Intel driver 2.10 & KMS issue

DrMonkey · 2010-02-12 05:58:30

Hi,
I run Arch Linux on a netbook, with a Intel 945GME graphic chipset. I don't use a Xorg.conf, worked fine without it so far. I don't use KMS too, because else, shit happened. The last update I did bring the Intel driver 2.10.0-1, which actually requires KMS. Then shit happens, which is :
KMS not active => Xorg fails. A look at the logs says "Fatal server error: no screens found", which seems triggered by this error "(EE) intel(0): No kernel modesetting driver detected."
KMS active, as explained on the wiki => Boot fails, got black screen very early, with nothing to do but rebooting and desactivating manually KMS with GRUB

How can I fix that ?
[edit : ho, and almost bricked netbook now, no display at all past the init. I feel anger and frustration, each damn kernel/gfx driver update ends up in a nightmare]

Last edited by DrMonkey (2010-02-13 00:31:26)

combuster · 2010-02-12 08:20:13

Boot with KMS, it blacks out, restart, add nomodeset kernel parameter as before and then read the log file, see what is wrong... /var/log/everything.log is a lil' bulky but contains all the information.

renato_garcia · 2010-02-12 08:54:34

I have a Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller, and this same problem. This is not a real solution, but downgrade xf86-video-intel to 2.9.1 was the only workaround that I found.

There are a bug report in http://bugs.archlinux.org/task/18298

redada · 2010-02-12 15:39:21

The same here on my old Vaio with i830m graphics.
I need to downgrade this evening too.

knedlyk · 2010-02-12 16:32:59

I don't think it is a bug, read http://www.archlinux.org/news/484/ - xf86-video-intel supports only KMS, so KMS must be enabled by default, in the opposite case xorg simply doesn't work, because it required KMS. That's it.

kgas · 2010-02-12 16:43:58

any one read this. same as previous post

Last edited by kgas (2010-02-12 16:46:19)

redada · 2010-02-12 16:56:38

Yes, but as CrMonkey said, i also get this:

KMS active, as explained on the wiki => Boot fails, got black screen very early, with nothing to do but rebooting and desactivating manually KMS with GRUB

jpt · 2010-02-12 17:25:21

Also having the same problems. Please if someone solves this keep us in the loop!

sctincman · 2010-02-12 18:52:00

Perhaps try late KMS, have the kernel boot with i915.modeset=0, then in modprobe.conf add in "options i915 modeset=1" to enable it once the module is loaded

I think that should work how I think it would I just remember when KMS was first introduced, there were ways to have it delay enabling it until it finished booting.

And like combuster said, boot with KMS enabled, then reboot without it and check that log file, could give clues to what's not working.

swiftscythe · 2010-02-12 19:46:21

I also have the same problem, if I start without i915.modeset=0, the screen goes black. If I start with that option, can't start X.

Here's the output of startx:

This is a pre-release version of the X server from The X.Org Foundation.
It is not supported in any way.
Bugs may be filed in the bugzilla at http://bugs.freedesktop.org/.
Select the "xorg" product for bugs you find in this release.
Before reporting bugs in pre-release versions please check the
latest version in the X.Org Foundation git repository.
See http://wiki.x.org/wiki/GitPage for git access instructions.

X.Org X Server 1.7.4.901 (1.7.5 RC 1)
Release Date: 2010-01-23
X Protocol Version 11, Revision 0
Build Operating System: Linux 2.6.32-ARCH i686
Current Operating System: Linux arch-tablet 2.6.32-ARCH #1 SMP PREEMPT Tue Feb 9 14:46:08 UTC 2010 i686
Kernel command line: root=/dev/disk/by-uuid/ca95a549-213b-4b3d-bcba-ba71383fd3a3 ro vga=773 i915.modeset=0
Build Date: 31 January 2010  11:40:13AM

Current version of pixman: 0.16.4
    Before reporting problems, check http://wiki.x.org
    to make sure that you have the latest version.
Markers: (--) probed, (**) from config file, (==) default setting,
    (++) from command line, (!!) notice, (II) informational,
    (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(==) Log file: "/var/log/Xorg.0.log", Time: Fri Feb 12 20:51:50 2010
(==) Using config file: "/etc/X11/xorg.conf"
(EE) Failed to load module "synaptics" (module does not exist, 0)
(EE) intel(0): No kernel modesetting driver detected.
(EE) Screen(s) found, but none have a usable configuration.

Fatal server error:
no screens found

Please consult the The X.Org Foundation support
     at http://wiki.x.org
 for help.
Please also check the log file at "/var/log/Xorg.0.log" for additional information.

giving up.
xinit:  No such file or directory (errno 2):  unable to connect to X server
xinit:  No such process (errno 3):  Server error.

