Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
xorg-server-1.13.1 crash on X-configure
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Gentoo on PPC
View previous topic :: View next topic  
Author Message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Thu Jan 31, 2013 7:40 pm    Post subject: xorg-server-1.13.1 crash on X-configure Reply with quote

These are the components installed
Quote:
sys-kernel/gentoo-sources-3.2.1-r2
x11-base/xorg-server-1.13.1 USE="nptl suid udev xorg"
x11-drivers/xf86-video-ati-7.0.0-r1 USE="udev (-glamor)"


When I run X -configure
Quote:
[ 872.574] (II) [KMS] drm report modesetting isn't supported.
[ 872.574] (EE)
[ 872.575] (EE) Backtrace:
[ 872.575] (EE) 0: X (xorg_backtrace+0x50) [0x1015e9f0]
[ 872.575] (EE) 1: X (0x10000000+0x162638) [0x10162638]
[ 872.575] (EE) 2: linux-vdso32.so.1 (__kernel_sigtramp_rt32+0x0) [0x100350]
[ 872.576] (EE) 3: X (DoConfigure+0xa40) [0x1005f17c]
[ 872.576] (EE) 4: X (InitOutput+0x1e8) [0x1006c6c4]
[ 872.576] (EE) 5: X (0x10000000+0x1e888) [0x1001e888]
[ 872.577] (EE) 6: /lib/libc.so.6 (0xfaee000+0x23410) [0xfb11410]
[ 872.577] (EE) 7: /lib/libc.so.6 (0xfaee000+0x235a0) [0xfb115a0]
[ 872.577] (EE)
[ 872.577] (EE) Segmentation fault at address 0x0
[ 872.580]
Fatal server error:
[ 872.581] Caught signal 11 (Segmentation fault). Server aborting


Code:
grep FB_ .config
Quote:

CONFIG_FB_DDC=y
CONFIG_FB_CFB_FILLRECT=y
CONFIG_FB_CFB_COPYAREA=y
CONFIG_FB_CFB_IMAGEBLIT=y
CONFIG_FB_MACMODES=y
CONFIG_FB_BACKLIGHT=y
CONFIG_FB_MODE_HELPERS=y
CONFIG_FB_RADEON=y
CONFIG_FB_RADEON_I2C=y
CONFIG_FB_RADEON_BACKLIGHT=y


Does anyone know what I missed? Seems to me this should work, although since I am running on PPC, it may not be as common as most others running the same components.
Back to top
View user's profile Send private message
Gusar
Advocate
Advocate


Joined: 09 Apr 2005
Posts: 2643
Location: Slovenia

PostPosted: Thu Jan 31, 2013 7:55 pm    Post subject: Reply with quote

X -configure is completely outdated. Though in this case it does at least show an error. Don't use FB_RADEON, use KMS. There's several threads about it at the forums, so you should find what you need. Pay extra attention to setting up the firmware properly.
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Thu Jan 31, 2013 8:06 pm    Post subject: Reply with quote

Gusar,

Thank you very much for pointing me in a direction. At this point I have no idea. Maybe I will be able to figure it out by researching KMS as you suggested.

I believe I may not need an additional firmware to drive the card:

Code:
lspci -v
Quote:
0000:00:10.0 VGA compatible controller: Advanced Micro Devices [AMD] nee ATI RV280 [Radeon 9200] (rev 01) (prog-if 00 [VGA controller])
Subsystem: Advanced Micro Devices [AMD] nee ATI RV280 [Radeon 9200]
Flags: bus master, 66MHz, medium devsel, latency 255, IRQ 48
Memory at 98000000 (32-bit, prefetchable) [size=128M]
I/O ports at 0400 [size=256]
Memory at 90000000 (32-bit, non-prefetchable) [size=64K]
Expansion ROM at 90020000 [size=128K]
Capabilities: [58] AGP version 2.0
Capabilities: [50] Power Management version 2
Kernel driver in use: radeonfb


