[arch-general] Anybody else have problems with the new Nvidia drivers (275.09.07-1)
Hi all, As the title of this email suggests I'm having trouble with the new Nvidia driver. I've attached my Xorg log, as I'm sure that there are a log of people who understand the ouput better than I do. I've had to revert to the previous driver (270.41.19-3) to get my system up and running again. What seems odd to me from the log is that it doesn't even try to load the nvidia driver. I'm using Xorg without any explicit configuration. My card is a GeForce 210 and I'm running a 32 bit arch. Did I forget something? Anybody have any suggestions? Regards, Pico
On 06/17/2011 08:46 PM, Pico Geyer wrote:
Hi all,
As the title of this email suggests I'm having trouble with the new Nvidia driver. I've attached my Xorg log, as I'm sure that there are a log of people who understand the ouput better than I do.
attachments doesn't work on this list
I've had to revert to the previous driver (270.41.19-3) to get my system up and running again. What seems odd to me from the log is that it doesn't even try to load the nvidia driver.
if it doesn't load the driver it means you don't use our xorg-server. for 275.09.07 i dropped the configuration that comes with it because now is loaded by xorg-server !next -- Ionuț
On Friday, June 17, 2011 12:55:01 PM Ionut Biru wrote:
On 06/17/2011 08:46 PM, Pico Geyer wrote:
Hi all,
As the title of this email suggests I'm having trouble with the new Nvidia driver. I've attached my Xorg log, as I'm sure that there are a log of people who understand the ouput better than I do.
attachments doesn't work on this list
I've had to revert to the previous driver (270.41.19-3) to get my system up and running again. What seems odd to me from the log is that it doesn't even try to load the nvidia driver.
if it doesn't load the driver it means you don't use our xorg-server.
for 275.09.07 i dropped the configuration that comes with it because now is loaded by xorg-server
!next
Would be nice if you could actually be specific about your problem. Is Xorg not working? Is it freezing? Also, no attachments on here.
On Fri, Jun 17, 2011 at 7:55 PM, Ionut Biru <ibiru@archlinux.org> wrote:
On 06/17/2011 08:46 PM, Pico Geyer wrote:
Hi all,
As the title of this email suggests I'm having trouble with the new Nvidia driver. I've attached my Xorg log, as I'm sure that there are a log of people who understand the ouput better than I do.
attachments doesn't work on this list
Ok, I did not realize that. I'll put my log inline at the bottom of this mail,
I've had to revert to the previous driver (270.41.19-3) to get my system up and running again. What seems odd to me from the log is that it doesn't even try to load the nvidia driver.
if it doesn't load the driver it means you don't use our xorg-server.
I'm not sure if I understand you correctly, but I didn't install a custom xorg-server, it comes from extra. pacman -Qs xorg-server local/xorg-server 1.10.2-1 (xorg) Xorg X server local/xorg-server-common 1.10.2-1 Xorg server common files local/xorg-server-devel 1.10.2-1 Development files for the X.Org X server local/xorg-server-utils 7.6-2 Transition package depending on xorg server utilities
for 275.09.07 i dropped the configuration that comes with it because now is loaded by xorg-server
I did try to manually load the nvidia module and then start X but that does not change the situation. I'm afraid I do not have the machine with me this weekend, so I'll need to follow up on this tomorrow.