Last edited by swiftscythe (2010-02-12 19:57:50)

DrMonkey · 2010-02-13 00:29:16

By the way
- the "late" or "early" way of triggering KMS ends up the same way : once the system try to go KMS, black screen. The netbook still works, in console mode (Xorg failure, same as posted previously here), I can type commands though without screen it's not very handy.
- The only fix so far is downgrading >_<

sctincman · 2010-02-13 00:31:36

@swiftcycle
KMS is now required for the xf86-video-intel driver. http://www.archlinux.org/news/484/

@DrMonkey

I was just reading this thread, http://bbs.archlinux.org/viewtopic.php?id=90857 and was thinking that perhaps with KMS, the driver might be setting the wrong resolution or maybe attempting to display on another adapter

Perhaps you could try appending this to your boot line, with KMS enabled

video=LVDS-1:1280x800

with your desired resolution of course.

you may need to change LVDS-1, depending on your setup, check the output of xrandr -q to see what your display is named

Last edited by sctincman (2010-02-13 00:32:44)

swiftscythe · 2010-02-13 00:36:46

@sctincman
Yes, I know that, but when I enable it, the screen goes blank... I can only use the computer with "nomodset". How can I fix that?

Thanks for the response

combuster · 2010-02-13 00:45:54

Will someone with this problem attach it's /var/log/everything.log on pastebin or any filesharing site?

swiftscythe · 2010-02-13 12:56:35

@combuster: Hi! Here's my /var/log/everything.log
It's quite big... Hope it can help solve this problem
http://tinyurl.com/yl22hb3
Thanks in advance

patroclo7 · 2010-02-13 14:33:06

swiftscythe wrote:

http://tinyurl.com/yl22hb3

Perhaps it is not the only problem but vesafb is kicking in, and it shouldn't. Did you remove any vga and video parameter from grub.conf? Could you post it?

Also, is /etc/modprobe.d/framebuffer_blacklist.conf in place?

swiftscythe · 2010-02-13 15:14:31

I don't think i removed anyvga and video parametre. I added two:
# (0) Arch Linux
title Arch Linux
root (hd0,0)
kernel /vmlinuz26 root=/dev/disk/by-uuid/ca95a549-213b-4b3d-bcba-ba71383fd3a3 ro nomodeset vga=773
initrd /kernel26.img

Yes, /etc/modprobe.d/framebuffer_blacklist.conf is in place:

blacklist atyfb
blacklist aty128fb
blacklist radeonfb
blacklist sisfb
blacklist viafb
blacklist i810fb
blacklist kyrofb
blacklist rivafb
blacklist vt8623fb
blacklist pm2fb
blacklist cyber2000fb
blacklist uvesafb
blacklist tdfxfb
blacklist s3fb
blacklist neofb
blacklist nvidiafb
blacklist savagefb
blacklist cirrusfb
blacklist sstfb
blacklist vmlfb
blacklist tridentfb

Thanks for helping

patroclo7 · 2010-02-13 17:16:04

I am missing something. With the latest intel driver you need KMS. With kms you have to remove both vga=773 and nomodeset and to activate modesetting.

With that grub.cof you are hopeless, I fear. There is no way to make the latest intel driver without KMS.

If KMS (withut vga=773 in grub.conf) does not work for you, then what we need is the log in this case

Last edited by patroclo7 (2010-02-13 17:18:26)

swiftscythe · 2010-02-13 17:31:43

Yes, I know that. I tried to start without that two options but the screen just goes blank and there's nothing I can do.
This is the only way I got X working:
- Use the vesa driver instead of intel
- add nomodeset to the grub options

To enable modesetting I tried by adding i915.modeset=1, but nothing

Thanks

patroclo7 · 2010-02-13 17:35:57

Try to boot with i915.modeset=1 and no vga, let the screen go blank. After that reboot with nomodeset. You should be able to get the log of the time when you tried to boot with KMS, so we can see what is broken in this case.

swiftscythe · 2010-02-13 17:39:34

which log is the one you need?

patroclo7 · 2010-02-13 17:45:54

everything.log or everything.log.1 but just the portion relative to the time you boot with KMS (look at the timestamps)

In the everything.log you posted only one time you have booted without vga=773, but in that case you used i917.modeset=1 (it is i915 ). In another case you used i915.modeset=1, but together with vga=773, which can't work.

