Re: i915GM 2D+3D intel driver regression

Re: i915GM 2D+3D intel driver regression

[Sergio Monteiro Basto]

[-- Attachment #1.1: Type: text/plain, Size: 2669 bytes --]

Why don't send to the list , and to Marc Deop i Argemí.
I think you try compile intel-drv , but it is not easy , got kernel,
lib-drm, mesa , xserver and intel "Synced".
thanks, 

On Tue, 2010-05-04 at 18:45 -0700, SD wrote: 
> Actually, last update in Fedora 13:
> kernel-PAE-2.6.33.3-72.fc13.i686
> xorg-x11-drv-intel-2.11.0-3.fc13.i686
> made 2D works quick enough.
> 
> I can confirm - there was big improvement.
> 
> Thanks.
> 
> --- On Wed, 5/5/10, Sergio Monteiro Basto <sergio@sergiomb.no-ip.org> wrote:
> 
> > From: Sergio Monteiro Basto <sergio@sergiomb.no-ip.org>
> > Subject: Re: [Intel-gfx] i915GM 2D+3D intel driver regression
> > To: "Marc Deop i Argemí" <damnshock@gmail.com>
> > Cc: intel-gfx@lists.freedesktop.org
> > Date: Wednesday, May 5, 2010, 5:31 AM
> > On Sun, 2010-05-02 at 21:48 +0200,
> > Marc Deop i Argemí wrote: 
> > > On Friday April 30 2010 22:56:12 Clemens Eisserer
> > wrote:
> > > > And although so much has changed, I still
> > > > - Can't scroll through my gmail inbox "fluently",
> > not with firefox
> > > > nore with Opera
> > > > - Can't play even simple OpenGL games at
> > 1280x800
> > > > - Still have to disable Composition because it
> > dramatically slows down
> > > > window-resize performance.
> > > > 
> > > > So yep, it still sucks. However it doesn't crash
> > every 2 hours as it
> > > > did with intel-2.7/2.8.
> > > 
> > > I see I'm not the only one with this really bad
> > performance. I thought it was 
> > > my system which was buggy or somethings :S
> > > 
> > > My experience with an intel 945GM on linux has been
> > *TERRIBLE*. I was even 
> > > thinkig to switch back to Windows for some time (in
> > fact, I even got to 
> > > install Windows 7 although it didn't last long...).
> > > 
> > > However, I must thank the devs an intel because I feel
> > they are headed in the 
> > > right direction. Even though the performance is still
> > terrible, I see that 
> > > there are improvements with every driver release and
> > kernel update.
> > > 
> > > I hope I'll get a really usable driver in short :)
> > 
> > If you use intel-drv shipped on distros, should have
> > problems , do you
> > use ubuntu ?
> > 
> > I got a desktop with 945GM , don't have performance
> > problems , neither
> > locks or crashes .
> > 
> > 
> > 
> > -- 
> > Sérgio M. B.
> > 
> > -----Inline Attachment Follows-----
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx@lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/intel-gfx
> > 
> 
> 
>       

-- 
Sérgio M. B.

[-- Attachment #1.2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 3159 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[Sergio Monteiro Basto]

[-- Attachment #1.1: Type: text/plain, Size: 710 bytes --]

On Thu, 2010-05-06 at 10:47 +0200, Marc Deop i Argemí wrote: 
> > 
> > xorg-x11-drv-intel > n require kernel > n and libdrm > n .
> 
> What does that "n" mean? I don't understand that sentence.

n is a number 


> > So the order should be:
> > 1st kernel
> > 2nd libdrm
> > 3 mesa.
> > 4 Xserver
> > 5 all drives.
> 
> I've done all that as my distro packages are not that updated yet. I even got 
> the source from git. The problemes were still there...

Getting git doesn't solve the problem ... , 
Try for example "Intel 2010Q1 release"
Please check http://intellinuxgraphics.org/2010Q1.html for the
recommended stack and what's new in this release.

Regards, 
-- 
Sérgio M. B.

