Re: amdgpu.ko crashes on Radeon Vega

From: none <lojas_at_arroway.org>
Date: Fri, 21 May 2021 23:18:18 -0300
On Fri, May 21, 2021 22:29, none wrote:
> On Sat, May 15, 2021 14:50, George Mitchell wrote:
>> On 5/15/21 1:29 PM, Nenhum_de_Nos wrote:
>>>[...]
>>> George,
>>>
>>> I'm about to try this on Ryzen 3400G here, where can I find this step
> by step to try to repeat what you did? My box has just basic FreeBSD
> install,
>>> cause I couldn't use Xorg at all. When I finish something on another OS
>> I
>>> will try this.
>>>
>>> Thanks,
>>>
>>> matheus
>>>
>>>
>> 1. Install FreeBSD 12.
>> 2. Install drm-fbsd12.0-kmod.
>> 3. Boot into single-user mode.
>> 4. Type "kldload amdgpu".
>
> Hi,
>
> I did just that and no crash here. The system is:
>
> FreeBSD xxx 12.2-RELEASE-p6 FreeBSD 12.2-RELEASE-p6 GENERIC  amd64
>
> % kldstat
> Id Refs Address                Size Name
>  1   63 0xffffffff80200000  227ae98 kernel
>  2    1 0xffffffff82a11000   24f9e4 amdgpu.ko
>  3    2 0xffffffff82c61000    75e80 drm.ko
>  4    5 0xffffffff82cd7000    12d30 linuxkpi.ko
>  5    4 0xffffffff82cea000    13f30 linuxkpi_gplv2.ko
>  6    2 0xffffffff82cfe000      6d0 debugfs.ko
>  7    1 0xffffffff82cff000     f0e1 ttm.ko
>  8    1 0xffffffff82d0f000    16bf0 if_iwm.ko
>  9    1 0xffffffff82d26000     2698 intpm.ko
> 10    1 0xffffffff82d29000      b40 smbus.ko
> 11    1 0xffffffff82d2a000   28debf iwm9260fw.ko
> 12    1 0xffffffff82fb8000     4260 ng_ubt.ko
> 13    6 0xffffffff82fbd000     9bd0 netgraph.ko
> 14    2 0xffffffff82fc7000     9128 ng_hci.ko
> 15    3 0xffffffff82fd1000      9b0 ng_bluetooth.ko
> 16    1 0xffffffff82fd2000     1860 uhid.ko
> 17    1 0xffffffff82fd4000     2908 ums.ko
> 18    1 0xffffffff82fd7000     1a40 wmt.ko
> 19    1 0xffffffff82fd9000     caf0 ng_l2cap.ko
> 20    1 0xffffffff82fe6000    1af20 ng_btsocket.ko
> 21    1 0xffffffff83001000     2150 ng_socket.ko
>
>
>> When you say you can't use Xorg at all, does that mean VESA mode is not
> working?                                                       -- George
>
> Yep, but I had quite no time to  research for the right xorg.conf to use
> it. I tried startx here zero xorg.conf fiddling, and server error. Its so
> many time since I last did this, I am here quite a first timer.
>
> thanks,
>
> matheus

I tried it on rc.conf for loading, no crashes on two reboots. The console
won't change resolution though, as it does on another box here that uses
some old radeon gpu.

startx dies in front of me:

