AMD Catalyst 9.7 For Linux Released

[phoronix]

Phoronix: AMD Catalyst 9.7 For Linux Released

It's a bit late in the month, but AMD has just released the ATI Catalyst 9.7 driver update for Linux. Officially, the only new feature in Catalyst 9.7 is production support for Red Flag DT 7.0, but there are some bug-fixes too...

http://www.phoronix.com/vr.php?view=NzQwNA

[m4rgin4l]

If AMD keeps this trend, the next Catalyst release will have LESS features than the previous one...and support for Corel Linux.

[blindfrog]

FRAK another useless release. Still no 2.6.29 support WTF!? .30 has been out for ages. Hopefully the 9.6's introduced video playback lock ups are gone...This is just so fraking frustrating...

MEGAGRRRRRR

And it's going to take at least a year until free drivers have any form of GL2 support (optimistic estimate, yeah been following this scene enough to know). This is just horrible situation...At the moment there're two choices neither works. Just horrible situation we AMD users are forced to live in.

This new driver was supposed to fix all and it's already been over a year and half and something as ordinary as video playback is still SHITE!

You know actually this is not horrible. It was horrible a year ago and now it's just sad. *sigh*

[poofyyoda]

Been waiting a while for this, I literally just woke up and there it is.

Aside from kernel support, these drivers are getting better each time. The last 3 catalyst releases, have each allowed higher shader graphics settings in wine games without slowdowns. Now RandR is fixed, another bonus.

[rbmorse]

FRAK another useless release. Still no 2.6.29 support WTF!? .30 has been out for ages. Hopefully the 9.6's introduced video playback lock ups are gone...This is just so fraking frustrating...

MEGAGRRRRRR
Let the bashing begin!

[midol]

In another thread (yesterday) I was able to get directions about how to recover from a mismatch between kernel versions supported by fglrx 9.6 and the kernel installed. Fine. Thanks to those who replied.

But in looking over the release notes for this (9.7) driver I see the kernel requirements specified at 2.6 or higher. Well that isn't very helpful. If it said between 2.6.0 and 2.6.X where X is the highest kernel version actually supported, wouldn't that be useful?

Also, when I ran the installer, it seems that it would be elementary to check the version number and refuse to install a non-working driver.

Is that too naive? or what? It seems to me that it would be simple and useful to do that.

Dave

[A_Modest_Proposal]

this version also doesnt seem to like the current 2.6.30 patches from the gentoo bug trackers.
what a dissapointment.

[TrentZ]

But in looking over the release notes for this (9.7) driver I see the kernel requirements specified at 2.6 or higher. Well that isn't very helpful. If it said between 2.6.0 and 2.6.X where X is the highest kernel version actually supported, wouldn't that be useful?

Also, when I ran the installer, it seems that it would be elementary to check the version number and refuse to install a non-working driver.

Is that too naive? or what? It seems to me that it would be simple and useful to do that.

Dave

Hi Dave. The official "explanation" for the kernel version requirement noted as "2.6 or above" is here (http://www.phoronix.com/forums/showpost.php?p=83220&postcount=51).

Of course it would be useful to know explicitely which kernels are supported, but consider that AMD's website *still* claims you must install 32-bits libs to use Catalyst if you have a x86_64 system (oh, but that means if only if you want to use 32-bit apps, go figure! :D).

Btw, is not worth to patch KCL under Fedora 11. It will hang your Xserver randomly.

[bridgman]

I guess the key point is that the support statements should be "AND-ed" not "OR-ed". We'll try to make that more clear.

[tball]

I actually like this release very much.
The slow xv in window-mode is fixed :-)

Wine runs better -> No crashes anymore.

Keep up the good work AMD/ATI. Well, the we would like to have kernel 2.6.30, but fortunately there is good patches around.

[BlackStar]

I actually like this release very much.
The slow xv in window-mode is fixed :-)

Wow, this is awesome! Thumbs up :)

[acreda]

I have to agree, the driver does seem to get better and better, yes its a slow effort but as fglrx is closed sourced AMD can only justify so many resources to the project, while leaps and bounds are being made by the Mesa drivers because the can call on bigger resources around the world even if its ad-hoc.

If you ever need reminding that fglrx is always improving, I went back to OpenSUSE 11.1 from Ubuntu 9.04 due to a recent system upgrade and the driver inc with 11.1 I think is now 9.2 and let me tell you, compositing was in a mess - I have Google earth breaking though all the windows etc

It will take time but the more we can show support to the developers maybe rather than whinging, they may appricate it more aferall, We dont know how many people are assigned to this - they are propably doing the best they can.

Mind you, for some reason I think my UT2004 has improved in this release. I can't get PTS to run on my install (it is a bit ropey,) anyone fancy running a few benchmarks????

[TrentZ]

I have to agree, the driver does seem to get better and better, yes its a slow effort but as fglrx is closed sourced AMD can only justify so many resources to the project, while leaps and bounds are being made by the Mesa drivers because the can call on bigger resources around the world even if its ad-hoc.

:confused:

Just two or three people are working actively on the Mesa DRI driver for r6xx-r7xx. The driver being closed source seems totally irrelevant for justifying the amount of resources that AMD destines to it. I really doubt Mesa developers "can call resources around the world", driver development is hard and requires certain knowledge even to do "ad-hoc" contributions.

[Ant P.]

:confused:

Just two or three people are working actively on the Mesa DRI driver for r6xx-r7xx.

Is there something preventing more people from getting involved? Or is it just apathy...

[PuckPoltergeist]

this version also doesnt seem to like the current 2.6.30 patches from the gentoo bug trackers.
what a dissapointment.

As far as I can see, it will only need some minor changes to the patches...

[lowlands]

I just tried building it on F11 i386. At some point it just hangs. It would have been *really* nice if this 9.7 release had support for 2.6.29+ and Fedora 11. Now it's another month of waiting and hoping. Unless someone succeeds in making this driver build & install on F11. In that case I sure would appreciate a hint or two. Thanks!

[rafi]

I can confirm that Xvideo + Compositor improved a lot in this release. That really nice considering it is not mentioned in the release note. Thank you very much ATI :)

Tested on Xubuntu 9.04 with HD4870

[poofyyoda]

After finally trying it out, it appears this driver is badly broken. RandR is not fixed, killing a newly spawned X causes a black screen, and the system now crashes more regularly. Other than non-laggy video playback, this driver doesn't really offer anything. Glxgears also flickers with composite.

This driver is unusable to me without multiple x screens. Someone do

X :3 -ac -terminate

if you want to do a restart.

[saxonww]

consider that AMD's website *still* claims you must install 32-bits libs to use Catalyst if you have a x86_64 system (oh, but that means if only if you want to use 32-bit apps, go figure! :D).

Which is a very disappointing thing, because Gentoo still requires multilib to install these drivers [1]. And once you go no-multilib, you can't go back [2].

I'm happy that AMD is releasing drivers for Linux at all - they help many more ATI/linux users than not - but the continued ambiguity with this is a real bummer.

[1] http://sources.gentoo.org/viewcvs.py/gentoo-x86/x11-drivers/ati-drivers/ati-drivers-9.6.ebuild?view=markup

[2] http://www.gentoo.org/doc/en/gentoo-amd64-faq.xml#multilib

[samile]

I can confirm that Xvideo + Compositor improved a lot in this release. That really nice considering it is not mentioned in the release note. Thank you very much ATI :)

Tested on Xubuntu 9.04 with HD4870

Are you not experiencing the resizing delays any more?

[A_Modest_Proposal]

As far as I can see, it will only need some minor changes to the patches...

yep.

just putting the warning out there that the current patches arnt enough for .30 compat and will leave you with broken drivers under that kernel.

[hpestilence]

Catalyst 9.7 gave me a huge boost in fps in wine + eve-online when using fbo offscreen rendering mode. I was running 30 - 55 fps in space and now I'm running 65 - 90 fps. Over 100 in a station :D

I'm guess it's because of the new vertex array bgra extension.

[Erikina]

Wow, just wow. They can't support a 4 month old kernel. I guess looking at the community patches would be too much work..

[Laughing1]

Does this patch still work for 9.7?

http://www.phoronix.com/forums/showpost.php?p=79084&postcount=153

[SyXbiT]

I can't remember the last time there was a catalyst released and the title didn't read "Catalyst x.x released. nothing much new"

[kensai]

I guess the key point is that the support statements should be "AND-ed" not "OR-ed". We'll try to make that more clear.
No, I think the key is, AMD/ATI developers are doing absolutely nothing, I really hope AMD CEO can see the inefficiency of his team. I really doubt he has been able to appreciate this situation. So I guess, it will require a more aggressive approach by the users. One month to get such a worthless piece of bad software without any new features, just a few very very poor bug fixes, and support for a distribution that is as outdated as can be. Guys, come on, start supporting Linux, not some distributions.

Is a shame all my computers and all my hardware I have lying around is from AMD. But I tell you, no more I will buy AMD ever again, and will enthusiastically tell others to do the same.

I know it is worthless to direct any message to you guys, since you never hear the consumer, and never will. But soon, more than it has happened thus far, it will hurt the company.

Still concentrating in a very bad OSS driver with no 3D support, so we only have two half-backed alternatives. Can't we have one fully backed driver?

