Discussion: the ati/radeon open source drivers & radeon repository

bwat47 · 2010-05-09 11:31:14

Perry3D wrote:

@bwat47: It's not compatible with xorg 1.8. And PM should be enabled by default. With the latest kernel update there were some changes: http://www.phoronix.com/scan.php?page=n … &px=ODIyOA
@handy: Did you try to switch PM to performance? echo high > /sys/class/drm/card0/device/power_profile

Guess that explians why my latest arch install went awry, I tried to use this repo and xorg 1.8 and just got horrible artifacts right before gdm should come up.

handy · 2010-05-09 12:46:53

Sorry bwat47, I should've kept my mouth shut.

[Edit:] I don't use Compiz or the like, if that makes any difference?

Last edited by handy (2010-05-10 02:05:16)

megawebmaster · 2010-05-09 19:35:47

Lol? I'm using Xorg 1.8 and [radeon] repo and all is working fine. Excluding xf86-video-ati-git, because of ABI incompatibility. BTW - I was really surprised when I turn on my laptop and the screen was blank and KMS off - I'm using Compiz, so I'm loading it on Gnome start - there was white screen without everything

Crozepp · 2010-05-09 20:55:27

With the last kernel update from core (2.6.33.3-2) I get a kernel panic if I have "radeon.modeset=1" in grub line, with radeon-initrd from AUR installed to get early KMS start (as is explained in the wiki and the first post of this thread). If I delete "radeon.modeset=1" I get early KMS start without boot problems. Has anyone else noticed?

Megawebmaster => I tried Xorg 1.8 with all the testing respective updates and I have problems too. With fusion-icon & compiz my screen freezes and input devices stops working (no blank screen, however).

esdaniel · 2010-05-09 21:18:32

Crozepp wrote:

With the last kernel update from core (2.6.33.3-2) I get a kernel panic if I have "radeon.modeset=1"

See here: http://bbs.archlinux.org/viewtopic.php? … 14#p756214

bwat47 · 2010-05-09 23:17:33

Ok so I got the drm-radeon-testing kernel working. So since dynpm has been removed how do I enable it? If I am reading this right its not enabled:
[brandon@arch ~]$ dmesg | grep drm
[drm] Initialized drm 1.1.0 20060810
[drm] radeon defaulting to kernel modesetting.
[drm] radeon kernel modesetting enabled.
[drm] initializing kernel modesetting (RV630 0x1002:0x9581).
[drm] register mmio base: 0xF3000000
[drm] register mmio size: 65536
[drm] Clocks initialized !
[drm] Detected VRAM RAM=512M, BAR=256M
[drm] RAM width 128bits DDR
[drm] radeon: 512M of VRAM memory ready
[drm] radeon: 512M of GTT memory ready.
[drm] radeon: using MSI.
[drm] radeon: irq initialized.
[drm] GART: num cpu pages 131072, num gpu pages 131072
[drm] Loading RV630 Microcode
[drm] ring test succeeded in 1 usecs
[drm] radeon: ib pool ready.
[drm] ib test succeeded in 0 usecs
[drm] Enabling audio support
[drm] Unknown TV standard; defaulting to NTSC
[drm] Radeon Display Connectors
[drm] Connector 0:
[drm] LVDS
[drm] DDC: 0xac0 0xac0 0xac4 0xac4 0xac8 0xac8 0xacc 0xacc
[drm] Encoders:
[drm] LCD1: INTERNAL_LVTM1
[drm] Connector 1:
[drm] VGA
[drm] DDC: 0x7e40 0x7e40 0x7e44 0x7e44 0x7e48 0x7e48 0x7e4c 0x7e4c
[drm] Encoders:
[drm] CRT1: INTERNAL_KLDSCP_DAC1
[drm] Connector 2:
[drm] HDMI-A
[drm] HPD1
[drm] DDC: 0x7e50 0x7e50 0x7e54 0x7e54 0x7e58 0x7e58 0x7e5c 0x7e5c
[drm] Encoders:
[drm] DFP1: INTERNAL_KLDSCP_TMDS1
[drm] Internal thermal controller without fan control
[drm] radeon: power management initialized
[drm] fb mappable at 0xD0141000
[drm] vram apper at 0xD0000000
[drm] size 4096000
[drm] fb depth is 24
[drm] pitch is 5120
fb0: radeondrmfb frame buffer device
[drm] Initialized radeon 2.3.0 20080528 for 0000:01:00.0 on minor 0

