GPU hang ups (Radeon X1400)

Question

Hey folks,
I'm experiencing problems with the internal graphics adapter of my laptop (ATI Radeon Mobility X1400) for some days now. I think it has to do with the system (that is, XFCE's power manager) trying to put the GPU to sleep when the computer isn't in use. The desktop kind of freezes, but I can still move the mouse cursor. Trying to switch to a VT results in black screen with grey stripes. Well, at least I'm able to a soft reboot via ctrl + alt + del.
This happened twice to me in the last days, the first time with a much more verbose message. After the first time, I was able to restart X, but afterwards, video acceleration was broken (playing a video just gave a black square for example).
May 1 21:49:41 localhost kernel: [drm] radeon: 1 quad pipes, 1 z pipes initialized.
May 1 21:49:41 localhost kernel: [drm] PCIE GART of 512M enabled (table at 0x0000000000040000).
May 1 21:49:41 localhost kernel: radeon 0000:01:00.0: WB enabled
May 1 21:49:41 localhost kernel: [drm] fence driver on ring 0 use gpu addr 0x08000000 and cpu addr 0xff81a000
May 1 21:49:41 localhost kernel: [drm] radeon: ring at 0x0000000008001000
May 1 21:49:41 localhost kernel: [drm:r100_ring_test] *ERROR* radeon: ring test failed (scratch(0x15E4)=0xCAFEDEAD)
May 1 21:49:41 localhost kernel: [drm:r100_cp_init] *ERROR* radeon: cp isn't working (-22).
May 1 21:49:41 localhost kernel: radeon 0000:01:00.0: failed initializing CP (-22).
May 1 21:49:42 localhost kernel: [drm:radeon_ib_schedule] *ERROR* radeon: couldn't schedule IB(4).
May 1 21:49:42 localhost kernel: [drm:radeon_cs_ib_chunk] *ERROR* Failed to schedule IB !
May 1 21:49:42 localhost kernel: ------------[ cut here ]------------
May 1 21:49:42 localhost kernel: WARNING: at drivers/gpu/drm/radeon/radeon_fence.c:204 radeon_fence_signaled+0xc5/0xf0 [radeon]()
May 1 21:49:42 localhost kernel: Hardware name: Satellite A100
May 1 21:49:42 localhost kernel: Querying an unemitted fence : d501be00 !
May 1 21:49:42 localhost kernel: Modules linked in: nls_utf8 nls_cp437 vfat fat usb_storage uas snd_seq_midi snd_seq_midi_even4> [<c1321f35>] ? __slab_free+0xda/0x236
May 1 21:49:42 localhost kernel: [<c10d0929>] ? tlb_finish_mmu+0x9/0x30
May 1 21:49:42 localhost kernel: [<f8384880>] ? drm_copy_field+0x70/0x70 [drm]
May 1 21:49:42 localhost kernel: [<c110690a>] ? do_vfs_ioctl+0x7a/0x560
May 1 21:49:42 localhost kernel: [<c10ed33a>] ? kmem_cache_free+0xda/0xf0
May 1 21:49:42 localhost kernel: [<c10d5d63>] ? remove_vma+0x43/0x60
May 1 21:49:42 localhost kernel: [<c10d5d63>] ? remove_vma+0x43/0x60
May 1 21:49:42 localhost kernel: [<c10d5d63>] ? remove_vma+0x43/0x60
May 1 21:49:42 localhost kernel: [<c10d6dca>] ? do_munmap+0x15a/0x2d0
May 1 21:49:42 localhost kernel: [<c1106e5a>] ? sys_ioctl+0x6a/0x80
May 1 21:49:42 localhost kernel: [<c132715f>] ? sysenter_do_call+0x12/0x28
May 1 21:49:42 localhost kernel: ---[ end trace 7a6af2bef415be9e ]---
May 1 21:57:19 localhost kernel: ------------[ cut here ]------------
May 1 21:57:19 localhost kernel: WARNING: at drivers/gpu/drm/radeon/radeon_fence.c:204 radeon_fence_signaled+0xc5/0xf0 [radeon]()
May 1 21:57:19 localhost kernel: Hardware name: Satellite A100
May 1 21:57:19 localhost kernel: Querying an unemitted fence : d501be00 !
May 1 21:57:19 localhost kernel: Modules linked in: nls_utf8 nls_cp437 vfat fat usb_storage uas snd_seq_midi snd_seq_midi_event snd_seq_dummy snd_seq binfmt_misc fuse aes_i586 cryptd aes_generic af_packet ext2 snd_usb_audio snd_usbmidi_lib snd_hda_codec_si3054 snd_rawmidi snd_seq_device arc4 snd_hda_codec_realtek joydev iwl3945 iwlegacy snd_hda_intel snd_hda_codec snd_hwdep mac80211 snd_pcm snd_page_alloc sdhci_pci sdhci e100 cfg80211 snd_timer snd mmc_core tifm_7xx1 tifm_core soundcore rtc_cmos button mii toshiba_acpi psmouse evdev battery intel_agp video thermal fan ac sparse_keymap i2c_i801 intel_gtt rng_core nfs nfs_acl lockd auth_rpcgss sunrpc fscache ipv6 autofs4 ext4 crc16 jbd2 mbcache usbhid hid sd_mod pata_acpi ata_generic ata_piix libata scsi_mod uhci_hcd ehci_hcd radeon i2c_algo_bit drm_kms_helper ttm drm agpgart i2c_core
May 1 21:57:19 localhost kernel: Pid: 17718, comm: X Tainted: G W 3.3.3-0-lqx #1
May 1 21:57:19 localhost kernel: Call Trace:
May 1 21:57:19 localhost kernel: [<c102eb38>] ? warn_slowpath_common+0x78/0xb0
May 1 21:57:19 localhost kernel: [<f8513795>] ? radeon_fence_signaled+0xc5/0xf0 [radeon]
May 1 21:57:19 localhost kernel: [<f8513795>] ? radeon_fence_signaled+0xc5/0xf0 [radeon]
May 1 21:57:19 localhost kernel: [<c102ec03>] ? warn_slowpath_fmt+0x33/0x40
May 1 21:57:19 localhost kernel: [<f8513795>] ? radeon_fence_signaled+0xc5/0xf0 [radeon]
May 1 21:57:19 localhost kernel: [<f83d3e67>] ? ttm_bo_wait+0x117/0x1a0 [ttm]
May 1 21:57:19 localhost kernel: [<f838e75f>] ? drm_mm_put_block+0x5f/0x70 [drm]
May 1 21:57:19 localhost kernel: [<f838e75f>] ? drm_mm_put_block+0x5f/0x70 [drm]
May 1 21:57:19 localhost kernel: [<f83d44d3>] ? ttm_bo_release+0xa3/0x1e0 [ttm]
May 1 21:57:19 localhost kernel: [<f83d4643>] ? ttm_bo_unref+0x33/0x50 [ttm]
May 1 21:57:19 localhost kernel: [<f8515a43>] ? radeon_bo_unref+0x33/0x70 [radeon]
May 1 21:57:19 localhost kernel: [<f8525a56>] ? radeon_gem_object_free+0x16/0x20 [radeon]
May 1 21:57:19 localhost kernel: [<f838600d>] ? drm_gem_object_free+0x1d/0x30 [drm]
May 1 21:57:19 localhost kernel: [<f83863ff>] ? drm_gem_handle_delete+0xbf/0xf0 [drm]
May 1 21:57:19 localhost kernel: [<f8386880>] ? drm_gem_destroy+0x40/0x40 [drm]
May 1 21:57:19 localhost kernel: [<f8384a22>] ? drm_ioctl+0x1a2/0x3d0 [drm]
May 1 21:57:19 localhost kernel: [<f8386880>] ? drm_gem_destroy+0x40/0x40 [drm]
May 1 21:57:19 localhost kernel: [<c1058f0c>] ? update_cpu_clock+0x4dc/0x550
May 1 21:57:19 localhost kernel: [<c1059a75>] ? scheduler_tick+0x2d5/0x3f0
May 1 21:57:19 localhost kernel: [<f8384880>] ? drm_copy_field+0x70/0x70 [drm]
May 1 21:57:19 localhost kernel: [<c110690a>] ? do_vfs_ioctl+0x7a/0x560
May 1 21:57:19 localhost kernel: [<c104f706>] ? enqueue_hrtimer+0x16/0x80
May 1 21:57:19 localhost kernel: [<c101d063>] ? lapic_next_event+0x13/0x20
May 1 21:57:19 localhost kernel: [<c106adbf>] ? clockevents_program_event+0x9f/0x150
May 1 21:57:19 localhost kernel: [<c106bf01>] ? tick_program_event+0x21/0x30
May 1 21:57:19 localhost kernel: [<c1050614>] ? hrtimer_interrupt+0x164/0x260
May 1 21:57:19 localhost kernel: [<c10f804a>] ? fput+0x15a/0x1d0
May 1 21:57:19 localhost kernel: [<c1106e5a>] ? sys_ioctl+0x6a/0x80
May 1 21:57:19 localhost kernel: [<c132715f>] ? sysenter_do_call+0x12/0x28
May 1 21:57:19 localhost kernel: ---[ end trace 7a6af2bef415be9f ]---
May 6 18:59:26 localhost kernel: [drm:r100_ring_test] *ERROR* radeon: ring test failed (scratch(0x15E4)=0xCAFEDEAD)
May 6 18:59:26 localhost kernel: [drm:r100_cp_init] *ERROR* radeon: cp isn't working (-22).
May 6 18:59:26 localhost kernel: radeon 0000:01:00.0: failed initializing CP (-22).
I'm using a custom 3.3.4 liquorix kernel (3.3.3 at the time of the first error) and ati-dri 8.0.2 drivers. Does anyone have an idea what the problem might be?