This is an old mac-mini and the Radeon card in it is not one of the newer ones that requires an additional firmware. I am basing this on the table from this page:

https://wiki.gentoo.org/wiki/Radeon#Feature_support

I believe the hardware I have corresponds to the R200 Family and is supported without an additional firmware.
I may be misinterpreting, is this so?
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Thu Jan 31, 2013 9:07 pm    Post subject: Hmmm... KMS seems to be selected but doesn't work Reply with quote

Code:
grep KMS .config
Quote:
CONFIG_DRM_KMS_HELPER=m
CONFIG_DRM_RADEON_KMS=y

The output of dmesg, I removed lines that had no relation to the problem
Quote:
/i2c@f8001000
channel 0 bus <multibus>
channel 1 bus <multibus>
KeyWest i2c @0x80018000 irq 26 /pci@f2000000/mac-io@17/i2c@18000
channel 0 bus <multibus>
PMU i2c /pci@f2000000/mac-io@17/via-pmu@16000/pmu-i2c
channel 1 bus <multibus>
channel 2 bus <multibus>
PCI: Probing PCI hardware
pci 0000:00:0b.0: [106b:0034] type 0 class 0x000600
pci 0000:00:10.0: [1002:5962] type 0 class 0x000300
pci 0000:00:10.0: reg 10: [mem 0x98000000-0x9fffffff pref]
pci 0000:00:10.0: reg 14: [io 0x0400-0x04ff]
pci 0000:00:10.0: reg 18: [mem 0x90000000-0x9000ffff]
pci 0000:00:10.0: reg 30: [mem 0x90020000-0x9003ffff pref]
pci 0000:00:10.0: supports D1 D2
pci 0001:10:0b.0: [106b:0035] type 0 class 0x000600
pci 0001:10:12.0: [14e4:4320] type 0 class 0x000280
pci 0001:10:12.0: reg 10: [mem 0x80084000-0x80085fff]
pci 0001:10:12.0: supports D1 D2
pci 0001:10:12.0: PME# supported from D0 D1 D2 D3hot D3cold
pci 0001:10:12.0: PME# disabled
pci 0001:10:17.0: [106b:003e] type 0 class 0x00ff00
pci 0001:10:17.0: reg 10: [mem 0x80000000-0x8007ffff]
pci 0001:10:18.0: [106b:003f] type 0 class 0x000c03
pci 0001:10:18.0: reg 10: [mem 0x00000000-0x00000fff]
pci 0001:10:19.0: [106b:003f] type 0 class 0x000c03
pci 0001:10:19.0: reg 10: [mem 0x00000000-0x00000fff]
pci 0001:10:1a.0: [106b:003f] type 0 class 0x000c03
pci 0001:10:1a.0: reg 10: [mem 0x80083000-0x80083fff]
pci 0001:10:1b.0: [1033:0035] type 0 class 0x000c03
pci 0001:10:1b.0: reg 10: [mem 0x80082000-0x80082fff]
pci 0001:10:1b.0: supports D1 D2
pci 0001:10:1b.0: PME# supported from D0 D1 D2 D3hot D3cold
pci 0001:10:1b.0: PME# disabled
pci 0001:10:1b.1: [1033:0035] type 0 class 0x000c03
pci 0001:10:1b.1: reg 10: [mem 0x80081000-0x80081fff]
pci 0001:10:1b.1: supports D1 D2
pci 0001:10:1b.1: PME# supported from D0 D1 D2 D3hot D3cold
pci 0001:10:1b.1: PME# disabled
pci 0001:10:1b.2: [1033:00e0] type 0 class 0x000c03
pci 0001:10:1b.2: reg 10: [mem 0x80080000-0x800800ff]
pci 0001:10:1b.2: supports D1 D2
pci 0001:10:1b.2: PME# supported from D0 D1 D2 D3hot D3cold
pci 0001:10:1b.2: PME# disabled
pci 0002:20:0b.0: [106b:0036] type 0 class 0x000600
pci 0002:20:0d.0: [106b:003b] type 0 class 0x00ff00
pci 0002:20:0d.0: reg 10: [mem 0xf5004000-0xf5007fff]
pci 0002:20:0e.0: [106b:0031] type 0 class 0x000c00
pci 0002:20:0e.0: reg 10: [mem 0xf5000000-0xf5000fff]
pci 0002:20:0e.0: supports D1 D2
pci 0002:20:0e.0: PME# supported from D0 D1 D2 D3hot
pci 0002:20:0e.0: PME# disabled
pci 0002:20:0f.0: [106b:0032] type 0 class 0x000200
pci 0002:20:0f.0: reg 10: [mem 0xf5200000-0xf53fffff]
pci 0002:20:0f.0: reg 30: [mem 0xf5100000-0xf51fffff pref]
PCI 0000:00 Cannot reserve Legacy IO [io 0x802000-0x802fff]
PCI: max bus depth: 0 pci_try_num: 1
pci_bus 0000:00: resource 0 [io 0x802000-0x1001fff]
pci_bus 0000:00: resource 1 [mem 0xf1000000-0xf1ffffff]
pci_bus 0000:00: resource 2 [mem 0x90000000-0x9fffffff]
pci_bus 0001:10: resource 0 [io 0x0000-0x7fffff]
pci_bus 0001:10: resource 1 [mem 0xf3000000-0xf3ffffff]
pci_bus 0001:10: resource 2 [mem 0x80000000-0x8fffffff]
pci_bus 0002:20: resource 0 [io 0xff7fe000-0xffffdfff]
pci_bus 0002:20: resource 1 [mem 0xf5000000-0xf5ffffff]
bio: create slab <bio-0> at 0
vgaarb: device added: PCI:0000:00:10.0,decodes=io+mem,owns=mem,locks=none
vgaarb: loaded
vgaarb: bridge control possible 0000:00:10.0
ssb: Core 0 found: ChipCommon (cc 0x800, rev 0x04, vendor 0x4243)
ssb: Core 1 found: IEEE 802.11 (cc 0x812, rev 0x05, vendor 0x4243)
ssb: Core 2 found: PCMCIA (cc 0x80D, rev 0x02, vendor 0x4243)
ssb: Core 3 found: V90 (cc 0x807, rev 0x02, vendor 0x4243)
ssb: Core 4 found: PCI (cc 0x804, rev 0x09, vendor 0x4243)
ssb: Sonics Silicon Backplane found on PCI device 0001:10:12.0
PCI: CLS mismatch (32 != 1020), using 32 bytes
Linux agpgart interface v0.103
radeonfb 0000:00:10.0: enabling device (0006 -> 0007)
radeonfb 0000:00:10.0: Invalid ROM contents
radeonfb (0000:00:10.0): Invalid ROM signature 0 should be 0xaa55
radeonfb: Retrieved PLL infos from Open Firmware
radeonfb: Reference=27.00 MHz (RefDiv=12) Memory=190.00 Mhz, System=250.00 MHz
radeonfb: PLL min 12000 max 35000
i2c i2c-2: unable to read EDID block.
i2c i2c-2: unable to read EDID block.
i2c i2c-2: unable to read EDID block.
radeonfb: Monitor 1 type CRT found
radeonfb: EDID probed
radeonfb: Monitor 2 type CRT found
radeonfb: EDID probed
Console: switching to colour frame buffer device 240x67
radeonfb (0000:00:10.0): ATI Radeon 5962 "Yb"
[drm] Initialized drm 1.1.0 20060810
i2c /dev entries driver
PowerMac i2c bus pmu 2 registered
PowerMac i2c bus pmu 1 registered
PowerMac i2c bus mac-io 0 registered
PowerMac i2c bus uni-n 1 registered
PowerMac i2c bus uni-n 0 registered
i2c i2c-0: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-0: Please use another way to instantiate your i2c_client
i2c i2c-1: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-1: Please use another way to instantiate your i2c_client
i2c i2c-2: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-2: Please use another way to instantiate your i2c_client
i2c i2c-3: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-3: Please use another way to instantiate your i2c_client
i2c i2c-4: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-4: Please use another way to instantiate your i2c_client
i2c i2c-5: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-5: Please use another way to instantiate your i2c_client
i2c i2c-6: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-6: Please use another way to instantiate your i2c_client
i2c i2c-7: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-7: Please use another way to instantiate your i2c_client
i2c i2c-8: aoa_codec_onyx: attach_adapter method is deprecated
i2c i2c-8: Please use another way to instantiate your i2c_client
i2c i2c-0: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-0: Please use another way to instantiate your i2c_client
i2c i2c-1: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-1: Please use another way to instantiate your i2c_client
i2c i2c-2: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-2: Please use another way to instantiate your i2c_client
i2c i2c-3: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-3: Please use another way to instantiate your i2c_client
i2c i2c-4: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-4: Please use another way to instantiate your i2c_client
i2c i2c-5: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-5: Please use another way to instantiate your i2c_client
i2c i2c-6: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-6: Please use another way to instantiate your i2c_client
i2c i2c-7: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-7: Please use another way to instantiate your i2c_client
i2c i2c-8: aoa_codec_tas: attach_adapter method is deprecated
i2c i2c-8: Please use another way to instantiate your i2c_client
devtmpfs: mounted


