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

flamelab · 2010-06-23 20:46:41

@all: Is somebody gonna try to edit the powersaving section in the ATi wiki please ?

Perry3D · 2010-06-23 21:00:52

flamelab wrote:

@all: Is somebody gonna try to edit the powersaving section in the ATi wiki please ?

There is no sysfs interface to control the power managment with the stock kernel. Did some try appending radeon.dynpm=1 to the kernel parameters?

gtklocker · 2010-06-23 21:46:32

I updated the powersaving part in the wiki. Tell me if something is wrong.

Perry3D · 2010-06-23 21:58:28

gtklocker wrote:

I updated the powersaving part in the wiki. Tell me if something is wrong.

It was just a speculation that radeon.dynpm=1 works. Someone has to confirm this

lowie82ph · 2010-06-24 01:58:27

Perry3D wrote:

gtklocker wrote:
I updated the powersaving part in the wiki. Tell me if something is wrong.
It was just a speculation that radeon.dynpm=1 works. Someone has to confirm this

lowie ~  $  dmesg | grep radeon
Command line: root=/dev/disk/by-uuid/44dab59a-4768-4439-b7af-b2a7e7c816f2 resume=/dev/sda2 ro radeon.dynpm=1
Kernel command line: root=/dev/disk/by-uuid/44dab59a-4768-4439-b7af-b2a7e7c816f2 resume=/dev/sda2 ro radeon.dynpm=1
[drm] radeon defaulting to kernel modesetting.
[drm] radeon kernel modesetting enabled.
radeon 0000:01:05.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
radeon 0000:01:05.0: setting latency timer to 64
[drm] radeon: dynamic power management enabled
[drm] radeon: power management initialized
radeon 0000:01:05.0: VRAM: 128M 0xC0000000 - 0xC7FFFFFF (128M used)
radeon 0000:01:05.0: GTT: 512M 0xA0000000 - 0xBFFFFFFF
[drm] radeon: 128M of VRAM memory ready
[drm] radeon: 512M of GTT memory ready.
[drm] radeon: irq initialized.
platform radeon_cp.0: firmware: requesting radeon/RS780_pfp.bin
platform radeon_cp.0: firmware: requesting radeon/RS780_me.bin
platform radeon_cp.0: firmware: requesting radeon/R600_rlc.bin
[drm] radeon: ib pool ready.
fb0: radeondrmfb frame buffer device
[drm] Initialized radeon 2.3.0 20080528 for 0000:01:05.0 on minor 0

is it working already?

gtklocker · 2010-06-24 04:35:54

It seems so lowie82ph.

Laserpithium · 2010-06-24 11:44:50

Well, so I decided to start tests since the beginning to solve my upgrading issue.

First issue: with my RV730 (Mobility HD4650) I can't manage to start X in KMS mode. Everything OK in UMS (even 3D)
I tried both early and late start, following word by word the wiki. I tried too with intel_agp in the module line: idem.
I tried to pass the option "pci=nomsi" to the kernel at start-up: same result.
The problem is always the same: KMS is well enable during boot but Impossible to start X: black screen and system freez (impossible to switch the tty).
For information, I am not using any xorg.conf.

xorg-server:1.7.6-3
xf86-video-ati: 6.12.192-1
Kernel 2.6.34, 64 bits from core repo.
Mesa 7.7.1-1
ati-dri and libgl 7.7-1

My mkinitcpio.conf:

MODULES=""
HOOKS="base udev autodetect pata scsi sata filesystems keymap"

My rc.conf

MOD_AUTOLOAD="yes"
#MOD_BLACKLIST=() #deprecated
MODULES=(!pcspkr acpi_cpufreq dm_mod intel_agp radeon)
DAEMONS=(syslog-ng !network netfs @crond acpid cpufreq @alsa hal iptables ip6tables @privoxy @tor @wicd kdm @ntpd)

And the parameters to the kernel:

kernel /boot/vmlinuz26 root=/dev/sdb2 ro quiet

Last edited by Laserpithium (2010-06-24 11:46:30)

Perry3D · 2010-06-24 11:52:27

Did you look at Xorg.0.log? If you are not able to look to Xorg.0.log because of the system freeze, restart in UMS mode and look at Xorg.1.log.

Laserpithium · 2010-06-24 12:16:09

Perry3D wrote:

Did you look at Xorg.0.log? If you are not able to look to Xorg.0.log because of the system freeze, restart in UMS mode and look at Xorg.1.log.

Hi,

I just did it.
But Xorg.1.log contains information from december 2009... It doesn't contain any information from the previous (missed) start.
Some configuration mistake somewhere ?

I did some search on the forum, and it appears that I have no .xinitrc file in my home directory. It's normal ?