!next
-- Ionuț
Xorg log: [ 20.621] X.Org X Server 1.10.2 Release Date: 2011-05-28 [ 20.621] X Protocol Version 11, Revision 0 [ 20.621] Build Operating System: Linux 2.6.38-ARCH i686 [ 20.621] Current Operating System: Linux naix 2.6.39-ARCH #1 SMP PREEMPT Tue Jun 7 05:49:02 UTC 2011 i686 [ 20.621] Kernel command line: root=/dev/disk/by-uuid/37a133fd-bd99-435a-967a-546463f8afcc ro [ 20.621] Build Date: 30 May 2011 08:16:10AM [ 20.621] [ 20.621] Current version of pixman: 0.22.0 [ 20.621] Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. [ 20.621] Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. [ 20.622] (==) Log file: "/var/log/Xorg.0.log", Time: Fri Jun 17 19:20:37 2011 [ 20.668] (==) Using config file: "/etc/X11/xorg.conf" [ 20.668] (==) Using config directory: "/etc/X11/xorg.conf.d" [ 20.700] (==) No Layout section. Using the first Screen section. [ 20.700] (==) No screen section available. Using defaults. [ 20.700] (**) |-->Screen "Default Screen Section" (0) [ 20.700] (**) | |-->Monitor "<default monitor>" [ 20.700] (==) No device specified for screen "Default Screen Section". Using the first device section listed. [ 20.700] (**) | |-->Device "Card0" [ 20.700] (==) No monitor specified for screen "Default Screen Section". Using a default monitor configuration. [ 20.700] (==) Automatically adding devices [ 20.700] (==) Automatically enabling devices [ 20.710] (WW) The directory "/usr/share/fonts/OTF/" does not exist. [ 20.710] Entry deleted from font path. [ 20.752] (==) FontPath set to: /usr/share/fonts/misc/, /usr/share/fonts/TTF/, /usr/share/fonts/Type1/, /usr/share/fonts/100dpi/, /usr/share/fonts/75dpi/ [ 20.752] (==) ModulePath set to "/usr/lib/xorg/modules" [ 20.752] (II) The server relies on udev to provide the list of input devices. If no devices become available, reconfigure udev or disable AutoAddDevices. [ 20.752] (II) Loader magic: 0x822e120 [ 20.752] (II) Module ABI versions: [ 20.752] X.Org ANSI C Emulation: 0.4 [ 20.752] X.Org Video Driver: 10.0 [ 20.752] X.Org XInput driver : 12.2 [ 20.752] X.Org Server Extension : 5.0 [ 20.752] (--) PCI:*(0:1:0:0) 10de:0a65:1458:34d5 rev 162, Mem @ 0xf9000000/16777216, 0xd0000000/268435456, 0xee000000/33554432, I/O @ 0x0000bf00/128, BIOS @ 0x????????/524288 [ 20.752] (WW) Open ACPI failed (/var/run/acpid.socket) (No such file or directory) [ 20.752] (II) LoadModule: "extmod" [ 20.777] (II) Loading /usr/lib/xorg/modules/extensions/libextmod.so [ 20.784] (II) Module extmod: vendor="X.Org Foundation" [ 20.784] compiled for 1.10.2, module version = 1.0.0 [ 20.784] Module class: X.Org Server Extension [ 20.784] ABI class: X.Org Server Extension, version 5.0 [ 20.784] (II) Loading extension MIT-SCREEN-SAVER [ 20.784] (II) Loading extension XFree86-VidModeExtension [ 20.784] (II) Loading extension XFree86-DGA [ 20.784] (II) Loading extension DPMS [ 20.784] (II) Loading extension XVideo [ 20.784] (II) Loading extension XVideo-MotionCompensation [ 20.784] (II) Loading extension X-Resource [ 20.784] (II) LoadModule: "dbe" [ 20.784] (II) Loading /usr/lib/xorg/modules/extensions/libdbe.so [ 20.791] (II) Module dbe: vendor="X.Org Foundation" [ 20.791] compiled for 1.10.2, module version = 1.0.0 [ 20.791] Module class: X.Org Server Extension [ 20.791] ABI class: X.Org Server Extension, version 5.0 [ 20.791] (II) Loading extension DOUBLE-BUFFER [ 20.791] (II) LoadModule: "glx" [ 20.791] (II) Loading /usr/lib/xorg/modules/extensions/libglx.so [ 21.517] (II) Module glx: vendor="NVIDIA Corporation" [ 21.517] compiled for 4.0.2, module version = 1.0.0 [ 21.517] Module class: X.Org Server Extension [ 21.517] (II) NVIDIA GLX Module 275.09.07 Wed Jun 8 16:01:21 PDT 2011 [ 21.517] (II) Loading extension GLX [ 21.517] (II) LoadModule: "record" [ 21.517] (II) Loading /usr/lib/xorg/modules/extensions/librecord.so [ 21.529] (II) Module record: vendor="X.Org Foundation" [ 21.529] compiled for 1.10.2, module version = 1.13.0 [ 21.529] Module class: X.Org Server Extension [ 21.529] ABI class: X.Org Server Extension, version 5.0 [ 21.529] (II) Loading extension RECORD [ 21.529] (II) LoadModule: "dri" [ 21.529] (II) Loading /usr/lib/xorg/modules/extensions/libdri.so [ 21.547] (II) Module dri: vendor="X.Org Foundation" [ 21.547] compiled for 1.10.2, module version = 1.0.0 [ 21.547] ABI class: X.Org Server Extension, version 5.0 [ 21.547] (II) Loading extension XFree86-DRI [ 21.547] (II) LoadModule: "dri2" [ 21.554] (II) Loading /usr/lib/xorg/modules/extensions/libdri2.so [ 21.555] (II) Module dri2: vendor="X.Org Foundation" [ 21.555] compiled for 1.10.2, module version = 1.2.0 [ 21.555] ABI class: X.Org Server Extension, version 5.0 [ 21.555] (II) Loading extension DRI2 [ 21.555] (II) LoadModule: "nouveau" [ 21.599] (WW) Warning, couldn't open module nouveau [ 21.599] (II) UnloadModule: "nouveau" [ 21.599] (II) Unloading nouveau [ 21.599] (EE) Failed to load module "nouveau" (module does not exist, 0) [ 21.599] (EE) No drivers available. [ 21.599] Fatal server error: [ 21.599] no screens found [ 21.599] Please consult the The X.Org Foundation support at http://wiki.x.org for help. [ 21.599] Please also check the log file at "/var/log/Xorg.0.log" for additional information. [ 21.599] Regards, Pico
On 06/19/2011 05:56 PM, Pico Geyer wrote:
[ 20.668] (==) Using config file: "/etc/X11/xorg.conf"
isn't that obvious ?
[ 21.555] (II) LoadModule: "nouveau" [ 21.599] (WW) Warning, couldn't open module nouveau [ 21.599] (II) UnloadModule: "nouveau" [ 21.599] (II) Unloading nouveau [ 21.599] (EE) Failed to load module "nouveau" (module does not exist, 0) [ 21.599] (EE) No drivers available. [ 21.599] Fatal server error:
rm /etc/X11/xorg.conf -- Ionuț
On Sun, Jun 19, 2011 at 5:03 PM, Ionut Biru <ibiru@archlinux.org> wrote:
On 06/19/2011 05:56 PM, Pico Geyer wrote:
[ 20.668] (==) Using config file: "/etc/X11/xorg.conf"
isn't that obvious ?
Indeed, it is. Sorry for the noise!
[ 21.555] (II) LoadModule: "nouveau" [ 21.599] (WW) Warning, couldn't open module nouveau [ 21.599] (II) UnloadModule: "nouveau" [ 21.599] (II) Unloading nouveau [ 21.599] (EE) Failed to load module "nouveau" (module does not exist, 0) [ 21.599] (EE) No drivers available. [ 21.599] Fatal server error:
rm /etc/X11/xorg.conf
-- Ionuț
Pico
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver. ATTENTION: This can crash your X server and corrupt memory! Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it. [1] http://www.abload.de/img/nvbugy7cd.jpg -- Regards, Richard Schütz
On Friday, June 17, 2011 04:10:01 PM Richard Schütz wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
Are we sure ths is a bug with the nVidia driver? Can someone load that image who isn't using it?
275.09.07 here, no problems loading big images. I do use a custom kernel with with a recompiled nvidia drivers, but I doubt that would have anything to do with it. 2011/6/17 Yaro Kasear <yaro@marupa.net>:
On Friday, June 17, 2011 04:10:01 PM Richard Schütz wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
Are we sure ths is a bug with the nVidia driver? Can someone load that image who isn't using it?