The contents of /etc/yaboot.conf
Quote:
boot=/dev/hda2
device=/pci@f4000000/ata-6@d/disk@0:
partition=3
root=/dev/hda3
timeout=30
install=/usr/lib/yaboot/yaboot
magicboot=/usr/lib/yaboot/ofboot
append="radeon.modeset=1"

image=/boot/vmlinux
label=Linux
read-only


As you can see, everything is configured to use KMS, but it does not appear to be enabled.
Back to top
View user's profile Send private message
Gusar
Advocate
Advocate


Joined: 09 Apr 2005
Posts: 2643
Location: Slovenia

PostPosted: Thu Jan 31, 2013 11:42 pm    Post subject: Reply with quote

That dmesg output shows radeonfb being loaded. So you're still screwing up somewhere (a common problem is not having the /boot partition mounted when installing a new kernel). The goal is to get the KMS radeondrmfb driver.

Firmware is needed for all radeon cards, though for yours it might already be part of the kernel source, if that's really the case then all you need is CONFIG_FIRMWARE_IN_KERNEL=y
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Fri Feb 01, 2013 4:47 am    Post subject: Reply with quote

Gusar wrote:
That dmesg output shows radeonfb being loaded. So you're still screwing up somewhere (a common problem is not having the /boot partition mounted when installing a new kernel). The goal is to get the KMS radeondrmfb driver.


The kernel being loaded is the one I configured, the date-stamp on it bears this out.

Code:
ls -l /boot
Quote:
total 15736
-rwxr-xr-x 1 root root 8676789 Jan 31 11:17 vmlinux


Also, the way the hard-disk is partitioned, there is no boot partition. Macs have a mac-specific bootblock that can pretty much only be modified by editing the /etc/yaboot.conf file and then executing ybin -v to apply the changes and this is for specifying grub-like options. The usual mount-the-boot-partition-copy-the-new-kernel procedure we are used to on x86 and amd64 does not apply.

Code:
mac-fdisk -l
Quote:
/dev/hda
# type name length base ( size ) system
/dev/hda1 Apple_partition_map Apple 63 @ 1 ( 31.5k) Partition map
/dev/hda2 Apple_Bootstrap untitled 1954 @ 64 (977.0k) NewWorld bootblock
/dev/hda3 Apple_UNIX_SVR2 untitled 150251954 @ 2018 ( 71.6G) Linux native
/dev/hda4 Apple_UNIX_SVR2 swap 6047516 @ 150253972 ( 2.9G) Linux swap

Block size=512, Number of Blocks=156301488
DeviceType=0x0, DeviceId=0x0


This is also reflected in the fstab, which does not have a boot entry:

Code:
cat /etc/fstab
Quote:
# <fs> <mountpoint> <type> <opts> <dump/pass>
/dev/hda3 / ext3 noatime 0 1
/dev/hda4 none swap sw 0 0
/dev/cdrom /mnt/cdrom auto noauto,ro 0 0


Gusar wrote:
Firmware is needed for all radeon cards, though for yours it might already be part of the kernel source, if that's really the case then all you need is CONFIG_FIRMWARE_IN_KERNEL=y


Code:
grep CONFIG_FIRMWARE_IN_KERNEL .config
Quote:
CONFIG_FIRMWARE_IN_KERNEL=y


Everything seems to be in place but X still does not start.

