On Fri, 7 Sep 2001, François Gouget wrote:
I had a report of a crash and it turned out to happen during the XVidMode initialization. What happened is that for some unknown reason XF86VidModeGetAllModeLines returned a single mode with a width and/or height of zero (htotal/vtotal). This caused a division by zero in convert_modeline.
Last time this happened, we considered this a bug in the X server, as under these circumstances, xvidtune also crashes.
Then will come the work of trying to reproduce the problem and see how best to deal with this server (XFree86-SVGA 4.1.0 apparently).
I don't think such a server exists. There's either XFree86 4.x with loadable driver modules, or there's XFree86-SVGA 3.x. The latter sounds more likely to have this kind of bug.
On Sat, 8 Sep 2001, Ove Kaaven wrote:
On Fri, 7 Sep 2001, François Gouget wrote:
I had a report of a crash and it turned out to happen during the XVidMode initialization. What happened is that for some unknown reason XF86VidModeGetAllModeLines returned a single mode with a width and/or height of zero (htotal/vtotal). This caused a division by zero in convert_modeline.
Last time this happened, we considered this a bug in the X server, as under these circumstances, xvidtune also crashes.
It might still be nice to automatically detect such buggy drivers and then either completely disable XVidMode, or ignore such modelines. Can we get three modelines with just one such invalid modeline in the middle? In the example I got there was only one modeline anyway.
Then will come the work of trying to reproduce the problem and see how best to deal with this server (XFree86-SVGA 4.1.0 apparently).
I don't think such a server exists. There's either XFree86 4.x with loadable driver modules, or there's XFree86-SVGA 3.x. The latter sounds more likely to have this kind of bug.
Yes, I meant the SVGA driver of XFree86 4.1.0 (from Debian SID). I was explained that this is so because the laptop's graphics card is not supported by an accelerated driver.
-- Francois Gouget fgouget@free.fr http://fgouget.free.fr/ Hiroshima '45 - Czernobyl '86 - Windows '95
On Sat, 8 Sep 2001, Francois Gouget wrote:
It might still be nice to automatically detect such buggy drivers and then either completely disable XVidMode, or ignore such modelines. Can we get three modelines with just one such invalid modeline in the middle? In the example I got there was only one modeline anyway.
I think a buggy driver is likely to give invalid data for all modelines.
I don't think such a server exists. There's either XFree86 4.x with loadable driver modules, or there's XFree86-SVGA 3.x. The latter sounds more likely to have this kind of bug.
Yes, I meant the SVGA driver of XFree86 4.1.0 (from Debian SID).
That still doesn't exist.
ovek@acroyali:~$ ls /usr/X11R6/lib/modules/drivers apm_drv.o fbdev_drv.o neomagic_drv.o tdfx_drv.o ark_drv.o glide_drv.o nv_drv.o tga_drv.o ati_drv.o glint_drv.o r128_drv.o trident_drv.o atimisc_drv.o i128_drv.o radeon_drv.o tseng_drv.o chips_drv.o i740_drv.o rendition_drv.o vesa_drv.o cirrus_alpine.o i810_drv.o s3virge_drv.o vga_drv.o cirrus_drv.o imstt_drv.o savage_drv.o vmware_drv.o cirrus_laguna.o linux siliconmotion_drv.o cyrix_drv.o mga_drv.o sis_drv.o
There's no svga driver there. However...
ovek@acroyali:~$ apt-cache show xserver-svga Package: xserver-svga Priority: optional Section: x11 Installed-Size: 3236 Maintainer: Branden Robinson branden@debian.org Architecture: i386 Source: xfree86v3 Version: 3.3.6-38 Replaces: xbase (<< 3.3.2.1-1) Provides: xserver Depends: debconf, xserver-common-v3 (>= 3.3.6-17), libc6 (>= 2.2.1-2), zlib1g (>= 1:1.1.3) Filename: pool/main/x/xfree86v3/xserver-svga_3.3.6-38_i386.deb Size: 1350144 MD5sum: 0a96d8350bd9fe9a0ca0f2ef67d51f0c Description: X server for SVGA graphics cards
does exist in Debian sid, mostly for users with cards not supported by XFree86 4.x - but it's clearly 3.3.6, not 4.1. All the XF86 4.x stuff is in xserver-xfree86.
-
Francois Gouget -
Ove Kaaven