[-- Attachment #1.2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 3159 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[Marc Deop i Argemí]

On Wednesday May 5 2010 19:17:51 Sergio Monteiro Basto wrote:
> On Wed, 2010-05-05 at 15:59 +0200, Marc Deop i Argemí wrote:
> > On Wednesday May 5 2010 03:31:51 Sergio Monteiro Basto wrote:
> > > If you use intel-drv shipped on distros, should have problems , do you
> > > use ubuntu ?
> > 
> > Intel-drv? Don't have such a package... I should have problems? :S
> 
> xorg-x11-drv-intel > n require kernel > n and libdrm > n .

What does that "n" mean? I don't understand that sentence.

> 
> > I'm happily running Archlinux
> 
> happily ?

Yeah, happily. I love this distro :D

> So maybe is yours fault , because 945GM works since early drives without
> problems.

It does work. However, not fine performancewise. And I am not the only one to 
say that...

> Update only Xserver , the most probably that things stop work , after
> that you should recompile all drives (including xorg-x11-drv-intel)  ,
> see if xorg-x11-drv-intel requires a new libdrm and if libdrm requires a
> new mesa etc . If so before compile Xserver , you should compile libdrm
> and after that mesa.
> So the order should be:
> 1st kernel
> 2nd libdrm
> 3 mesa.
> 4 Xserver
> 5 all drives.

I've done all that as my distro packages are not that updated yet. I even got 
the source from git. The problemes were still there...
 
> Top-posting is to read only the top of the posting , 

I'm affraid you are mistaken. Top-posting is to write with the "replied" text 
below

> If you know the
> message , is more efficient . that looking for the answers in meddle of
> the text.

I'll give you that.

However, I tend to get dozens (sometimes hundreds) of emails per day: I 
usually *don't* know the message ;)

Damnshock

-- 
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

Re: i915GM 2D+3D intel driver regression

[Sergio Monteiro Basto]

[-- Attachment #1.1: Type: text/plain, Size: 1888 bytes --]

On Wed, 2010-05-05 at 15:59 +0200, Marc Deop i Argemí wrote: 
> On Wednesday May 5 2010 03:31:51 Sergio Monteiro Basto wrote:
> > If you use intel-drv shipped on distros, should have problems , do you
> > use ubuntu ?
> > 
> 
> Intel-drv? Don't have such a package... I should have problems? :S

xorg-x11-drv-intel > n require kernel > n and libdrm > n .


> 
> I'm happily running Archlinux
happily ? 

> > I got a desktop with 945GM , don't have performance problems , neither
> > locks or crashes .
> 
> I do have performance problems, specially on Kwin (kde window manager) but not 
> only on it.

So maybe is yours fault , because 945GM works since early drives without
problems.

> 
> Locks or crashes? Well, fullscreen opengl apps don't seem to work on my 
> external monitor as the screen  turns black after a few seconds(although it 
> don't really crashes as I can go to VT and back to get the screen back, weird 
> :S)
> 
> By the way, I had to go  back to  Xorg 1.7 instead of 1.8 because I had plenty 
> of freezes on my laptop :(

Update only Xserver , the most probably that things stop work , after
that you should recompile all drives (including xorg-x11-drv-intel)  , 
see if xorg-x11-drv-intel requires a new libdrm and if libdrm requires a
new mesa etc . If so before compile Xserver , you should compile libdrm
and after that mesa.
So the order should be:
1st kernel 
2nd libdrm
3 mesa.
4 Xserver
5 all drives. 

> 
> Now everything seems ok.
> A: Because it messes up the order in which people normally read text.
> Q: Why is top-posting such a bad thing?
> A: Top-posting.
> Q: What is the most annoying thing in e-mail?

Top-posting is to read only the top of the posting , If you know the
message , is more efficient . that looking for the answers in meddle of
the text.   

Regards, 
-- 
Sérgio M. B.

[-- Attachment #1.2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 3159 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[Marc Deop i Argemí]

On Wednesday May 5 2010 03:31:51 Sergio Monteiro Basto wrote:
> If you use intel-drv shipped on distros, should have problems , do you
> use ubuntu ?
> 

Intel-drv? Don't have such a package... I should have problems? :S

I'm happily running Archlinux

> I got a desktop with 945GM , don't have performance problems , neither
> locks or crashes .

I do have performance problems, specially on Kwin (kde window manager) but not 
only on it.

Locks or crashes? Well, fullscreen opengl apps don't seem to work on my 
external monitor as the screen  turns black after a few seconds(although it 
don't really crashes as I can go to VT and back to get the screen back, weird 
:S)

By the way, I had to go  back to  Xorg 1.7 instead of 1.8 because I had plenty 
of freezes on my laptop :(

Now everything seems ok.
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

Re: i915GM 2D+3D intel driver regression

[Sergio Monteiro Basto]

[-- Attachment #1.1: Type: text/plain, Size: 1384 bytes --]

On Sun, 2010-05-02 at 21:48 +0200, Marc Deop i Argemí wrote: 
> On Friday April 30 2010 22:56:12 Clemens Eisserer wrote:
> > And although so much has changed, I still
> > - Can't scroll through my gmail inbox "fluently", not with firefox
> > nore with Opera
> > - Can't play even simple OpenGL games at 1280x800
> > - Still have to disable Composition because it dramatically slows down
> > window-resize performance.
> > 
> > So yep, it still sucks. However it doesn't crash every 2 hours as it
> > did with intel-2.7/2.8.
> 
> I see I'm not the only one with this really bad performance. I thought it was 
> my system which was buggy or somethings :S
> 
> My experience with an intel 945GM on linux has been *TERRIBLE*. I was even 
> thinkig to switch back to Windows for some time (in fact, I even got to 
> install Windows 7 although it didn't last long...).
> 
> However, I must thank the devs an intel because I feel they are headed in the 
> right direction. Even though the performance is still terrible, I see that 
> there are improvements with every driver release and kernel update.
> 
> I hope I'll get a really usable driver in short :)

If you use intel-drv shipped on distros, should have problems , do you
use ubuntu ?

I got a desktop with 945GM , don't have performance problems , neither
locks or crashes .



-- 
Sérgio M. B.

[-- Attachment #1.2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 3159 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[Marc Deop i Argemí]

On Friday April 30 2010 22:56:12 Clemens Eisserer wrote:
> And although so much has changed, I still
> - Can't scroll through my gmail inbox "fluently", not with firefox
> nore with Opera
> - Can't play even simple OpenGL games at 1280x800
> - Still have to disable Composition because it dramatically slows down
> window-resize performance.
> 
> So yep, it still sucks. However it doesn't crash every 2 hours as it
> did with intel-2.7/2.8.

I see I'm not the only one with this really bad performance. I thought it was 
my system which was buggy or somethings :S

My experience with an intel 945GM on linux has been *TERRIBLE*. I was even 
thinkig to switch back to Windows for some time (in fact, I even got to 
install Windows 7 although it didn't last long...).

However, I must thank the devs an intel because I feel they are headed in the 
right direction. Even though the performance is still terrible, I see that 
there are improvements with every driver release and kernel update.

I hope I'll get a really usable driver in short :)


-- 
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

Re: i915GM 2D+3D intel driver regression

[Sergio Monteiro Basto]

[-- Attachment #1.1: Type: text/plain, Size: 1648 bytes --]

On Fri, 2010-04-30 at 22:56 +0200, Clemens Eisserer wrote: 
> > Oh, I get it.  You don't understand software.
> 
> Well although the whole discussion isn't really focussed, he has one point.
> The overall desktop experience is way better on the old terminals
> running Fedora-7 (intel-2.2 / XAA) than on my i945GM based laptop
> running Fedora-12 + intel-2.11.
> 
> And although so much has changed, I still
> - Can't scroll through my gmail inbox "fluently", not with firefox
> nore with Opera
> - Can't play even simple OpenGL games at 1280x800
> - Still have to disable Composition because it dramatically slows down
> window-resize performance.
> 
> So yep, it still sucks. However it doesn't crash every 2 hours as it
> did with intel-2.7/2.8.