Thank you very much for your help, Gusar. It is very perplexing. Unless we can think of some other avenue, I believe I may have to file a bug... Possibly with the upstream developers, as it may be a programming issue related to ppc32 and R200-family video cards.

If you have any idea at all, please suggest!
Thank you again for your help
Back to top
View user's profile Send private message
Gusar
Advocate
Advocate


Joined: 09 Apr 2005
Posts: 2643
Location: Slovenia

PostPosted: Fri Feb 01, 2013 9:26 am    Post subject: Reply with quote

If radeonfb loads, it means the kernel is not correctly configured. Post the full kernel config, but not directly here, use pastebin please.
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Fri Feb 01, 2013 12:51 pm    Post subject: Reply with quote

Gusar wrote:
If radeonfb loads, it means the kernel is not correctly configured. Post the full kernel config

http://pastebin.ca/2309607

Thanks!
Back to top
View user's profile Send private message
Gusar
Advocate
Advocate


Joined: 09 Apr 2005
Posts: 2643
Location: Slovenia

PostPosted: Fri Feb 01, 2013 1:20 pm    Post subject: Reply with quote

Yep, you still have FB_RADEON in there. Remove all framebuffer stuff except FRAMEBUFFER_CONSOLE
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Fri Feb 01, 2013 2:07 pm    Post subject: Reply with quote

OK, now the only things in the config that have FB_ set in them are these

Code:
grep FB_ .config
Quote:
CONFIG_FB_CFB_FILLRECT=m
CONFIG_FB_CFB_COPYAREA=m
CONFIG_FB_CFB_IMAGEBLIT=m
CONFIG_FB_BACKLIGHT=y
CONFIG_FB_MODE_HELPERS=y


I tried it and X still segfaults

http://pastebin.ca/2309620

I downgraded xf86-video-ati-7.0.0-r1 to xf86-video-ati-7.0.0
I had listed the driver in package.keywords in case there was a bug in it that might have been fixed in the unstable version but this is apparently not the cause.
All components are from the stable tree.
Back to top
View user's profile Send private message
Gusar
Advocate
Advocate


Joined: 09 Apr 2005
Posts: 2643
Location: Slovenia

PostPosted: Fri Feb 01, 2013 10:01 pm    Post subject: Reply with quote

Ouch, driver crash, not good. Does dmesg show any errors regarding radeon? If not, xf86-video-ati-7.1.0 was recently released, you could try that. And if it still segfaults, make a bug report upstream.
Back to top
View user's profile Send private message
chithanh
Developer
Developer


Joined: 05 Aug 2006
Posts: 1674
Location: Berlin, Germany

PostPosted: Fri Feb 01, 2013 10:37 pm    Post subject: Reply with quote

Before starting X, double check that radeonfb is indeed not "Kernel driver in use" according to lspci -k
/proc/fb should contain radeondrmfb, not radeonfb.
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Fri Feb 01, 2013 10:37 pm    Post subject: Reply with quote

Just to make sure I had not skipped over anything, I re-read the document on this page
https://wiki.gentoo.org/wiki/Radeon

Even though I originally thought that my card was supported without a blob, I checked the lib/firmware folder of the kernel.

It turns out I had misunderstood, the R200-family cards DO require an external firmware blob, but it is distributed with the kernel, so it is already on the filesystem and I don't have to emerge radeon-ucode, although the blob-file must still be added to the kernel config as explained in the document.

The reason I thought it was not necessary to do this is that all versions of xorg-server before 1.13 worked without requiring me to add an external blob.

After adding the radeon/R200_cp.bin file to the config, rebuilding the kernel and copying it to /boot, X STILL crashes, but now in the X log it says that kernel modesetting IS enabled.

So I am a bit closer to getting X working.

Here are both the X log
http://pastebin.ca/2309805

and the dmesg output
http://pastebin.ca/2309806

if you have any suggestion, please say.