% cat /var/log/Xorg.0.log
[    36.115]
X.Org X Server 1.20.9
X Protocol Version 11, Revision 0
[    36.115] Build Operating System: FreeBSD 12.1-RELEASE-p12 amd64
[    36.115] Current Operating System: FreeBSD elita 12.2-RELEASE-p6
FreeBSD 12.2-RELEASE-p6 GENERIC amd64
[    36.115] Build Date: 05 January 2021  02:48:32PM
[    36.115]
[    36.115] Current version of pixman: 0.40.0
[    36.115] 	Before reporting problems, check http://wiki.x.org
	to make sure that you have the latest version.
[    36.115] Markers: (--) probed, (**) from config file, (==) default
setting,
	(++) from command line, (!!) notice, (II) informational,
	(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[    36.116] (==) Log file: "/var/log/Xorg.0.log", Time: Fri May 21
23:06:24 2021
[    36.118] (==) Using system config directory
"/usr/local/share/X11/xorg.conf.d"
[    36.119] (==) No Layout section.  Using the first Screen section.
[    36.120] (==) No screen section available. Using defaults.
[    36.120] (**) |-->Screen "Default Screen Section" (0)
[    36.120] (**) |   |-->Monitor "<default monitor>"
[    36.120] (==) No monitor specified for screen "Default Screen Section".
	Using a default monitor configuration.
[    36.120] (==) Automatically adding devices
[    36.120] (==) Automatically enabling devices
[    36.120] (==) Not automatically adding GPU devices
[    36.121] (==) Max clients allowed: 256, resource mask: 0x1fffff
[    36.126] (==) FontPath set to:
	/usr/local/share/fonts/misc/,
	/usr/local/share/fonts/TTF/,
	/usr/local/share/fonts/OTF/,
	/usr/local/share/fonts/Type1/,
	/usr/local/share/fonts/100dpi/,
	/usr/local/share/fonts/75dpi/,
	catalogue:/usr/local/etc/X11/fontpath.d
[    36.126] (==) ModulePath set to "/usr/local/lib/xorg/modules"
[    36.126] (II) The server relies on udev to provide the list of input
devices.
	If no devices become available, reconfigure udev or disable AutoAddDevices.
[    36.126] (II) Loader magic: 0x42f020
[    36.126] (II) Module ABI versions:
[    36.126] 	X.Org ANSI C Emulation: 0.4
[    36.126] 	X.Org Video Driver: 24.1
[    36.126] 	X.Org XInput driver : 24.1
[    36.126] 	X.Org Server Extension : 10.0
[    36.126] (--) PCI:*(10_at_0:0:0) 1002:15d8:1002:15d8 rev 200, Mem _at_
0xe0000000/268435456, 0xf0000000/2097152, 0xfca00000/524288, I/O _at_
0x0000e000/256, BIOS _at_ 0x????????/65536
[    36.126] (II) LoadModule: "glx"
[    36.127] (II) Loading /usr/local/lib/xorg/modules/extensions/libglx.so
[    36.138] (II) Module glx: vendor="X.Org Foundation"
[    36.138] 	compiled for 1.20.9, module version = 1.0.0
[    36.139] 	ABI class: X.Org Server Extension, version 10.0
[    36.139] (==) Matched ati as autoconfigured driver 0
[    36.139] (==) Matched modesetting as autoconfigured driver 1
[    36.139] (==) Matched scfb as autoconfigured driver 2
[    36.139] (==) Matched vesa as autoconfigured driver 3
[    36.139] (==) Assigned the driver to the xf86ConfigLayout
[    36.139] (II) LoadModule: "ati"
[    36.139] (WW) Warning, couldn't open module ati
[    36.139] (EE) Failed to load module "ati" (module does not exist, 0)
[    36.139] (II) LoadModule: "modesetting"
[    36.140] (II) Loading
/usr/local/lib/xorg/modules/drivers/modesetting_drv.so
[    36.141] (II) Module modesetting: vendor="X.Org Foundation"
[    36.141] 	compiled for 1.20.9, module version = 1.20.9
[    36.141] 	Module class: X.Org Video Driver
[    36.141] 	ABI class: X.Org Video Driver, version 24.1
[    36.141] (II) LoadModule: "scfb"
[    36.141] (II) Loading /usr/local/lib/xorg/modules/drivers/scfb_drv.so
[    36.141] (II) Module scfb: vendor="X.Org Foundation"
[    36.141] 	compiled for 1.20.9, module version = 0.0.5
[    36.141] 	ABI class: X.Org Video Driver, version 24.1
[    36.141] (II) LoadModule: "vesa"
[    36.141] (II) Loading /usr/local/lib/xorg/modules/drivers/vesa_drv.so
[    36.142] (II) Module vesa: vendor="X.Org Foundation"
[    36.142] 	compiled for 1.20.9, module version = 2.5.0
[    36.142] 	Module class: X.Org Video Driver
[    36.142] 	ABI class: X.Org Video Driver, version 24.1
[    36.142] (II) modesetting: Driver for Modesetting Kernel Drivers: kms
[    36.142] (II) scfb: driver for wsdisplay framebuffer: scfb
[    36.142] (II) VESA: driver for VESA chipsets: vesa
[    36.142] (--) Using syscons driver with X support (version 2.0)
[    36.142] (--) using VT number 9

[    36.142] (EE) open /dev/dri/card0: No such file or directory
[    36.142] (WW) Falling back to old probe method for modesetting
[    36.142] (EE) open /dev/dri/card0: No such file or directory
[    36.142] (WW) Falling back to old probe method for scfb
[    36.142] scfb trace: probe start
[    36.142] (II) scfb(1): using default device
[    36.142] scfb trace: probe done
[    36.142] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card
support
[    36.142] (EE) Screen 0 deleted because of no matching config section.
[    36.142] (II) UnloadModule: "modesetting"
[    36.142] (EE)
Fatal server error:
[    36.142] (EE) Cannot run in framebuffer mode. Please specify busIDs   
    for all framebuffer devices
[    36.142] (EE)
[    36.142] (EE)
Please consult the The X.Org Foundation support
	 at http://wiki.x.org
 for help.
[    36.142] (EE) Please also check the log file at "/var/log/Xorg.0.log"
for additional information.
[    36.142] (EE)
[    36.142] (EE) Server terminated with error (1). Closing log file.

I tried it without any pre-config, this is the error.

I tried Xorg -configure and the file created by it. Error no screens found.

Am I missing something?

The handbook told me not much more.

Thanks,

matheus

PS: I couldn't answer that first mail using my other email address, as the
mail list software told me several times there was an access error:
The message from <xxxx> with subject "Re: amdgpu.ko
crashes on Radeon Vega" was unable to be delivered to the list because of
an access rule set up by the list administrator.

(The denied message is below.)

The mail is subscribed, is receiving messages fine (got that one from new
address). I didn't get this error at all.


-- 
"We will call you Cygnus,
the God of balance you shall be."
Received on Sat May 22 2021 - 02:18:18 UTC

Original text of this message