[AdrenalineJunky]

yup, this months driver release is absolutely worthless - with no improvements.

which perfectly explains why several people posted about big improvements with this driver. wait....

yes newer kernel support would be great, and yes, they should get on that, however, i had absolutely no problems installing this on 64-bit arch with 2.6.30, and as noted, video playback with compositing is better (though not perfect) and i'm getting better frame-rates in some games.

[nanonyme]

yup, this months driver release is absolutely worthless - with no improvements.Yeah, I suspect if every new version had enough new features to keep Michael happy, they'd soon start including everything including the kitchen sink and neighbour's gold fish. I don't personally see what's wrong with just pushing bug fixes and performance fixes in new versions.

[LavosPhoenix]

No, I think the key is, AMD/ATI developers are doing absolutely nothing, I really hope AMD CEO can see the inefficiency of his team. I really doubt he has been able to appreciate this situation. So I guess, it will require a more aggressive approach by the users. One month to get such a worthless piece of bad software without any new features, just a few very very poor bug fixes, and support for a distribution that is as outdated as can be. Guys, come on, start supporting Linux, not some distributions.

Is a shame all my computers and all my hardware I have lying around is from AMD. But I tell you, no more I will buy AMD ever again, and will enthusiastically tell others to do the same.

I know it is worthless to direct any message to you guys, since you never hear the consumer, and never will. But soon, more than it has happened thus far, it will hurt the company.

Still concentrating in a very bad OSS driver with no 3D support, so we only have two half-backed alternatives. Can't we have one fully backed driver?

Yeah, this is basically the bitching that I've been doing for nearly two years now. Yet fglrx is still shitty, and they can't seem to spend the couple of hours, if even that it would take to get the driver to use the latest kernel. Yes, people do USE the latest kernels, because they don't want to be stuck in the past like the other operating systems.

And no, I'm not bitching about "something that's free". I'm bitching because of the problems involved in getting a STABLE driver that supports all of the features of the hardware that was paid for. I can't do shit with the hardware if there isn't driver support for it.

[highlandsun]

Would be nice to actually be able to use the accelerated video decoding someday. And doesn't it seem strange that video decoding, which is pretty squarely a consumer issue, is only supported in fglrx which is supposedly a workstation driver?

It's time to tell the MPAA to FOAD and quit with all this DRM foolishness. It's a waste of hardware, and a waste of intellect...

[Pulfer]

Still concentrating in a very bad OSS driver with no 3D support

OSS driver is not as bad as Cataclysm/fglrx. It misses some very important features (3D, fan speed control) but at least it doesn't freeze my system after doing this or that. In 2D and video playback OSS driver radeon is much better than fglrx. If it had same quality support for 3D, it could be really great driver. But the development goes slow, I don't expect to see full 3D support until at least 2010. So I already ordered GeForce GTS 250 card to replace 4850 in my home PC.

[BlackStar]

Good release, here.

XV issues with Compiz have been completely fixed and 3d performance seems to be slightly higher than before.

Unfortunately, vsync under OpenGL 3.0 is still not fixed...

[Yfrwlf]

I have to agree, the driver does seem to get better and better, yes its a slow effort but as fglrx is closed sourced AMD can only justify so many resources to the project, while leaps and bounds are being made by the Mesa drivers because the can call on bigger resources around the world even if its ad-hoc.

Yep, I guess fglrx is still too far ahead of the open source drivers to just can the project yet, but I hope they do eventually and just focus on the open source goodness. Working on something yourself is just dumb when you can work with others, in this case at least.

[rafi]

Are you not experiencing the resizing delays any more?

I don't know what you mean by delays, but if it is resizing, then yes, resizing vlc is a bit slow (like "6 fps"). On the other hand maximizing is quite good (don't have to wait 5 second to have the image).

There is a funny bug with this driver. When you move the player window on the left of the screen, if the video is not entirely displayed, the image become blurry. I don't have this problem on the other edges (top, bottom and right are ok).

[Heiko]

For me this seems a good release. I see improvements in OpenGL (a beta, probably incomplete, OpenGL 3.1 context is available). So far I haven't noticed any regressions.

But I do hope we will see better kernel support in the future. If AMD doesn't want to give full support on newer kernels, please provide beta support for them or something like that.

I'm also hoping we will see OpenCL support soon... And OpenGL 3.2 will probably be announced soon, so I hope AMD will provide support for these new techniques soon.

All in all, I still have the feeling the driver is on average improving, so I have hope we will have a very good driver some day. And perhaps the massive trolling in these forums will decrease at some point...

[Hasenpfote]

Wine runs better -> No crashes anymore.Is Team Fortress 2 working?

[lowlands]

Yeah, I suspect if every new version had enough new features to keep Michael happy, they'd soon start including everything including the kitchen sink and neighbour's gold fish. I don't personally see what's wrong with just pushing bug fixes and performance fixes in new versions.

I absolutely agree with you that there is nothing wrong with shipping bug- and performance fixes. They are most welcome and most appreciated. However it is disappointing that for the umpteenth time there is no support for kernel 2.6.30 (or 2.6.29 in Fedora land which contains 2.6.30 stuff) and no support for Fedora 11. That is hardly the kitchen sink and to me sounds quite reasonable. If the fglx stuff was Open Source, support for 2.6.30 and Fedora 11 would probably have been committed the day 2.6.30 and F11 were released.

Sigh, another month of waiting and absolutely no certainty that the next release contains support for 2.6.30 and Fedora 11. ATI: how about a roadmap or confirmation that the next release will 100% sure contain support for 2.6.30 and Fedora 11?

[val-gaav]

Still concentrating in a very bad OSS driver with no 3D support, so we only have two half-backed alternatives. Can't we have one fully backed driver?

Just so you know open source driver has 3d support for all earlier then r600 cards ... Before the doc drops from AMD open source driver had good support for r200 and r300 and not that good support for r400 cards...
Thanks to the drops support for those improved and r500 support was added.

[Mr. Hilarious]

For me, this driver is the best one in ages (after some very rudimentary testing). Firstly, setting DisplaySize didn't work in 9.6 for me (could have been just my system - I guess something had changed in xorg-server and some other part was lacking behind, be that fglrx or some bug in Gentoo). But also, OepnGL performance is better. I can play DosBOX games without hiccups now in opengl mode - I doesn't crash when I toggle between windowed and fullscreen :D

Thanks ATI, this is the right direction. Though, I would of course like to see the OSS driver improving also :)

[xeizo]

Well, for me I've given up temporarily. I bought a very cheap Nvidia 9500GT-card(<40$) w. HDMI + audio passthrough for my Linux-box. I'm bound to using 2.6.31+ kernels for certain features I use, and no fglrx there yet.
Besides, I'm in for Beta-testing Voddler and wan't quality videodrivers with good videoacceleration as my Linux-box will be used as a HTPC for Voddler with my FullHD LCD in the livingroom.
I do not need high performance 3D, I have another Windows gamingbox for that, I just want basic working 3D and good video but low power consumption and noice, so I believe 9500GT will suffice(almost worthless for gaming).

Eventually I will switch back to ATI when drivers are more mature, but for now I'm looking forward to investigating the very fullfeatured 190.18 Nvidia-drivers :)

[mirv]

I've not tried the latest drivers yet (might plug away at an ebuild tonight) but I notice that the windows version contains some additional opengl support (EXT_provoking_vertex and EXT_vertex_array_bgra) - anyone know if this is also in the linux version (I would suspect so, given they moved to a common opengl base), and if anything else new is listed?

[NeoBrain]

Just downloading the driver (90 Mb, c'mon), did anyone try out the Compiz blur plugin with this driver, yet?

Also, does xinit xterm -- :2 work finally? This one keeps segfaulting for me everytime I try it...

Another issue I was having with > 9.2 was that I could alt+tab out of games (which support that feature), but then everything would just be overdrawn by the game graphics (i.e. same visual result as if alt+tab would not work), although everything was working as if the game was really minimized.

[kensai]

Eventually I will switch back to ATI when drivers are more mature, but for now I'm looking forward to investigating the very fullfeatured 190.18 Nvidia-drivers :)
So sad, then we won't see you again in ATI camp, you honestly think they will mature?

Every nvidia driver is full of many new features anticipated by many, almost every ATI driver has no new features and very poor bug fixes.

[tball]

Is Team Fortress 2 working?

Sorry I don't have TF2 :( But source games run.

[lbcoder]

I think that most of you are missing the strategy....

Catalyst should and probably WILL be discontinued as far as end users go. It will probably be maintained *enough* for some of their embedded clients/other weird proprietary configurations/etc. For end users, it WILL be replaced by the open source driver (as it has been already for everything up to and including R500).

With the condition of the open source support for R600/R700, I *really* doubt that there will be any major new features added to fglrx. The beginnings of usable 3D in open source will hopefully happen within the next few weeks.


And for everybody complaining about how bad fglrx is compared to nvidia... I remember (not too long ago) having a regression SO BAD that their driver was absolutely and completely unusable. It started with the first beta 165.xx version and it lasted for the better part of a YEAR. The result of this bug was that DPMS/mode switching would put the GPU into a really nice "limbo" state that required a radical reboot. Combine that with the poor manufacturing quality for 8000 models and higher (you know, the one where the temperature fluctuations will cause the GPU to flex and ultimately self destruct), and I'm afraid that the last nvidia I have and most likely will EVER own is my 7800 (which I've been meaning to yank for several months, but house construction has interfered).

[McDuck]

Is Team Fortress 2 working?

Wine
TF2 runs, kind of. Enabling multicore doesn't work and it is crash prone.