I will try upgrading to xf86-video-ati-7.0.0-r1 in the meantime.
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Fri Feb 01, 2013 10:41 pm    Post subject: Reply with quote

chithanh wrote:
Before starting X, double check that radeonfb is indeed not "Kernel driver in use" according to lspci -k
/proc/fb should contain radeondrmfb, not radeonfb.

Code:
lspci -k
Quote:
0000:00:0b.0 Host bridge: Apple Inc. UniNorth 2 AGP
0000:00:10.0 VGA compatible controller: Advanced Micro Devices [AMD] nee ATI RV280 [Radeon 9200] (rev 01)
Subsystem: Advanced Micro Devices [AMD] nee ATI RV280 [Radeon 9200]
Kernel driver in use: radeon
0001:10:0b.0 Host bridge: Apple Inc. UniNorth 2 PCI
0001:10:12.0 Network controller: Broadcom Corporation BCM4306 802.11b/g Wireless LAN Controller (rev 03)
Subsystem: Apple Inc. AirPort Extreme
Kernel driver in use: b43-pci-bridge
0001:10:17.0 Unassigned class [ff00]: Apple Inc. KeyLargo/Intrepid Mac I/O
Kernel driver in use: macio
0001:10:18.0 USB controller: Apple Inc. KeyLargo/Intrepid USB
0001:10:19.0 USB controller: Apple Inc. KeyLargo/Intrepid USB
0001:10:1a.0 USB controller: Apple Inc. KeyLargo/Intrepid USB
Kernel driver in use: ohci_hcd
0001:10:1b.0 USB controller: NEC Corporation OHCI USB Controller (rev 43)
Subsystem: NEC Corporation USB Controller
Kernel driver in use: ohci_hcd
0001:10:1b.1 USB controller: NEC Corporation OHCI USB Controller (rev 43)
Subsystem: NEC Corporation USB Controller
Kernel driver in use: ohci_hcd
0001:10:1b.2 USB controller: NEC Corporation uPD72010x USB 2.0 Controller (rev 04)
Subsystem: NEC Corporation uPD72010x USB 2.0 Controller
Kernel driver in use: ehci_hcd
0002:20:0b.0 Host bridge: Apple Inc. UniNorth 2 Internal PCI
0002:20:0d.0 Unassigned class [ff00]: Apple Inc. UniNorth/Intrepid ATA/100
Kernel driver in use: ide-pmac
0002:20:0e.0 FireWire (IEEE 1394): Apple Inc. UniNorth 2 FireWire (rev 81)
Subsystem: Apple Inc. iBook G4 2004
0002:20:0f.0 Ethernet controller: Apple Inc. UniNorth 2 GMAC (Sun GEM) (rev 80)
Kernel driver in use: gem


Last edited by Progman3K on Fri Feb 01, 2013 10:45 pm; edited 1 time in total
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Fri Feb 01, 2013 10:43 pm    Post subject: Reply with quote

chithanh wrote:
Before starting X, double check that radeonfb is indeed not "Kernel driver in use" according to lspci -k
/proc/fb should contain radeondrmfb, not radeonfb.

Code:
cat /proc/fb
Quote:
0 radeondrmfb
Back to top
View user's profile Send private message
Gusar
Advocate
Advocate


Joined: 09 Apr 2005
Posts: 2643
Location: Slovenia

PostPosted: Fri Feb 01, 2013 10:59 pm    Post subject: Reply with quote

Why is /root/xorg.conf.new being used? Are you logged in as root when you start X? You don't need any config files anyway, so remove them. Shouldn't really matter regarding the crash, but still, makes no sense having outdated config files around (like I said, X -configure is outdated).

Quote:
It turns out I had misunderstood, the R200-family cards DO require an external firmware blob, but it is distributed with the kernel, so it is already on the filesystem and I don't have to emerge radeon-ucode, although the blob-file must still be added to the kernel config as explained in the document.

No, the blob file does not need to be added as explained there, CONFIG_FIRMWARE_IN_KERNEL=y already did everything necessary. Only blobs that are not already in the kernel source need to be added like that, and even that only if radeon is built-in (CONFIG_DRM_RADEON=y) rather than a module (CONFIG_DRM_RADEON=m).
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Sat Feb 02, 2013 12:07 am    Post subject: Reply with quote

Gusar wrote:
Why is /root/xorg.conf.new being used? Are you logged in as root when you start X? You don't need any config files anyway, so remove them. Shouldn't really matter regarding the crash, but still, makes no sense having outdated config files around (like I said, X -configure is outdated).

