| View previous topic :: View next topic |
| Author |
Message |
coolsnowmen
Veteran
Joined: 30 Jun 2004
Posts: 1479
Location: No.VA
|
 Posted: Thu Jul 20, 2006 6:11 pm Post subject: |
 |
|
I have read this solution...but for somepeople disabling that acceleration is not considered "a solution" (people who need opengl acceleration ...)
I'm sticking to 7.0 for now, I've not heard of anything special in 7.1... |
|
| Back to top |
|
 |
nahpets
Veteran
Joined: 05 Oct 2003
Posts: 1178
Location: Montreal, Canada
|
| Posted: Thu Jul 20, 2006 7:10 pm Post subject: |
|
|
I did do a search and didn't find the threads that you linked to. The first thread you linked to has nothing do do with my problem... the second one is more useful though, thanks.
| coolsnowmen wrote: | I have read this solution...but for somepeople disabling that acceleration is not considered "a solution" (people who need opengl acceleration ...)
I'm sticking to 7.0 for now, I've not heard of anything special in 7.1... |
I upgraded to 7.1 because my ATI driver was broken... I think that having ATI and Nvidia cards living in the same box together isn't such a great idea...
_________________
Let me guess, you picked out yet another colorful box with a crank that I'm expected to turn and turn until OOP! big shock, a jack pops out and you laugh and the kids laugh and the dog laughs and I die a little inside. |
|
| Back to top |
|
|
coolsnowmen
Veteran
Joined: 30 Jun 2004
Posts: 1479
Location: No.VA
|
| Posted: Thu Jul 20, 2006 7:16 pm Post subject: |
|
|
| nahpets wrote: |
I upgraded to 7.1 because my ATI driver was broken... I think that having ATI and Nvidia cards living in the same box together isn't such a great idea... |
It makes you for an excellent test case.... 
"always look on the bright side of life" |
|
| Back to top |
|
|
nahpets
Veteran
Joined: 05 Oct 2003
Posts: 1178
Location: Montreal, Canada
|
| Posted: Thu Jul 20, 2006 7:40 pm Post subject: |
|
|
Tell me about it. I've had a lot of problems getting them to co-exist. I was never able to install the binary drivers for nvidia and ATI (flgrx) and have them work together. I opted for using the Nvidia binary driver on my primary display and the Xorg 'ati' drivers. Lately, with all the upgrades to Xorg, it seems that fixing one always breaks the other. For the time being I have both monitors working with 'nv' and 'ati' drivers.
GLX is working with MESA:
| Code: |
(==) AIGLX enabled
(II) Loading extension GLX
(II) GLX: Initialized MESA-PROXY GL provider for screen 0
(II) GLX: Initialized MESA-PROXY GL provider for screen 1
|
But it's WAY slower than NVidia glx.
_________________
Let me guess, you picked out yet another colorful box with a crank that I'm expected to turn and turn until OOP! big shock, a jack pops out and you laugh and the kids laugh and the dog laughs and I die a little inside. |
|
| Back to top |
|
|
tomk
Bodhisattva
Joined: 23 Sep 2003
Posts: 7221
Location: Sat in front of my computer
|
|
| Back to top |
|
|
slycordinator
Advocate
Joined: 31 Jan 2004
Posts: 3065
Location: Korea
|
| Posted: Thu Jul 20, 2006 11:44 pm Post subject: Re: No Reason For Unmasked? |
|
|
| s1n wrote: | | What are the rules that determines what packages are masked in the official portage tree? I do not understand why if so many people are reporting problems that the latest xorg server has not been masked. It may be many more moons before NVidia updates their drivers, so I cannot perform any updates unless I manually mask it myself? Someone really dropped the ball on this one; who's responsible for this?? |
So you want xorg-server masked because nvidia can't make a timely update?
That would be like blaming AMD if Zalman didn't carry a fan to fit on AMD's newest chip. |
|
| Back to top |
|
|
amdunlock
Guru
Joined: 24 Apr 2004
Posts: 409
Location: Darmstadt Germany
|
| Posted: Fri Jul 21, 2006 11:18 am Post subject: |
|
|
is there a statement of the part of nvidia when they will release their 9XXX driver?
