Re: [Dri-devel] problem with DRM/Radeon 7000
On Tue, Feb 11, 2003 at 03:07:57PM +0100, Michel Dänzer mentioned: The point is that this is either the wrong (too old) radeon module, or the radeon DRM is already built into the kernel (in which case you can load another one as a module unfortunately, but clients see the one built into the kernel). Ah. I should have thought of that! OK, I've a rebuild kernel. radeon.o and agpgart.o are modprobed. However, I get the same error. Is there a method I can use strings or some such to verify that all the required libraries are matching versions ? (Problem being I'm no longer sure what libraries are supposed to be used). glxinfo -v does output; OpenGL renderer string: Mesa DRI Radeon 20021125 AGP 1x x86/MMX/SSE NO-TCL OpenGL version string: 1.2 Mesa 5.0 Is this good ? John --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] problem with DRM/Radeon 7000
On Wed, Feb 12, 2003 at 04:33:03PM +0100, Michel Dänzer mentioned: OK, I've a rebuild kernel. radeon.o and agpgart.o are modprobed. However, I get the same error. What does dmesg | grep '\[drm\]' say? ratbert[11]:~% dmesg | grep '\[drm\]' [drm] AGP 0.99 on Intel i845 @ 0xf800 64MB [drm] Initialized radeon 1.7.0 20020828 on minor 0 glxinfo -v does output; OpenGL renderer string: Mesa DRI Radeon 20021125 AGP 1x x86/MMX/SSE NO-TCL OpenGL version string: 1.2 Mesa 5.0 Is this good ? Yes, it means the DRI should work once the DRM is new enough. Hmm. OK, so I've to match the DRM date and the renderer date. And it looks like the DRM is really old. OK, I'll try update that...properly. John --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] problem with DRM/Radeon 7000
On Tue, Feb 11, 2003 at 12:36:54AM +0100, Michel Dänzer mentioned: On Mon, 2003-02-10 at 19:31, John P. Looney wrote: Hi, I was using the old Radeon drivers that shipped with a dist-upgraded Woody, though they were mostly fine in Quake3, in Wolfenstien, there were severe problems with corrupted textures. After noting a few people were recommended to upgrade to get rid of these problems, I did. To upgrade, I used drm-trunk-module-2.4.18 from debian woody. Huh? Where exactly did you get that from? Get which - the code or the recommendation ? The recommendation was from this list, some months ago, and the code was from the URL below... The code seems to be dated Jan 18 2003. I got the .debs via apt from http://people.debian.org/~daenzer/dri-trunk/ Received signal 8, exiting... This bug should be fixed in the packages for sid at deb http://people.debian.org/~daenzer/dri-trunk-sid/./ I'll hopefully get around to updating the packages for woody/sarge soon (once those for sid have settled a bit), in the meantime I think setting MESA_NO_SSE=1 should serve as a workaround. I tried that, and got: wolf.x86: radeon_ioctl.h:165: radeonAllocCmdBuf: Assertion `rmesa-dri.drmMinor = 3' failed. Received signal 6, exiting... As soon as I get some broadband, I'll upgrade to sid. I'm still on ISDN here, so I didn't think upgrading to sid would be a good idea. Can the source packages for sid be compiled on woody, without much effort ? I'll give that a go, if it's possible. John --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] problem with DRM/Radeon 7000
On Mon, Feb 10, 2003 at 11:24:58PM +0100, Jacek Pop³awski mentioned: On Mon, Feb 10, 2003 at 06:31:37PM +, John P. Looney wrote: Hi, I was using the old Radeon drivers that shipped with a dist-upgraded Woody, though they were mostly fine in Quake3, in Wolfenstien, there were severe problems with corrupted textures. Could you try driver 20021022 (from dri.sf.net downloads)? As I wrote in other post - flickering textures are only visible in latest version of driver. I can't see any mention of older drivers, just a link to: http://www.xfree86.org/~alanh/ Is there an archive ? If the problems are known, and fixable, I don't mind waiting a while for new drivers to be released. John --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] problem with DRM/Radeon 7000
