Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
[ati-drivers] No supported adapters detected
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3, 4, 5  Next  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 6:16 pm    Post subject: Reply with quote

Quote:
configure and rebuild the kernel for the closed source driver


What exactly does this mean? For example, with the radeon, I had to set the firmware filenames and point to its directory.
I didn't do anything similar for the fglrx, maybe that's the problem? I relied solely on emerging ati-drivers and the changes you suggested earlier. Like tile-blitting, etc..

I'll paste the conf in a moment, but I'm afraid it is not so fresh as you might expect, because it says "using xorg.conf"... then although it says it's backing up the current one, my changes are still there. I do remember its changes from the first time I ran though, or at least what I noticed at the time, and it was mostly in the Device section, under a paragraph of comments explicitly from amd, though most options are commented out, maybe I have to set them all?

edit: There: [url]bpaste.net/show/198010[/url]
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sat Apr 05, 2014 7:23 pm    Post subject: Reply with quote

that xorg.conf is huge compared to mine and does not look like aticonfig --initial produced it.

A member named Gusar used to appear here and on the arch forum regularly and demonstrate great skill in solving fglrx problems. He often swore (in 2012) that a /etc/X11/xorg.conf containing only:
Quote:
Section "Device"
Identifier "Radeon Card"
Driver "fglrx"
EndSection
was sufficient to allow the xorg scripts to get X running with fglrx. I'm going to try it now, BRB.
_________________
Defund the FCC.
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sat Apr 05, 2014 7:38 pm    Post subject: Reply with quote

That minimal xorg.conf still works.
suggest
Code:
mv /etc/X11/xorg.conf /etc/X11/xorg.conf.backup
nano /etc/X11/xorg.conf
edit to
Quote:
Section "Device"
Identifier "Radeon Card"
Driver "fglrx"
EndSection
and
Code:
startx

_________________
Defund the FCC.
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 9:29 pm    Post subject: Reply with quote

cool! Will do.

brb...

edit for Results: Out of Sync....

I'm beginning to suspect it is indeed something in the kernel, even though it detects the card, the monitor and everything. Remember when I said I noticed before the thing booted, that the radeon was no longer present and the compilation had finally worked? That's because the radeon driver at least ws capable of booting into a framebuffer console, this does not change into it. Whatever was the option to make it do it, should still be there, no? :?
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54098
Location: 56N 3W

PostPosted: Sat Apr 05, 2014 9:45 pm    Post subject: Reply with quote

MarcoMarin,

No. You must disable the Radeon driver it the kernel or it will grab the grahics chip before fgrlx loads.
Then the binary blob will not find your graphics chip.

Sight of your /var/log/Xorg.0.log may help.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sat Apr 05, 2014 10:01 pm    Post subject: Reply with quote

Hi Neddy , welcome
http://bpaste.net/show/197230/ from back a day

@ Marco
fresh copy with the minimal xorg.conf would be good

@ everybody
Spent two pages not getting changes made to menuconfig into the kernel. Now aticonfig is not making changes to xorg.conf?
_________________
Defund the FCC.


Last edited by DONAHUE on Sat Apr 05, 2014 10:20 pm; edited 1 time in total
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 10:11 pm    Post subject: Reply with quote

Hi again Neddy ^_^

Yes, but I think the problem might not be fglrx not loading but conversely it trying to load too much :lol: no kidding, the thing is not only loaded at boot up but the xorg log you ask (below) states at least twice to be trying to load the thing, and there is a moment it even says it's going to do it old school style! It sounds rough! lol (not kidding! exact wording may vary)

http://bpaste.net/show/198104

excerpt: "Falling back to old probe method for fglrx".. it doesn't sound good at all... and it is right after it supposedly succeeded with loading it already.

A few lines later it tries to load fglrxdrm, which it had already done.

a few pages down (I'm on 25 chars vertical res), it does complain of an error about not finding a shared object, fglrx_dri, to load.

Does my .config here says radeon is on? :? I don't remember...

Edit: A few days back, donahue posted a statement saying it needed "firmware" to work. I've found strange but kept it. Now that you say that, I notice a slight inconsistency with that statement and his sequence of commands to change from radeon to fglrx. He did unmerge the firmware. Do you think it may be interfering, even though lspci -k says fglrx is loaded?
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 10:24 pm    Post subject: Reply with quote

Hi again Donahue.