Greetings, Oliver
_________________
-2500+ @ 3200+ ---- 512MB-Ram watercooled Htf X Triple @ 3 x 5V - Abit AN7 s*cking microguru - Geforce 6800 - Audigy 2 -=Gentoo Stage 2004.2 @ ext3=- |
|
| Back to top |
|
|
boroshan
l33t
Joined: 16 Apr 2003
Posts: 730
Location: upside down
|
| Posted: Fri Jul 21, 2006 10:14 pm Post subject: Re: No Reason For Unmasked? |
|
|
| slycordinator wrote: | | That would be like blaming AMD if Zalman didn't carry a fan to fit on AMD's newest chip. |
Not a bad analogy, but it needs a little bit of work before it will fly.
I mean, we have to suppose that AMD had a way of autmatically "upgrading" customer chips. We need to suppose that this was done via a routine and trusted maintenance process, one set up by AMD in the first place.
Then we need need to stipulate that the AMD devs used this process to change the characteristics of the chips their customers had installed so that all the Zalman fans stopped working, thereby breaking a sizeable percentage of their customers computers. If we postulate that, then the analogy just about works.
It's still not perfect though. NVidia and ATI cards are rather more widely deployed than Zalman fans, and while a Zalman fan is almost always the result of a deliberate user upgrade, an awfully large number of computers come with Nvidia or ATI pre installed. So we have to assume that Zalman fans are a lot more widely deployed than in fact they are. We also need to assume that they are often bonded to the motherboard, the chassis or both, since the vast majority of laptop users can't realistically change their graphics chipset. Let's add those qualifications to the list.
The last change I need to make to your analogy is that we need to assume that all the Gentoo users are using AMD chips. Unless of course you meant the AMD/Intel aspect of the analogy to refer to the stable/unstable split.
So yes. Given that you're proposing a scenario where AMD intentionally brought about a change that brought a significant percentage of their userbase to needless inconvenience due to an upgrade that was known in advance to have serious incompatibility issues, given that then yes. Yes, I would be very annoyed indeed. I don't think that would be unreasonable under the circumstances.
Of course, it's hardly likely that AMD would do such a thing. If they did though, I would want to know why. My desktop machine is getting a little elderly now, but the Athlon XP chip that drives it has never given me any noticable trouble. So if it suddenly started refusing to be cooled by the fan that's been perfectly fine for these past four years, and through no change on the part of the fan, well then I'd want to know why. I don't think that's particularly unreasonable either.
In the non-metaphorical case of Gentoo and X7.1, I also want to know why the decision was made. The devs knew that this would break unstable ATI and NVidia cards, and they did it anyway. So far I've seen no explanation of what was gained by this exercise that couldn't have been achieved by a hard mask and/or an overlay.
I searched the forums for an explanation, The closest I came were a few sneering remarks about closed source graphics drivers. It leads me to wonder: was this a purely political decision, or was there a useful purpose served by the exercise. Call me idealistic if yoiu must, but I would like to think I didn't lose two weeks from my development schedule purely because someone wanted to pressure two graphic card companies, and couldn't think of a way to do so that didn't involve pissing off their own users and hoping that we all blamed Nvidia.
I raised these issues once before on this thread. I didn't persue them before because I didn't want to start a flame war in a support thread. I still don't. But I do feel this decision should be examined publically, and I would very much like to be a part of that debate.
If anyone knows the proper forum for such issues, please tell me.
[edit - removed a minor redundancy ]
_________________
Don't let THEM immanentize the Eschaton! |
|
| Back to top |
|
|
slycordinator
Advocate
Joined: 31 Jan 2004
Posts: 3065
Location: Korea
|
| Posted: Sat Jul 22, 2006 12:27 am Post subject: Re: No Reason For Unmasked? |
|
|
| boroshan wrote: | | I mean, we have to suppose that AMD had a way of autmatically "upgrading" customer chips. We need to suppose that this was done via a routine and trusted maintenance process, one set up by AMD in the first place. |
My analogy is simply referring to the fact that one product changed and another group ignored those changes even though their product needed to work with the other.
| Quote: | | Then we need need to stipulate that the AMD devs used this process to change the characteristics of the chips their customers had installed so that all the Zalman fans stopped working, thereby breaking a sizeable percentage of their customers computers. If we postulate that, then the analogy just about works. |
Look at the analogy from the other end.
