Bug 172893 - Ati Radeon with Xorg and DRI activated freezes the Machine
Summary: Ati Radeon with Xorg and DRI activated freezes the Machine
Status: Open
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: Unspecified
Hardware: Any Any
: Normal Affects Only Me
Assignee: freebsd-bugs (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-10-20 04:10 UTC by M.
Modified: 2020-09-28 04:16 UTC (History)
2 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description M. 2012-10-20 04:10:00 UTC
# src and ports updated at 2012-10-16 from cvsup.uk.freebsd.org via csup/make

The usage of a PCI-Express ATI Radeon HD 4350 with the Xorg-Server
and DRI enabled (default) freezes the entire Machine imediatly after the start of the Xorg. The only running solution is to deactivate the acceleration with
'Option "NoAccel"  "True" in th xorg.conf. The Xserver does than shadowing the entire graphic memory on the graphic card, which reduces the available/usable memory by an half. Anything that needs at least 2D acceleration like playing a video or flash movie is horridbly slow. Specially with GLX, XV can't get used cause of the missing DRI support.

Further is the described freeze only moved at the end of the actual Xsession.
If i close fluxbox and exit him, the machine gets freezed than.
(slim is as login manager installed and the polkit with the consolekit as
 session management, this works flawless without HAL in the same configuration
 on a laptop with intel graphics )

For the fun: if i call reboot from a xterm inside the session, the machine gets rebooted clean without a freeze. :8~)

The Display/Monitor is connected to the VGA-Port and gets corretly found by the probing of Xorg.

The Mainboard is an Asus P5K, with 6GigByte of OCZ Value-Memory.
Memory tests are also ok and the hardware is known to work perfectly.

For the interest the irq 16 that is used by the graphics adapter is shared with the north bridge/acpi/usb/firewire. I have no idea how i can set any of both to a free _non shared_ irq.

This got tested in any way, with or without the kernel modules for radeon and drm/drm2.
the latter one should not be the approbiate cause iirc its only good for intel graphic adapters. the freeze is repeateable with the ati/radeon driver and with the radeonhd driver. ( i know the latter one should be outdated and not used anymore, if i have understood the wiki at xorg)

I had this grapic adapter on this mainboard already running earlier _under linux_ with 2D acceleration with the xorg drivers and 3D acceleration with the proprietary fglrx drivers from ati. both with dri. afaik that is neccessary for any hardware acceleration.

i addressed this bug into the kern category cause of:
under no circuumstances should a buggy X driver drag down the entire machine with all his mounted filesystems while this can cause a serious dataloss and corrupted filesystems. specially not if the default xorg.conf created by Xorg -configure comes up with non safe parameters.

if any help is needed to examine this further i can help if i get an guidance
how i can trace this down/back.

machine/software config following:

Kernel modules loaded at boot time wia loader.conf(just a few important):
(it makes no difference if with or without these modules loaded )
drm_load="YES"
radeon_load="YES"
#vesa_load="YES"
#acpi_video_load="YES"
aibs_load="YES"
coretemp_load="YES"

----8<---snip:dmesg-parts---8<---
Copyright (c) 1992-2012 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
        The Regents of the University of California. All rights reserved.
FreeBSD is a registered trademark of The FreeBSD Foundation.
FreeBSD 9.1-PRERELEASE #2: Wed Oct 17 03:55:20 CEST 2012
    root@bsdburner.michael-schuh.net:/usr/obj/usr/src/sys/GENERIC amd64
CPU: Intel(R) Core(TM)2 Duo CPU     E7300  @ 2.66GHz (2672.78-MHz K8-class CPU)
  Origin = "GenuineIntel"  Id = 0x10676  Family = 0x6  Model = 0x17  Stepping = 6
  Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CM
OV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
  Features2=0x8e39d<SSE3,DTES64,MON,DS_CPL,EST,TM2,SSSE3,CX16,xTPR,PDCM,SSE4.1>
  AMD Features=0x20100800<SYSCALL,NX,LM>
  AMD Features2=0x1<LAHF>
  TSC: P-state invariant, performance statistics
real memory  = 6442450944 (6144 MB)
avail memory = 6163501056 (5877 MB)
Event timer "LAPIC" quality 400
ACPI APIC Table: <A_M_I_ OEMAPIC >
FreeBSD/SMP: Multiprocessor System Detected: 2 CPUs
FreeBSD/SMP: 1 package(s) x 2 core(s)
 cpu0 (BSP): APIC ID:  0
 cpu1 (AP): APIC ID:  1
ioapic0 <Version 2.0> irqs 0-23 on motherboard
Cuse4BSD v0.1.26 @ /dev/cuse
kbd1 at kbdmux0
ctl: CAM Target Layer loaded
acpi0: <A_M_I_ OEMRSDT> on motherboard
acpi0: Power Button (fixed)
acpi0: reservation of 0, a0000 (3) failed
acpi0: reservation of 100000, cff00000 (3) failed
cpu0: <ACPI CPU> on acpi0
ACPI Warning: Incorrect checksum in table [OEMB] - 0x72, should be 0x69 (20110527/tbutils-282)
# i think the above line should ot be dramatical cause a warning
# and i have seen this already earlier on other fully stable machines.
# i think the failed messages above are the powerstates stuff which i didn't
# configured in the BIOS
Comment 1 Eitan Adler freebsd_committer freebsd_triage 2017-12-31 08:00:15 UTC
For bugs matching the following criteria:

Status: In Progress Changed: (is less than) 2014-06-01

Reset to default assignee and clear in-progress tags.

Mail being skipped
Comment 2 Alexey Dokuchaev freebsd_committer 2020-09-28 04:16:16 UTC
Do you still have this problem on recent FreeBSD versions?  Note that DRM support had moved out of the base system/kernel to the `graphics/drm-*-kmod' and `graphics/gpu-firmware-kmod' ports.

Your report was for 9.1-PRERELEASE which went out of support long time ago.  Please update to the supported FreeBSD version and reconduct your tests.

Adding x11@ to CC list so it can be tracked while filed under base system/kern.