Last edited by patroclo7 (2010-02-13 17:52:09)

swiftscythe · 2010-02-13 18:19:53

Here you are:
http://pastebin.ca/1795482

Thank you very much

combuster · 2010-02-13 18:57:39

Feb 13 19:02:24 arch-tablet kernel: ------------[ cut here ]------------
Feb 13 19:02:24 arch-tablet kernel: WARNING: at drivers/gpu/drm/drm_crtc_helper.c:1035 drm_helper_initial_config+0x5e/0x60 [drm_kms_helper]()
Feb 13 19:02:24 arch-tablet kernel: Hardware name: STYLISTIC ST503X
Feb 13 19:02:24 arch-tablet kernel: No connectors reported connected with modes
Feb 13 19:02:24 arch-tablet kernel: Modules linked in: joydev hid_logitech ff_memless btusb usbhid hid bluetooth sr_mod rfkill cdrom snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq firewire_sbp2 i915(+) snd_seq_device snd_intel8x0 snd_intel8x0m snd_pcm_oss snd_ac97_codec snd_mixer_oss drm_kms_helper sdhci_pci apanel ac97_bus pcmcia sdhci drm input_polldev ipw2200 snd_pcm firewire_ohci mmc_core i2c_algo_bit i2c_i801 uhci_hcd iTCO_wdt snd_timer libipw irda snd yenta_socket firewire_core intel_agp iTCO_vendor_support fujitsu_laptop rsrc_nonstatic tg3 ehci_hcd crc_ccitt soundcore led_class lib80211 video pcmcia_core crc_itu_t psmouse button battery usbcore i2c_core snd_page_alloc ac agpgart output thermal processor sg evdev pcspkr libphy serio_raw fuse rtc_cmos rtc_core rtc_lib ext4 mbcache jbd2 crc16 sd_mod ata_piix ata_generic pata_acpi libata scsi_mod
Feb 13 19:02:24 arch-tablet kernel: Pid: 723, comm: modprobe Not tainted 2.6.32-ARCH #1
Feb 13 19:02:24 arch-tablet kernel: Call Trace:
Feb 13 19:02:24 arch-tablet kernel: [<c104042e>] ? warn_slowpath_common+0x6e/0xb0
Feb 13 19:02:24 arch-tablet kernel: [<f8478b2e>] ? drm_helper_initial_config+0x5e/0x60 [drm_kms_helper]
Feb 13 19:02:24 arch-tablet kernel: [<c10404bb>] ? warn_slowpath_fmt+0x2b/0x30
Feb 13 19:02:24 arch-tablet kernel: [<f8478b2e>] ? drm_helper_initial_config+0x5e/0x60 [drm_kms_helper]
Feb 13 19:02:24 arch-tablet kernel: [<f8759b05>] ? i915_driver_load+0x1365/0x1530 [i915]
Feb 13 19:02:24 arch-tablet kernel: [<f8758780>] ? i915_vga_set_decode+0x0/0x20 [i915]
Feb 13 19:02:24 arch-tablet kernel: [<f8581e79>] ? drm_get_dev+0x299/0x4c0 [drm]
Feb 13 19:02:24 arch-tablet kernel: [<c11376db>] ? sysfs_addrm_start+0x5b/0xb0
Feb 13 19:02:24 arch-tablet kernel: [<c1188290>] ? pci_match_device+0xa0/0xc0
Feb 13 19:02:24 arch-tablet kernel: [<c118811b>] ? local_pci_probe+0xb/0x10
Feb 13 19:02:24 arch-tablet kernel: [<c1188f01>] ? pci_device_probe+0x61/0x80
Feb 13 19:02:24 arch-tablet kernel: [<c11fc64b>] ? driver_probe_device+0x7b/0x170
Feb 13 19:02:24 arch-tablet kernel: [<c11fc7b9>] ? __driver_attach+0x79/0x80
Feb 13 19:02:24 arch-tablet kernel: [<c11fc740>] ? __driver_attach+0x0/0x80
Feb 13 19:02:24 arch-tablet kernel: [<c11fbeb2>] ? bus_for_each_dev+0x52/0x80
Feb 13 19:02:24 arch-tablet kernel: [<c11fc4e6>] ? driver_attach+0x16/0x20
Feb 13 19:02:24 arch-tablet kernel: [<c11fc740>] ? __driver_attach+0x0/0x80
Feb 13 19:02:24 arch-tablet kernel: [<c11fb7a6>] ? bus_add_driver+0xc6/0x2b0
Feb 13 19:02:24 arch-tablet kernel: [<c1188e40>] ? pci_device_remove+0x0/0x40
Feb 13 19:02:24 arch-tablet kernel: [<c11fca53>] ? driver_register+0x63/0x120
Feb 13 19:02:24 arch-tablet kernel: [<f857d50d>] ? drm_init+0x2d/0xf0 [drm]
Feb 13 19:02:24 arch-tablet kernel: [<f8606000>] ? i915_init+0x0/0x48 [i915]
Feb 13 19:02:24 arch-tablet kernel: [<c118912d>] ? __pci_register_driver+0x3d/0xb0
Feb 13 19:02:24 arch-tablet kernel: [<c100112f>] ? do_one_initcall+0x2f/0x190
Feb 13 19:02:24 arch-tablet kernel: [<c1074d84>] ? sys_init_module+0xb4/0x220
Feb 13 19:02:24 arch-tablet kernel: [<c10039f3>] ? sysenter_do_call+0x12/0x28
Feb 13 19:02:24 arch-tablet kernel: ---[ end trace dae2bb5f39b9f00b ]---
Feb 13 19:02:24 arch-tablet kernel: [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
Feb 13 19:02:24 arch-tablet kernel: EXT4-fs (sda4): mounted filesystem with ordered data mode
Feb 13 19:02:24 arch-tablet kernel: Adding 1028152k swap on /dev/sda2.  Priority:-1 extents:1 across:1028152k
Feb 13 19:02:24 arch-tablet crond[1431]: /usr/sbin/crond 4.4 dillon's cron daemon, started with loglevel info
Feb 13 19:02:24 arch-tablet init: Entering runlevel: 5
Feb 13 19:02:29 arch-tablet kernel: [drm] DAC-6: set mode 640x480 0
Feb 13 19:02:29 arch-tablet kernel: [drm] DAC-6: set mode 640x480 0
Feb 13 19:02:30 arch-tablet kernel: [drm] TV-8: set mode NTSC 480i 0
Feb 13 19:02:30 arch-tablet kernel: [drm] TV-8: set mode NTSC 480i 0
Feb 13 19:02:30 arch-tablet kernel: [drm] DAC-6: set mode 640x480 0
Feb 13 19:02:30 arch-tablet kernel: [drm] DAC-6: set mode 640x480 0
Feb 13 19:02:30 arch-tablet kernel: [drm] TV-8: set mode NTSC 480i 0
Feb 13 19:02:31 arch-tablet kernel: [drm] TV-8: set mode NTSC 480i 0
Feb 13 19:02:31 arch-tablet kdm[1516]: X server died during startup
Feb 13 19:02:31 arch-tablet kdm[1516]: X server for display :0 cannot be started, session disabled