On Sat, Jun 18, 2011 at 12:09 AM, Yaro Kasear <yaro@marupa.net> wrote:
On Friday, June 17, 2011 04:10:01 PM Richard Schütz wrote:
ATTENTION: This can crash your X server and corrupt memory!
Are we sure ths is a bug with the nVidia driver? Can someone load that image who isn't using it?
I braved the danger. My X has remained a faithful ally. lspci -vv -s 00:02.0|grep modules Kernel modules: i915 -- Sébastien Luttringer www.seblu.net
On Fri, 2011-06-17 at 17:09 -0500, Yaro Kasear wrote:
On Friday, June 17, 2011 04:10:01 PM Richard Schütz wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
Are we sure ths is a bug with the nVidia driver? Can someone load that image who isn't using it?
Loads fine for me in firefox and midori, using intel driver. == John K Pate http://homepages.inf.ed.ac.uk/s0930006/ -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.
Am 18.06.2011 00:20, schrieb John K Pate:
On Fri, 2011-06-17 at 17:09 -0500, Yaro Kasear wrote:
On Friday, June 17, 2011 04:10:01 PM Richard Schütz wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
Are we sure ths is a bug with the nVidia driver? Can someone load that image who isn't using it?
Loads fine for me in firefox and midori, using intel driver.
I can confirm that i915 works fine in that case. -- Regards, Richard Schütz
<snip>
Are we sure ths is a bug with the nVidia driver? Can someone load that image who isn't using it?
confirming display screwage with firefox on an onboard nvidia chip. nasty. 02:00.0 VGA compatible controller: nVidia Corporation C77 [GeForce 8200] (rev a2) cheers!
On Friday, June 17, 2011 05:35:06 PM Martti Kühne wrote:
<snip>
Are we sure ths is a bug with the nVidia driver? Can someone load that image who isn't using it?
confirming display screwage with firefox on an onboard nvidia chip. nasty. 02:00.0 VGA compatible controller: nVidia Corporation C77 [GeForce 8200] (rev a2)
cheers!
Unfortunately I can confirm, though X didn't crash or flicker. What I saw was visual data from everything else *but* the image. I hope that might help with a bug report to nVidia besides "flicker and crash." Using latest nVidia driver with a GeForec GT 240.
works fine here with thinkpad - NVIDIA GPU Quadro FX 880M - latest linux and nvidia on kde. tested on rekonq and chromium.
I have nvidia 9800GT. After update nvidia driver KDE's desktop effects are chrash and disable :(
*before* nvidia update I had very strange artifacts and freezes with lxterminal. now it's all right with all kernels and applications.
Yaro Kasear [2011.06.17 1709 -0500]:
On Friday, June 17, 2011 04:10:01 PM Richard Schütz wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
Are we sure ths is a bug with the nVidia driver? Can someone load that image who isn't using it?
Loaded the image without problems on my laptop with Intel Core i7 built-in graphics. On my desktop with GeForce 7500 GT, the screen flickers and then X crashes. Window manager, firefox version, etc. are the same on the two machines. Cheers, Norbert
On 17 June 2011 23:10, Richard Schütz <r.schtz@t-online.de> wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
[1] http://www.abload.de/img/nvbugy7cd.jpg
-- Regards, Richard Schütz
Works fine using the nvidia 275.09.07-1 package using nVidia 9600 GT. However the image is black, which is, I suppose, a bug in Firefox. I'm seeing this issue with some images straight from the camera for a quite a long time and only in firefox. Actually, I can open 30+Mpix images in firefox without problems. Lukas
On Sat, Jun 18, 2011 at 7:10 AM, Richard Schütz <r.schtz@t-online.de> wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
Image loads fine in Chromium. In Firefox 4 it eventually displays a black rectangle after both of my screens flicker for several seconds. Packages: kernel26 2.6.39.1-1 nvidia 275.09.07-1 nvidia-utils 275.09.07-1
On Sat, Jun 18, 2011 at 4:43 PM, Mike Sampson <mike@sambodata.com> wrote:
On Sat, Jun 18, 2011 at 7:10 AM, Richard Schütz <r.schtz@t-online.de> wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
Image loads fine in Chromium. In Firefox 4 it eventually displays a black rectangle after both of my screens flicker for several seconds.