Quote:
It turns out I had misunderstood, the R200-family cards DO require an external firmware blob, but it is distributed with the kernel, so it is already on the filesystem and I don't have to emerge radeon-ucode, although the blob-file must still be added to the kernel config as explained in the document.

No, the blob file does not need to be added as explained there, CONFIG_FIRMWARE_IN_KERNEL=y already did everything necessary. Only blobs that are not already in the kernel source need to be added like that, and even that only if radeon is built-in (CONFIG_DRM_RADEON=y) rather than a module (CONFIG_DRM_RADEON=m).


X -configure does not use /root/xorg.conf.new, it creates it. Removing makes no difference. Yes I am invoking X -configure as root.
There is no xorg.conf anywhere else on the filesystem.

What command should I use to test X? The way it is set up now, GDM tries to start X at default runlevel. The screen goes dark, comes back to the console, goes dark, comes back to the console, goes dark and then comes back to the console saying GDM could not start X, offers to let me see the error logs and then goes back to the console log-in prompt.

Are you sure I don't need to include the firmware like I did? Before doing that it always reported KMS was not enabled and now it is...

Upgrading to xf86-video-ati-7.0.0-r1 has the same result:
Quote:
[ 236.383] (II) [KMS] Kernel modesetting enabled.
[ 236.383] (EE)
[ 236.384] (EE) Backtrace:
[ 236.384] (EE) 0: X (xorg_backtrace+0x50) [0x1015e9f0]
[ 236.384] (EE) 1: X (0x10000000+0x162638) [0x10162638]
[ 236.384] (EE) 2: linux-vdso32.so.1 (__kernel_sigtramp_rt32+0x0) [0x100350]
[ 236.384] (EE) 3: /usr/lib/xorg/modules/drivers/radeon_drv.so (0xf910000+0x2f614) [0xf93f614]
[ 236.384] (EE) 4: X (DoConfigure+0xa90) [0x1005f1cc]
[ 236.385] (EE) 5: X (InitOutput+0x1e8) [0x1006c6c4]
[ 236.385] (EE) 6: X (0x10000000+0x1e888) [0x1001e888]
[ 236.385] (EE) 7: /lib/libc.so.6 (0xfaee000+0x23410) [0xfb11410]
[ 236.385] (EE) 8: /lib/libc.so.6 (0xfaee000+0x235a0) [0xfb115a0]
[ 236.385] (EE)
[ 236.385] (EE) Segmentation fault at address 0x14
[ 236.385]
Fatal server error:
[ 236.385] Caught signal 11 (Segmentation fault). Server aborting
Back to top
View user's profile Send private message
Gusar
Advocate
Advocate


Joined: 09 Apr 2005
Posts: 2643
Location: Slovenia

PostPosted: Sat Feb 02, 2013 12:25 am    Post subject: Reply with quote

Progman3K wrote:
What command should I use to test X?

As the Gentoo X guide recommends, xterm and twm with startx. Definitely no desktop manager.

Progman3K wrote:
Are you sure I don't need to include the firmware like I did? Before doing that it always reported KMS was not enabled and now it is...

Very sure. If you look at all the logs you posted again, this one already says kernel modesetting is enabled, and that was before you did the additional firmware stuff.

Progman3K wrote:
Upgrading to xf86-video-ati-7.0.0-r1 has the same result:

I said 7.1.0. However, your configuration is now correct, so the crash is a driver bug which should be reported upstream.
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Sat Feb 02, 2013 3:07 pm    Post subject: Reply with quote

OK, I've filed a bug

https://bugs.freedesktop.org/show_bug.cgi?id=60180

An X developer has responded asking me for more info; rebuilding the suspect modules with debug info.