Answer 1

Quote from: SCRAPS on 13-October-08, 03:41:30
The usb flashing tool is alot better than using a floppy or any other way of booting the bios? In other words its less risky? I have never booted offa pendrive or uspdrive befor so it will be a new experiance for me. I dont have to install teh whole system to flash bios right? I can do it off the cold boot right outside the case right? I'll have to make the usbdrive bootable too right? Is all that information found in the flash tool in the forum or where should i look? Im currently out of town and me and that older processor should be back at the computer on tuesday. So im hopefull it will work.
BIOS flashing always has its risks, regardless of the method used.  As for mitigating the risks, some users will argue floppy is inherently risky (floppy media can have latent bad sectors that can corrupt the bios image file).  As for booting to the hdd, it's not a bad option if you don't mind the inconvenience of building a bootable dos partition.  Flashing a BIOS from windows (or any non threadbare OS) is akin to begging for trouble.  Hence, flashing from USB drive has gained much popularity.  (On a side note, please never attempt to auto-update your BIOS using the windows MSI-Update utilities.)
As for the process itself, you should have no problem booting from USB drive with the P7N Diamond.  You'll find the necessary options in your CMOS menu once you've replaced your E8400 with the Conroe D.  Please someone else correct me if I'm wrong, but I believe there is also a Boot Menu that is accessible by pressing F11 during POST.  If a USB storage device is detected it should be listed under the boot device choices (displayed as CMB Drive or something similar).
Hope this helps...