insanemal · 2010-05-10 03:26:14

Crozepp wrote:

With the last kernel update from core (2.6.33.3-2) I get a kernel panic if I have "radeon.modeset=1" in grub line, with radeon-initrd from AUR installed to get early KMS start (as is explained in the wiki and the first post of this thread). If I delete "radeon.modeset=1" I get early KMS start without boot problems. Has anyone else noticed?
Megawebmaster => I tried Xorg 1.8 with all the testing respective updates and I have problems too. With fusion-icon & compiz my screen freezes and input devices stops working (no blank screen, however).

This sounds like my issues.. Have you found any solution? or did you downgrade also?

Crozepp · 2010-05-10 05:37:36

insanemal wrote:

This sounds like my issues.. Have you found any solution? or did you downgrade also?

Sorry insanemal, I only did a downgrade. Too lazy or inexperienced user to have [testing] active all the time . This time I have been very curious about xorg 1.8, and my cat almost died.
I have a partition apart with KDE (99% of the time I use gnome) and Kwin doesn't have any troubles.
So this can be an incompability with compiz code and mesa or xorg or evdev or I don't know. Or merely can be a bad configuration.

Perry3D · 2010-05-10 08:37:35

bwat47 wrote:

Ok so I got the drm-radeon-testing kernel working. So since dynpm has been removed how do I enable it? If I am reading this right its not enabled:
...
[drm] Internal thermal controller without fan control
[drm] radeon: power management initialized
...

It is enabled. After the last kernel update even with voltage control . But they removed the debug messages.

I read this thread on phoronix: http://www.phoronix.com/forums/showthread.php?t=23686
There is a way to enable it again, but i don't know how. Maybe it depends on the kernel config?
I know that you can switch the profiles with /sys/class/drm/card0/device/power_profile (high, low, auto or default).

megawebmaster · 2010-05-10 14:02:50

Hmmm, after latest mesa upgrade my problems with compiz just disappeard. Any everything is now fine. Sometimes I've got little freezes i.e. when trying to use "Open file" window, but in this time everything else is working, so I think it's problem with Gnome. I've noticed also that sometimes one of a buttons (Close, Minimize, Maximize) have got something weird below them, but minimizing and then maximizing window fixes this.

Perry3D · 2010-05-10 14:05:44

How do you start compiz? With KMS/DRI2 you can disable indirect rendering.

bwat47 · 2010-05-10 15:19:29

I tried enabling dynpm via echo dynpm > /sys/class/drm/card0/device/power_method. The screen flickered once so i assume it activated, problem it it works for 1 min then my pc gets insanely laggy.

I am not a fan of the profile approach at all cause I use my laptop plugged in most of the time (and the profile mode runs your clocks max all the time unless you are on battery), my issue is in linux my video card gets really, really, really, hot using the oss drivers.

So right now I am just using profile mode set to low until dynpm gets fixed.

Last edited by bwat47 (2010-05-10 15:20:57)

megawebmaster · 2010-05-11 18:42:43

I'm starting it via required components in GConf (http://wiki.archlinux.org/index.php/Com … on.22.29_2) Now I've noticed that sometimes some letters in some font turns into a square, but only in that size - another sizes (i.e. while I'm changing size in Firefox) are good. I remember that something like this sometimes happend some time ago, when I was switching from Xorg 1.6 to 1.7 so I think it's driver related. Maybe git version will be better, I will try this from AUR.