I noticed something on the xorg log. Remember I said there was a -xinerama USE option in some package I merged during those experiments? In the xorg log it say it goes right ahead and enable xinerama 8O

Another thing, in your sequence of commands to change between radeon and fglrx, there is a consequence of doing it twice that I may not get here. Namely, you unmerge fglrx, then re-emerge xorg for the radeon, then do that again.
A symptom I have here while I re-emerged ati-drivers and xorg was to have it state a new version was available but it was not going to get that because ati-drivers needed the installed version. Maybe it then didn't think it should apply some new USE var? i dunno.... should I un-merge ati-drivers, then re-emerge xorg, then re-emerge ati-drivers?? :?
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sat Apr 05, 2014 10:30 pm    Post subject: Reply with quote

if
Code:
ls /lib/firmware/radeon/PITCAIRN*
shows
Quote:
/lib/firmware/radeon/PITCAIRN_ce.bin /lib/firmware/radeon/PITCAIRN_pfp.bin
/lib/firmware/radeon/PITCAIRN_mc.bin /lib/firmware/radeon/PITCAIRN_rlc.bin
/lib/firmware/radeon/PITCAIRN_me.bin /lib/firmware/radeon/PITCAIRN_smc.bin
and
Code:
 ls /lib/firmware/radeon/TAHITI_uvd.bin
shows
Quote:
/lib/firmware/radeon/TAHITI_uvd.bin
then firmware is good
_________________
Defund the FCC.
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 10:35 pm    Post subject: Reply with quote

Quote:
then firmware is good

It does show them.

Question is: is good good or is good bad. :) I mean, should firmware be good there or is it interfering? I guess you say it should be there? thx.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54098
Location: 56N 3W

PostPosted: Sat Apr 05, 2014 10:58 pm    Post subject: Reply with quote

MarcoMarin,

Code:
[   204.002] (EE) Backtrace:
[   204.002] (EE) 0: /usr/bin/X (xorg_backtrace+0x36) [0x585aa6]
[   204.002] (EE) 1: /usr/bin/X (0x400000+0x189749) [0x589749]
[   204.003] (EE) 2: /lib64/libpthread.so.0 (0x7f06b93c3000+0x10b10) [0x7f06b93d3b10]
[   204.003] (EE) 3: /usr/lib64/libpciaccess.so.0 (0x7f06b95e0000+0x51fe) [0x7f06b95e51fe]
[   204.003] (EE) 4: /usr/lib64/xorg/modules/drivers/fglrx_drv.so (amd_xs113_int10_x_inb+0x46) [0x7f06b6a15e26]
...
[   204.012] (EE)
[   204.013] (EE) Segmentation fault at address 0x0
means that something tried to access a region of memory that it doesn't own, so the kernel killed it.
This is a very bad thing.
As fgrlx is a binary blob, it tends to get out of step with Xorg and the kernel. Try the testing fglrx.

You can leave firmware lying around. It will only be used if something loads it. You will remove the something (Radeon) when you switch tu the binary blob.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sat Apr 05, 2014 11:02 pm    Post subject: Reply with quote