On Die, 2003-02-11 at 10:56, John P. Looney wrote: On Tue, Feb 11, 2003 at 12:36:54AM +0100, Michel Dänzer mentioned: On Mon, 2003-02-10 at 19:31, John P. Looney wrote: Hi, I was using the old Radeon drivers that shipped with a dist-upgraded Woody, though they were mostly fine in Quake3, in Wolfenstien, there were severe problems with corrupted textures. After noting a few people were recommended to upgrade to get rid of these problems, I did. To upgrade, I used drm-trunk-module-2.4.18 from debian woody. Huh? Where exactly did you get that from? Get which - the code or the recommendation ? The recommendation was from this list, some months ago, and the code was from the URL below... Well, that's the source of my confusion - I don't provide a drm-trunk-module-2.4.18 package, and neither does Debian woody. I'll just assume you built it yourself from drm-trunk-module-src. Received signal 8, exiting... This bug should be fixed in the packages for sid at deb http://people.debian.org/~daenzer/dri-trunk-sid/./ I'll hopefully get around to updating the packages for woody/sarge soon (once those for sid have settled a bit), in the meantime I think setting MESA_NO_SSE=1 should serve as a workaround. I tried that, and got: wolf.x86: radeon_ioctl.h:165: radeonAllocCmdBuf: Assertion `rmesa-dri.drmMinor = 3' failed. Received signal 6, exiting... Now you need to make sure the new DRM is actually used. The easiest way to achieve that is probably to disable the DRM in the kernel. Can the source packages for sid be compiled on woody, without much effort ? I'll give that a go, if it's possible. Building shouldn't be a problem, but I'm not sure how the packaging will work out in woody. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] problem with DRM/Radeon 7000
On Tue, Feb 11, 2003 at 02:28:58PM +0100, Michel Dänzer mentioned: Well, that's the source of my confusion - I don't provide a drm-trunk-module-2.4.18 package, and neither does Debian woody. I'll just assume you built it yourself from drm-trunk-module-src. Ah yes, I did build it from your drm-trunk-module-src package. wolf.x86: radeon_ioctl.h:165: radeonAllocCmdBuf: Assertion `rmesa-dri.drmMinor = 3' failed. Received signal 6, exiting... Now you need to make sure the new DRM is actually used. The easiest way to achieve that is probably to disable the DRM in the kernel. I don't understand...I've got these loaded at the moment: ratbert:/var/www# lsmod Module Size Used byNot tainted i810 64472 0 (unused) radeon107896 0 (unused) radeonfb 17888 0 (unused) fbcon-cfb24 4320 0 [radeonfb] fbcon-cfb8 3392 0 [radeonfb] fbcon-cfb32 3744 0 [radeonfb] fbcon-cfb16 4032 0 [radeonfb] Should the radeon.o not be loaded ? What would then provide the DRM functionality ? John --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] problem with DRM/Radeon 7000
On Die, 2003-02-11 at 14:55, John P. Looney wrote: On Tue, Feb 11, 2003 at 02:28:58PM +0100, Michel Dänzer mentioned: wolf.x86: radeon_ioctl.h:165: radeonAllocCmdBuf: Assertion `rmesa-dri.drmMinor = 3' failed. Received signal 6, exiting... Now you need to make sure the new DRM is actually used. The easiest way to achieve that is probably to disable the DRM in the kernel. I don't understand...I've got these loaded at the moment: ratbert:/var/www# lsmod Module Size Used byNot tainted i810 64472 0 (unused) radeon107896 0 (unused) radeonfb 17888 0 (unused) fbcon-cfb24 4320 0 [radeonfb] fbcon-cfb8 3392 0 [radeonfb] fbcon-cfb32 3744 0 [radeonfb] fbcon-cfb16 4032 0 [radeonfb] Should the radeon.o not be loaded ? What would then provide the DRM functionality ? The point is that this is either the wrong (too old) radeon module, or the radeon DRM is already built into the kernel (in which case you can load another one as a module unfortunately, but clients see the one built into the kernel). -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] problem with DRM/Radeon 7000