Last edited by Laserpithium (2010-06-24 13:09:01)

gabriel9 · 2010-06-24 23:14:23

Can someone tell me is it posible to play quake live with Free drivers?
Ehen i try to play it it brings me this:
http://www.freeimagehosting.net/image.p … 641c3e.png

Find the solution: http://www.quakelive.com/forum/showpost … stcount=16

Last edited by gabriel9 (2010-06-25 00:42:36)

del_diablo · 2010-06-25 17:03:20

Interisting, I think i got some "lag" with the open drivers.
Putting on powersave seems to make it a tad more noticeable
Mobility 3470

squiddy · 2010-07-02 11:11:23

I'm sorry if this thread is not the right place to ask. Using radeon 3200 HD (RS780M/RS780MN). from what i've read here, the proper driver to get the power management working is to use xf86-video-ati then add the line radeon.dynpm=1 to kernel options and i will get a cooler laptop?

CMIIW

Last edited by squiddy (2010-07-02 11:12:12)

BigLouis87 · 2010-07-12 10:01:52

any news from radeon driver/team?
repo's packages and last posts are quite old for a bleeding edge driver

Have a nice day

Perry3D · 2010-07-12 11:26:42

BigLouis87 wrote:

any news from radeon driver/team?
repo's packages and last posts are quite old for a bleeding edge driver
Have a nice day

Hi and sorry.
i was very busy the last weeks. I did the last test for my master (Diplom) today. Now i have some free time to care of the repo. But i think there were no exciting new features, only some fixes.

BigLouis87 · 2010-07-12 12:00:06

I was only wondering if there was updates, I have some little problems with actual version...

Break a leg!!!! Don't care for the repo... master is more important!

mixi · 2010-07-13 07:39:16

BigLouis87 wrote:

I was only wondering if there was updates, I have some little problems with actual version...

If you want the packages being rebuilt more often you could always try the nightly repo (rebuilt every night and it was working last time I tried).

Perry3D · 2010-07-13 12:29:57

mixi wrote:

If you want the packages being rebuilt more often you could always try the nightly repo (rebuilt every night and it was working last time I tried).

But don't use both repo at the same time. There are conflicting packages: libdrm-git for example. And there are some differences: the nightly repo is the split version of mesa while i prefer the mesa-full package in my repo.

By the way, i did a full update yesterday.

lman · 2010-07-17 10:24:37

Hi, what happened to the kernel26-git package? I thought it was in the radean repository too

handy · 2010-07-17 14:50:52

I've been away since the 23rd of June, returned today & upgraded this system (over 200Mb), the upgrade has caused me to loose my network. When I reboot using the current stable kernel all is fine.

Anyone else had similar problems using
- kernel26-drm-radeon-testing ?

roughL · 2010-07-17 15:08:44

handy wrote:

I've been away since the 23rd of June, returned today & upgraded this system (over 200Mb), the upgrade has caused me to loose my network. When I reboot using the current stable kernel all is fine.
Anyone else had similar problems using
- kernel26-drm-radeon-testing ?

I have exactly the same problem here with kernel26-drm-radeon-testing 20100712-1 , network does not start. But by typing

sudo dhcpcd eth0

after boot I get the network working.

handy · 2010-07-18 03:12:34

roughL wrote:

handy wrote:
I've been away since the 23rd of June, returned today & upgraded this system (over 200Mb), the upgrade has caused me to loose my network. When I reboot using the current stable kernel all is fine.
Anyone else had similar problems using
- kernel26-drm-radeon-testing ?
I have exactly the same problem here with kernel26-drm-radeon-testing 20100712-1 , network does not start. But by typing
sudo dhcpcd eth0
after boot I get the network working.

Thanks for your reply.

I'll reboot into ...radeon-testing & give it a shot.

I'm sure this bug won't last too long.

handy · 2010-07-22 00:17:47

I just upgraded to kernel26-drm-radeon-testing 20100721-1, & I'm still experiencing the same network bug.

Fortunately the easy fix of just entering:

sudo dhcpcd eth0

Still works. I made an alias for it. I've forgotten which file in /etc to enter it in to run it automatically.

On another note, the OpenGL speed of this testing kernel is dreadful on my machine:

OpenGL vendor string: Advanced Micro Devices, Inc.
OpenGL renderer string: Mesa DRI R600 (RV630 9583) 20090101 TCL DRI2
OpenGL version string: 2.0 Mesa 7.9-devel
OpenGL shading language version string: 1.10
OpenGL extensions:
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.

304 frames in 5.0 seconds = 60.778 FPS
302 frames in 5.0 seconds = 60.324 FPS
302 frames in 5.0 seconds = 60.324 FPS
302 frames in 5.0 seconds = 60.324 FPS
302 frames in 5.0 seconds = 60.324 FPS
302 frames in 5.0 seconds = 60.324 FPS

Last edited by handy (2010-07-22 00:39:01)

Vi0L0 · 2010-07-22 01:28:07

handy wrote:

I just upgraded to kernel26-drm-radeon-testing 20100721-1, & I'm still experiencing the same network bug.
Fortunately the easy fix of just entering:
sudo dhcpcd eth0
Still works. I made an alias for it. I've forgotten which file in /etc to enter it in to run it automatically.