Packages:
kernel26 2.6.39.1-1 nvidia 275.09.07-1 nvidia-utils 275.09.07-1
Sorry, forgot to mention I'm using a GeForce 8600 GT Mike
On Friday, June 17, 2011 23:10:01 Richard Schütz wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
I can confirm crash of Xorg with Firefox 4 on Nvidia GT218 with 275.09.7 drivers. Epiphany displays weird artefacts, but does not crash. Opera and Chrome are OK. Looks like only Gecko has these problems. Dan -- --- Dan Vrátil dan@progdan.cz Tel.: +420732326870 Jabber: progdan@jabber.cz Tento email neobsahuje žádné viry, protože odesílatel nepoužívá Windows. / This email does not contain any viruses because the sender does not use Windows.
Am 18.06.2011 09:08, schrieb Dan Vratil:
On Friday, June 17, 2011 23:10:01 Richard Schütz wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
I can confirm crash of Xorg with Firefox 4 on Nvidia GT218 with 275.09.7 drivers. Epiphany displays weird artefacts, but does not crash. Opera and Chrome are OK. Looks like only Gecko has these problems.
I don't think so. Epiphany is based on WebKit these days. And because I can trigger the bug in EOG (Eye of GNOME) it doesn't look like a browser problem for me at all. Perhaps it is a problem of GTK together with nvidia 275.09.07. I tried feh (a simple X11 image viewer that doesn't use GTK or Qt) and it works fine without flickering, artifacts or something else strange. -- Regards, Richard Schütz
Image loads fine here with nvidia 7600GT. I tried firefox-pgo4.0.1(without xulrunner), firefox-beta-bin-5b7(with xulrunner), chromium, konqueror, eog, gpicview, gwenview. Using KDE atm with: nvidia 275.09.07-1 kernel26 2.6.39.1-1 On Sat, Jun 18, 2011 at 11:03 AM, Richard Schütz <r.schtz@t-online.de> wrote:
Am 18.06.2011 09:08, schrieb Dan Vratil:
On Friday, June 17, 2011 23:10:01 Richard Schütz wrote:
The new driver really seems to have a major problem. I figured out that you just need an image with 2047px width to screw up the driver.
ATTENTION: This can crash your X server and corrupt memory!
Example: [1], an otherwise harmless picture. Every picture with the same width will work, too. I could trigger the bug at least with Firefox, Midori, Epiphany and EOG. It looks like some applications like Chromium alter the size, so they don't trigger it.
I can confirm crash of Xorg with Firefox 4 on Nvidia GT218 with 275.09.7 drivers. Epiphany displays weird artefacts, but does not crash. Opera and Chrome are OK. Looks like only Gecko has these problems.
I don't think so. Epiphany is based on WebKit these days. And because I can trigger the bug in EOG (Eye of GNOME) it doesn't look like a browser problem for me at all.
Perhaps it is a problem of GTK together with nvidia 275.09.07. I tried feh (a simple X11 image viewer that doesn't use GTK or Qt) and it works fine without flickering, artifacts or something else strange.
-- Regards, Richard Schütz
-- (\_ /) copy the bunny to your profile (0.o ) to help him achieve world domination. (> <) come join the dark side. /_|_\ (we have cookies.)
On Sat, Jun 18, 2011 at 10:03 AM, Richard Schütz <r.schtz@t-online.de> wrote:
Am 18.06.2011 09:08, schrieb Dan Vratil: I don't think so. Epiphany is based on WebKit these days. And because I can trigger the bug in EOG (Eye of GNOME) it doesn't look like a browser problem for me at all.
Perhaps it is a problem of GTK together with nvidia 275.09.07. I tried feh (a simple X11 image viewer that doesn't use GTK or Qt) and it works fine without flickering, artifacts or something else strange.