L4D runs, but flashlight doesn't work and mat_hdr_level 0 is well, not an option :P. Enabling multicore makes it extremely dark.

Cedega
TF2 doesn't work with DX9, only DX8.1.

L4D doesn't run at all.

Edit: With hd4850 catalyst 9.7, Wine 1.1.26, Cedega 7.3.1

[d2kx]

Seriously, I think Cedega is a mess.

[bulletxt]

I'm quite sure fglrx source code is so dirty and messed up that it is the team itself that has serious problems adding new features to the driver, or even actually fixing bugs without generating 200 new ones.

Fglrx is a lost case, and now that they are writing an oss driver fglrx is more obsolete than Windows 3.1 . Maybe phoronix should even just stop publishing articles of fglrx since it's funny to read them.


"but there are some bug-fixes too."

more than saying lol to that I'm not sure what else one could say :)

[McDuck]

Seriously, I think Cedega is a mess.

Sure it is, agreed. But those games does work with nvidia and cedega.
Afaik the games runs without major problems with nvidia&wine aswell.

[PuckPoltergeist]

the gentoo-patches, adapted for catalyst-9.7:

2.6.29-cat9.7.patch

diff -Nparu catalyst-9.7/common/lib/modules/fglrx/build_mod/firegl_public.c catalyst-9.7-2.6.29/common/lib/modules/fglrx/build_mod/firegl_public.c
--- catalyst-9.7/common/lib/modules/fglrx/build_mod/firegl_public.c 2009-07-11 01:02:14.000000000 +0200
+++ catalyst-9.7-2.6.29/common/lib/modules/fglrx/build_mod/firegl_public.c 2009-07-24 15:51:30.946052492 +0200
@@ -1448,7 +1448,11 @@ KCL_TYPE_Pid ATI_API_CALL KCL_GetTgid(vo
*/
KCL_TYPE_Uid ATI_API_CALL KCL_GetEffectiveUid(void)
{
+#if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,28)
+ return current->cred->euid;
+#else
return current->euid;
+#endif
}

/** /brief Delay execution for the specified number of microseconds
@@ -1820,15 +1824,30 @@ int ATI_API_CALL KCL_PosixSecurityCapChe
*/
void ATI_API_CALL KCL_PosixSecurityCapSetIPCLock(unsigned int lock)
{
+
+#if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,28)
+ struct cred *new = prepare_creds();
+ if (!new) {
+ printk(KERN_ERR "fglrx: could not allocate memory\n");
+ return;
+ }
+#else
+#define new current
+#endif
if (lock == 0 )
{
- cap_lower(current->cap_effective, CAP_IPC_LOCK);
+ cap_lower(new->cap_effective, CAP_IPC_LOCK);
}
else
{
- cap_raise(current->cap_effective, CAP_IPC_LOCK);
+ cap_raise(new->cap_effective, CAP_IPC_LOCK);
}
- return;
+
+#if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,28)
+ commit_creds(new);
+#else
+#undef new
+#endif
}

/** \brief Get number of available RAM pages
diff -Nparu catalyst-9.7/common/lib/modules/fglrx/build_mod/firegl_public.h catalyst-9.7-2.6.29/common/lib/modules/fglrx/build_mod/firegl_public.h
--- catalyst-9.7/common/lib/modules/fglrx/build_mod/firegl_public.h 2009-07-11 01:02:14.000000000 +0200
+++ catalyst-9.7-2.6.29/common/lib/modules/fglrx/build_mod/firegl_public.h 2009-07-24 15:51:30.946052492 +0200
@@ -18,6 +18,7 @@
#define _FIREGL_PUBLIC_H_

#include <stdarg.h>
+#include <asm/pgtable.h>
#include "kcl_pci.h"
#include "kcl_io.h"

@@ -603,6 +604,11 @@ extern unsigned long KCL_SYSINFO_
#define cpu_has_pge test_bit(X86_FEATURE_PGE, &boot_cpu_data.x86_capability)
#endif

+#if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,28)
+#undef pgprot_writecombine
+#undef pgprot_noncached
+#endif
+
#ifndef pgprot_writecombine
#define pgprot_writecombine(prot) __pgprot((pgprot_val(prot) & ~(_PAGE_PCD)) | _PAGE_PWT)
#endif
@@ -611,6 +617,7 @@ extern unsigned long KCL_SYSINFO_
#define pgprot_noncached(prot) __pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT)
#endif

+
#endif //FIREGL_USWC_SUPPORT


diff -Nparu catalyst-9.7/common/lib/modules/fglrx/build_mod/kcl_acpi.c catalyst-9.7-2.6.29/common/lib/modules/fglrx/build_mod/kcl_acpi.c
--- catalyst-9.7/common/lib/modules/fglrx/build_mod/kcl_acpi.c 2009-07-11 01:02:14.000000000 +0200
+++ catalyst-9.7-2.6.29/common/lib/modules/fglrx/build_mod/kcl_acpi.c 2009-07-24 15:51:30.946052492 +0200
@@ -18,6 +18,12 @@
#include <linux/autoconf.h>
#include <linux/acpi.h>

+#if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,28)
+#include <../drivers/acpi/acpica/acconfig.h>
+#include <../drivers/acpi/acpica/aclocal.h>
+#include <../drivers/acpi/acpica/acobject.h>
+#endif
+
#include "kcl_config.h"
#include "kcl_type.h"
#include "kcl_acpi.h"


2.6.30-enable_msi-cat9.7.patch

diff -Nparu catalyst-9.7-2.6.29/common/lib/modules/fglrx/build_mod/kcl_pci.c catalyst-9.7-2.6.30-enable_msi/common/lib/modules/fglrx/build_mod/kcl_pci.c
--- catalyst-9.7-2.6.29/common/lib/modules/fglrx/build_mod/kcl_pci.c 2009-07-11 01:02:14.000000000 +0200
+++ catalyst-9.7-2.6.30-enable_msi/common/lib/modules/fglrx/build_mod/kcl_pci.c 2009-07-24 15:55:04.586051435 +0200
@@ -310,3 +310,9 @@ void ATI_API_CALL KCL_PCI_FreeDmaCoheren
}
#endif //__x86_64__

+#undef pci_enable_msi
+int pci_enable_msi(struct pci_dev *pdev)
+{
+ return pci_enable_msi_block(pdev, 1);
+}
+


2.6.30-irqreturn_t-cat9.7.patch

diff -Nparu catalyst-9.7-2.6.30-enable_msi/common/lib/modules/fglrx/build_mod/drm_os_linux.h catalyst-9.7-2.6.30-irqreturn_t/common/lib/modules/fglrx/build_mod/drm_os_linux.h
--- catalyst-9.7-2.6.30-enable_msi/common/lib/modules/fglrx/build_mod/drm_os_linux.h 2009-07-11 01:02:14.000000000 +0200
+++ catalyst-9.7-2.6.30-irqreturn_t/common/lib/modules/fglrx/build_mod/drm_os_linux.h 2009-07-24 16:01:19.354051852 +0200
@@ -42,10 +42,12 @@
#define DRM_IRQ_ARGS int irq, void *arg, struct pt_regs *regs
/** backwards compatibility with old irq return values */
#ifndef IRQ_HANDLED
+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,30)
typedef void irqreturn_t;
#define IRQ_HANDLED /* nothing */
#define IRQ_NONE /* nothing */
#endif
+#endif

/** AGP types */
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,5,70)


2.6.30-rt_compat-9.7.patch

diff -Nparu catalyst-9.7-2.6.30-irqreturn_t/common/lib/modules/fglrx/build_mod/firegl_public.c catalyst-9.7-2.6.30-rt_compat/common/lib/modules/fglrx/build_mod/firegl_public.c
--- catalyst-9.7-2.6.30-irqreturn_t/common/lib/modules/fglrx/build_mod/firegl_public.c 2009-07-24 15:51:30.946052492 +0200
+++ catalyst-9.7-2.6.30-rt_compat/common/lib/modules/fglrx/build_mod/firegl_public.c 2009-07-24 16:04:37.254053559 +0200
@@ -1284,8 +1284,12 @@ unsigned int ATI_API_CALL KCL_ProcessIsT
struct task_struct *p;
int process_terminated = 1;

-#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17)
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17)
+#if !defined(CONFIG_PREEMPT_RT)
rcu_read_lock();
+#else
+ preempt_disable();
+#endif
#else
read_lock(&tasklist_lock);
#endif
@@ -1306,7 +1310,11 @@ unsigned int ATI_API_CALL KCL_ProcessIsT
}
}
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17)
+#if !defined(CONFIG_PREEMPT_RT)
rcu_read_unlock();
+#else
+ preempt_disable();
+#endif
#else
read_unlock(&tasklist_lock);
#endif


2.6.30-tlb_flush-cat9.7.patch

diff -Nparu catalyst-9.7-2.6.30-rt_compat/common/lib/modules/fglrx/build_mod/firegl_public.c catalyst-9.7-2.6.30-tlb_flush/common/lib/modules/fglrx/build_mod/firegl_public.c
--- catalyst-9.7-2.6.30-rt_compat/common/lib/modules/fglrx/build_mod/firegl_public.c 2009-07-24 16:04:37.254053559 +0200
+++ catalyst-9.7-2.6.30-tlb_flush/common/lib/modules/fglrx/build_mod/firegl_public.c 2009-07-24 16:08:07.778051481 +0200
@@ -2342,7 +2342,7 @@ void ATI_API_CALL KCL_flush_tlb_onepage(
* kernel < 2.6.27, on_each_cpu has 4 parameters.
* kernel >= 2.6.27, on_each_cpu has 3 parameters (removed the "retry" parameter)
*/
-#if defined(__x86_64__) && (defined(__SMP__) || defined(CONFIG_SMP)) && (LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,25))
+#if defined(CONFIG_SMP) && (LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,25))
# if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,27))
on_each_cpu(KCL_flush_tlb_one, &va, 1, 1);
# else

[PuckPoltergeist]

I actually like this release very much.
The slow xv in window-mode is fixed :-)


No it isn't. It's still very sluggish with compositing enabled (kwin). :(

[Joe Sixpack]

I don't know what you mean by delays, but if it is resizing, then yes, resizing vlc is a bit slow (like "6 fps"). On the other hand maximizing is quite good (don't have to wait 5 second to have the image).

There is a funny bug with this driver. When you move the player window on the left of the screen, if the video is not entirely displayed, the image become blurry. I don't have this problem on the other edges (top, bottom and right are ok).

Do you experience this same problem with SMPlayer? From my experiences VLC is noticeably slower than any MPlayer based player. I tried using VLC 0.9.7 - 0.9.9. Gnome-MPlayer was always faster when resizing windows with compositing turned on.

[skywarp04]

Fglrx is a mess. It needs to be thrown to the bottom a scrap heap as a failure. I don't think it is the fglrx driver team's fault. I'm not going to sit here and call someone's work crap without knowing what goes on behind the scenes. I feel that AMD/ATI doesn't allocate the appropriate amount or resources to the Linux driver team. It is a left over remnant from ATI. When AMD acquired them they decided to open source stuff finally and not bother allocating more resources to fglrx. I have a feeling that it will eventually disappear.

I don't know why they even bother with the monthly releases. I haven't seen a change since Catalyst 9.4 or 5. Notice that Phoronix didn't even bother to post something about 9.7 being released. I have a problem since Catalyst 9.4 was released. I can use my computer for a day or two and I will be in the middle of doing something and then whatever window is in focus suddenly becomes unusable. Then whatever is in either the top or bottom half of the window duplicates itself inside that window. Once this starts the GUI becomes useless. I have to switch to a tty terminal and restart from there. Just tried 9.7 last night and the same problem still exists, matter of fact it seems to happen more often.

So what do I do? Go report a bug to http://ati.cchtml.com/. Then what, is someone going to actually grab a PC put a Radeon 3850 in it and then install Kubuntu to test what I claim? I doubt it. Instead I'll just hear some excuse as to such and such causing the problem and it's not the driver's fault. Whatever. I give up on fglrx. I'm not even going to bother installing the next release. I'll just wait for Karmic to get released which is rumored to use the latest HD Radeon driver by default.

[skywarp04]

Ooops, missed it. Here's a short blurb. http://www.phoronix.com/scan.php?page=news_item&px=NzQwNA

[rafi]

Do you experience this same problem with SMPlayer? From my experiences VLC is noticeably slower than any MPlayer based player. I tried using VLC 0.9.7 - 0.9.9. Gnome-MPlayer was always faster when resizing windows with compositing turned on.

There is no real difference, resizing VLC is as slow as resizing SMPlayer.
Actually, resizing is slow, even resizing Thunar is slow. I compared with the catalyst 9.1 on my x1600 mobility, and resizing Thunar is much smoother.

[bridgman]

Fglrx is a mess. It needs to be thrown to the bottom a scrap heap as a failure. I don't think it is the fglrx driver team's fault. I'm not going to sit here and call someone's work crap without knowing what goes on behind the scenes. I feel that AMD/ATI doesn't allocate the appropriate amount or resources to the Linux driver team. It is a left over remnant from ATI. When AMD acquired them they decided to open source stuff finally and not bother allocating more resources to fglrx. I have a feeling that it will eventually disappear.

The fglrx driver is primarily written for the professional workstation market (CAD etc..), which is the main market for Linux graphics right now. It is unlikely that the open source drivers will ever be able to replace fglrx in that segment.

The open source driver, on the other hand, is aimed primarily at the consumer and enterprise client market, and is intended to allow distros to tailor the out-of-box experience for users who are installing Linux on a system which was not configured and shipped with Linux by the HW vendor.

In between those two segments you have OEM systems which are designed with Linux support in mind, and then configured/shipped from the factory with Linux pre-installed. Again, fglrx is a great fit there.

I don't know why they even bother with the monthly releases. I haven't seen a change since Catalyst 9.4 or 5. Notice that Phoronix didn't even bother to post something about 9.7 being released. I have a problem since Catalyst 9.4 was released. I can use my computer for a day or two and I will be in the middle of doing something and then whatever window is in focus suddenly becomes unusable. Then whatever is in either the top or bottom half of the window duplicates itself inside that window. Once this starts the GUI becomes useless. I have to switch to a tty terminal and restart from there. Just tried 9.7 last night and the same problem still exists, matter of fact it seems to happen more often.

I don't think we have seen a report on this or observed it in our testing. Are you seeing signs of a memory leak ?

So what do I do? Go report a bug to http://ati.cchtml.com/. Then what, is someone going to actually grab a PC put a Radeon 3850 in it and then install Kubuntu to test what I claim? I doubt it.

The purpose of the ati.cchtml.com tracker is to collect and organize enough information that a developer will be able to reproduce the problem in house. For the problem you are describing we would need more information than just "3850" and "Kubuntu", of course, eg which applications we should be running to make the problem appear.

Instead I'll just hear some excuse as to such and such causing the problem and it's not the driver's fault. Whatever.

In cases where the same problem appeared at the same time with multiple vendors and multiple drivers I don't think it's fair to call that an "excuse". Your call though.

I give up on fglrx. I'm not even going to bother installing the next release. I'll just wait for Karmic to get released which is rumored to use the latest HD Radeon driver by default.

Jaunty also shipped with the open source drivers by default. If they do everything you want then I think everyone would agree that staying with the default drivers is your best bet.

[DogWizard]

Does this patch still work for 9.7?

http://www.phoronix.com/forums/showpost.php?p=79084&postcount=153

Still works together with #203 for 2.6.31-rc4 (patch can handle the minor differences in the files)

[TrentZ]

bridgman, is there any hope to see OpenCL implemented in Linux? Brook+ seems pretty dead.

http://forums.amd.com/forum/messageview.cfm?catid=328&threadid=116379

[daponz]

I don't think we have seen a report on this or observed it in our testing. Are you seeing signs of a memory leak ?

I observed the same behavior on my computer running a 4850 on ubuntu.
I've seen my Xorg memory usage increase constantly so there might be a memory leak related to FGLRX but I never had the time to investigate more :/.
It started for me with 9.6 catalyst.

But I don't really care anymore, got bored with FGLRX issues... I think I'll switch back to Nvidia

[energyman]

this version also doesnt seem to like the current 2.6.30 patches from the gentoo bug trackers.
what a dissapointment.

I used the patches from the 9.6 ebuild - in fact all I did was renaming the ebuild - and it built and installed fine.

[ 72.995213] fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, Starnberg, GERMANY' taints kernel.
[ 72.995217] Disabling lock debugging due to kernel taint
[ 73.014722] [fglrx] Maximum main memory to use for locked dma buffers: 5765 MBytes.
[ 73.014807] [fglrx] vendor: 1002 device: 9501 count: 1
[ 73.014985] [fglrx] ioport: bar 4, base 0xb000, size: 0x100
[ 73.014997] pci 0000:02:00.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
[ 73.015001] pci 0000:02:00.0: setting latency timer to 64
[ 73.015140] [fglrx] Kernel PAT support is enabled
[ 73.015161] [fglrx] module loaded - fglrx 8.63.2 [Jul 2 2009] with 1 minors
[ 73.204402] fglrx_pci 0000:02:00.0: irq 30 for MSI/MSI-X
[ 73.204737] [fglrx] Firegl kernel thread PID: 4056
[ 75.225921] [fglrx] Gart USWC size:1279 M.
[ 75.225924] [fglrx] Gart cacheable size:508 M.
[ 75.225929] [fglrx] Reserved FB block: Shared offset:0, size:1000000
[ 75.225931] [fglrx] Reserved FB block: Unshared offset:fbff000, size:401000
[ 75.225932] [fglrx] Reserved FB block: Unshared offset:1fffc000, size:4000

[rklrkl]

I think the problem with the Catalyst drivers is actually the policy of only supporting enterprise Linux distros, which is utterly ludicrous because as a percentage of Linux desktops, enterprise distros probably cover less than 1%!

The people who really *need* 3D drivers are those running home Linux desktops - why on earth can't ATI support them in a timely fashion? If they did, then the enterprise distros would automatically get support too (since they always lag behind w.r.t. versions).

Hence, anything non-enterprise (OpenSUSE, Fedora, Ubuntu non-LTS) Linux will inevitably use a later kernel and X server than enterprise distros and Catalyst inevitably becomes immediately unusable upon release of the latest non-enterprise distro version.

The biggest Catalyst disaster of the last few years was the release of Fedora 9 - I can guarantee you that there were more users of F9 within a month of its release than *all* the enterprise Linux distros in the world combined.

And did ATI care? Nope, they sat on their fat arses for month after month, not supporting the kernel/X server combo that F9 shipped with. It took 4.5 months (yes, months) before a Catalyst driver finally came out that worked - so long that I stuck with F8 and jumped to F10 a few months after that.

Now we're in the same vicious cycle with Fedora 11 - one of the best Linux desktop distro users out there (particularly for programmers) and completely unsupported for multiple months *again* by ATI. This time, it looks like a kernel 2.6.29 issue (F9 mostly had the X server to blame) and it isn't the only non-enterprise distro suffering this problem.

And what do we read in the release notes for Catalyst 9.7? Not support for a later X server release, not support for a newer kernel (leaving Catalyst now 3 versions behind, which is an absolute disgrace), but the addition of support for an obscure enterprise Linux distro (Red Flag DT 7.0 - barely used in Western countries where I suspect the bulk of ATI's Linux market is) that's probably running a crusty old X server/kernel combo and not used by many desktop users anyway (can you see a lot of Chinese people paying top Yen for an obscure enterprise Linux distro on their desktop? Nope, I can't either).

Basically, ATI have badly let down Linux desktop users in the last couple of years and this latest 9.7 release shows no sign of fixing that either. At this rate, I'm wondering if Fedora 12 will be out before ATI supports Fedora 11, it's getting that ridiculous!

Before you mention the "radeon" and "radeonhd" drivers, I've tried them both and they simply don't work well - monitor alignments are wrong, refresh rates aren't set right and the 3D acceleration is either poor or non-existent. The Catalyst driver, when they eventually update it for the X server/kernel I run, is the only one that does its job properly, albeit sadly many months late.

Get your arses into gear, ATI! You may have a monthly release schedule, but it's virtually worthless if your latest release doesn't work on the latest versions of world's most popular Linux destkop distros (Ubuntu, Fedora and OpenSUSE).

[Kamikaze321]

LOL!!!!!!!!

Open Source driver works with .31 on my karmic alpha. And look at fglrx LOL!!

3D is better at fglrx? install it on my mashine with x1400 and .31 and we will se what driver has better 3d performance. LOL

[mirv]

Ahh, I was wondering where the trolls were lurking....

[AdrenalineJunky]

LOL!!!!!!!!

Open Source driver works with .31 on my karmic alpha. And look at fglrx LOL!!

3D is better at fglrx? install it on my mashine with x1400 and .31 and we will se what driver has better 3d performance. LOL

thats with an x1400.... alot of people have newer cards than that which have less full featured OSS drivers....

[poofyyoda]

For anybody with 9.7, who sincerely believes that this driver release is progress please run

X :3 -ac -terminate &

and then kill the server and then enjoy the frozen machine:o


Please tell me if it actually works for someone without a reboot, so I can give it another go assuming my setup was wrong.

I pretty much rely on multiple x screens for every game I launch so that they aren't affected by the compositing manager, and apps setting my desktop to non-preferred resolution after quiting. I also use them for running virtualbox - very handy for switching between tasks.

[bridgman]

I think the problem with the Catalyst drivers is actually the policy of only supporting enterprise Linux distros, which is utterly ludicrous because as a percentage of Linux desktops, enterprise distros probably cover less than 1%!

When did Ubuntu become only an enterprise distro ? Based on feedback from OEMs and end users RHEL, SuSE and Ubuntu together cover at least 60% of our market, much more if you factor in the higher Linux marketshare in the workstation segment. For SuSE, our test focus is actually on OpenSUSE (the consumer distro) not SLES/SLED (the enterprise distros).

The people who really *need* 3D drivers are those running home Linux desktops - why on earth can't ATI support them in a timely fashion?

Can I suggest that the primary people who *need* 3D drivers are those running commercial workstation (eg CAD) applications ? They run enterprise distros almost exclusively, and represent a significant part of the Linux graphics business. I agree that home desktop users who class themselves as "gamers" (perhaps 30% of that market) are the next in line.

If they did, then the enterprise distros would automatically get support too (since they always lag behind w.r.t. versions). Hence, anything non-enterprise (OpenSUSE, Fedora, Ubuntu non-LTS) Linux will inevitably use a later kernel and X server than enterprise distros and Catalyst inevitably becomes immediately unusable upon release of the latest non-enterprise distro version.

For Ubuntu and SuSE we focus testing on the consumer releases (non-LTS Ubuntu and OpenSUSE) since the consumer and enterprise versions are close enough that testing the consumer version is usually sufficient to ensure the driver will work on the enterprise version. In the case of RHEL/Fedora, the two distros are much further apart, to the point that testing on Fedora is not sufficient to ensure operation on RHEL.

The biggest Catalyst disaster of the last few years was the release of Fedora 9 - I can guarantee you that there were more users of F9 within a month of its release than *all* the enterprise Linux distros in the world combined.

And did ATI care? Nope, they sat on their fat arses for month after month, not supporting the kernel/X server combo that F9 shipped with. It took 4.5 months (yes, months) before a Catalyst driver finally came out that worked - so long that I stuck with F8 and jumped to F10 a few months after that.

OK, help me here. We don't claim to support Fedora, and we recommend the open source drivers for Fedora (which RH agrees with). How can we have a "Catalyst disaster" with Fedora ?

Now we're in the same vicious cycle with Fedora 11 - one of the best Linux desktop distro users out there (particularly for programmers) and completely unsupported for multiple months *again* by ATI. This time, it looks like a kernel 2.6.29 issue (F9 mostly had the X server to blame) and it isn't the only non-enterprise distro suffering this problem.

And what do we read in the release notes for Catalyst 9.7? Not support for a later X server release, not support for a newer kernel (leaving Catalyst now 3 versions behind, which is an absolute disgrace), but the addition of support for an obscure enterprise Linux distro (Red Flag DT 7.0 - barely used in Western countries where I suspect the bulk of ATI's Linux market is) that's probably running a crusty old X server/kernel combo and not used by many desktop users anyway (can you see a lot of Chinese people paying top Yen for an obscure enterprise Linux distro on their desktop? Nope, I can't either).

China is actually a huge market for Linux, particularly for OEM preloads.

Basically, ATI have badly let down Linux desktop users in the last couple of years and this latest 9.7 release shows no sign of fixing that either. At this rate, I'm wondering if Fedora 12 will be out before ATI supports Fedora 11, it's getting that ridiculous!

Before you mention the "radeon" and "radeonhd" drivers, I've tried them both and they simply don't work well - monitor alignments are wrong, refresh rates aren't set right and the 3D acceleration is either poor or non-existent. The Catalyst driver, when they eventually update it for the X server/kernel I run, is the only one that does its job properly, albeit sadly many months late.

Have you filed bugs for the display issues ? I don't think I remember hearing those.

Get your arses into gear, ATI! You may have a monthly release schedule, but it's virtually worthless if your latest release doesn't work on the latest versions of world's most popular Linux destkop distros (Ubuntu, Fedora and OpenSUSE).

I believe the latest releases of Ubuntu and OpenSuSE are both supported.

[ptiboli]

Some improvements with my HD3200 running Ubuntu 9.04 : more frames in glxgears, even with compiz on... video decoding still works, we'll see if the occasional image freeze still happens.

[energyman]

before you get pissed because AMD is concentrating on CAD users:
just compare the prize of a FireGL card with a normal desktop product.

It is basically the same hardware. You are paying the premium for the drivers and support.

If you are willing to shell out the money for a firegl card (and you buy, say 100 or 1000 of them) I am sure AMD will be a lot quicker helping you out.

At the moment supporting 'us' is a nice bonus. A fallout of the firegl support. And it is not that bad. Sure, dmesg is still spammed, but 9.6 has some nice bugfixes. Supporting the open source driver development with documentation AND manpower is something AMD should be lauded for.

So, please, calm down a bit, ok? AMD is not as big as Intel and unlike nvidia plays nicely.

[McDuck]

OK, help me here. We don't claim to support Fedora, and we recommend the open source drivers for Fedora (which RH agrees with). How can we have a "Catalyst disaster" with Fedora ?


I don't like that "we don't claim to support" statement, why aren't you even trying to make it work?

The point is that you _should_ support fedora, a very common distribution that has a lot of users. Hiding behind some statement that fedora is too far apart from rhel is not a good reason for me, I want support for fedora itself, not rhel.

That you decide on a few distributions and support them is perhaps enough for you, enough for any big corpration, but not for an end user like me and many others.

Remember that you are competing against nVidia, so people will always want you to have more or less equal support.

I just had to change distribution from ArchLinux to Ubuntu because you don't support any new kernels not in these selected distributions.
I am fine with this, which is why I've never complained, but alot of people are not, and with right.

And also, please don't say that you support through your oss driver until it actually is good. Full 3D with very good performance (think I read an estimate of ~80% of fglrx in some thread), and more or less feature parity.

[Erikina]

OK, help me here. We don't claim to support Fedora, and we recommend the open source drivers for Fedora (which RH agrees with). How can we have a "Catalyst disaster" with Fedora ? Because you're ignoring a distribution with many ati users? Fedora / RH will always push the opensource (and patent unencumbered) solution -- but it doesn't mean that magically it suits all users (that's why there's RPM fusion etc.). To put it in perspective, there are an estimate ~16 million fedora machines (albeit the vast majority are too smart to have an ATI card). The amount of bad will this is generating is astounding, all for something you're going to add in a couple month anyway. (and is seriously not that big of a deal to add)

Considering the quality of all the previous fglrx releases, I don't see why fglrx is reduced to bare-maintenance releases and all effort put into a solution that doesn't suck (radeon drivers) including adding all the CAD stuff that you say opensource will never support ;D.

[skywarp04]

Posted by bridgman
The fglrx driver is primarily written for the professional workstation market (CAD etc..), which is the main market for Linux graphics right now. It is unlikely that the open source drivers will ever be able to replace fglrx in that segment.

I'm so tired of hearing this. Does that change the fact that I paid $200 for a video card that is not fully supported by the company that made it? You don't hear this kind of lame statement from your competition. I have products from AMD and Nvidia in my home and I can testify to the support of both companies. AMD and Nvidia are on even ground when it comes to Windows in my opinion even though there are still tons of people who claim otherwise. But when it comes to Linux Nvidia mops the floor with AMD. I don't hear Nvidia claiming that they only support their workstation clients. There Linux driver isn't called quadro. So what you are telling me is that I should just use Windows. Oh, there's the open source drivers you say. Have you tried them lately? Apparently not because if you had you would know that they are still not that usable for my hardware. Unless the latest version is being used with the 2.6.30 kernel the Radeon 3850 doesn't have any 3D acceleration. That's not happening until Ubuntu 9.10 is released.

Posted by bridgman
Jaunty also shipped with the open source drivers by default. If they do everything you want then I think everyone would agree that staying with the default drivers is your best bet.

Obviously they don't do what I want, otherwise I wouldn't be trying to use fglrx. The version of the radeon driver that ships with Jaunty only provides 2D acceleration for my card. Why shouldn't I have be able to access my $200 video cards 3D capabilities? I constantly can't do compositing or play simple games because fglrx is too damn flaky. I've seen you mention using XRender before for doing compositing because compositing doesn't require full OpenGL acceleration. Have you ever tried using XRender? It's slow as molasses and doesn't always work right.

Posted by bridgman
The purpose of the ati.cchtml.com tracker is to collect and organize enough information that a developer will be able to reproduce the problem in house. For the problem you are describing we would need more information than just "3850" and "Kubuntu", of course, eg which applications we should be running to make the problem appear.

Yes, it does seem to be a memory leak. As far as what applications I use, it doesn't matter. It happens with different programs all the time. Firefox, Konsole, VirtualBox, Amarok, whatever. I'll give you information like my motherboard model, CPU, Ram, whatever all day long if I know someone is actually going to do something with it. But like you said fglrx is for CAD workstations, average consumers don't matter to AMD. Unless you can tell me someone is actually going to investigate something I'm not going to waste my time putting together my specs for a company who doesn't care unless I'm using CAD.

I hate that you have to take the brunt of everyone's anger here in this forum and I commend you for keeping your cool as well as you do with some of the posts I've seen directed towards you. I wish you would drop the "we only support CAD users" defense though. I hope AMD pays you well for monitoring this forum. :)

[bridgman]

Because you're ignoring a distribution with many ati users? Fedora / RH will always push the opensource (and patent unencumbered) solution -- but it doesn't mean that magically it suits all users (that's why there's RPM fusion etc.). To put it in perspective, there are an estimate ~16 million fedora machines (albeit the vast majority are too smart to have an ATI card).

With respect, if you add the installed base estimates from all the major OSes, then compare the total with the estimated number of PCs in the world, something seems seriously out of whack. Most estimates indicate that roughly 1% of the billion PCs in the world are running Linux, or 10 million total. You can find higher estimates but if you dig a bit you find the numbers are things like "PCs sold without an OS, and we assume the user will run Linux rather than pirating Windows".

I don't think anyone knows what the real numbers are, but please understand that there is a huge range of estimates out there and the numbers you are citing sit out near the edge of the range. Also note that Fedora is probably the most "different" from upstream kernel & X server of any distro out there, so the idea that supporting Fedora would magically bring support for other distros is a bit of a myth. I do think you will see faster support for other distros over time, but we do want to make sure that we have a solid solution in place for the current range of distros first.

The amount of bad will this is generating is astounding, all for something you're going to add in a couple month anyway. (and is seriously not that big of a deal to add)

Two years ago everyone told us that writing drivers was easy and that if only we would open up the hardware specs then the community would write better drivers in no time. A year later that shifted to loud complaints that *we* weren't spending enough money writing the open source drivers ourselves, and what I'm hearing today is "we demand proprietary drivers, writing open source drivers is too hard and the open source drivers don't do enough". Despite all that, you are still telling us that adding features and support to the proprietary drivers is "easy" and that we're obviously stupid for not doing it.

The choice is pretty simple -- we can spread our efforts out and attemt to support all distros equally but make slower progress overall, or focus on a representative subset of distros and make progress more quickly, *then* focus on improving / speeding up support for the rest. The first approach means our overall quality of Linux support will be lower but we won't piss anyone off; the second approach means we can offer a good Linux user experience more quickly but will some distro users will feel insulted by our choices.

Considering the quality of all the previous fglrx releases, I don't see why fglrx is reduced to bare-maintenance releases and all effort put into a solution that doesn't suck (radeon drivers) including adding all the CAD stuff that you say opensource will never support ;D.

Simple. The workstation market is very fast moving and competitive, but it also represents the largest share of the Linux graphics market, so stopping is not an option. The whole point of fglrx is to give Linux users access to the proprietary code we share across all OSes, which we can't put in the open source drivers for a number of reasons.

[bridgman]

I'm so tired of hearing this. Does that change the fact that I paid $200 for a video card that is not fully supported by the company that made it?

You're taking my words out of context. I was responding to your comment that "Fglrx is a mess. It needs to be thrown to the bottom a scrap heap as a failure." - I said that we need fglrx for the workstation market, but you're responding as if I said "we don't care about you, only about the workstation market".

You don't hear this kind of lame statement from your competition.

Tell them that they should scrap their driver and see what they say :D

I have products from AMD and Nvidia in my home and I can testify to the support of both companies. AMD and Nvidia are on even ground when it comes to Windows in my opinion even though there are still tons of people who claim otherwise. But when it comes to Linux Nvidia mops the floor with AMD.

Yep, and what we've been doing is bringing that same core code to Linux users and then knocking off the Linux-specific issues that result. I know you would like all the work to be finished in a month, but it doesn't work that way.

I don't hear Nvidia claiming that they only support their workstation clients.

You don't hear us claiming it either :D

There Linux driver isn't called quadro. So what you are telling me is that I should just use Windows. Oh, there's the open source drivers you say. Have you tried them lately? Apparently not because if you had you would know that they are still not that usable for my hardware. Unless the latest version is being used with the 2.6.30 kernel the Radeon 3850 doesn't have any 3D acceleration. That's not happening until Ubuntu 9.10 is released.

I actually use the open source drivers all the time on rv570, rv620 and rv770 so I think I have a pretty good idea what they do ;)

2.6.30 is for 2D acceleration, but that code was backported into the Ubuntu kernel for 9.04. Kernel support for 3D will probably go into 2.6.32.

The main thing gating 3D driver availability is writing the %^$&%@! driver. We released programming info and sample code at the end of 2008 but it turned out that all of the community developers were still working on other important projects, so it's really just been our guys writing the 3D driver. It's making pretty good progress though; there's still a "mystery problem" with textures and some intermittent corruption that looks like something we don't understand about the new radeon-rewrite code, but other than that it seems to be coming together pretty well.

Obviously they don't do what I want, otherwise I wouldn't be trying to use fglrx. The version of the radeon driver that ships with Jaunty only provides 2D acceleration for my card. Why shouldn't I have be able to access my $200 video cards 3D capabilities? I constantly can't do compositing or play simple games because fglrx is too damn flaky. I've seen you mention using XRender before for doing compositing because compositing doesn't require full OpenGL acceleration. Have you ever tried using XRender? It's slow as molasses and doesn't always work right.

It worked pretty well for me, although I was mostly using Metacity. I can try it again on the weekend if you like.

Yes, it does seem to be a memory leak. As far as what applications I use, it doesn't matter. It happens with different programs all the time. Firefox, Konsole, VirtualBox, Amarok, whatever. I'll give you information like my motherboard model, CPU, Ram, whatever all day long if I know someone is actually going to do something with it. But like you said fglrx is for CAD workstations, average consumers don't matter to AMD.

No, that's what *you* said. What I said was that we could not "throw it to the bottom of the scrap heap" because we needed it for the workstation market.

Unless you can tell me someone is actually going to investigate something I'm not going to waste my time putting together my specs for a company who doesn't care unless I'm using CAD.

Again, the CAD thing is your statement not mine. I"m only suggesting that you fill out a bug ticket so our devs can reproduce the problem and have a chance of fixing it. The only promise I can make for sure is that if we *can't* reproduce it then it's only likely to get fixed by accident since nobody will be able to work on it.

I hate that you have to take the brunt of everyone's anger here in this forum and I commend you for keeping your cool as well as you do with some of the posts I've seen directed towards you. I wish you would drop the "we only support CAD users" defense though.

I'm not saying that. I am saying that workstation users represent the largest part of the market and that we need to allocate resources in a way that is at least approximately based on market size (be it current or future). Any company who says they don't do that is probably lying to you.

I hope AMD pays you well for monitoring this forum. :)

My "job" here is as the open source guy, but I try to help out with fglrx questions where I can. Everyone should have a hobby :D

[Erikina]

With respect, if you add the installed base estimates from all the major OSes, then compare the total with the estimated number of PCs in the world, something seems seriously out of whack.

Agreed, but Fedora figures are backed up by statistics and methdology (http://jspaleta.livejournal.com/42464.html?thread=104160) (something I don't think any other distro has provided.) Also, keep in mind all those cheap One-Laptop-Per-Child (>1 mill) are running Fedora, as a fair few servers, and it's a favourite at universities.


Most estimates indicate that roughly 1% of the billion PCs in the world are running Linux, or 10 million total. You can find higher estimates but if you dig a bit you find the numbers are things like "PCs sold without an OS, and we assume the user will run Linux rather than pirating Windows". The 1-2% estimates are generally derived from looking at statistics from web browser traffic. Something that will under-represent Linux as a whole .. although probably a decent metric for looking at desktop usage (as surfing the net net is a pretty desktop oriented activity).


Two years ago everyone told us that writing drivers was easy and that if only we would open up the hardware specs then the community would write better drivers in no time.

For what it's worth, opening the specs is the reason I have a ATI card now.


A year later that shifted to loud complaints that *we* weren't spending enough money writing the open source drivers ourselves, and what I'm hearing today is "we demand proprietary drivers, writing open source drivers is too hard and the open source drivers don't do enough".

What you're hearing is the complaints of people who want their hardware to work. I wouldn't exactly take it as a technical guide to writing drivers.


Despite all that, you are still telling us that adding features and support to the proprietary drivers is "easy" and that we're obviously stupid for not doing it.
Oh please. I've seen patches sitting in your bug tracker for months without getting applied. And I'm not even asking for a million new features, I just want it to work in a _4 month_ old kernel that you claim to support. The patch to get 9.6/9.7 to compile with later kernels is relatively minor (although, it looks like there are a couple other changes needed in your binary).

The website and release notes don't even mention the fact, that your driver will not work. The installer is all too happy to break the system. If this the best you can do, then maybe there is something wrong with the dev team..


The choice is pretty simple -- we can spread our efforts out and attemt to support all distros equally but make slower progress overall, or focus on a representative subset of distros and make progress more quickly, *then* focus on improving / speeding up support for the rest. The first approach means our overall quality of Linux support will be lower but we won't piss anyone off; the second approach means we can offer a good Linux user experience more quickly but will some distro users will feel insulted by our choices. Again, I don't buy it. The way any decent team would do it, is make sure it works with the latest stable software (kernel, xorg). Then focus on the supported distros and features. Unsupported distros / build-it-yourself-stuff would just need to figure out how to package it themselves. That way, you're not increasing the amount of work (supporting later software is inevitable) while not completely shafting users.

[skywarp04]

Ok, maybe I took it a little out of context, but just to clarify these were your words.

Posted by bridgman
The fglrx driver is primarily written for the professional workstation market (CAD etc..), which is the main market for Linux graphics right now. It is unlikely that the open source drivers will ever be able to replace fglrx in that segment.


Go to page 6. :) Instead of saying CAD, maybe I should workstation, either way your not referring to the average consumer.

I am waiting for the open source driver to mature more. It just sucks not being able to fully utilize what I paid for. Surely anyway can understand that. The main reason I purchased an AMD card was because they were open sourcing their driver. I think a lot of progress has been made on it and I don't claim to think it will happen in a short amount of time. I'm just tired of being forced to use Windows to fully utilize my hardware just because hardware manufacturers choose not to support Linux. I'm know they are things like market share and other such business crap that drive where support is given, but I'm a computer science major. I hate business. :) I just like my crap to work the way it's supposed to.

[doubledr]

fairly speaking, 9.7 solves the slow xv problem and I am very happy.

[bridgman]

Ok, maybe I took it a little out of context, but just to clarify these were your words. Go to page 6. :) Instead of saying CAD, maybe I should workstation, either way your not referring to the average consumer.

I had to go back to page 6 the first time to remember the context ;)

The point though is that I was responding to your suggestion that we scrap fglrx, and trying to make the point that even if it wasn't making consumer users totally happy yet scrapping it didn't make sense because fglrx had another market to serve.

I am waiting for the open source driver to mature more. It just sucks not being able to fully utilize what I paid for. Surely anyway can understand that.

You can fully utilize what you paid for today. You bought a card which advertised specific features and functions, and which said you required Windows in order to use those features and functions.

We are *also* bringing most of those features to Linux.

The main reason I purchased an AMD card was because they were open sourcing their driver. I think a lot of progress has been made on it and I don't claim to think it will happen in a short amount of time. I'm just tired of being forced to use Windows to fully utilize my hardware just because hardware manufacturers choose not to support Linux.

If you bought into the myth that we were just a bunch of idiots and that the open source community would be able to write a faster and more feature-rich driver in no time I'm sorry; I made a point of setting expectations low but there were a lot of wild statements floating around at the time. The reality is that modern graphics drivers are large and complicated, and that they take time to mature. We started working on consumer support about 18 months ago and have made a lot of progress, but it doesn't happen overnight and insulting the developers and their management doesn't actually seem to help, strangely enough ;)

I'm know they are things like market share and other such business crap that drive where support is given, but I'm a computer science major. I hate business. :) I just like my crap to work the way it's supposed to.

Welcome to the club. There's a reason I put up with the abuse :D

[Joe Sixpack]

I'm going to say something controversial, but I've been feeling like this for a long time now...

Most ATI users on this website need to have a tall glass of STFU! Yes, the driver has issues. Yes, you have to right to feel however you choose after you spend your money on hardware. But what annoys me is the lack of maturity amongst ATI users on this website. Even though fglrx has issues, the driver has had issues for years now (way before AMD bought ATI). So to expect a flawless driver experience is completely unrealistic. That's not making excuses for them - that's stating the obvious. When I bought my HD 3870 I intended on playing games in Windows, and I expected Linux support to be a work in progress. Which brings me to my second point:

Just because the feature you want isn't working doesn't make the new driver a complete waste. I'm sorry, but that line of reasoning is completely fucking childish. Catalyst 9.3 & 9.4 improved compositing support. Catalyst 9.4 - 9.6 each fixed bugs and significantly improved wine support (perfect example is HL 2 supports DirectX 9 now). So just like you can make the argument that theses drivers are worthless, I can make the argument that they're excellent because they fixed the bugs *I* was waiting on.

[Wintervenom]

Woo-hoo:
Looks like the linux driver was leaked for x86_64 people. It builds cleanly against 2.6.30 kernels and latest stable xorg. Performance is better than the recently released 9.7 driver.

Some more info:
OpenGL version string: 2.1.8975
2D driver version: 8.65.2
New extension: GLX_SGI_swap_control

http://www.simpleupload.net/download/448487/fglrx-8.650-ubuntu9.04-amd64only.7z.html

Source (http://forums.guru3d.com/showthread.php?t=301172) | OP Thread (http://bbs.archlinux.org/viewtopic.php?pid=590086#p590086)

[nanonyme]

Egh, so since not knowing what's inside closed drivers wasn't bad enough, now you also download unofficial releases of them from third-party websites. :D

[energyman]

I'm going to say something controversial, but I've been feeling like this for a long time now...

Most ATI users on this website need to have a tall glass of STFU! Yes, the driver has issues. Yes, you have to right to feel however you choose after you spend your money on hardware. But what annoys me is the lack of maturity amongst ATI users on this website. Even though fglrx has issues, the driver has had issues for years now (way before AMD bought ATI). So to expect a flawless driver experience is completely unrealistic. That's not making excuses for them - that's stating the obvious. When I bought my HD 3870 I intended on playing games in Windows, and I expected Linux support to be a work in progress. Which brings me to my second point:

Just because the feature you want isn't working doesn't make the new driver a complete waste. I'm sorry, but that line of reasoning is completely fucking childish. Catalyst 9.3 & 9.4 improved compositing support. Catalyst 9.4 - 9.6 each fixed bugs and significantly improved wine support (perfect example is HL 2 supports DirectX 9 now). So just like you can make the argument that theses drivers are worthless, I can make the argument that they're excellent because they fixed the bugs *I* was waiting on.

so true.
And even if 9.8 is going to work with 2.6.30, has perfect 2d, perfect xv and never locks up, people would still complain.

[hpestilence]

I created a diff showing the differences between the beta and 9.7. Maybe it can be used so that the 9.7 driver can support .29 or .30 kernels.

http://pastebin.com/f3e07714e

[Pfanne]

its been a while since i used my onboard ati gpu (6 months) but i remember terrible 2D performance with enabled composite...
and though i read the comments and news on every new fglrx release i dont know how well this works by now...
can someone enlighten me? ;)

[ZedDB]

I created a diff showing the differences between the beta and 9.7. Maybe it can be used so that the 9.7 driver can support .29 or .30 kernels.

http://pastebin.com/f3e07714e
It compiles and starts up X but it seems like i'm getting software rendering. Glxinfo get me this:
name of display: :0.0
X Error of failed request: BadRequest (invalid request code or no such operation)
Major opcode of failed request: 134 (GLX)
Minor opcode of failed request: 19 (X_GLXQueryServerString)
Serial number of failed request: 12
Current serial number in output stream: 12

[Fran]

Meh, after this release (still no support for 2.6.29? WTF?) and seeing as the open source drivers will get support for the remaining features in r6xx-7xx cards starting with 2.6.32 (KMS/power throttling/suspend/3D) I decided to remove "fglrx" from my VIDEO_CARDS variable and leave just "radeon radeonhd".

Goodbye forever, fglrx. I won't miss you.

[tball]

Meh, after this release (still no support for 2.6.29? WTF?) and seeing as the open source drivers will get support for the remaining features in r6xx-7xx cards starting with 2.6.32 (KMS/power throttling/suspend/3D) I decided to remove "fglrx" from my VIDEO_CARDS variable and leave just "radeon radeonhd".

Goodbye forever, fglrx. I won't miss you.

I guess you didn't see the former posts about a leaked 9.8 driver, which corfirms 2.6.30 in the next fglrx official release?

[NeoBrain]

I guess you didn't see the former posts about a leaked 9.8 driver, which corfirms 2.6.30 in the next fglrx official release?

well, then they still have to fix S2Disk and implement KMS in fglrx (which they don't want to do though)...

I wonder how hard it can be to fix S2Disk, it never worked for me with any fglrx release I tested, not even before 8.42.

[PuckPoltergeist]

I created a diff showing the differences between the beta and 9.7. Maybe it can be used so that the 9.7 driver can support .29 or .30 kernels.

http://pastebin.com/f3e07714e

Can this fix the binary blob? (rhetorical question)

[BlackStar]

For the record, 9.8 beta includes a package called xvba (guess what that is :p) and seems to improve video and 3d performance. Still no VSync on OpenGL 3 contexts, despite the introduction of GLX_SGI_swap_control.

I'll test suspend now.

Edit: (one reboot cycle later) No, suspend is thoroughly broken. With 9.7, you can suspend but resume hardlocks. With 9.8 you hardlock before suspending.

[Fran]

Edit: (one reboot cycle later) No, suspend is thoroughly broken. With 9.7, you can suspend but resume hardlocks. With 9.8 you hardlock before suspending.
And you can scratch suspend from my previous list: I've just tested using the gnome suspend dialog (instead of the custom script I used to use) and it works flawlessly with the OS driver :). So it's just KMS, power throttling and 3D (all of them very close to be supported).

[BlackStar]

And you can scratch suspend from my previous list: I've just tested using the gnome suspend dialog (instead of the custom script I used to use) and it works flawlessly with the OS driver :). So it's just KMS, power throttling and 3D (all of them very close to be supported).

I can't wait, actually. Unfortunately, it will probably take a few years to see any real 3d support (OpenGL 1.3/1.5 is pretty useless nowadays, I am talking about 2.1/3.0+).

[Muad'Dib]

For me the latest 9.7 drivers showed some progress.

Less Tearing, XV works a way better with composite on.

I am using Ubuntu 9.04 and I could fix a few problems with compiz by turning off "unredirect_fullscreen_windows" with gconf-edtior found in apps->compiz->general->screen0->options.

X-Server with no-backfill for Ubuntu 9.04, delivers a workaround for the resize problem. Downside is, Gnome/Compiz startup is really slow, but works fine after hidung/unhiding a few windows.

deb http://ppa.launchpad.net/ubuntu-x-swat/xserver-no-backfill/ubuntu jaunty main
deb-src http://ppa.launchpad.net/ubuntu-x-swat/xserver-no-backfill/ubuntu jaunty main

sudo apt-key adv --recv-keys --keyserver keyserver.ubuntu.com 0x643dc6bd56580ceb1ab4a9f63b22ab97af1cdfa9

The only real regressions I am experiencing now are:

broken standby/suspend
about 30% lower OpenGL performance compared to XP in ET: Quake Wars Still tearing
I can't enable PC-levels for XV
black screen protions
hiding/unhiding XV causes Video going black with compiz on


In general I can say about this release progress is there but it takes too long.

[skywarp04]

For those of you who are saying that people just need to STFU and quit complaining maybe you should take your own advice. Some people do have unreasonable expectations, however I haven't seen a lot of unreasonable expectations in this thread. To me an unreasonable expectation is someone expecting whatever game to work flawlessly in WINE. That is such a small segment of the Linux consumer market. I don't expect to be able to play games in Linux, yet. Here's hoping for the future though. I simply want a usable desktop. I'm glad fglrx works fine for you, but it does't for me. It doesn't even provide a decent desktop experience, with or without compositing. For me when fglrx is installed and compositing is off I can't even move a simple window around my screen without it jerking all over the place.

I don't think my expectations are that high. I simply want the following.

-Windows that behave the way they are supposed to. Moving them around and resizing them is smooth.

-Smooth xv video playback.

-Compositing working so I can have some simple Kwin OpenGL not XRender desktop effects. You can't even use half of Kwin's effects with XRender.

-Stability. No funky mysterious black artifacts that randomly appear on my screen or random self duplicating windows.

As it stands I have none of this when using fglrx. I know the open source driver is in development and when Karmic is released will probably fulfill all of my above expectations. I'm not bashing anyone's work or calling anyone incompetent when I state my complaints. I know some people do in their posts, but they shouldn't. I'm studying computer science and I'm a senior. I have lots of programming experience and knowledge of how hardware works. I know it's not a simple thing to get it working. A lot of people do not understand the amount of work that goes into making something simple work. Once I'm a graduate and no longer constantly tied up with frigain homework I will find ways to contribute to Linux then I'll be taking criticism for my work. :)

To summarize here's what I want that I can't do. Have my $200 video card work with Linux well enough that I can use Linux on a daily basis and then only boot into Windows when I want to play a game. Seeing as how fglrx is meant for workstations you would think that I could meet those expectations, except for the compositing. Anyway, everyone reading this that works on any driver for Linux keep up the good work.

[mirv]

just in case you haven't checked - is dri enabled (glxinfo | grep direct)? Because, if you're the only one with those issues, there could always be the possibility that your system isn't set up correctly.
Just a thought.

[t.s.]

..I'm not bashing anyone's work or calling anyone incompetent when I state my complaints. I know some people do in their posts, but they shouldn't. I'm studying computer science and I'm a senior. I have lots of programming experience and knowledge of how hardware works. I know it's not a simple thing to get it working. A lot of people do not understand the amount of work that goes into making something simple work. Once I'm a graduate and no longer constantly tied up with frigain homework I will find ways to contribute to Linux then I'll be taking criticism for my work. :)

you're looks funny, mister :D
I've seen somebody make comments on fglrx (that it unusable and is a piece of crap). Well, he's half right. When you buy something, you want to use it right at that moment. Not to use it 2 or 3 years from the time you bought that stuff, right? And for hardware like graphic card, when you bought the card, you too, received the driver, and of course the one that bought that item want to have a _working_ driver. If he's not, he should complaint. Oh, and people don't want to care/to know about how hard ones have to make ones product to work. They bought from you, and you're responsible for what you sold, that simple; The half that's not right is, he is to be blamed too, for not doing his homework, gathering information about the product that he want to buy.

But then again, if we have zero people like that, companies like AMD/ATI will give up on open sourcing their stuff.

And yeah, something like this to can be better because there's someone complaining, someone bicthing about it (all we have now partially because of that, right?)

So? what is right? what is wrong? what is shouldn't, what is should?

[ObiWan]

Woo-hoo:

Source (http://forums.guru3d.com/showthread.php?t=301172) | OP Thread (http://bbs.archlinux.org/viewtopic.php?pid=590086#p590086)
The leaked driver still needs EXPORT_SYMBOL(find_task_by_vpid); in kernel/pid.c to work with 2.6.31.
But other than that it seems to work quite well.

[skywarp04]

just in case you haven't checked - is dri enabled (glxinfo | grep direct)? Because, if you're the only one with those issues, there could always be the possibility that your system isn't set up correctly.
Just a thought.

Yes, I've been through all that. Glxinfo show the correct information. Fgl_glxgears gives good frame rates. I can even run Unigine benchmarks and get decent framerates, but using my computer for normal everyday tasks sucks when using fglrx. Like I said I can't even move a window around or resize it without problems.

At any rate I've given up using fglrx. I don't play games in Linux. I reverted back to the standard Jaunty radeon driver. It gives me everything I want except full OpenGL compositing. I have to use XRender which means I can only use a hand full of simple compositing effects in Kwin. This is acceptable for now. Thanks for the suggestion though mirv.

[Koorac]

Sadly this driver is no good for me. Random X hangups, resizing still slow, but finally less tearing in videos. Had to go back to 9.5, the only driver that works okay on my machine. Running openSuSE(x86_64) with xserver 1.6.2

[mirv]

Yes, I've been through all that. Glxinfo show the correct information. Fgl_glxgears gives good frame rates. I can even run Unigine benchmarks and get decent framerates, but using my computer for normal everyday tasks sucks when using fglrx. Like I said I can't even move a window around or resize it without problems.

At any rate I've given up using fglrx. I don't play games in Linux. I reverted back to the standard Jaunty radeon driver. It gives me everything I want except full OpenGL compositing. I have to use XRender which means I can only use a hand full of simple compositing effects in Kwin. This is acceptable for now. Thanks for the suggestion though mirv.

Ah well, it was worth a try!

I still can't open amdcccle (!dpy->xcb->reply_data failed is the latest error message) but I'm quite certain there's something different about my system that causes. Buggered if I know what it is though.
Might reinstall everything a-fresh one day, just to clean out my system and make sure the lot is up to date.

[TrentZ]

Woo-hoo:

Source (http://forums.guru3d.com/showthread.php?t=301172) | OP Thread (http://bbs.archlinux.org/viewtopic.php?pid=590086#p590086)

The download link is dead. Would anybody be kind enough to re-upload it? thanks!