Umm... do you have network in DAEMONS array in /etc/rc.conf?
Plus dont forget about gateway, ie.:

gateway="default gw 192.168.1.1"
ROUTES=(gateway)

and ofcourse eth0, ie:

eth0="eth0 192.168.1.71 netmask 255.255.255.0 broadcast 192.168.1.255"

Maybe /etc/resolv.conf is causing some problems? In most cases gateway is a resolver, ie:

nameserver 192.168.1.1

You definitely have module for your eth card, so it must be something with a config.
You can also try to add your eth card module to MODULES array in rc.conf.

Anyhow - for dhcp on eth0 just edit /etc/rc.conf :

lo="lo 127.0.0.1"
eth0="dhcp"
INTERFACES=(lo eth0)

(well maybe you dont 'need' lo, but its good to have it )

handy wrote:

On another note, the OpenGL speed of this testing kernel is dreadful on my machine:
OpenGL vendor string: Advanced Micro Devices, Inc.
OpenGL renderer string: Mesa DRI R600 (RV630 9583) 20090101 TCL DRI2
OpenGL version string: 2.0 Mesa 7.9-devel
OpenGL shading language version string: 1.10
OpenGL extensions:
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.
304 frames in 5.0 seconds = 60.778 FPS
302 frames in 5.0 seconds = 60.324 FPS
302 frames in 5.0 seconds = 60.324 FPS
302 frames in 5.0 seconds = 60.324 FPS
302 frames in 5.0 seconds = 60.324 FPS
302 frames in 5.0 seconds = 60.324 FPS

Looks like vsync to vblank works good :-)

BTW Is there finally a switch to disable it?

Last edited by Vi0L0 (2010-07-22 02:02:20)

handy · 2010-07-22 03:42:51

Vi0L0 wrote:

handy wrote:
I just upgraded to kernel26-drm-radeon-testing 20100721-1, & I'm still experiencing the same network bug.
Fortunately the easy fix of just entering:
sudo dhcpcd eth0
Still works. I made an alias for it. I've forgotten which file in /etc to enter it in to run it automatically.
Umm... do you have network in DAEMONS array in /etc/rc.conf?

yes

Vi0L0 wrote:

Plus dont forget about gateway, ie.:
gateway="default gw 192.168.1.1"
ROUTES=(gateway)

I've never used those settings but I have turned them on out of interest.

Vi0L0 wrote:

and ofcourse eth0, ie:

eth0="eth0 192.168.1.71 netmask 255.255.255.0 broadcast 192.168.1.255"

I always use those.

Vi0L0 wrote:

Maybe /etc/resolv.conf is causing some problems? In most cases gateway is a resolver, ie:
nameserver 192.168.1.1

I use resolve.conf.head & point to my IPCop box with nameserver 192.168.1.1

Vi0L0 wrote:

You definitely have module for your eth card, so it must be something with a config.
You can also try to add your eth card module to MODULES array in rc.conf.

The interesting thing is, everything was working fine until I installed the current & previous versions of the kernel26-radeon-testing.

When I boot with the current [stable] Arch kernel I have no problems at all? So it seems to me that this is a problem with the radeon-testing kernel.

sky2 is already in my MODULES array.

Vi0L0 wrote:

Anyhow - for dhcp on eth0 just edit /etc/rc.conf :
lo="lo 127.0.0.1"
eth0="dhcp"
INTERFACES=(lo eth0)
(well maybe you dont 'need' lo, but its good to have it )

Yes, I always run those settings also.

Last edited by handy (2010-07-22 03:51:58)

ijanos · 2010-07-22 07:38:05

handy wrote:

On another note, the OpenGL speed of this testing kernel is dreadful on my machine:
OpenGL vendor string: Advanced Micro Devices, Inc.
OpenGL renderer string: Mesa DRI R600 (RV630 9583) 20090101 TCL DRI2
OpenGL version string: 2.0 Mesa 7.9-devel
OpenGL shading language version string: 1.10
OpenGL extensions:
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.
304 frames in 5.0 seconds = 60.778 FPS

glxgears is NOT a benchmark and these value are perfectly fine.

Arch Linux

#1451 2010-06-23 20:46:41

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

#1452 2010-06-23 21:00:52

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

#1453 2010-06-23 21:46:32

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

#1454 2010-06-23 21:58:28

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

#1455 2010-06-24 01:58:27

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

#1456 2010-06-24 04:35:54

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

#1457 2010-06-24 11:44:50

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

#1458 2010-06-24 11:52:27

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

#1459 2010-06-24 12:16:09

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

#1460 2010-06-24 23:14:23

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

#1461 2010-06-25 17:03:20

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

#1462 2010-07-02 11:11:23

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

#1463 2010-07-12 10:01:52

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

#1464 2010-07-12 11:26:42

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

#1465 2010-07-12 12:00:06

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

#1466 2010-07-13 07:39:16

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

#1467 2010-07-13 12:29:57

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

#1468 2010-07-17 10:24:37

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

#1469 2010-07-17 14:50:52

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

#1470 2010-07-17 15:08:44

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

#1471 2010-07-18 03:12:34

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

#1472 2010-07-22 00:17:47

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

#1473 2010-07-22 01:28:07

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

#1474 2010-07-22 03:42:51

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

#1475 2010-07-22 07:38:05

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

Board footer