-- Regards, Richard Schütz
Update, I also see anything but the picture when the bug occurs, namely random garbage. I can confirm that feh, launched the regular way does not trigger the bug, although, also I have to confirm that display (imagemagick-6.6.9.8-1) actually does trigger the bug. Actually my guess is feh restricts itself to become larger than the screen's resolution (which is smaller than 2047 here) and thus is just not storing the full resolution image in video memory. I just managed to something like trigger the bug, X hang for about 4 seconds here and will display some garbage, with $ feh --geometry 2047x1529 nvbugy7cd.jpg heh, after trying this several times, the hang disappears. Might this be related to caching? cheers!
Am 18.06.2011 11:29, schrieb Martti Kühne:
On Sat, Jun 18, 2011 at 10:03 AM, Richard Schütz<r.schtz@t-online.de> wrote:
Am 18.06.2011 09:08, schrieb Dan Vratil: I don't think so. Epiphany is based on WebKit these days. And because I can trigger the bug in EOG (Eye of GNOME) it doesn't look like a browser problem for me at all.
Perhaps it is a problem of GTK together with nvidia 275.09.07. I tried feh (a simple X11 image viewer that doesn't use GTK or Qt) and it works fine without flickering, artifacts or something else strange.
-- Regards, Richard Schütz
Update, I also see anything but the picture when the bug occurs, namely random garbage. I can confirm that feh, launched the regular way does not trigger the bug, although, also I have to confirm that display (imagemagick-6.6.9.8-1) actually does trigger the bug. Actually my guess is feh restricts itself to become larger than the screen's resolution (which is smaller than 2047 here) and thus is just not storing the full resolution image in video memory. I just managed to something like trigger the bug, X hang for about 4 seconds here and will display some garbage, with
$ feh --geometry 2047x1529 nvbugy7cd.jpg
heh, after trying this several times, the hang disappears. Might this be related to caching?
cheers!
Yes, you are right: display crashed my X server right now. So this seems to be even unrelated to the toolkit. -- Regards, Richard Schütz
On Saturday, June 18, 2011 04:57:07 AM Richard Schütz wrote:
Am 18.06.2011 11:29, schrieb Martti Kühne:
On Sat, Jun 18, 2011 at 10:03 AM, Richard Schütz<r.schtz@t-online.de> wrote:
Am 18.06.2011 09:08, schrieb Dan Vratil: I don't think so. Epiphany is based on WebKit these days. And because I can trigger the bug in EOG (Eye of GNOME) it doesn't look like a browser problem for me at all.
Perhaps it is a problem of GTK together with nvidia 275.09.07. I tried feh (a simple X11 image viewer that doesn't use GTK or Qt) and it works fine without flickering, artifacts or something else strange.
-- Regards, Richard Schütz
Update, I also see anything but the picture when the bug occurs, namely random garbage. I can confirm that feh, launched the regular way does not trigger the bug, although, also I have to confirm that display (imagemagick-6.6.9.8-1) actually does trigger the bug. Actually my guess is feh restricts itself to become larger than the screen's resolution (which is smaller than 2047 here) and thus is just not storing the full resolution image in video memory. I just managed to something like trigger the bug, X hang for about 4 seconds here and will display some garbage, with
$ feh --geometry 2047x1529 nvbugy7cd.jpg
heh, after trying this several times, the hang disappears. Might this be related to caching?
cheers!
Yes, you are right: display crashed my X server right now. So this seems to be even unrelated to the toolkit.
Okay, so, has anyone filed a bug report with nVidia?
Loads fine here with Opera. By the way, nice picture ;) Nvidia 9600M GT kernel26-ck-core2 1:2.6.39.1-3 nvidia-ck-core2 275.09.07-1 KDE 4.6.3 On Sat, 18 Jun 2011 10:56:06 -0400, Yaro Kasear <yaro@marupa.net> wrote:
On Saturday, June 18, 2011 04:57:07 AM Richard Schütz wrote:
On Sat, Jun 18, 2011 at 10:03 AM, Richard Schütz<r.schtz@t-online.de> wrote:
Am 18.06.2011 09:08, schrieb Dan Vratil: I don't think so. Epiphany is based on WebKit these days. And because I can trigger the bug in EOG (Eye of GNOME) it doesn't look like a browser problem for me at all.
