The AMD/ATI Bar & Grill

ChemBro · 2010-05-28 21:10:04

*vitali* wrote:

I was only able to enable composting in kwin after I enabled Direct2D.
(sudo aticonfig --set-pcs-str=DDX,Direct2DAccel,TRUE; sudo reboot).

Worked. Thanks.

HoboJ · 2010-05-28 23:26:30

Hmm seems catalyst 10.5 makes my fps worse. With 10.4 I was able to hold a steady 60 fps in savage: the battle for newerth, now I get frequent drops down to 50fps with 10.5. Not sure If I'll downgrade yet though.

Cdh · 2010-05-28 23:32:09

Vi0L0 wrote:

- resolved issues pertaining to Hybrid CrossFire, kernel panics on hibernation, and suspend-

Do I have anything special to do for hibernation?
Right now I click on the hibernation button in the xfce4-session-logout but it doesn't work. The screen is getting black for a second or so and then I am back to my desktop...

Also I noticed with compositing enabled the vaapi:gl output is garbage while without compositing it works well.

timong · 2010-05-29 11:18:22

Vi0L0 wrote:

I have prepared wine 1.2 rc1 packages.
i686: http://dl.dropbox.com/u/4321633/wine-1. … pkg.tar.xz
x86_64: http://dl.dropbox.com/u/4321633/bin32-w … pkg.tar.xz
Feel free to test it. I decided not to add this pkgs to repos cuz im not sure will they work well...
Please forgive bad versioning (is 1.2-1, should be 1.2-rc1-1) - i have noticed it after half hour of compilation...
And PKGBUILDs i used:
i686 - here
x86_64 - here

rc2 is out any update?

Radioactiveman · 2010-05-29 13:25:12

Radioactiveman wrote:

If someone has 2D-acceleration enabled, do you have also strange rendering of http://hardware-infos.com/ with Firefox? Bugreport

Could someone test it? Thanks for answer.

Vi0L0 · 2010-05-29 14:39:40

timong wrote:

rc2 is out any update?

ok, so wine 1.2 rc2:
i686: heres my package, and PKGBUILD i used
x86_64 version (bin32-wine) is already in [arch-games] repo, pacman.conf section:

[arch-games]
Server = http://repo.archlinux-gaming.org/x86_64

Cdh · 2010-05-29 23:51:03

Radioactiveman wrote:

Radioactiveman wrote:
If someone has 2D-acceleration enabled, do you have also strange rendering of http://hardware-infos.com/ with Firefox? Bugreport
Could someone test it? Thanks for answer.

For me it seems to be ok: http://omploader.org/vNGZ5ag
I have compositing disabled. Xorg 1.8 with catalyst 10.5 and I think d2d enabled (run the command above and restarted).

Edit: Hibernation problem:

PM: Marking nosave pages: 0000000000002000 - 0000000000010000
PM: Marking nosave pages: 000000000009b000 - 0000000000100000
PM: Basic memory bitmaps created
PM: Syncing filesystems ... done.
Freezing user space processes ... (elapsed 0.01 seconds) done.
Freezing remaining freezable tasks ... (elapsed 0.01 seconds) done.
PM: Preallocating image memory... done (allocated 126075 pages)
PM: Allocated 504300 kbytes in 0.05 seconds (10086.00 MB/s)
Suspending console(s) (use no_console_suspend to debug)
ACPI handle has no context!
sd 2:0:0:0: [sda] Synchronizing SCSI cache
parport_pc 00:09: disabled
serial 00:08: disabled
HDA Intel 0000:02:00.1: PCI INT B disabled
ACPI handle has no context!
[fglrx] IRQ 27 Disabled
[fglrx] Preparing suspend fglrx in kernel.
ACPI handle has no context!
ata2: port disabled. ignoring.
HDA Intel 0000:00:05.0: PCI INT B disabled
ACPI handle has no context!
sata_nv 0000:00:08.0: PCI INT A disabled
[fglrx] Suspending fglrx in kernel completed.
[fglrx] Power down the ASIC .
PM: freeze of devices complete after 2533.043 msecs
PM: late freeze of devices complete after 0.230 msecs
ACPI: Preparing to enter system sleep state S4
PM: Saving platform NVS memory
Disabling non-boot CPUs ...
CPU 1 is now offline
SMP alternatives: switching to UP code
PM: Creating hibernation image:
PM: Need to copy 134556 pages
PM: Normal pages needed: 133825 + 1024, available pages: 126689
PM: Not enough free memory
PM: Error -12 creating hibernation image
Enabling non-boot CPUs ...
SMP alternatives: switching to SMP code
Booting Node 0 Processor 1 APIC 0x1
Initializing CPU#1
CPU1 is up
ACPI: Waking up from system sleep state S4
PM: early recover of devices complete after 0.065 msecs
pci 0000:00:04.0: setting latency timer to 64
pata_amd 0000:00:06.0: restoring config space at offset 0x1 (was 0xb00001, writing 0xb00005)
pata_amd 0000:00:06.0: setting latency timer to 64
sata_nv 0000:00:08.0: restoring config space at offset 0x1 (was 0xb00003, writing 0xb00007)
sata_nv 0000:00:08.0: PCI INT A -> Link[APSI] -> GSI 23 (level, low) -> IRQ 23
sata_nv 0000:00:08.0: setting latency timer to 64
fglrx_pci 0000:02:00.0: setting latency timer to 64
[fglrx] Power up the ASIC
sd 2:0:0:0: [sda] Starting disk
ata2: port disabled. ignoring.
[fglrx] Preparing resume fglrx in kernel.
serial 00:08: activated
parport_pc 00:09: activated
saa7134 0000:01:07.0: BAR 0: set to [mem 0xfdeff000-0xfdeff3ff] (PCI address [0xfdeff000-0xfdeff3ff]
saa7134[0]: board init: gpio is c806000
HDA Intel 0000:02:00.1: BAR 0: set to [mem 0xfddfc000-0xfddfffff 64bit] (PCI address [0xfddfc000-0xfddfffff]
HDA Intel 0000:02:00.1: PCI INT B -> Link[APC5] -> GSI 16 (level, low) -> IRQ 16
HDA Intel 0000:02:00.1: setting latency timer to 64
HDA Intel 0000:02:00.1: irq 25 for MSI/MSI-X
HDA Intel 0000:00:05.0: BAR 0: set to [mem 0xfe028000-0xfe02bfff] (PCI address [0xfe028000-0xfe02bfff]
HDA Intel 0000:00:05.0: restoring config space at offset 0xf (was 0x5020200, writing 0x502020a)
HDA Intel 0000:00:05.0: restoring config space at offset 0x1 (was 0xb00000, writing 0xb00002)
HDA Intel 0000:00:05.0: PCI INT B -> Link[AAZA] -> GSI 20 (level, low) -> IRQ 20
HDA Intel 0000:00:05.0: setting latency timer to 64
forcedeth 0000:00:07.0: BAR 0: set to [mem 0xfe02d000-0xfe02dfff] (PCI address [0xfe02d000-0xfe02dfff]
forcedeth 0000:00:07.0: BAR 1: set to [io  0xec00-0xec07] (PCI address [0xec00-0xec07]
forcedeth 0000:00:07.0: restoring config space at offset 0xf (was 0x14010100, writing 0x1401010a)
forcedeth 0000:00:07.0: restoring config space at offset 0x1 (was 0xb00000, writing 0xb00007)
[fglrx] Resuming fglrx in kernel completed.
[fglrx] IRQ 27 Enabled
ata1.00: ACPI cmd ef/03:42:00:00:00:a0 (SET FEATURES) filtered out
ata1.00: ACPI cmd ef/03:0c:00:00:00:a0 (SET FEATURES) filtered out
ata1: nv_mode_filter: 0x739f&0x739f->0x739f, BIOS=0x7000 (0xc0000000) ACPI=0x701f (60:600:0x13)
ata1.00: configured for UDMA/33
ata3: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
ata3.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES) filtered out
ata3.00: configured for UDMA/133
forcedeth 0000:00:07.0: irq 26 for MSI/MSI-X
eth0: no link during initialization.
PM: recover of devices complete after 530.208 msecs
Restarting tasks ... done.
PM: Basic memory bitmaps freed
ata4: SATA link down (SStatus 0 SControl 300)
eth0: link up.

I have 1 gb ram and 2.26 gb swap partition. Should be enaugh...?

I'm trying to use pm-hibernate. I'm not alone:
http://ubuntuforums.org/showpost.php?p= … stcount=10
http://newyork.ubuntuforums.org/showpos … stcount=17
http://comments.gmane.org/gmane.linux.s … neral/9309

Last edited by Cdh (2010-05-30 00:08:26)

okraits · 2010-05-30 17:23:14

I have two issues with the catalyst driver:

- the system almost freezes when flash videos are played fullscreen
- composite (xcompmgr) doesn't work (windows disappear)

My setup:

- radeon hd 5770
- catalyst 10.5 (the issues already existed with 10.4)
- xorg-server-1.7-catalyst-maximize-fix 1.7.6-3

Section "Device"
        Identifier  "aticonfig-Device[0]-0"
        Driver      "fglrx"
        Option      "UseInternalAGPGART" "no"
        BusID       "PCI:1:0:0"
        Option      "AGPMode" "8"
        Option      "EnablePageFlip" "true"
        Option      "backingstore" "true"
        Option      "AllowGLXWithComposite" "true"
        Option      "OpenGLOverlay" "1"
        Option      "XAANoOffscreenPixmaps" "true"
        Option      "no_accel" "no"
        Option      "no_dri" "no"
        Option      "NoTV" "yes"
EndSection

Any solutions/ideas to fix the issues?

Thanks in advance.

subzero316 · 2010-05-30 18:59:20

okraits wrote:

I have two issues with the catalyst driver:
- the system almost freezes when flash videos are played fullscreen
- composite (xcompmgr) doesn't work (windows disappear)

Fullscreen flash has always been a trouble. It works fine in 32bit linux OS that I used earlier but any 64 bit Full screen is bad.
BUt I don really care anymore Youtube now has most videos on HTML5.
I ve noticed that my GPU usage is 0% for flash videos and my cpu shoots to 50% on both cores. So my guess is that Flash 64bit plugin doesnot access my GPU at all.

Is that the file autogenerated by aticonfig --initial command? Mine is pretty simple and I love the 10.5 driver as my suspend is back on and kwin has been blistering fast since 10.4. Havent tried xcompmgr so no clues about that.

Edit: Deleted the BIG quotes

Last edited by subzero316 (2010-05-30 19:00:05)

okraits · 2010-05-30 22:07:58

subzero316 wrote:

Fullscreen flash has always been a trouble. It works fine in 32bit linux OS that I used earlier but any 64 bit Full screen is bad.
BUt I don really care anymore Youtube now has most videos on HTML5.
I ve noticed that my GPU usage is 0% for flash videos and my cpu shoots to 50% on both cores. So my guess is that Flash 64bit plugin doesnot access my GPU at all.
Is that the file autogenerated by aticonfig --initial command? Mine is pretty simple and I love the 10.5 driver as my suspend is back on and kwin has been blistering fast since 10.4. Havent tried xcompmgr so no clues about that.
Edit: Deleted the BIG quotes

Well, i think that fullscreen flash worked in ubuntu x86_64 when i tried it about 2 months ago, that's why i'm a little bit confused that it doesn't work now on arch.
Does youtube really already deliver pages in HTML5? I haven't seen a single video on youtube yet that didn't use flash.

Yeah, i think that flash on x86_64 does only software rendering, that's why it is so slow.

The xorg.conf is based on the autogenerated file. The issues already existed without my customizations.

Last edited by okraits (2010-05-30 22:09:38)

dcc24 · 2010-05-31 04:17:00

okraits wrote:

Does youtube really already deliver pages in HTML5? I haven't seen a single video on youtube yet that didn't use flash.

You need to go to http://www.youtube.com/html5 and enable it for your account.

okraits · 2010-05-31 17:31:08

dcc24 wrote:

You need to go to http://www.youtube.com/html5 and enable it for your account.

Thanks for the hint!

okraits · 2010-05-31 18:29:29

Edit: conky was the problem - some conky option doesn't work together with xcompmgr...

Edit: now flash in fullscreen runs fast again - again, conky was the issue:

Before:

own_window no
own_window_type desktop

After:

own_window yes
own_window_type desktop

Last edited by okraits (2010-05-31 19:59:31)

pnutzh4x0r · 2010-06-01 11:26:15

Arathis wrote:

Does anyone else have problems with xRandR via HDMI with catalyst 10.5? I have defined a virtual Desktop size of 1920x1080 in my xorg.conf, but aRandR only shows a 1440x1440 virtual desktop and I cannot let it output a bigger resolution. xrandr -q also just shows my TFT capable of 1400x1050.
I fell back to catalyst 10.4 and everything's fine here.

I have the same problem with catalyst 10.5. With HDMI I could no longer do 1680x1050. Switching back to catalyst-test (10.4) works for me.

Last edited by pnutzh4x0r (2010-06-01 11:26:36)

Gooler · 2010-06-08 18:25:23

I'm just curious, how much RAM does Xorg use after 1 hour working with the computer?

I've a HD 5850 and a KDE4 desktop where i mostly write web apps with kate and do some design with gimp and Xorg uses >700MB while on my notebook it's around 100-200 MB.
Can this be some catalyst related issue?

Packages versions are:
xorg-server-catalyst-maximize-fix 1.7.6-1
catalyst 10.5-1

Last edited by Gooler (2010-06-08 18:35:51)

dcc24 · 2010-06-08 18:28:20

700MB certainly looks like a memory leak. I'm using xorg-server 1.8 / Catalyst 10.5 / HD4330 and uptime is more than 3 days. X consumes around 150MB RAM.

Vi0L0 · 2010-06-09 13:02:11

Gooler wrote:

I'm just curious, how much RAM does Xorg use after 1 hour working with the computer?
I've a HD 5850 and a KDE4 desktop where i mostly write web apps with kate and do some design with gimp and Xorg uses >700MB while on my notebook it's around 100-200 MB.
Can this be some catalyst related issue?
Packages versions are:
xorg-server-catalyst-maximize-fix 1.7.6-1
catalyst 10.5-1

So it's not only my problem... Not good... I noticed that ~ 2 months ago with first catalyst 10.4 ubuntu's. Well not 700MB/h(!), and not directly by Xorg processes (X process is eating around 400MB after all day working) - but something is eating my memory - i can see that clearly in kde system monitor when im moving window on desktop and im loosing 10MB/s(!) of memory. This eaten memory isn't linked with X process - more like its buffered. And yes - after some time memory goes back to me, but not all. Problem exist only with kde and "only" when Kwin is enabled. Killing Xes (in example while relogin if you set such option in kdmrc) always helps with retrieving memory. Sometimes applying some kwin effects also helps.
I haven't found any solutions for this problem - but i stopped searching long time ago - got 4GB of ram, its not big problem for me cuz still i can use kde for all day long doing such memory engaged processing like working on a lot of virtual machines whose taking ~2,5GB ram.
But yes - it is a problem.
I have checked oss drivers and with them this problem doesnt exist.
Edit: @Gooler: Although my problem is a little bit different than your, i have found 1 year old post exactly like yours http://chakra-project.org/bbs/viewtopic.php?id=791
Edit2: Quoting https://bugs.launchpad.net/ubuntu/+sour … bug/353800:

got this also in catalyst 10.4 and 10.5 and ubuntu 10.4
however memory leaked can be reclaimed when running the command 'glxinfo'.

Last edited by Vi0L0 (2010-06-09 13:30:11)

venky80 · 2010-06-09 16:16:34

does anyone know if hardware acceleration is possible for mpeg4/h264 videos with catalyst and mobility radeon 3670?

maevius · 2010-06-09 17:10:04

venky80 wrote:

does anyone know if hardware acceleration is possible for mpeg4/h264 videos with catalyst and mobility radeon 3670?

From http://ati.amd.com/products/catalyst/linux.html

The ATI Proprietary Linux driver currently supports Radeon 8500 and later AGP or PCI Express graphics products, as well as ATI FireGL 8700 and later products. We do not currently plan to include support for any products earlier than this. Drivers for earlier products should already be available from the DRI Project or Utah-GLX project.

Check also here
http://www.phoronix.com/scan.php?page=a … aapi&num=1
http://www.splitted-desktop.com/~gbeauchesne/

Eit · 2010-06-11 15:41:32

I'm trying to switch to catalyst to get something more from my hd4890, I've managed following the wiki to install them but I have doubts about it....I'm actually not using a xorg.conf , should I use one for exploit catalyst drivers? if not , how can I check that I actually get 3D acceleration and all is working smoothly ?