EDIT: Nope - git version driver isn't better than the one from testing repo.

Last edited by megawebmaster (2010-05-12 08:27:08)

bwat47 · 2010-05-11 21:55:53

Im just waitin patiently for a new kernel in this repo, hoping I can get dynamic power management working in a newer one, this laptop gets so damn hot in linux.

megawebmaster · 2010-05-12 07:20:33

Profile mode and low method isn't working for you?

EDIT: Hmm, I've put profile mode and low method but my system is hotter than before upgrading kernel. How can I check actual clocks? And I've got only "power management initialized" in my dmesg...

Last edited by megawebmaster (2010-05-12 08:29:01)

bwat47 · 2010-05-12 16:02:41

Profile mode set to low keeps my temps 5+ degrees C cooler but it makes scrolling in chrome and ff laggy. So far the best power management ive had was the rc6 kernel with dynpm enabled. It only brought temps down a couple degrees but there was no speed issues or flickering.

Last edited by bwat47 (2010-05-12 16:03:30)

megawebmaster · 2010-05-12 16:24:27

Yeah, I can agree. I couldn't believe that in this version there is voltage managing, because my temp raised up... And for now - dynpm is unusable because of flickering, which didn't exist in previous kernel version.

bwat47 · 2010-05-12 16:31:04

megawebmaster wrote:

Yeah, I can agree. I couldn't believe that in this version there is voltage managing, because my temp raised up... And for now - dynpm is unusable because of flickering, which didn't exist in previous kernel version.

My laptops temp goes up amazingly fast, I just set up conky and it has the temp displayed all the time, and I booted my laptop up. Within 1 minute it went from 39 degrees to 50. Being on for more than 10 mins it stays around 60-64 which is noticeably hot when I touch my laptop.

megawebmaster · 2010-05-12 17:54:30

Hmmm, that's the same I've got. It is not really noticeable (I'm putting my laptop on a desk everytime), but it is not normal. I have to check if normal kernel has got the same temps as drm-radeon-testing.

EDIT: Nobody answered, so I'll ask again: how to check used frequencies with the newest drm-radeon-testing?

Last edited by megawebmaster (2010-05-13 20:55:45)

bwat47 · 2010-05-12 21:06:01

Id still rather burn my lap off then deal with fglrx ever again

Perry3D · 2010-05-14 16:09:53

Hi guys, i got a little present for you. I added lib32-libdrm-git and lib32-mesa-full to the x86_64 repo and tested it with googleearth.
It works fine (before it didn't). Currently i am installing bin32-wine .

/edit: I played Half-Life 2 for a few minutes. There were some graphic glitches in dx8.1 mode . But it was playable.

@megawebmaster: ok, i revert my statement. Voltage control is only implemented for chips < r600. For my laptop with an x300 card it is working fine.
From the phoronix forum:

agd5f wrote:

A couple quick notes regarding the current pm state:
- Voltage drop is already implemented for r1xx-r5xx chips in the new pm code
- Some additional asic features (like dynamic sclk and dynamic voltage where the asic scales the clocks/voltage rather than the driver doing it) are also already implemented in the driver for r1xx-r5xx asics. Whether or not these get enabled depends on the flags in the power state entry.
- Most r6xx/r7xx desktop cards have power saving modes

Last edited by Perry3D (2010-05-14 17:50:06)

Cdh · 2010-05-15 08:30:23

Would you mind to post a screenshot of half life 2? "Some graphic glitches" doesn't say much. I'm interested in how well it is rendering (all effects working?).
With catalyst I also had "some" graphic glitches also but that was in dx9 mode: http://stashbox.org/896104/ep2_outland_010000.jpg
Also what does playable mean? Good framerate with all/some/no effects enabled?