Perhaps it is a problem of GTK together with nvidia 275.09.07. I
Am 18.06.2011 11:29, schrieb Martti Kühne: tried
feh (a simple X11 image viewer that doesn't use GTK or Qt) and it works fine without flickering, artifacts or something else strange.
-- Regards, Richard Schütz
Update, I also see anything but the picture when the bug occurs, namely random garbage. I can confirm that feh, launched the regular way does not trigger the bug, although, also I have to confirm that display (imagemagick-6.6.9.8-1) actually does trigger the bug. Actually my guess is feh restricts itself to become larger than the screen's resolution (which is smaller than 2047 here) and thus is just not storing the full resolution image in video memory. I just managed to something like trigger the bug, X hang for about 4 seconds here and will display some garbage, with
$ feh --geometry 2047x1529 nvbugy7cd.jpg
heh, after trying this several times, the hang disappears. Might this be related to caching?
cheers!
Yes, you are right: display crashed my X server right now. So this seems to be even unrelated to the toolkit.
Okay, so, has anyone filed a bug report with nVidia?
Em 18-06-2011 05:03, Richard Schütz escreveu:
I don't think so. Epiphany is based on WebKit these days. And because I can trigger the bug in EOG (Eye of GNOME) it doesn't look like a browser problem for me at all.
Perhaps it is a problem of GTK together with nvidia 275.09.07. I tried feh (a simple X11 image viewer that doesn't use GTK or Qt) and it works fine without flickering, artifacts or something else strange.
I think you're right. I've downloaded and opened the file with some image viewers: geeqie / fotoxx / gimp are fine, but gthumb froze (eventually freezing xorg with it). By the way, opening other images, even bigger than yours, on gthumb worked without a glitch. It seems that there is something related specifically to that image involved in the freezing. nvidia 275.09.07-1 nvidia-utils 275.09.07-1 geeqie 1.0-5 fotoxx 11.06.1-1 gimp 2.6.11-5 GeForce 8600 GT Armando
Am 19.06.2011 04:07, schrieb Armando M. Baratti:
Em 18-06-2011 05:03, Richard Schütz escreveu:
I don't think so. Epiphany is based on WebKit these days. And because I can trigger the bug in EOG (Eye of GNOME) it doesn't look like a browser problem for me at all.
Perhaps it is a problem of GTK together with nvidia 275.09.07. I tried feh (a simple X11 image viewer that doesn't use GTK or Qt) and it works fine without flickering, artifacts or something else strange.
I think you're right. I've downloaded and opened the file with some image viewers: geeqie / fotoxx / gimp are fine, but gthumb froze (eventually freezing xorg with it).
Not all viewers trigger the bug, that's right. There surely is a small difference in the way the image is shown.
By the way, opening other images, even bigger than yours, on gthumb worked without a glitch. It seems that there is something related specifically to that image involved in the freezing.
As I wrote in my first post: it's the width of 2047px. You could take every picture with that width. The effect gets stronger with more height.
nvidia 275.09.07-1 nvidia-utils 275.09.07-1
geeqie 1.0-5 fotoxx 11.06.1-1 gimp 2.6.11-5
GeForce 8600 GT
nvidia 275.09.07 (x86_64) on GeForce 8600 GT here. I heard that older versions of the nvidia driver should be fine, but had no time to test it myself.
Armando
-- Regards, Richard Schütz
participants (17)
-
Adriano Moura
-
Armando M. Baratti
-
Axilleas P
-
Bernardo Barros
-
Dan Vratil
-
Ionut Biru
-
John K Pate
-
Joker-jar
-
Lukáš Jirkovský
-
Martin
-
Martti Kühne
-
Mike Sampson
-
Norbert Zeh
-
Pico Geyer
-
Richard Schütz
-
Seblu
-
Yaro Kasear