Has a owner of one 915GM

cat /var/log/Xorg.0.log |  grep -i chipset
(II) intel: Driver for Intel Integrated Graphics Chipsets: i810,
(II) intel(0): Integrated Graphics Chipset: Intel(R) 915GM
(--) intel(0): Chipset: "915GM"

I use Fedora 12 + intel-2.9.1 (rpm -qi xorg-x11-drv-intel)
Name        : xorg-x11-drv-intel
Version     : 2.9.1
Vendor: Fedora Project
Release     : 1.fc12                     
Build Date: Mon 26 Oct 2009 03:30:37 PM WET

And things work pretty well, composite, quake , 3D games etc. 

Just fedora updated (out of the box)
kernel-PAE-2.6.32.11-99.fc12.i686
libdrm-2.4.18-2.fc12.i686
xorg-x11-server-Xorg-1.7.6-3.fc12.i686


> - Clemens
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Sérgio M. B.

[-- Attachment #1.2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 3159 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[Sven Arvidsson]

[-- Attachment #1.1: Type: text/plain, Size: 418 bytes --]

Why this need to assume malignancy and throw a lot of accusations
around?

I just gave the examples you mentioned a try on my i915 (with up to date
drivers) and I can't reproduce this at all. Performance is good with
gears, and CPU usage when playing video low.

Chill, count to ten, punch a pillow, and then file a proper bug report?

-- 
Cheers,
Sven Arvidsson
http://www.whiz.se
PGP Key ID 760BDD22


[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[Adam Jackson]

[-- Attachment #1.1: Type: text/plain, Size: 739 bytes --]

On Fri, 2010-04-30 at 22:56 +0200, Clemens Eisserer wrote:
> > Oh, I get it.  You don't understand software.
> 
> Well although the whole discussion isn't really focussed, he has one point.

That's as may be true.  But the leap from "it's worse" to "because it
supports so many devices" is just nonsense.

> - Still have to disable Composition because it dramatically slows down
> window-resize performance.

That's really a property of the server.  Every resize is an allocate /
copy / free cycle when compositing.  Which is then made worse because
the app inevitably ends up repainting the whole thing, so the copy was a
waste of time.

A modest amount of hysteresis on window sizing would go a long way here.

- ajax

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[Clemens Eisserer]

> Oh, I get it.  You don't understand software.

Well although the whole discussion isn't really focussed, he has one point.
The overall desktop experience is way better on the old terminals
running Fedora-7 (intel-2.2 / XAA) than on my i945GM based laptop
running Fedora-12 + intel-2.11.

And although so much has changed, I still
- Can't scroll through my gmail inbox "fluently", not with firefox
nore with Opera
- Can't play even simple OpenGL games at 1280x800
- Still have to disable Composition because it dramatically slows down
window-resize performance.

So yep, it still sucks. However it doesn't crash every 2 hours as it
did with intel-2.7/2.8.

- Clemens
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[Adam Jackson]

[-- Attachment #1.1: Type: text/plain, Size: 539 bytes --]

On Fri, 2010-04-30 at 12:25 -0700, SD wrote:

> But, why not to split intel driver between old cards and new ones?
> Every one knows that it is impossible to provide full support for:
>  810, i810-dc100, i810e, i815, i830M, 845G, 852GM/855GM, 865G, 915G,
>  E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM, Pineview G,
>  965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33, GM45,
>  4 Series, G45/G43, Q45/Q43, G41, B43, Clarkdale, Arrandale
> in one driver.

Oh, I get it.  You don't understand software.

- ajax

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: i915GM 2D+3D intel driver regression

[SD]

Dear All.

I perfectly understand this situation. Of cause everything changes and new chipsets will appear every next year. I also like how kernel .33 works.

But, why not to split intel driver between old cards and new ones? Every one knows that it is impossible to provide full support for:
 810, i810-dc100, i810e, i815, i830M, 845G, 852GM/855GM, 865G, 915G,
 E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM, Pineview G,
 965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33, GM45,
 4 Series, G45/G43, Q45/Q43, G41, B43, Clarkdale, Arrandale
in one driver.

Why it was difficult to compile two or may be three driver, and one of them especially with XAA module. Why it is necessary to ruin everything what left behind then move forward. Stupidest idea. And now we have nothing - awful 2D, and the same 3D.

So I have a question to everyone:
It it possible to compile old X.org with old intel driver with XAA for 6.33 kernel?



About my post, regarding Fedora 13: that problem was as well in Fedora 12, and Fedora it self is not a problem, problem UXA and new intel driver.
And I think new Ubuntu or OpenSuse (which I use right now) will have completely same behavior.


Thank you.




--- On Fri, 4/30/10, Alan W. Irwin <irwin@beluga.phys.uvic.ca> wrote:

> From: Alan W. Irwin <irwin@beluga.phys.uvic.ca>
> Subject: Re: [Intel-gfx] i915GM 2D+3D intel driver regression
> To: "SD" <sd.domrep@yahoo.com>
> Cc: intel-gfx@lists.freedesktop.org
> Date: Friday, April 30, 2010, 10:28 PM
> On 2010-04-30 08:45-0700 SD wrote:
> 
> > Dear all.
> >
> > I have been using linux for 2 years already. And I use
> it on intel 915 GM video card on Lenovo laptop. With:
> > (II) Loading
> /usr/lib/xorg/modules//drivers/intel_drv.so
> > (II) Module intel: vendor="X.Org Foundation"
> >     compiled for 1.5.2, module version
> = 2.5.0
> >     Module class: X.Org Video Driver
> >     ABI class: X.Org Video Driver,
> version 4.1
> > And acceleration module XAA when I watch Xvid movie on
> full screen, according to TOP:
> > X uses 4% of CPU
> > SMplayer 10-12% of CPU.
> >
> > GLXgear gives me:
> > 3216 frames in 5.0 seconds = 643.117 FPS
> >
> >
> >
> > Now I tried Fedora13 (test) with intel driver:
> > [    27.854] (II) Loading
> /usr/lib/xorg/modules/drivers/intel_drv.so
> > [    27.855] (II) Module intel:
> vendor="X.Org Foundation"
> > [    27.855]  compiled for 1.8.0,
> module version = 2.11.0
> > [    27.855]  Module class: X.Org Video
> Driver
> > [    27.855]  ABI class: X.Org Video
> Driver, version 7.0
> >
> > And can you imagine GLXgear gives me:
> > 165 frames in 5.1 seconds = 32.612 FPS
> >
> > Watching Xvid movie:
> > X uses ~40% of CPU
> > SMplayer ~10-20% of CPU.
> > Even when I switch workspaces X11 uses ~20% of CPU -
> there is no any 2D acceleration at all
> >
> > After all of this I would like to ask:
> > Do you respect customer who use linux?
> > Does any one check your driver and UXA with i915?
> > Why, just why developers through away XAA from driver,
> your UXA works the same as EXA did - awful. Awful with 3D
> and more important, awful with 2D.
> >
> > Why dev. can't just leave what was done good for
> i915?
> > Why it was necessary to screw everything. Looks like
> you just put EXA to UXA.
> >
> > I do not know about other intel chipsets, but i915
> works really slow with new and previous driver on UXA.
> >
> > So, for i915GM new driver is BIG BIG REGRESSION and
> big step backward.
> 
> Personally, I think you were a little hard on the Intel
> developers.  I think
> we should all give them some slack so they have the freedom
> to get on with
> the job of the huge X stack changes that have been
> necessary over the last
> several years to deal with the capabilities of modern video
> chipsets
> (including Intel ones).
> 
> However, I think those developers are entirely on your side
> that
> _eventually_ these large X stack changes should be refined
> to the point that
> they will not severely impact older hardware
> performance.  For example,
> there have been reassurances in the past from the Intel
> developers on
> exactly this point.  Clearly, from your xvid and
> smplayer numbers (they will
> dismiss glxgears numbers for reasons that have been stated
> many times in the
> past) they are currently doing poorly at this job, and that
> is quite
> worrying.  For example, I am sticking to XAA for my
> older g33 Intel video
> chipset using the Debian stable X stack because of speed
> and stability
> concerns with the new X stack and new intel driver, and
> your post has
> reaffirmed that decision.  But both of us (and all the
> other users of older
> Intel hardware out there) cannot use old distributions
> forever so I hope
> that the Intel developers are reassuring once again in
> answer to your post
> that _soon_ (rather than "eventually") they will address
> the real-world (as
> opposed to glxgears) performance regressions compared to
> the old X stack and
> XAA.
> 
> Alan
> __________________________
> Alan W. Irwin
> 
> Astronomical research affiliation with Department of
> Physics and Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
> 
> Programming affiliations with the FreeEOS equation-of-state
> implementation
> for stellar interiors (freeeos.sf.net); PLplot scientific
> plotting software
> package (plplot.org); the libLASi project
> (unifont.org/lasi); the Loads of
> Linux Links project (loll.sf.net); and the Linux Brochure
> Project
> (lbproject.sf.net).
> __________________________
> 
> Linux-powered Science
> __________________________
> 


      

Re: i915GM 2D+3D intel driver regression

[Alan W. Irwin]

On 2010-04-30 08:45-0700 SD wrote:

> Dear all.
>
> I have been using linux for 2 years already. And I use it on intel 915 GM video card on Lenovo laptop. With:
> (II) Loading /usr/lib/xorg/modules//drivers/intel_drv.so
> (II) Module intel: vendor="X.Org Foundation"
> 	compiled for 1.5.2, module version = 2.5.0
> 	Module class: X.Org Video Driver
> 	ABI class: X.Org Video Driver, version 4.1
> And acceleration module XAA when I watch Xvid movie on full screen, according to TOP:
> X uses 4% of CPU
> SMplayer 10-12% of CPU.
>
> GLXgear gives me:
> 3216 frames in 5.0 seconds = 643.117 FPS
>
>
>
> Now I tried Fedora13 (test) with intel driver:
> [    27.854] (II) Loading /usr/lib/xorg/modules/drivers/intel_drv.so
> [    27.855] (II) Module intel: vendor="X.Org Foundation"
> [    27.855]  compiled for 1.8.0, module version = 2.11.0
> [    27.855]  Module class: X.Org Video Driver
> [    27.855]  ABI class: X.Org Video Driver, version 7.0
>
> And can you imagine GLXgear gives me:
> 165 frames in 5.1 seconds = 32.612 FPS
>
> Watching Xvid movie:
> X uses ~40% of CPU
> SMplayer ~10-20% of CPU.
> Even when I switch workspaces X11 uses ~20% of CPU - there is no any 2D acceleration at all
>
> After all of this I would like to ask:
> Do you respect customer who use linux?
> Does any one check your driver and UXA with i915?
> Why, just why developers through away XAA from driver, your UXA works the same as EXA did - awful. Awful with 3D and more important, awful with 2D.
>
> Why dev. can't just leave what was done good for i915?
> Why it was necessary to screw everything. Looks like you just put EXA to UXA.
>
> I do not know about other intel chipsets, but i915 works really slow with new and previous driver on UXA.
>
> So, for i915GM new driver is BIG BIG REGRESSION and big step backward.

Personally, I think you were a little hard on the Intel developers.  I think
we should all give them some slack so they have the freedom to get on with
the job of the huge X stack changes that have been necessary over the last
several years to deal with the capabilities of modern video chipsets
(including Intel ones).

However, I think those developers are entirely on your side that
_eventually_ these large X stack changes should be refined to the point that
they will not severely impact older hardware performance.  For example,
there have been reassurances in the past from the Intel developers on
exactly this point.  Clearly, from your xvid and smplayer numbers (they will
dismiss glxgears numbers for reasons that have been stated many times in the
past) they are currently doing poorly at this job, and that is quite
worrying.  For example, I am sticking to XAA for my older g33 Intel video
chipset using the Debian stable X stack because of speed and stability
concerns with the new X stack and new intel driver, and your post has
reaffirmed that decision.  But both of us (and all the other users of older
Intel hardware out there) cannot use old distributions forever so I hope
that the Intel developers are reassuring once again in answer to your post
that _soon_ (rather than "eventually") they will address the real-world (as
opposed to glxgears) performance regressions compared to the old X stack and
XAA.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the libLASi project (unifont.org/lasi); the Loads of
Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: i915GM 2D+3D intel driver regression

[Felix Miata]

On 2010/04/30 08:45 (GMT-0700) SD composed:

> I have been using linux for 2 years already. And I use it on intel 915 GM video card:
> X uses 4% of CPU
> SMplayer 10-12% of CPU.

> GLXgear gives me:
> 3216 frames in 5.0 seconds = 643.117 FPS

> Now I tried Fedora13 (test) with intel driver:

> And can you imagine GLXgear gives me:
> 165 frames in 5.1 seconds = 32.612 FPS

> Watching Xvid movie:
> X uses ~40% of CPU
> SMplayer ~10-20% of CPU.

> So, for i915GM new driver is BIG BIG REGRESSION and big step backward.

This regressive behavior is well known. F13 has not been released, so it's
betaware. If you can't deal with bugs then don't replace what's working OK
with development versions. Newer isn't always better. If it's working, don't
try to "fix" it by "upgrading" to betaware. Meanwhile, someone must install
the betaware to find the bugs. If you think your experience is unique, report
in the appropriate manner. Did you look in bugs.freedesktop.org and/or
bugzilla.redhat.com before installing F13 or comming here? Complaining here
about a "big regression" is unlikely to be much help in getting it fixed.
-- 
"Suppos [sic] a nation in some distant region, should
take the Bible for their only law book, and every member
should regulate his conduct by the precepts there
exhibited. . . . What a Eutopa, What a paradise would
this region be!"            John Adams, 2nd US President

 Team OS/2 ** Reg. Linux User #211409

Felix Miata  ***  http://fm.no-ip.com/

i915GM 2D+3D intel driver regression

[SD]

Dear all.

I have been using linux for 2 years already. And I use it on intel 915 GM video card on Lenovo laptop. With:
(II) Loading /usr/lib/xorg/modules//drivers/intel_drv.so
(II) Module intel: vendor="X.Org Foundation"
	compiled for 1.5.2, module version = 2.5.0
	Module class: X.Org Video Driver
	ABI class: X.Org Video Driver, version 4.1
And acceleration module XAA when I watch Xvid movie on full screen, according to TOP:
X uses 4% of CPU
SMplayer 10-12% of CPU.

GLXgear gives me:
3216 frames in 5.0 seconds = 643.117 FPS



Now I tried Fedora13 (test) with intel driver:
[    27.854] (II) Loading /usr/lib/xorg/modules/drivers/intel_drv.so
[    27.855] (II) Module intel: vendor="X.Org Foundation"
[    27.855]  compiled for 1.8.0, module version = 2.11.0
[    27.855]  Module class: X.Org Video Driver
[    27.855]  ABI class: X.Org Video Driver, version 7.0

And can you imagine GLXgear gives me:
165 frames in 5.1 seconds = 32.612 FPS

Watching Xvid movie:
X uses ~40% of CPU
SMplayer ~10-20% of CPU.
Even when I switch workspaces X11 uses ~20% of CPU - there is no any 2D acceleration at all

After all of this I would like to ask:
Do you respect customer who use linux?
Does any one check your driver and UXA with i915?
Why, just why developers through away XAA from driver, your UXA works the same as EXA did - awful. Awful with 3D and more important, awful with 2D.

Why dev. can't just leave what was done good for i915?
Why it was necessary to screw everything. Looks like you just put EXA to UXA.

I do not know about other intel chipsets, but i915 works really slow with new and previous driver on UXA.

So, for i915GM new driver is BIG BIG REGRESSION and big step backward.