(I'm running arch x64 with awesome)

Thanks in advance

ChemBro · 2010-06-11 15:50:46

glxinfo and fglrxinfo could give you the information you need. Also you could try a 3D-game like extremetuxracer.

Eit · 2010-06-11 16:42:07

ChemBro wrote:

glxinfo and fglrxinfo could give you the information you need. Also you could try a 3D-game like extremetuxracer.

I guess I've a problem

 glxinfo
name of display: :0.0
zsh: segmentation fault  glxinfo

 fglrxinfo
zsh: segmentation fault  fglrxinf

any ideas ?

Vi0L0 · 2010-06-11 21:58:34

@Eit:
- have you added fglrx to MODULES array in /etc/rc.conf ? (if yes - then what modprobe fglrx command is telling? nothing means good)
- are you using [catalyst] repo or AUR?
You may use aticonfig --initial command to build basic xorg.conf

Eit · 2010-06-12 07:21:07

Vi0L0 wrote:

@Eit:
- have you added fglrx to MODULES array in /etc/rc.conf ? (if yes - then what modprobe fglrx command is telling? nothing means good)
- are you using [catalyst] repo or AUR?
You may use aticonfig --initial command to build basic xorg.conf

I've added fglrx to my modules, modprobe fglrx gives as output

 modprobe fglrx 
WARNING: All config files need .conf: /etc/modprobe.d/blacklist, it will be ignored in a future release

I've installed catalyst from aur using yaourt

Is having a xorg config mandatory to use 3d acceleration from catalyst ? actually I don't have one...:|

Thanks for the replies

Cdh · 2010-06-12 08:33:33

xorg will use proprietary drivers only when being told so.
Therefore you need a xorg.conf
Easiest is to just do aticonfig --inital which will create it fully automatically.

The Warning you get is not related to fglrx.
It just says that your
/etc/modprobe.d/blacklist
should be called
/etc/modprobe.d/blacklist.conf

Arch Linux

#1876 2010-05-28 21:10:04

Re: The AMD/ATI Bar & Grill

#1877 2010-05-28 23:26:30

Re: The AMD/ATI Bar & Grill

#1878 2010-05-28 23:32:09

Re: The AMD/ATI Bar & Grill

#1879 2010-05-29 11:18:22

Re: The AMD/ATI Bar & Grill

#1880 2010-05-29 13:25:12

Re: The AMD/ATI Bar & Grill

#1881 2010-05-29 14:39:40

Re: The AMD/ATI Bar & Grill

#1882 2010-05-29 23:51:03

Re: The AMD/ATI Bar & Grill

#1883 2010-05-30 17:23:14

Re: The AMD/ATI Bar & Grill

#1884 2010-05-30 18:59:20

Re: The AMD/ATI Bar & Grill

#1885 2010-05-30 22:07:58

Re: The AMD/ATI Bar & Grill

#1886 2010-05-31 04:17:00

Re: The AMD/ATI Bar & Grill

#1887 2010-05-31 17:31:08

Re: The AMD/ATI Bar & Grill

#1888 2010-05-31 18:29:29

Re: The AMD/ATI Bar & Grill

#1889 2010-06-01 11:26:15

Re: The AMD/ATI Bar & Grill

#1890 2010-06-08 18:25:23

Re: The AMD/ATI Bar & Grill

#1891 2010-06-08 18:28:20

Re: The AMD/ATI Bar & Grill

#1892 2010-06-09 13:02:11

Re: The AMD/ATI Bar & Grill

#1893 2010-06-09 16:16:34

Re: The AMD/ATI Bar & Grill

#1894 2010-06-09 17:10:04

Re: The AMD/ATI Bar & Grill

#1895 2010-06-11 15:41:32

Re: The AMD/ATI Bar & Grill

#1896 2010-06-11 15:50:46

Re: The AMD/ATI Bar & Grill

#1897 2010-06-11 16:42:07

Re: The AMD/ATI Bar & Grill

#1898 2010-06-11 21:58:34

Re: The AMD/ATI Bar & Grill

#1899 2010-06-12 07:21:07

Re: The AMD/ATI Bar & Grill

#1900 2010-06-12 08:33:33

Re: The AMD/ATI Bar & Grill

Board footer