Can you try without i915.modeset=1, just leave the default kernel line ending with ro. KMS is enabled by default in the latest kernel and there is no need to specify i915.modeset=1. If that doesn't help you could report this error to the arch bug tracking system.

swiftscythe · 2010-02-13 19:04:03

Okay thanks! I shoud use the intel driver to test it? Or with the vesa it's ok?

Arch Linux

#1 2010-02-12 05:58:30

Intel driver 2.10 & KMS issue

#2 2010-02-12 08:20:13

Re: Intel driver 2.10 & KMS issue

#3 2010-02-12 08:54:34

Re: Intel driver 2.10 & KMS issue

#4 2010-02-12 15:39:21

Re: Intel driver 2.10 & KMS issue

#5 2010-02-12 16:32:59

Re: Intel driver 2.10 & KMS issue

#6 2010-02-12 16:43:58

Re: Intel driver 2.10 & KMS issue

#7 2010-02-12 16:56:38

Re: Intel driver 2.10 & KMS issue

#8 2010-02-12 17:25:21

Re: Intel driver 2.10 & KMS issue

#9 2010-02-12 18:52:00

Re: Intel driver 2.10 & KMS issue

#10 2010-02-12 19:46:21

Re: Intel driver 2.10 & KMS issue

#11 2010-02-13 00:29:16

Re: Intel driver 2.10 & KMS issue

#12 2010-02-13 00:31:36

Re: Intel driver 2.10 & KMS issue

#13 2010-02-13 00:36:46

Re: Intel driver 2.10 & KMS issue

#14 2010-02-13 00:45:54

Re: Intel driver 2.10 & KMS issue

#15 2010-02-13 12:56:35

Re: Intel driver 2.10 & KMS issue

#16 2010-02-13 14:33:06

Re: Intel driver 2.10 & KMS issue

#17 2010-02-13 15:14:31

Re: Intel driver 2.10 & KMS issue

#18 2010-02-13 17:16:04

Re: Intel driver 2.10 & KMS issue

#19 2010-02-13 17:31:43

Re: Intel driver 2.10 & KMS issue

#20 2010-02-13 17:35:57

Re: Intel driver 2.10 & KMS issue

#21 2010-02-13 17:39:34

Re: Intel driver 2.10 & KMS issue

#22 2010-02-13 17:45:54

Re: Intel driver 2.10 & KMS issue

#23 2010-02-13 18:19:53

Re: Intel driver 2.10 & KMS issue

#24 2010-02-13 18:57:39

Re: Intel driver 2.10 & KMS issue

#25 2010-02-13 19:04:03

Re: Intel driver 2.10 & KMS issue

Board footer