I'll try to be exact this time, configure and rebuild the kernel for the closed source driver:
Code:
genkernel --menuconfig all
edit menuconfig to:
Quote:
Device Drivers --->
Graphics support --->
<M> /dev/agpgart (AGP Support) --->
< > AMD Opteron/Athlon64 on-CPU GART support
< > Intel 440LX/BX/GX, I8xx and E7x05 chipset support
< > SiS chipset support
< > VIA chipset support
-*- VGA Arbitration
(2) Maximum number of GPUs
[ ] Laptop Hybrid Graphics - GPU switching support
< > Direct Rendering Manager (XFree86 4.1.0 and higher DRI support)
<*> Support for frame buffer devices --->
--- Support for frame buffer devices
[*] Enable firmware EDID
[ ] Framebuffer foreign endianness support ----
[*] Enable Video Mode Handling Helpers
[*] Enable Tile Blitting Support
*** Frame buffer hardware drivers ***
< > Cirrus Logic support
< > Permedia2 support
< > CyberPro 2000/2010/5000 support
< > Arc Monochrome LCD board support
[ ] Asiliant (Chips) 69000 display support
[ ] IMS Twin Turbo display support
< > VGA 16-color graphics support
< > Userspace VESA VGA graphics support
[*] VESA VGA graphics support
[*] EFI-based Framebuffer Support
< > N411 Apollo/Hecuba devkit support
< > Hercules mono graphics support
< > Epson S1D13XXX framebuffer support
< > nVidia Framebuffer Support
< > nVidia Riva support
< > Intel740 support
< > Intel LE80578 (Vermilion) support
< > Matrox acceleration
< > ATI Radeon display support
< > ATI Rage128 display support
< > ATI Mach64 display support
< > S3 Trio/Virge support
< > S3 Savage support
< > SiS/XGI display support
< > VIA UniChrome (Pro) and Chrome9 display support
< > NeoMagic display support
< > IMG Kyro support
< > 3Dfx Banshee/Voodoo3/Voodoo5 display support
< > 3Dfx Voodoo Graphics (sst1) support
< > VIA VT8623 support
< > Trident/CyberXXX/CyberBlade support
< > ARK 2000PV support
< > Permedia3 support
< > Fujitsu carmine frame buffer support
< > SMSC UFX6000/7000 USB Framebuffer support
< > Displaylink USB Framebuffer support
< > Goldfish Framebuffer
< > Virtual Frame Buffer support (ONLY FOR TESTING!)
< > E-Ink Metronome/8track controller support
< > Fujitsu MB862xx GDC support
< > E-Ink Broadsheet/Epson S1D13521 controller support
< > AUO-K190X EPD controller support
[ ] Simple framebuffer support
[ ] Exynos Video driver support ----
[*] Backlight & LCD device support --->
--- Backlight & LCD device support
<*> Lowlevel LCD controls
< > Platform LCD controls
<*> Lowlevel Backlight controls
< > Generic (aka Sharp Corgi) Backlight Driver
< > Apple Backlight Driver
< > Tabletkiosk Sahara Touch-iT Backlight Driver
< > Backlight Driver for ADP8860/ADP8861/ADP8863 using WLED
< > Backlight Driver for ADP8870 using WLED
< > Backlight Driver for LM3630A
< > Backlight Driver for LM3639
< > Backlight driver for TI LP855X
< > Sanyo LV5207LP Backlight
< > Rohm BD6107 Backlight
Console display driver support --->
-*- VGA text console
[*] Enable Scrollback Buffer in System RAM
(256) Scrollback Buffer Size (in KB)
<*> Framebuffer Console support
[ ] Map the console to the primary display device
[ ] Framebuffer Console Rotation
[*] Bootup logo --->
--- Bootup logo
[ ] Standard black and white Linux logo
[ ] Standard 16-color Linux logo
[*] Standard 224-color Linux logo
then exit saving the configuration. then
Code:
nano /etc/portage/make.conf
edit to
Quote:
VIDEO_CARDS="fglrx" INPUT_DEVICES="evdev"
then
Code:
emerge --unmerge xf86-video-ati
emerge linux-firmware xorg-server ati-drivers xf86-input-evdev
nano /etc/X11/xorg.conf
edit to (should still be)
Quote:

Section "Device"
Identifier "Radeon Card"
Driver "fglrx"
EndSection
Code:
reboot
when backup
Code:
startx

_________________
Defund the FCC.
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 11:06 pm    Post subject: Reply with quote

NeddySeagoon wrote:
means that something tried to access a region of memory that it doesn't own, so the kernel killed it.
This is a very bad thing.
As fgrlx is a binary blob, it tends to get out of step with Xorg and the kernel. Try the testing fglrx.

Couldn't it be my bailing out of it with Ctrl-C? (alt-ctrl-bksp not working)

check out the time jump at the left side. : )

Quote:
You can leave firmware lying around. It will only be used if something loads it. You will remove the something (Radeon) when you switch tu the binary blob.


Great! :D [/quote]
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54098
Location: 56N 3W

PostPosted: Sat Apr 05, 2014 11:08 pm    Post subject: Reply with quote

DONAHUE,

Code:
< > ATI Rage 128
< > ATI Radeon
must both be off too.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54098
Location: 56N 3W

PostPosted: Sat Apr 05, 2014 11:14 pm    Post subject: Reply with quote

MarcoMarin,

Ctl-C gets you a different signal. Thats SIGTERM, not SIGSEG. However, the only people that know what fglrx does with it signals are not telling :)

Now that you pointed out the time jump, it looks like Xorg ran normally until it was killed.
What were you expecting to see on the screen and why ?
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sat Apr 05, 2014 11:15 pm    Post subject: Reply with quote

@neddy
absolutely right
are you seeing them on in my submitted menuconfig? they are off in what i see.

edit:
rephrasing, set off (not listed) by not having < > Direct Rendering Manager (XFree86 4.1.0 and higher DRI support enabled
_________________
Defund the FCC.
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 11:26 pm    Post subject: Reply with quote

@ donahue

unmerging xf86-video-ati sounds promising. at the end it said << !needed sym (obj on the next line) /usr/lib64/libdrm_radeon.so.1

one of the re-loaded mods i mentioned above at xorg.log... I suppose (and hope) it did remove, as emerge says depclean is the safer way to do it.

emerging stuff...
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 11:34 pm    Post subject: Reply with quote

NeddySeagoon wrote:
MarcoMarin,

Ctl-C gets you a different signal. Thats SIGTERM, not SIGSEG. However, the only people that know what fglrx does with it signals are not telling :)

God knows what happens after I Ctrl-C such house of cards :lol:

Quote:
Now that you pointed out the time jump, it looks like Xorg ran normally until it was killed.
What were you expecting to see on the screen and why ?