Hi, I was using the old Radeon drivers that shipped with a dist-upgraded Woody, though they were mostly fine in Quake3, in Wolfenstien, there were severe problems with corrupted textures. After noting a few people were recommended to upgrade to get rid of these problems, I did. To upgrade, I used drm-trunk-module-2.4.18 from debian woody. The code seems to be dated Jan 18 2003. I got the .debs via apt from http://people.debian.org/~daenzer/dri-trunk/ When I try Wolfenstien now, it doesn't even start: Started tty console (use +set ttycon 0 to disable) ^5PunkBuster Client: PunkBuster Client (v1.017 | A0) Enabled ^3PunkBuster Server: PunkBuster Server (v0.993 | A0 C0.0) **DISABLED** Resolving wolfmotd.idsoftware.com wolfmotd.idsoftware.com resolved to 192.246.40.65:27951 Received signal 8, exiting... Shutdown tty console strace reports: 6882 write(2, Using 4/4/4 Color bits, 24 depth..., 53) = 53 6882 write(33, \221\24 \0\1\0\0\0\2\0\0\0m\0\0\0GL_ARB_multitext..., 376) = 376 6882 write(33, +\24\1\0, 4) = 4 6882 read(33, \23\0\'\0\2\0`\0\2\0`\0\1\0\0\0\22\0\0\0\1\0\0\0\254\315..., 32) = 32 6882 read(33, \17\367\'\0\2\0`\0\0E\253\10\230\367\340\1\200\2\200\2..., 32) = 32 6882 read(33, \1\2)\0\0\0\0\0Y\0\200\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0..., 32) = 32 6882 old_mmap(NULL, 266240, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x48ca1000 6882 write(33, \200\5\4\0\0\0\0\0\'\0\0\0\3\0`\0, 16) = 16 6882 read(33, \1\236*\0\0\0\0\0\7\0\0\0\0\0\0\0\214\370\377\277\1\0\0..., 32) = 32 6882 write(33, \200\5\4\0\0\0\0\0\'\0\0\0\4\0`\0, 16) = 16 6882 read(33, \1\236+\0\0\0\0\0\10\0\0\0\0\0\0\0\214\370\377\277\1\0..., 32) = 32 6882 brk(0xa514000)= 0xa514000 6882 rt_sigaction(SIGILL, NULL, {0x80d5014, [ILL], SA_RESTART|0x400}, 8) = 0 6882 rt_sigaction(SIGFPE, NULL, {0x80d5014, [FPE], SA_RESTART|0x400}, 8) = 0 6882 rt_sigaction(SIGILL, {0x464d26b4, [ILL], SA_RESTART|0x400}, {0x80d5014, [ILL], SA_RESTART|0x400}, 8) = 0 6882 rt_sigaction(SIGFPE, {0x464d26d4, [FPE], SA_RESTART|0x400}, {0x80d5014, [FPE], SA_RESTART|0x400}, 8) = 0 6882 --- SIGFPE (Floating point exception) --- Could this be a problem with the DRM upgrade ? I'd upgraded all the packages in that apt repository mentionedcould their be more i'm missing ? John --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] problem with DRM/Radeon 7000
On Mon, Feb 10, 2003 at 06:31:37PM +, John P. Looney wrote: Hi, I was using the old Radeon drivers that shipped with a dist-upgraded Woody, though they were mostly fine in Quake3, in Wolfenstien, there were severe problems with corrupted textures. Could you try driver 20021022 (from dri.sf.net downloads)? As I wrote in other post - flickering textures are only visible in latest version of driver. -- Free Software - find interesting programs and change them NetHack - meet interesting creatures, kill them and eat their bodies Usenet - meet interesting people from all over the world and flame them Decopter - unrealistic helicopter simulator, get it from http://decopter.sf.net --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] problem with DRM/Radeon 7000
On Mon, 2003-02-10 at 19:31, John P. Looney wrote: Hi, I was using the old Radeon drivers that shipped with a dist-upgraded Woody, though they were mostly fine in Quake3, in Wolfenstien, there were severe problems with corrupted textures. After noting a few people were recommended to upgrade to get rid of these problems, I did. To upgrade, I used drm-trunk-module-2.4.18 from debian woody. Huh? Where exactly did you get that from? The code seems to be dated Jan 18 2003. I got the .debs via apt from http://people.debian.org/~daenzer/dri-trunk/ When I try Wolfenstien now, it doesn't even start: Started tty console (use +set ttycon 0 to disable) ^5PunkBuster Client: PunkBuster Client (v1.017 | A0) Enabled ^3PunkBuster Server: PunkBuster Server (v0.993 | A0 C0.0) **DISABLED** Resolving wolfmotd.idsoftware.com wolfmotd.idsoftware.com resolved to 192.246.40.65:27951 Received signal 8, exiting... This bug should be fixed in the packages for sid at deb http://people.debian.org/~daenzer/dri-trunk-sid/./ I'll hopefully get around to updating the packages for woody/sarge soon (once those for sid have settled a bit), in the meantime I think setting MESA_NO_SSE=1 should serve as a workaround. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast --- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel