fri, Dec 9, 2011 at 22:39:40 +0100, Thomas Hahn wrote: If I configure desktop effects within KDE's system settings, there is a compositing combo box where you can switch between OpenGL and XRender. If I switch to OpenGL the desktop is being inverted when I press the Apply pushbutton. Choosing the XRender there is no such effect. For the time being I have switched of Desktop Effects in KDE. Hope that helps. Please attach your X log, and the output of glxinfo. Also make sure kwin uses direct rendering. Cheers, Julien Attached you will find the X log, glxinfo and the diff of .xsession-errors before and after switching on opengl. Please note that OpenGL options have Enable direct rendering clicked, but the info from .xsession-errors tells otherwise. from your glxinfo: direct rendering: No (If you want to find out why, try setting LIBGL_DEBUG=verbose) So what does LIBGL_DEBUG=verbose glxinfo say? (Don't compress the attachments it just makes it more painful to read.) Cheers, Julien -- 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/20111210145359.gd2...@radis.cristau.org
Bug#650962: kdebase-workspace-bin: desktop is inverted
On Sat, Dec 10, 2011 at 03:53:59PM +0100, Julien Cristau wrote: On Sat, Dec 10, 2011 at 15:49:03 +0100, Thomas Hahn wrote: On Sat, Dec 10, 2011 at 02:21:59PM +0100, Julien Cristau wrote: On Fri, Dec 9, 2011 at 22:39:40 +0100, Thomas Hahn wrote: If I configure desktop effects within KDE's system settings, there is a compositing combo box where you can switch between OpenGL and XRender. If I switch to OpenGL the desktop is being inverted when I press the Apply pushbutton. Choosing the XRender there is no such effect. For the time being I have switched of Desktop Effects in KDE. Hope that helps. Please attach your X log, and the output of glxinfo. Also make sure kwin uses direct rendering. Cheers, Julien Attached you will find the X log, glxinfo and the diff of .xsession-errors before and after switching on opengl. Please note that OpenGL options have Enable direct rendering clicked, but the info from .xsession-errors tells otherwise. from your glxinfo: direct rendering: No (If you want to find out why, try setting LIBGL_DEBUG=verbose) So what does LIBGL_DEBUG=verbose glxinfo say? (Don't compress the attachments it just makes it more painful to read.) Cheers, Julien I did exactly that, but the glxinfo output is identical. (Diff'd the output of both glxinfo run's ...) ? Thomas -- 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/20111210152143.ga6...@hversa.hahn.homeunix.net
Bug#650962: kdebase-workspace-bin: desktop is inverted
On Sat, Dec 10, 2011 at 16:21:43 +0100, Thomas Hahn wrote: On Sat, Dec 10, 2011 at 03:53:59PM +0100, Julien Cristau wrote: On Sat, Dec 10, 2011 at 15:49:03 +0100, Thomas Hahn wrote: On Sat, Dec 10, 2011 at 02:21:59PM +0100, Julien Cristau wrote: On Fri, Dec 9, 2011 at 22:39:40 +0100, Thomas Hahn wrote: If I configure desktop effects within KDE's system settings, there is a compositing combo box where you can switch between OpenGL and XRender. If I switch to OpenGL the desktop is being inverted when I press the Apply pushbutton. Choosing the XRender there is no such effect. For the time being I have switched of Desktop Effects in KDE. Hope that helps. Please attach your X log, and the output of glxinfo. Also make sure kwin uses direct rendering. Cheers, Julien Attached you will find the X log, glxinfo and the diff of .xsession-errors before and after switching on opengl. Please note that OpenGL options have Enable direct rendering clicked, but the info from .xsession-errors tells otherwise. from your glxinfo: direct rendering: No (If you want to find out why, try setting LIBGL_DEBUG=verbose) So what does LIBGL_DEBUG=verbose glxinfo say? (Don't compress the attachments it just makes it more painful to read.) Cheers, Julien I did exactly that, but the glxinfo output is identical. (Diff'd the output of both glxinfo run's ...) Nevermind, it also says client glx vendor string: NVIDIA Corporation which means you have nvidia's closed libGL installed. You need to get rid of that. Cheers, Julien -- 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/20111210153735.gf2...@radis.cristau.org
xorg-server: Changes to 'debian-unstable'