A xterm square without decorations at the screen and certainly not my monitors out-of-sync message floating around..
using the same .xinitrc when I was trying to fix razorqt. It worked with radeon.

It seems the drivers finds the card, finds the monitor, but it seems confused about 2 different ports (though it may have changed now with the different xorg.conf), a previous post here describes this better.

edit: Though I don't believe anymore it is confused at all, it uses the dvi port and detects the crt monitor as it should (with adapter). I have another box here which uses the same monitor, and it was just a matter of lowering the frequency of the monitor to solve the same symptom. However the probed mode it seems to select seem to have already a low enough freq (60hz) and I did try to set it on another xorg.conf (copied those fields verbatum) so the cause is probably different.
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sat Apr 05, 2014 11:54 pm    Post subject: Reply with quote

Output CRT1 using initial mode 1600x1200
fglrx(0): Modeline "1600x1200"x60.0 162.00 1600 1664 1856 2160 1200 1201 1204 1250 +hsync +vsync (75.0 kHz eP)

oops??

need to set preferred mode?
_________________
Defund the FCC.
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sat Apr 05, 2014 11:58 pm    Post subject: Reply with quote

@ donahue

Nope. Nothing changed. And even my idea of first removing updating xorg (by removing in order, ati-drivers, xorg, and whatever needs it), then remerging xorg (so it gets the new version) and only then merging ati-drivers, would not work either as ati-drivers seem to explicitly require <=xorg-server-1.14.49

and it should work with it anyway...

well, I'll be out of town tomorrow and probb monday, checking out how they've destroyed (they call it a "rebuild") the most famous brazilian stadium to accommodate the world cup. I won't be able to continue this charade, tell me if there are anymore ideas after I return I'll give it one last shot : ) thanks all.
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sun Apr 06, 2014 12:06 am    Post subject: Reply with quote

DONAHUE wrote:
Output CRT1 using initial mode 1600x1200
fglrx(0): Modeline "1600x1200"x60.0 162.00 1600 1664 1856 2160 1200 1201 1204 1250 +hsync +vsync (75.0 kHz eP)

oops??

need to set preferred mode?

I tried setting the preferred mode remember? But the port confusion I described had the thing reading the field for the DFP1 only to disconnect DFP1 and claim to have CRT1 connected and say "preferredMode" not used.

That's why I asked you how to direct CRT1 to the Monitor0 section so it could try that in the first place before DFP1 used it. And I'd still be happy to try that if you can tell me how... :D
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sun Apr 06, 2014 12:17 am    Post subject: Reply with quote

ati-drivers and kernel and xorg-server and evdev must be compatible

The following are compatible here:
ati-drivers-14.3_beta ~amd64
xorg-server-1.15.0 amd64
xf86-input-evdev-2.8.2 amd64
gentoo-sources ~amd64

Code:
echo "=x11-drivers/ati-drivers-14.3_beta ~amd64" >> /etc/portage/package.keywords
emerge -av xorg-server ati-drivers xf86-input-evdev


watch for compatibility complaints early in each package's emerge
_________________
Defund the FCC.


Last edited by DONAHUE on Sun Apr 06, 2014 12:25 am; edited 1 time in total
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sun Apr 06, 2014 12:24 am    Post subject: Reply with quote

I suppose the requirement for ati-driver which prevents me from updating xorg to 1.15 is ensuring compatibility?

But how do I check that?

My idea was exactly for this purpose, get the newer xorg, maybe the one I got when installing the desktop environment was not ideal, then hope ati-driver will use that or even better get their newest version... maybe my ati-driver had to be emerged as an older version because of this? It is 13.12:1 here! 8O
Back to top
View user's profile Send private message
DONAHUE
Watchman
Watchman


Joined: 09 Dec 2006
Posts: 7651
Location: Goose Creek SC

PostPosted: Sun Apr 06, 2014 12:29 am    Post subject: Reply with quote

suggest up ati-drivers as suggested in edit above.
_________________
Defund the FCC.
Back to top
View user's profile Send private message
MarcoMarin
Apprentice
Apprentice


Joined: 12 Mar 2014
Posts: 151

PostPosted: Sun Apr 06, 2014 12:31 am    Post subject: Reply with quote

Quote:


Code:
echo "=x11-drivers/ati-drivers-14.3_beta ~amd64" >> /etc/portage/package.keywords
emerge -av xorg-server ati-drivers xf86-input-evdev


watch for compatibility complaints early in each package's emerge


wow, I didn't have that file.. only "accept_keywords" :o

Wouldn't that break portage with some kind of circular dependency it can't resolve?

Ohh... apparently not, sounds promising!!!! :!: 11 :!: I'll finally have the newest version of the driver![/quote]
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page Previous  1, 2, 3, 4, 5  Next
Page 4 of 5

 
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