I've read the debugging guide
http://wiki.x.org/wiki/Development/Documentation/ServerDebugging

My question is, what is the clean or 'Gentoo' way of doing that?
Do I modify my compiler options /etc/make.conf and rebuild the modules or is there a better way?
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Mon Feb 04, 2013 5:31 pm    Post subject: Reply with quote

Investigation concludes that it may be a Gentoo-specific modification causing the bug.
https://bugs.gentoo.org/show_bug.cgi?id=455468
Back to top
View user's profile Send private message
chithanh
Developer
Developer


Joined: 05 Aug 2006
Posts: 1674
Location: Berlin, Germany

PostPosted: Mon Feb 04, 2013 5:53 pm    Post subject: Reply with quote

X -configure never works on KMS. Do not use it. For finding problems, it is best to not use any xorg.conf at all and only create individual xorg.conf.d/ snippets as needed.

Another option besides the xf86-video-ati driver would be xf86-video-modesetting.
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Tue Feb 05, 2013 3:00 am    Post subject: Reply with quote

chithanh wrote:
X -configure never works on KMS. Do not use it. For finding problems, it is best to not use any xorg.conf at all and only create individual xorg.conf.d/ snippets as needed.

Another option besides the xf86-video-ati driver would be xf86-video-modesetting.


Right now, X starts but GDM isn't able to start Gnome.

Even when X starts though, the mouse is incredibly sluggish, it takes it a full minute to react when I move it.

How is xf86-video-modesetting used? Do I need to uninstall the Radeon driver and change my VIDEO_CARDS variable in make.conf? What do I set the variable to?
Back to top
View user's profile Send private message
chithanh
Developer
Developer


Joined: 05 Aug 2006
Posts: 1674
Location: Berlin, Germany

PostPosted: Tue Feb 05, 2013 4:05 am    Post subject: Reply with quote

xf86-video-modesetting is a generic unaccelerated driver for KMS. It is not very useful for anything that uses 3D acceleration (e.g. GNOME 3).

You have some AGP related warnings in dmesg, maybe they will go away if you boot with radeon.agpmode=-1 kernel parameter.
Back to top
View user's profile Send private message
Progman3K
Guru
Guru


Joined: 03 Jan 2004
Posts: 535

PostPosted: Tue Feb 05, 2013 5:48 am    Post subject: Reply with quote

OK I added the kernel parameters you suggested to /etc/yaboot.conf
Quote:
boot=/dev/hda2
device=/pci@f4000000/ata-6@d/disk@0:
partition=3
root=/dev/hda3
timeout=30
install=/usr/lib/yaboot/yaboot
magicboot=/usr/lib/yaboot/ofboot
append="radeon.modeset=1 radeon.agpmode=-1"

image=/boot/vmlinux
label=Linux
read-only

Ran ybin -v, rebooted.

Here is the resultant dmesg output
http://gentoo.pastebin.ca/2310809

I don't see and AGP warnings. The line
[drm:radeon_agp_init] *ERROR* Unable to acquire AGP: -19
has gone.

There might be more than one problem; xdm is unable to start GDM or Gnome (I can see a process called XKeepsCrashing appearing in the tasks) but I can still run startx and even though it takes a very long time, eventually X does start.

I can also create an ~/.xinitrc that invokes gnome-session, which also takes a very long time to load but eventually does start.

When either the generic X session or the Gnome session are running though, the mouse pointer takes hours to move across the screen, although when typing, sometimes the characters appears at regular speed...

I've created an xorg.conf file
Quote:

Section "Device"
Identifier "Configured Video Device"
BusID "PCI:0:16:0"
Option "UseFBDev" "true"
Option "IgnoreEDID" "True"
Option "DDCMode" "False"
# Option "ForceTVOut" "True"
# Option "TVStandard" "pal"
EndSection

Section "Monitor"
Identifier "Configured Monitor"
DisplaySize 240 135
HorizSync 31.5-37.9
VertRefresh 59.9-60.1
EndSection

Section "Screen"
Identifier "Default Screen"
Monitor "Configured Monitor"
Device "Configured Video Device"
SubSection "Display"
Depth 24
Modes "1920x1080"
EndSubSection
EndSection


Maybe setting some of the options could fix the slowness problem
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Gentoo on PPC All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum