You are not logged in.
Hallo... the answer is quite easy:
Ati drivers need all the agp-related stuff to be built as a module to work and to compile agpgart as a module is needed to compile FB_I810 and FB_I810_GTF as modules otherwise you are not allowed to select agpgart as module. That's all...
Anyway, new ati drivers are out!!!
I'm gonna try 'em as soon as possible. C yaaaaa
Release notes : http://www2.ati.com/drivers/linux/linux_8.16.20.html
Download at : https://support.ati.com/ics/support...ge&folderID=300
Offline
They've at least fixed a memory leak that has been pesting me and yet again they're promising that they've improved the performance. The interesting part is that how long it'll take for those to end up in [community] this time..? Place your bets, now! :twisted:
Offline
stfu - you are lucky they are even in there seeing as I don't even use them
The new ati-drivers will be built against the new kernel26archck kernel in [community] - this is a replacement pkg for kernel26cko as the -cko patchset was discontinued
new drivers are in [community]
Offline
I've installed the ati-drivers from community along with the kernel. The only difference I can see is glxgears shows my frame rate is about 10x faster, am I missing something. :?:
Norm
Offline
U're lucky... for me the only difference is that kde freeze whenever i try to start an application...
Offline
I was looking on ATI's site for information on how to set up a PCI card with fglrx, and found I was supposed to add Option "BusType" "PCI" to xorg.conf
I installed the kernel26archck package, along with the latest ATI drivers, added the above option to xorg.conf, and now my /var/log/Xorg.0.log shows DRI to be enabled, which before it wasn't because it was looking for AGP.
However, glxinfo still shows dri disabled, and fglrxinfo shows the vendor string as mesa. Is there a permissions problem perhaps with DRI, since the driver output shows it to be ok?
Any ideas on what I should check?
Thanks,
Chris
Offline
Can someone let me know how to uninstall the old drivers before install the new ones? I used the giant sh driver installer.
Offline
i figured out how to uninstall the driver; but, the new one is using a PCI transfer mode instead of AGP. The old one detected AGP flawlessly. I do have agpgart and sis_agp (my motherboard uses a sis agp chipset) loaded as modules as well as fglrx. Anyone have the same problem with 8.16.20 driver?
Offline
My problem with this new driver is that I can't seem to set the resolution to 1680x1050. I've never had a problem with this before. Now, it just starts up in whatever secondary resolution is in xorg.conf or in 1400x1050.
anyone else experiencing similar problems? This is driving me nuts.
Offline
I updated to the last version of the kernel from community and now the sound card is not found when I reboot:?:
When all else fails read the wiki!!
Norm
Offline
I updated to the last version of the kernel from community and now the sound card is not found when I reboot:?:
I have similar problem with the kernel in current repo. In /etc/rc.d/alsa, put
sleep 5
before the
/usr/sbin/alsactl restore
line.
Offline
Snowman I tried your suggestion, didn't help me
Norm
Offline
Using suggestions of wiki http://wiki.archlinux.org/index.php/ATI … Kernel_2.6
I have no success using both the mega ati install way, or community package way.
in the first case the result is a kernel error like this (when i try a "glxinfo" command or any access to opengl libraries).
[fglrx] module loaded - fglrx 8.21.7 [Jan 14 2006] on minor 0
[fglrx] Maximum main memory to use for locked dma buffers: 429 MBytes.
[fglrx] Internal AGP support requested, but kernel AGP support active.
[fglrx] Have to use kernel AGP support to avoid conflicts.
[fglrx] AGP detected, AgpState = 0x1f000217 (hardware caps of chipset)
agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
agpgart: Putting AGP V2 device at 0000:00:00.0 into 4x mode
agpgart: Putting AGP V2 device at 0000:01:00.0 into 4x mode
[fglrx] AGP enabled, AgpCommand = 0x1f000314 (selected caps)
Bad page state at free_hot_cold_page (in process 'X', page c11c0b0c)
flags:0x80000414 mapping:00000000 mapcount:0 count:0
Backtrace:
[<c0150a47>] bad_page+0x87/0xc0
[<c01513dc>] free_hot_cold_page+0x4c/0x140
[<c015d2c8>] zap_pte_range+0x238/0x2f0
[<c015d476>] unmap_page_range+0xf6/0x130
[<c015d580>] unmap_vmas+0xd0/0x250
[<c0162092>] unmap_region+0xb2/0x160
[<c0162439>] do_munmap+0x109/0x150
[<c01624d1>] sys_munmap+0x51/0x80
[<c0103487>] sysenter_past_esp+0x54/0x75
Trying to fix it up, but a reboot is needed
[fglrx] free AGP = 256126976
[fglrx] max AGP = 256126976
[fglrx] free LFB = 114274304
[fglrx] max LFB = 114274304
[fglrx] free Inv = 0
[fglrx] max Inv = 0
[fglrx] total Inv = 0
[fglrx] total TIM = 0
[fglrx] total FB = 0
[fglrx] total AGP = 65536
------------[ cut here ]------------
kernel BUG at mm/swap.c:49!
invalid operand: 0000 [#1]
PREEMPT SMP
Modules linked in: fglrx sbp2 ohci1394 eth1394 ieee1394 uhci_hcd ehci_hcd usbhid usblp usb_storage yenta_socket rsrc_nonstatic pcmcia_core tg3 ppp_generic slhc ide_cd snd_intel8x0m snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_pcm snd_timer snd_page_alloc snd_mixer_oss snd soundcore shpchp pci_hotplug serio_raw evdev i2c_i801 i2c_core rtc intel_agp agpgart cdrom sn9c102 videodev ipw2200 ieee80211 ieee80211_crypt cpufreq_ondemand acpi_cpufreq freq_table usbcore xfs reiserfs jfs ext3 jbd ide_floppy ide_disk piix generic ide_core
CPU: 0
EIP: 0060:[<c0157ea3>] Tainted: P B VLI
EFLAGS: 00010256 (2.6.15-ARCH)
EIP is at put_page+0x63/0x80
eax: 00000000 ebx: d02c8604 ecx: c11c0b0c edx: c11c0b0c
esi: b7d81000 edi: c11c0b0c ebp: 00000020 esp: cda8de94
ds: 007b es: 007b ss: 0068
Process glxinfo (pid: 4437, threadinfo=cda8c000 task=cfd17070)
Stack: c015d2c8 c11c0b0c b7d81000 0c76b067 dd406ac0 ffffffff 00000000 c124642c
cf446b7c b7d82000 cda8df28 cf446b7c c015d476 c1482900 dd53122c cf446b7c
b7d81000 b7d82000 cda8df28 00000000 cf446b7c b7d81fff 00000001 b7d81000
Call Trace:
[<c015d2c8>] zap_pte_range+0x238/0x2f0
[<c015d476>] unmap_page_range+0xf6/0x130
[<c015d580>] unmap_vmas+0xd0/0x250
[<c0162092>] unmap_region+0xb2/0x160
[<c0162439>] do_munmap+0x109/0x150
[<c01624d1>] sys_munmap+0x51/0x80
[<c0103487>] sysenter_past_esp+0x54/0x75
Code: 00 8b 52 0c 8b 02 89 d1 f6 c4 40 74 03 8b 4a 0c 8b 41 04 40 74 1f f0 83 42 04 ff 0f 98 c0 84 c0 74 cc 8b 4a 38 89 54 24 04 ff e1 <0f> 0b 31 00 64 b1 3b c0 eb ad 0f 0b 29 00 64 b1 3b c0 eb d7 89
<6>note: glxinfo[4437] exited with preempt_count 2
scheduling while atomic: glxinfo/0x00000002/4437
[<c039e63d>] schedule+0x9cd/0xd60
[<c012108e>] __wake_up+0x3e/0x60
[<c0126dff>] release_console_sem+0xaf/0xc0
[<c0126cc5>] vprintk+0x305/0x310
[<c039f6e5>] rwsem_down_read_failed+0xa5/0x1b0
[<c012ab20>] .text.lock.exit+0x27/0x87
[<c0129708>] do_exit+0xf8/0x420
[<c010480b>] die+0x18b/0x190
[<c0104ba0>] do_invalid_op+0x0/0xd0
[<c0104c58>] do_invalid_op+0xb8/0xd0
[<c0157ea3>] put_page+0x63/0x80
[<c011772e>] flush_tlb_all+0x3e/0x60
[<c0165536>] __remove_vm_area+0x36/0x60
[<c01655df>] __vunmap+0x4f/0x130
[<c01656e8>] vfree+0x28/0x40
[<c0104007>] error_code+0x4f/0x54
[<e14c007b>] firegl_proc_init+0x5b/0x200 [fglrx]
[<e14c007b>] firegl_proc_init+0x5b/0x200 [fglrx]
[<c0157ea3>] put_page+0x63/0x80
[<c015d2c8>] zap_pte_range+0x238/0x2f0
[<c015d476>] unmap_page_range+0xf6/0x130
[<c015d580>] unmap_vmas+0xd0/0x250
[<c0162092>] unmap_region+0xb2/0x160
[<c0162439>] do_munmap+0x109/0x150
[<c01624d1>] sys_munmap+0x51/0x80
[<c0103487>] sysenter_past_esp+0x54/0x75
and notebook will hang after trying to escape from X session.
in the second way, after installing kernel package and ati drivers package from repo community (and modifying xorg.conf and loading fglrx module) the result is... nothing. no "direct rendering" and mesa opengl version (but i have not installed mesa package...)
someone have seen something of same type?
Offline
It has been solved using big ati self extracting script.
It is an issue documented here http://www.thinkwiki.org/wiki/Problems_with_fglrx
a trouble using kernel 2.6.15 and latest version of ati driver.
there is a patch here: http://marc.theaimsgroup.com/?l=linux-k … 515001&w=2
Offline
For those still having troubles with ati-drivers and kernel 2.6.15:
I've built a package [1] called ati-drivers-arch that includes the patch to fix the problems with the newest kernel.
The package is based on 2.6.15-ARCH, if you want to use it with the archck-kernel just change the _kernel parameter in the PKGBUILD.
Greez Sigi
Haven't been here in a while. Still rocking Arch.
Offline
For those still having troubles with ati-drivers and kernel 2.6.15:
I've built a package [1] called ati-drivers-arch that includes the patch to fix the problems with the newest kernel.
The package is based on 2.6.15-ARCH, if you want to use it with the archck-kernel just change the _kernel parameter in the PKGBUILD.
Greez Sigi
Did anyone tried to use drivers with success on Xorg 7.0.
I did but I had a problem with PKGBUILD. I did change path for new Xorg 7.0 and depends=('xorg>=6.8.0' 'kernel26') I changed to depends=('xorg>=7.0' 'kernel26') and then I got a message: request Xorg 7.0. How should I change, please?
Mitja
Offline
I haven't tried it out myself, but I read somewhere about serious problems with xorg 7.0. I think you will have to copy many files by hand into the right paths at the moment. But like I already mentioned, I haven't tried it yet (and I really got better thinks to do with my time than that ;-) )
Anyway, feel free to post you PKGBUILD if you success
Greez Sigi
Haven't been here in a while. Still rocking Arch.
Offline
updated the ati-drivers package in the community repo with the same 2.6.15 patch as distributed by Sigi, thanks man
iphitus
Offline
I'm having a problem that I've never had with ATI drivers before. It happens with both ati-drivers from community and the new 8.21 drivers installed using their installer on 2.6.15-ARCH.
Everything seems to work. GLX works. But when I run `glxinfo`, the terminal never returns to the command line after the output. After this, all the other tty's get screwed up. Running, for example, `ps x` shows the output, but never comes back to a shell prompt. ^C does nothing. Soon after, X freezes and I end up rebooting.
The following shows up in /var/log/messages.log:
Jan 31 13:49:30 casper <1>Fixing recursive fault but reboot is needed!
Anyone else experiencing similar problems?
Offline
Mazaev - sounds like the problem that was fixed by the patch... does glxgears freeze up too?
If so, wait a bit for your mirror to sync then update to iph's new ATI package.
Offline
Mazaev - sounds like the problem that was fixed by the patch... does glxgears freeze up too?
If so, wait a bit for your mirror to sync then update to iph's new ATI package.
Finally updated to the new ati-drivers along with the new archck and everything finally seems to work. Not only that, but my glxgears fps went from ~3800 to ~5000.
Thanks a bunch to those who've been working on this.
Offline
iphitus: you're welcome
Haven't been here in a while. Still rocking Arch.
Offline
Might be useful to people watching this thread:
Offline
is this BAD, GOOD or GREAT ?
That's about what I get for fgl_glxgears. Seems you've got everything set up correctly.
Offline