ChangeLog| 198 +++ Xext/saver.c |3 Xext/xvdisp.c|4 Xi/extinit.c |4 config/hal.c |1 configure.ac |4 debian/changelog |6 + hw/kdrive/src/kinput.c |6 - hw/vfb/InitOutput.c |4 hw/xfree86/fbdevhw/fbdevhw.c | 18 ++- hw/xfree86/loader/loadmod.c |2 hw/xfree86/xaa/xaaGC.c |1 randr/rrcrtc.c | 29 ++ 13 files changed, 237 insertions(+), 43 deletions(-) New commits: commit ba71b91eb3a8f116e39950299d136bb16a814fa5 Author: Julien Cristau jcris...@debian.org Date: Sat Dec 10 16:44:06 2011 +0100 Bump changelogs diff --git a/ChangeLog b/ChangeLog index bf744e1..e5a0887 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,201 @@ +commit 0ca8869e4537412f25bfd5610aba5e604952608c +Author: Jeremy Huddleston jerem...@apple.com +Date: Fri Dec 9 12:58:30 2011 -0800 + +configure.ac: 1.11.2.902 (1.11.3 RC2) + +Signed-off-by: Jeremy Huddleston jerem...@apple.com + +commit e76c397eb90291b048e0b2102111008e3d2e469e +Author: Dave Airlie airl...@redhat.com +Date: Mon Nov 28 16:37:59 2011 + + +kdrive: drop screen crossing code. + +The only kdrive server we probably care about anymore is Xephyr, +and this screen enable/disable code totally breaks it in multi-screen mode. + +When you are in one screen the other stops updating. + +Fixes https://bugzilla.redhat.com/show_bug.cgi?id=757457 + +Signed-off-by: Dave Airlie airl...@redhat.com +Reviewed-by: Peter Hutterer peter.hutte...@who-t.net +(cherry picked from commit 98c4a888a4428789386c7c47cecc81933b5999ba) + +commit db418ff7501d960c8de813a37b8bb37b1ba22175 +Author: Dave Airlie airl...@redhat.com +Date: Thu Oct 20 10:48:26 2011 +0100 + +hal: free tmp_val in one missing case + +Pointed out by coverity scan. + +Signed-off-by: Dave Airlie airl...@redhat.com +Reviewed-by: Daniel Stone dan...@fooishbar.org +(cherry picked from commit 8d3731a811e33e263920dd7c8ec63d02968cb56e) + +commit 6b11b18ab683e1f38e5b27066b35947b96956e5f +Author: Dave Airlie airl...@redhat.com +Date: Thu Oct 20 11:00:43 2011 +0100 + +xv: test correct number of requests. (v2) + +Pointed out by coverity. + +v2: fix swapped as well, as pointed out by Alan + +Signed-off-by: Dave Airlie airl...@redhat.com +Reviewed-by: Alan Coopersmith alan.coopersm...@oracle.com +(cherry picked from commit 41229392b790f30a0f0ef1f4ed95647c5bca4001) + +commit df16b789aaf5fe7025e5661117477a5f6f9129ee +Author: Dave Airlie airl...@redhat.com +Date: Wed Oct 19 16:57:13 2011 +0100 + +xaa: avoid possible freed pointer reuse in epilogue + +If the pGCPriv-flags == 2, then we try to assign the freed pGCPriv-XAAOps +avoid this by clearing the flags in to be destroyed pGCPriv. + +Reported by coverity. + +Signed-off-by: Dave Airlie airl...@redhat.com +Reviewed-by: Jeremy Huddleston jerem...@apple.com +(cherry picked from commit 1049139499d9132a20cd6d4d156fe9da9cddb6c2) + +commit befa8a3b836e8b5781491aff4e659ae856beb494 +Author: Dave Airlie airl...@redhat.com +Date: Wed Oct 19 16:22:31 2011 +0100 + +Xi: avoid overrun of callback array. + +This code had an off-by-one and would allow writing one past the end of +the callbacks array. + +Pointed out by coverity. + +Signed-off-by: Dave Airlie airl...@redhat.com +Reviewed-by: Jeremy Huddleston jerem...@apple.com +(cherry picked from commit 682c09a2cedd234b005334cc01247d859dd7f26a) + +commit b22783f483d440a005ff76d5c7894e2cf2a95dcb +Author: Dave Airlie airl...@redhat.com +Date: Wed Oct 19 16:21:26 2011 +0100 + +xext: don't free uninitialised pointer when malloc fails. (v2) + +Initialise the pAttr-values to values so if the values allocation +fails it just ends up as free(NULL). + +Pointed out by coverity. + +v2: use Alan's suggestion. + +Signed-off-by: Dave Airlie airl...@redhat.com +Reviewed-by: Alan Coopersmith alan.coopersm...@oracle.com +Reviewed-by: Jeremy Huddleston jerem...@apple.com +(cherry picked from commit b62dc4fcbcffd10de16650bee284702c8608bb60) + +commit 083599c5597276289af9adaf8a8571cbe743c3fd +Author: Adam Jackson a...@redhat.com +Date: Thu Apr 28 13:34:28 2011 +1000 + +fbdevhw: iterate over all modes that match a mode. (v3) + +So on RHEL5 anaconda sets an xorg.conf with a fixed 800x600 mode in it, +we run radeonfb and fbdev since ati won't work in userspace due to domain +issues in the older codebase. + +On certain pseries blades the built-in KVM can't accept an 800x600-43 mode, +it requires the 800x600-60 mode, so we have to have the kernel radeonfb +driver reject the 800x600-43 mode when it sees it. However then fbdev +doesn't try
xorg-server: Changes to 'upstream-unstable'
Xext/saver.c |3 +--
Xext/xvdisp.c|4 ++--
Xi/extinit.c |4 ++--
config/hal.c |1 +
configure.ac |4 ++--
hw/kdrive/src/kinput.c |6 --
hw/vfb/InitOutput.c |4 +++-
hw/xfree86/fbdevhw/fbdevhw.c | 18 ++
hw/xfree86/loader/loadmod.c |2 +-
hw/xfree86/xaa/xaaGC.c |1 +
randr/rrcrtc.c | 29 ++---
11 files changed, 33 insertions(+), 43 deletions(-)
New commits:
commit 0ca8869e4537412f25bfd5610aba5e604952608c
Author: Jeremy Huddleston jerem...@apple.com
Date: Fri Dec 9 12:58:30 2011 -0800
configure.ac: 1.11.2.902 (1.11.3 RC2)
Signed-off-by: Jeremy Huddleston jerem...@apple.com
diff --git a/configure.ac b/configure.ac
index 9237686..817c907 100644
--- a/configure.ac
+++ b/configure.ac
@@ -26,8 +26,8 @@ dnl
dnl Process this file with autoconf to create configure.
AC_PREREQ(2.57)
-AC_INIT([xorg-server], 1.11.2.901,
[https://bugs.freedesktop.org/enter_bug.cgi?product=xorg], xorg-server)
-RELEASE_DATE=2011-11-28
+AC_INIT([xorg-server], 1.11.2.902,
[https://bugs.freedesktop.org/enter_bug.cgi?product=xorg], xorg-server)
+RELEASE_DATE=2011-12-09
AC_CONFIG_SRCDIR([Makefile.am])
AM_INIT_AUTOMAKE([foreign dist-bzip2])
AM_MAINTAINER_MODE
commit e76c397eb90291b048e0b2102111008e3d2e469e
Author: Dave Airlie airl...@redhat.com
Date: Mon Nov 28 16:37:59 2011 +
kdrive: drop screen crossing code.
The only kdrive server we probably care about anymore is Xephyr,
and this screen enable/disable code totally breaks it in multi-screen mode.
When you are in one screen the other stops updating.
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=757457
Signed-off-by: Dave Airlie airl...@redhat.com
Reviewed-by: Peter Hutterer peter.hutte...@who-t.net
(cherry picked from commit 98c4a888a4428789386c7c47cecc81933b5999ba)
diff --git a/hw/kdrive/src/kinput.c b/hw/kdrive/src/kinput.c
index c14dd82..500a3d8 100644
--- a/hw/kdrive/src/kinput.c
+++ b/hw/kdrive/src/kinput.c
@@ -2135,12 +2135,6 @@ KdCursorOffScreen(ScreenPtr *ppScreen, int *x, int *y)
static void
KdCrossScreen(ScreenPtr pScreen, Bool entering)
{
-#ifndef XIPAQ
-if (entering)
- KdEnableScreen (pScreen);
-else
- KdDisableScreen (pScreen);
-#endif
}
int KdCurScreen; /* current event screen */
commit db418ff7501d960c8de813a37b8bb37b1ba22175
Author: Dave Airlie airl...@redhat.com
Date: Thu Oct 20 10:48:26 2011 +0100
hal: free tmp_val in one missing case
Pointed out by coverity scan.
Signed-off-by: Dave Airlie airl...@redhat.com
Reviewed-by: Daniel Stone dan...@fooishbar.org
(cherry picked from commit 8d3731a811e33e263920dd7c8ec63d02968cb56e)
diff --git a/config/hal.c b/config/hal.c
index 297520a..a4141d5 100644
--- a/config/hal.c
+++ b/config/hal.c
@@ -355,6 +355,7 @@ device_added(LibHalContext *hal_ctx, const char *udi)
if (!strcasecmp(tmp, .options)
(!xkb_opts.options))
xkb_opts.options = strdup(tmp_val);
}
+free(tmp_val);
}
}
}
commit 6b11b18ab683e1f38e5b27066b35947b96956e5f
Author: Dave Airlie airl...@redhat.com
Date: Thu Oct 20 11:00:43 2011 +0100
xv: test correct number of requests. (v2)
Pointed out by coverity.
v2: fix swapped as well, as pointed out by Alan
Signed-off-by: Dave Airlie airl...@redhat.com
Reviewed-by: Alan Coopersmith alan.coopersm...@oracle.com
(cherry picked from commit 41229392b790f30a0f0ef1f4ed95647c5bca4001)
diff --git a/Xext/xvdisp.c b/Xext/xvdisp.c
index b968431..84b10b4 100644
--- a/Xext/xvdisp.c
+++ b/Xext/xvdisp.c
@@ -1265,7 +1265,7 @@ ProcXvDispatch(ClientPtr client)
UpdateCurrentTime();
- if (stuff-data xvNumRequests) {
+ if (stuff-data = xvNumRequests) {
SendErrorToClient(client, XvReqCode, stuff-data, 0, BadRequest);
return BadRequest;
}
@@ -1589,7 +1589,7 @@ SProcXvDispatch(ClientPtr client)
UpdateCurrentTime();
- if (stuff-data xvNumRequests) {
+ if (stuff-data = xvNumRequests) {
SendErrorToClient(client, XvReqCode, stuff-data, 0, BadRequest);
return BadRequest;
}
commit df16b789aaf5fe7025e5661117477a5f6f9129ee
Author: Dave Airlie airl...@redhat.com
Date: Wed Oct 19 16:57:13 2011 +0100
xaa: avoid possible freed pointer reuse in epilogue
If the pGCPriv-flags == 2, then we try to assign the freed pGCPriv-XAAOps
avoid this by clearing the flags in to be destroyed pGCPriv.
Reported by coverity.
Signed-off-by: Dave Airlie airl...@redhat.com
Reviewed-by: Jeremy Huddleston jerem...@apple.com
(cherry picked from commit 1049139499d9132a20cd6d4d156fe9da9cddb6c2)
diff --git a/hw/xfree86/xaa/xaaGC.c b/hw/xfree86/xaa/xaaGC.c
index
Bug#650962: kdebase-workspace-bin: desktop is inverted
On Sat, Dec 10, 2011 at 04:37:35PM +0100, Julien Cristau wrote: On Sat, Dec 10, 2011 at 16:21:43 +0100, Thomas Hahn wrote: On Sat, Dec 10, 2011 at 03:53:59PM +0100, Julien Cristau wrote: On Sat, Dec 10, 2011 at 15:49:03 +0100, Thomas Hahn wrote: On Sat, Dec 10, 2011 at 02:21:59PM +0100, Julien Cristau wrote: On Fri, Dec 9, 2011 at 22:39:40 +0100, Thomas Hahn wrote: If I configure desktop effects within KDE's system settings, there is a compositing combo box where you can switch between OpenGL and XRender. If I switch to OpenGL the desktop is being inverted when I press the Apply pushbutton. Choosing the XRender there is no such effect. For the time being I have switched of Desktop Effects in KDE. Hope that helps. Please attach your X log, and the output of glxinfo. Also make sure kwin uses direct rendering. Cheers, Julien Attached you will find the X log, glxinfo and the diff of .xsession-errors before and after switching on opengl. Please note that OpenGL options have Enable direct rendering clicked, but the info from .xsession-errors tells otherwise. from your glxinfo: direct rendering: No (If you want to find out why, try setting LIBGL_DEBUG=verbose) So what does LIBGL_DEBUG=verbose glxinfo say? (Don't compress the attachments it just makes it more painful to read.) Cheers, Julien I did exactly that, but the glxinfo output is identical. (Diff'd the output of both glxinfo run's ...) Nevermind, it also says client glx vendor string: NVIDIA Corporation which means you have nvidia's closed libGL installed. You need to get rid of that. removed that. KWin issues a warning that is has crashed in the past, I ignored the warning and switched Desktop effects on. Maybe it would be a good idea if there is some check in these libglx-nvidia-alternatives packages which issues a warning if you have a laptop without nvidia glx hardware, so that one doesn't run into that problem. I surely haven't actively installed the nvidia packages. This is a i3 1st gen. Aldi laptop, cheep stuff, but good enough for my purposes. I know there is no real dedicated graphics on it ... Thanks and Regards Thomas Cheers, Julien -- 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/20111210160003.gb6...@hversa.hahn.homeunix.net
Bug#651316: libdrm-intel1: X.org crashes when I try to play a video
tags 651316 patch fixed-upstream thanks Hello, On penktadienis 09 Gruodis 2011 19:50:48 Pedro Antonio Neves wrote: After installing the patched versions of libva and xserver-xorg-video-intel_2.17.0-1+kibi1 I'm still unable to play video files. The windows show up but they're black. I had the same problem. The backtrace of X crashes was: (gdb) bt #0 0x7f9fd101b405 in *__GI_raise (sig=optimized out) at ../nptl/sysdeps/unix/sysv/linux/raise.c:64 #1 0x7f9fd101e680 in *__GI_abort () at abort.c:92 #2 0x7f9fd10145b1 in *__GI___assert_fail (assertion=0x7f9fceddb7a7 bo_gem-map_count == 0, file=optimized out, line=1016, function=0x7f9fceddbe70 drm_intel_gem_bo_map) at assert.c:81 #3 0x7f9fcedd8b10 in drm_intel_gem_bo_map (bo=0x7f9fd8010b90, write_enable=1) at ../../intel/intel_bufmgr_gem.c:1016 #4 0x7f9fceffcac3 in intel_alloc_and_map (name=optimized out, size=4096, bop=0x7fffe3bf6b30, virtualp=0x7fffe3bf6b38, intel=optimized out) at ../../src/i965_video.c:392 #5 0x7f9fceffe420 in I965DisplayVideoTextured (scrn=0x7f9fd39c7e80, adaptor_priv=optimized out, id=optimized out, dstRegion=0x7f9fd8013c80, width=optimized out, height=optimized out, video_pitch=312, video_pitch2=624, src_w=624, src_h=352, drw_w=884, drw_h=499, pixmap=0x7f9fd816d8f0) at ../../src/i965_video.c:1301 #6 0x7f9fceff57de in I830PutImageTextured (scrn=0x7f9fd39c7e80, src_x=0, src_y=optimized out, drw_x=optimized out, drw_y=optimized out, src_w=624, src_h=352, drw_w=884, drw_h=499, id=842094169, buf=0x7f9fc9aca000 '\020' repeats 200 times..., width=624, height=352, sync=1, clipBoxes=0x7fffe3bf6da0, data=0x7f9fd5542f90, drawable=0x7f9fd8240380) at ../../src/intel_video.c:1579 #7 0x7f9fd30599ce in xf86XVPutImage (client=optimized out, pDraw=0x7f9fd8240380, pPort=0x7f9fd5543d10, pGC=optimized out, src_x=optimized out, src_y=optimized out, src_w=624, src_h=352, drw_x=0, drw_y=23, drw_w=884, drw_h=499, format=0x7f9fd5543a90, data=0x7f9fc9aca000 '\020' repeats 200 times..., sync=1, width=624, height=352) at ../../../../hw/xfree86/common/xf86xv.c:1865 #8 0x7f9fd00f1e02 in ProcXvShmPutImage (client=0x7f9fd8240740) at ../../Xext/xvdisp.c:1091 #9 0x7f9fd3004fc9 in Dispatch () at ../../dix/dispatch.c:432 #10 0x7f9fd2ff422a in main (argc=8, argv=optimized out, envp=optimized out) at ../../dix/main.c:287 (gdb) q I upgraded libdrm to the master branch as of writing ( dd9a5b4f7fb07c78db4e7481bedca1b981030e3f ) and the problem is gone now. $ git log 2.4.28..master | cat commit dd9a5b4f7fb07c78db4e7481bedca1b981030e3f Author: Chris Wilson ch...@chris-wilson.co.uk Date: Tue Dec 6 13:12:37 2011 + intel: Evict cached VMA in order to make room for new mappings As the max number of VMA mappings is a hard per-process limit, we need to include the number of currently active mappings when evicting in order to make room for a new mmap. Signed-off-by: Chris Wilson ch...@chris-wilson.co.uk commit e4b60f29609e9993dc7268993da509530862aa78 Author: Chris Wilson ch...@chris-wilson.co.uk Date: Mon Dec 5 21:29:05 2011 + intel: Add an interface to limit vma caching There is a per-process limit on the number of vma that the process can keep open, so we cannot keep an unlimited cache of unused vma's (besides keeping track of all those vma in the kernel adds considerable overhead). However, in order to work around inefficiencies in the kernel it is beneficial to reuse the vma, so keep a MRU cache of vma. Signed-off-by: Chris Wilson ch...@chris-wilson.co.uk commit 902ee661f1864aaf8325621085f6a1b5a6a3673a Author: Dave Airlie airl...@redhat.com Date: Mon Dec 5 21:24:48 2011 + test/radeon: add missing files for dist commit 5c5332bbc38ff25c06081ac53a15ad583ad4cbc4 Author: Chris Wilson ch...@chris-wilson.co.uk Date: Mon Dec 5 10:39:49 2011 + intel: Clean up mmaps on freeing the buffer As a precautionary measure munmap on buffer free so that we never leak the vma. Also include a warning during debugging. Signed-off-by: Chris Wilson ch...@chris-wilson.co.uk I suggest pulling the relevant patches into the debian package as the problem is pretty serious. I was not able to get any video to play due this crash. signature.asc Description: This is a digitally signed message part.
Processed: Re: libdrm-intel1: X.org crashes when I try to play a video
Processing commands for cont...@bugs.debian.org: tags 651316 patch fixed-upstream Bug #651316 [libdrm-intel1] libdrm-intel1: X.org crashes when I try to play a video Added tag(s) fixed-upstream and patch. thanks Stopping processing here. Please contact me if you need assistance. -- 651316: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=651316 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems -- 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/handler.s.c.132353877132242.transcr...@bugs.debian.org
Bug#651370: libgl1-mesa-glx: need close on exec for dri device
Set the close on exec flag when opening dri character devices, so they
will be closed and free any resouces allocated in exec.
---
Just for fun I included a git bundle of the commit.
Please Cc me on any replies as I'm not subscribed to the mesa list.
src/egl/drivers/dri2/platform_wayland.c| 11 ++-
src/egl/drivers/dri2/platform_x11.c|9 +
src/gallium/state_trackers/egl/drm/native_drm.c| 11 ++-
.../state_trackers/egl/fbdev/native_fbdev.c| 11 ++-
.../state_trackers/egl/wayland/native_drm.c| 10 +-
src/gallium/state_trackers/egl/x11/x11_screen.c| 10 +-
src/glx/dri2_glx.c | 10 +-
7 files changed, 66 insertions(+), 6 deletions(-)
diff --git a/src/egl/drivers/dri2/platform_wayland.c
b/src/egl/drivers/dri2/platform_wayland.c
index 7a70d8d..2dfde69 100644
--- a/src/egl/drivers/dri2/platform_wayland.c
+++ b/src/egl/drivers/dri2/platform_wayland.c
@@ -747,7 +747,16 @@ drm_handle_device(void *data, struct wl_drm *drm, const
char *device)
if (!dri2_dpy-device_name)
return;
- dri2_dpy-fd = open(dri2_dpy-device_name, O_RDWR);
+#ifdef O_CLOEXEC
+ dri2_dpy-fd = open(dri2_dpy-device_name, O_RDWR | O_CLOEXEC);
+ if (dri2_dpy-fd == -1 errno == EINVAL)
+#endif
+ {
+ dri2_dpy-fd = open(dri2_dpy-device_name, O_RDWR);
+ if (dri2_dpy-fd != -1)
+ fcntl(dri2_dpy-fd, F_SETFD, fcntl(dri2_dpy-fd, F_GETFD) |
+FD_CLOEXEC);
+ }
if (dri2_dpy-fd == -1) {
_eglLog(_EGL_WARNING, wayland-egl: could not open %s (%s),
dri2_dpy-device_name, strerror(errno));
diff --git a/src/egl/drivers/dri2/platform_x11.c
b/src/egl/drivers/dri2/platform_x11.c
index 8dd231a..a31a587 100644
--- a/src/egl/drivers/dri2/platform_x11.c
+++ b/src/egl/drivers/dri2/platform_x11.c
@@ -998,7 +998,16 @@ dri2_initialize_x11_dri2(_EGLDriver *drv, _EGLDisplay
*disp)
if (!dri2_load_driver(disp))
goto cleanup_conn;
+#ifdef O_CLOEXEC
dri2_dpy-fd = open(dri2_dpy-device_name, O_RDWR | O_CLOEXEC);
+ if (dri2_dpy-fd == -1 errno == EINVAL)
+#endif
+ {
+ dri2_dpy-fd = open(dri2_dpy-device_name, O_RDWR);
+ if (dri2_dpy-fd != -1)
+ fcntl(dri2_dpy-fd, F_SETFD, fcntl(dri2_dpy-fd, F_GETFD) |
+FD_CLOEXEC);
+ }
if (dri2_dpy-fd == -1) {
_eglLog(_EGL_WARNING,
DRI2: could not open %s (%s), dri2_dpy-device_name,
diff --git a/src/gallium/state_trackers/egl/drm/native_drm.c
b/src/gallium/state_trackers/egl/drm/native_drm.c
index c013769..a762f8f 100644
--- a/src/gallium/state_trackers/egl/drm/native_drm.c
+++ b/src/gallium/state_trackers/egl/drm/native_drm.c
@@ -312,7 +312,16 @@ native_create_display(void *dpy, boolean use_sw)
gbm = dpy;
if (gbm == NULL) {
- fd = open(/dev/dri/card0, O_RDWR);
+ const char *device_name=/dev/dri/card0;
+#ifdef O_CLOEXEC
+ fd = open(device_name, O_RDWR | O_CLOEXEC);
+ if (fd == -1 errno == EINVAL)
+#endif
+ {
+ fd = open(device_name, O_RDWR);
+ if (fd != -1)
+fcntl(fd, F_SETFD, fcntl(fd, F_GETFD) | FD_CLOEXEC);
+ }
/* FIXME: Use an internal constructor to create a gbm
* device with gallium backend directly, without setenv */
setenv(GBM_BACKEND, gbm_gallium_drm.so, 1);
diff --git a/src/gallium/state_trackers/egl/fbdev/native_fbdev.c
b/src/gallium/state_trackers/egl/fbdev/native_fbdev.c
index e126888..b45ab5c 100644
--- a/src/gallium/state_trackers/egl/fbdev/native_fbdev.c
+++ b/src/gallium/state_trackers/egl/fbdev/native_fbdev.c
@@ -515,7 +515,16 @@ native_create_display(void *dpy, boolean use_sw)
/* well, this makes fd 0 being ignored */
if (!dpy) {
- fd = open(/dev/fb0, O_RDWR);
+ const char *device_name=/dev/fb0;
+#ifdef O_CLOEXEC
+ fd = open(device_name, O_RDWR | O_CLOEXEC);
+ if (fd == -1 errno == EINVAL)
+#endif
+ {
+ fd = open(device_name, O_RDWR);
+ if (fd != -1)
+fcntl(fd, F_SETFD, fcntl(fd, F_GETFD) | FD_CLOEXEC);
+ }
}
else {
fd = dup((int) pointer_to_intptr(dpy));
diff --git a/src/gallium/state_trackers/egl/wayland/native_drm.c
b/src/gallium/state_trackers/egl/wayland/native_drm.c
index 5618f3e..92e73ed 100644
--- a/src/gallium/state_trackers/egl/wayland/native_drm.c
+++ b/src/gallium/state_trackers/egl/wayland/native_drm.c
@@ -134,7 +134,15 @@ drm_handle_device(void *data, struct wl_drm *drm, const
char *device)
if (!drmdpy-device_name)
return;
- drmdpy-fd = open(drmdpy-device_name, O_RDWR);
+#ifdef O_CLOEXEC
+ drmdpy-fd = open(drmdpy-device_name, O_RDWR | O_CLOEXEC);
+ if (drmdpy-fd == -1 errno == EINVAL)
+#endif
+ {
+ drmdpy-fd = open(drmdpy-device_name, O_RDWR);
+ if (drmdpy-fd != -1)
+ fcntl(drmdpy-fd, F_SETFD, fcntl(drmdpy-fd, F_GETFD) | FD_CLOEXEC);
+ }
if (drmdpy-fd == -1) {
_eglLog(_EGL_WARNING,
Debian testing: alt-tab doesn't switch windows on LXDE
Platform: Intel P4 Intel graphics card debian testing gdm3 3.0.4-4 lxde 0.5.5-1 openbox 3.4.11.2-2 I installed the lxde desktop, and it works fine, except that Alt-Tab doesn't work for switching windows, and Alt-F2 doesn't pop up a run command dialog box. As far as I can tell by googling, both shortcuts are supposed to work in the openbox WM. The ~/.config/openbox/lxde-rc.xml config file, has the following: ... keybind key=A-Tab action name=NextWindow/ /keybind keybind key=A-S-Tab action name=PreviousWindow/ /keybind keybind key=C-A-Tab action name=NextWindow panelsyes/panels desktopyes/desktop /action /keybind !-- Keybindings for running applications -- ... keybind key=A-F2 action name=Execute commandlxpanelctl run/command /action /keybind ... So the config should be OK. The Alt-Tab event is picked up by the applications. Eg. emacs tries to do expand, in many modes, and indent in some. So there is nothing sitting on top grabbing them. Has anyone else seen this? Do you know any tricks to try? Any debugging tricks? Thanks! - Steinar -- 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/87zkf01a7a@dod.no
Bug#651623: intel_uxa_prepare_access: bo map failed: Cannot allocate memory
On Sat, Dec 10, 2011 at 08:50:50PM +0300, Roman Sakal wrote:
Package: xserver-xorg-video-intel
Version: 2:2.17.0-1
Severity: important
I've the same probleme since the 2.17 upgrade. Here a backtrace:
libdrm-intel12.4.28-1
xserver-xorg-video-intel 2:2.17.0-1
xserver-xorg 1:7.6+9
xserver-xorg-core2:1.11.2.901-1
(gdb) bt full
#0 0x7f7b42c03405 in *__GI_raise (sig=optimized out) at
../nptl/sysdeps/unix/sysv/linux/raise.c:64
pid = optimized out
selftid = optimized out
#1 0x7f7b42c06680 in *__GI_abort () at abort.c:92
act = {__sigaction_handler = {sa_handler = 0x7f7b40c147a7, sa_sigaction
= 0x7f7b40c147a7}, sa_mask = {__val = {140167378731848, 140735620932240, 1016,
140735620932480,
140167377796054, 206158430232, 140735620932496, 140735620932272,
140167377707912, 206158430256, 140735620932520, 140167456580912, 131616,
3273383998825390688,
8462091486410927422, 140735620939424}}, sa_flags = 1121018783,
sa_restorer = 0x7f7b40c14a10}
sigs = {__val = {32, 0 repeats 15 times}}
#2 0x7f7b42bfc5b1 in *__GI___assert_fail (assertion=0x7f7b40c147a7
bo_gem-map_count == 0, file=optimized out, line=1016,
function=0x7f7b40c14e70 drm_intel_gem_bo_map)
at assert.c:81
buf = 0x7f7b47756530 X: ../../intel/intel_bufmgr_gem.c:1016:
drm_intel_gem_bo_map: Assertion `bo_gem-map_count == 0' failed.\n
#3 0x7f7b40c11b10 in ?? () from /usr/lib/x86_64-linux-gnu/libdrm_intel.so.1
No symbol table info available.
#4 0x7f7b44b2bc4a in i965_create_dst_surface_state (scrn=optimized out,
pixmap=0x7f7b475c2670, surf_bo=0x7f7b47aaae90, offset=0) at
../../src/i965_video.c:411
intel = 0x7f7b45a31490
dest_surf_state = optimized out
pixmap_bo = 0x7f7b45a42570
#5 0x7f7b44b2d17b in I965DisplayVideoTextured (scrn=0x7f7b45a30780,
adaptor_priv=0x7f7b475a82a0, id=optimized out, dstRegion=0x7fff90b16880,
width=optimized out,
height=optimized out, video_pitch=424, video_pitch2=848, src_w=848,
src_h=480, drw_w=1199, drw_h=679, pixmap=0x7f7b475c2670) at
../../src/i965_video.c:1188
intel = 0x7f7b45a31490
pbox = optimized out
nbox = optimized out
dxo = optimized out
dyo = optimized out
pix_xoff = optimized out
pix_yoff = optimized out
src_scale_x = optimized out
src_scale_y = optimized out
n_src_surf = 6
src_surf_format = 320
src_surf_base = {0, 0, 508800, 508800, 407040, 407040}
src_width = {848, 848, 424, 424, 424, 424}
src_height = {480, 480, 240, 240, 240, 240}
src_pitch = {848, 848, 424, 424, 424, 424}
surface_state_binding_table_bo = 0x7f7b47aaae90
#6 0x7f7b44b24cbe in I830PutImageTextured (scrn=0x7f7b45a30780, src_x=0,
src_y=optimized out, drw_x=optimized out, drw_y=optimized out, src_w=848,
src_h=480, drw_w=1199,
drw_h=679, id=808596553,
buf=0x7f7b3d5b5000 \200\200\200\200\200\200\200\200, '\177' repeats 14
times, ~~~}}~~\177\177\177\200\200\200\200\200\200\200\200, '\177' repeats
16 times\200,
\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\177\177\177\200\202\204\210\211\216\223\222\212{xoiht\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200\200~}~~}zz|}\177~}}\177\200\177\177\177\200\200\201\201\201\177~\177\177\177\177\177\177...,
width=848, height=480, sync=0, clipBoxes=0x7fff90b16880, data=0x7f7b475a82a0,
drawable=0x7f7b47aaa110) at ../../src/intel_video.c:1579
intel = 0x7f7b45a31490
adaptor_priv = 0x7f7b475a82a0
pixmap = 0x7f7b475c2670
dstPitch = 424
dstPitch2 = 848
dstBox = {x1 = 84, y1 = 41, x2 = 1283, y2 = 720}
crtc = 0x7f7b45a3e600
top = 0
left = 0
npixels = 848
---Type return to continue, or q return to quit---
nlines = 480
#7 0x7f7b44c419ce in xf86XVPutImage (client=optimized out,
pDraw=0x7f7b47aaa110, pPort=0x7f7b475a90d0, pGC=optimized out,
src_x=optimized out, src_y=optimized out, src_w=848,
src_h=480, drw_x=0, drw_y=0, drw_w=1199, drw_h=679, format=0x7f7b475a8ed0,
data=0x7f7b3d5b5000 \200\200\200\200\200\200\200\200, '\177' repeats 14
times, ~~~}}~~\177\177\177\200\200\200\200\200\200\200\200, '\177' repeats
16 times\200,
Bug#651623: intel_uxa_prepare_access: bo map failed: Cannot allocate memory
On Sun, Dec 11, 2011 at 01:14:42AM +0100, Gonéri Le Bouder wrote: On Sat, Dec 10, 2011 at 08:50:50PM +0300, Roman Sakal wrote: Package: xserver-xorg-video-intel Version: 2:2.17.0-1 Severity: important I've the same probleme since the 2.17 upgrade. Here a backtrace: FYI, There is no breakage anymore if I turn kms off in /etc/modprobe.d/i915-kms.conf -- Gonéri -- 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/20111211002347.GA7252@localhost
Bug#651316: libdrm-intel1: X.org crashes when I try to play a video
tag 651316 - patch fixed-upstream thanks Hi, Modestas Vainius mo...@debian.org (10/12/2011): tags 651316 patch fixed-upstream thanks please don't do that. There are several bugs here, plenty of reporters, at least 3 involved packages, and different causes. That's enough of a mess. I had the same problem. The backtrace of X crashes was: […] I upgraded libdrm to the master branch as of writing ( dd9a5b4f7fb07c78db4e7481bedca1b981030e3f ) and the problem is gone now. I suggest pulling the relevant patches into the debian package as the problem is pretty serious. I was not able to get any video to play due this crash. With latest master (libdrm+xxvintel) and patched libva, crashes seem to be gone, but playback is still failing, at least on a machine of mine. With latest libdrm master (and released xxvintel), I'm still getting crashes, possibly with an extra kernel bug, so I'm keeping this bug open with no tag for now. Mraw, KiBi. signature.asc Description: Digital signature
Processed: Re: Bug#651316: libdrm-intel1: X.org crashes when I try to play a video
Processing commands for cont...@bugs.debian.org: tag 651316 - patch fixed-upstream Bug #651316 [libdrm-intel1] libdrm-intel1: X.org crashes when I try to play a video Removed tag(s) fixed-upstream and patch. thanks Stopping processing here. Please contact me if you need assistance. -- 651316: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=651316 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems -- 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/handler.s.c.132356566328942.transcr...@bugs.debian.org
Bug#651623: intel_uxa_prepare_access: bo map failed: Cannot allocate memory
Gonéri Le Bouder gon...@rulezlan.org (11/12/2011): FYI, There is no breakage anymore if I turn kms off in /etc/modprobe.d/i915-kms.conf Of course. Look at Xorg.0.log, you'll see you're either using fbdev or vesa, since KMS is mandatory for intel (and has been for a while now). Mraw, KiBi. signature.asc Description: Digital signature
Re: Debian testing: alt-tab doesn't switch windows on LXDE
Hi Steinar. Steinar Bang s...@dod.no (10/12/2011): I installed the lxde desktop, and it works fine, except that Alt-Tab doesn't work for switching windows, and Alt-F2 doesn't pop up a run command dialog box. Alt vs. Meta? The Alt-Tab event is picked up by the applications. Eg. emacs tries to do expand, in many modes, and indent in some. So there is nothing sitting on top grabbing them. Has anyone else seen this? Do you know any tricks to try? Any debugging tricks? Check what happens in “xev”? I'd suggest trying debian-user@ or the LXDE maintainers/developers. I'm afraid it hasn't much to do with X itself. Mraw, KiBi. signature.asc Description: Digital signature
Bug#651623: intel_uxa_prepare_access: bo map failed: Cannot allocate memory
Gonéri Le Bouder gon...@rulezlan.org (11/12/2011): On Sat, Dec 10, 2011 at 08:50:50PM +0300, Roman Sakal wrote: Package: xserver-xorg-video-intel Version: 2:2.17.0-1 Severity: important I've the same probleme since the 2.17 upgrade. Here a backtrace: libdrm-intel12.4.28-1 xserver-xorg-video-intel 2:2.17.0-1 xserver-xorg 1:7.6+9 xserver-xorg-core2:1.11.2.901-1 I think we have patches for this on the driver side (in git master), but you should be getting a functional setup by downgrading libdrm* to 2.4.27*. See #651316. I'm still waiting for some feedback (see for example https://bugs.freedesktop.org/43704) to decide what to do with it. Mraw, KiBi. signature.asc Description: Digital signature
Processed: tagging 651316
Processing commands for cont...@bugs.debian.org: # http://blog.mraw.org/2011/12/11/Fixing_libdrm_for_intel/ tags 651316 + pending Bug #651316 [libdrm-intel1] libdrm-intel1: X.org crashes when I try to play a video Added tag(s) pending. thanks Stopping processing here. Please contact me if you need assistance. -- 651316: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=651316 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems -- 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/handler.s.c.132357127918264.transcr...@bugs.debian.org