BigLouis87 · 2010-05-15 11:11:23

Hi! I have just installed Arch and I followed the guide in the first post of this thread to get the latest experimental radeon driver; so I installed the kernel26-drm-radeon-testing and all the mentioned packets from the radeon repo... but now Xorg doesn't start anymore! Now I can only start arch with the vesa driver.
This is an interesting part of my Xorg log:

[   174.977] (II) Module extmod: vendor="X.Org Foundation"
[   174.977]     compiled for 1.8.0.902, module version = 1.0.0
[   174.977]     Module class: X.Org Server Extension
[   174.977]     ABI class: X.Org Server Extension, version 3.0
[   174.977] (II) Loading extension MIT-SCREEN-SAVER
[   174.977] (II) Loading extension XFree86-VidModeExtension
[   174.977] (II) Loading extension XFree86-DGA
[   174.977] (II) Loading extension DPMS
[   174.977] (II) Loading extension XVideo
[   174.977] (II) Loading extension XVideo-MotionCompensation
[   174.977] (II) Loading extension X-Resource
[   174.977] (II) LoadModule: "radeon"
[   174.977] (II) Loading /usr/lib/xorg/modules/drivers/radeon_drv.so
[   174.978] (II) Module radeon: vendor="X.Org Foundation"
[   174.978]     compiled for 1.7.6, module version = 6.13.99
[   174.978]     Module class: X.Org Video Driver
[   174.978]     ABI class: X.Org Video Driver, version 6.0
[   174.978] (EE) module ABI major version (6) doesn't match the server's version (7)
[   174.978] (II) UnloadModule: "radeon"
[   174.978] (II) Unloading /usr/lib/xorg/modules/drivers/radeon_drv.so
[   174.978] (EE) Failed to load module "radeon" (module requirement mismatch, 0)
[   174.978] (EE) No drivers available.
[   174.978] 
Fatal server error:
[   174.978] no screens found

It seems that arch can't load radeon driver because it is compiled for an older version of Xorg. What can I do?
Thank you

PS: I want to salute handy who helped me in the ubuntu forum and brought me to the Arch side of Linux!

Perry3D · 2010-05-15 11:17:01

Are you using the testing repo?

What is the output of pacman -Qs xorg-server?

i686 or x86_64?

Last edited by Perry3D (2010-05-15 11:18:14)

BigLouis87 · 2010-05-15 11:43:14

Perry3D wrote:

Are you using the testing repo?
What is the output of pacman -Qs xorg-server?
i686 or x86_64?

local/xorg-server 1.8.0.902-1 (xorg)
    X.Org X servers
local/xorg-server-utils 7.5-3 (xorg)
    X.Org utilities required by xorg-server

sorry... I'm i686 with a mobility radeon X1600
yes, I've enabled the testing repo... is it a problem?

thanks

Arch Linux

#1276 2010-05-09 11:31:14

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1277 2010-05-09 12:46:53

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1278 2010-05-09 19:35:47

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1279 2010-05-09 20:55:27

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1280 2010-05-09 21:18:32

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1281 2010-05-09 23:17:33

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1282 2010-05-10 03:26:14

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1283 2010-05-10 05:37:36

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1284 2010-05-10 08:37:35

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1285 2010-05-10 14:02:50

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1286 2010-05-10 14:05:44

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1287 2010-05-10 15:19:29

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1288 2010-05-11 18:42:43

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1289 2010-05-11 21:55:53

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1290 2010-05-12 07:20:33

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1291 2010-05-12 16:02:41

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1292 2010-05-12 16:24:27

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1293 2010-05-12 16:31:04

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1294 2010-05-12 17:54:30

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1295 2010-05-12 21:06:01

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1296 2010-05-14 16:09:53

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1297 2010-05-15 08:30:23

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1298 2010-05-15 11:11:23

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1299 2010-05-15 11:17:01

Re: Discussion: the ati/radeon open source drivers & radeon repository

#1300 2010-05-15 11:43:14

Re: Discussion: the ati/radeon open source drivers & radeon repository

Board footer