Lets postulate that AMD announced that they were making changes and gave notice as to exactly what these changes were. Then lets say Zalman didn't do anything to make it so their product worked with these changes.
This is what happened with xorg and ati/nvidia. They had to have known that ABI changes were coming. It's not like one day, the devs made 5,000 changes and there was no inkling of this coming.
| Quote: | | The last change I need to make to your analogy is that we need to assume that all the Gentoo users are using AMD chips. Unless of course you meant the AMD/Intel aspect of the analogy to refer to the stable/unstable split. |
So Matrox doesn't exist?
| Quote: | | So yes. Given that you're proposing a scenario where AMD intentionally brought about a change that brought a significant percentage of their userbase to needless inconvenience due to an upgrade that was known in advance to have serious incompatibility issues, given that then yes. Yes, I would be very annoyed indeed. I don't think that would be unreasonable under the circumstances. |
This is crazy.
This is what happened: Changes happened to xorg. The people at ati/nvidia knew that this would happen and knew which changes would be occurring and did nothing.
Once in a while changes happen to the kernel and the nvidia-kernel no longer will install. When that has happened people expected nvidia to make changes to their driver. Also, when that happens I don't go and blame Linus.
| Quote: | | In the non-metaphorical case of Gentoo and X7.1, I also want to know why the decision was made. The devs knew that this would break unstable ATI and NVidia cards, and they did it anyway. So far I've seen no explanation of what was gained by this exercise that couldn't have been achieved by a hard mask and/or an overlay. |
In other words:
You feel that because xorg-7.1 doesn't work with nvidia/ati drivers, it should be hardmasked.
And instead, what gentoo did was say:
There's nothing at all wrong with xorg-7.1 itself. ati/nvidia drivers don't work with it so we'll make it so the 2 of them can't be installed together.
| Quote: | | I searched the forums for an explanation, The closest I came were a few sneering remarks about closed source graphics drivers. It leads me to wonder: was this a purely political decision, or was there a useful purpose served by the exercise. Call me idealistic if yoiu must, but I would like to think I didn't lose two weeks from my development schedule purely because someone wanted to pressure two graphic card companies, and couldn't think of a way to do so that didn't involve pissing off their own users and hoping that we all blamed Nvidia. |
~x86 = testing branch for x86
Perhaps they put it in the testing branch to get more people testing it?
Furthermore, what exactly made you lose 2 weeks of work? Unless I'm mistaken, when they put xorg-7.1 into the testing branch they also put in the blocks for ati-drivers and nvidia-kernel. |
|
| Back to top |
|
|
boroshan
l33t
Joined: 16 Apr 2003
Posts: 730
Location: upside down
|
| Posted: Sat Jul 22, 2006 4:31 pm Post subject: |
|
|
| Quote: |
This is what happened with xorg and ati/nvidia. They had to have known that ABI changes were coming. It's not like one day, the devs made 5,000 changes and there was no inkling of this coming.
|
So what happened? Did ATI and NVidia say "we have no intention of ever support 7.1" or somesuch? I didn't see anything to that effect, but maybe I wasn't paying attention. If they did say that then I can understand why drastic action may have seemed appropriate. I would have appreciated the courtesy of a little advance warning, as I expect would many others, but I could see how such action might have seemed necessary. The only problem here is that if ATI and NVidia had refused to support 7.1, I think the flame wars would still be raging over on Slashdot and the like. So I'm guessing that this isn't actually the case
Now, if ATI and NVidia haven't refused to support 7.1 then don't you perhaps think the devs were a little premature in unmasking 7.1? That perhaps we could have waited until new drivers were ready before unleashing this sucker on the world. I mean it was only, what? three, four weeks since 7.0? It's not like we were lagging behind.
The gripping hand is that even if we were to agree that NVidia and ATI were in the wrong, I still don't see how that justifies releasing an update with known serious compatibility issues. Without warning at that. If they did this with network cards, there'd be hell to pay, Why should graphics cards be any different?
| slycordinator wrote: |
So Matrox doesn't exist?
|
Don't be silly, of course Matrox exists. So does Transmeta if you want to further burden an already shaky metaphor. What's your point?
| slycordinator wrote: |
Once in a while changes happen to the kernel and the nvidia-kernel no longer will install. When that has happened people expected nvidia to make changes to their driver. Also, when that happens I don't go and blame Linus.
|
Quite right. Getting annoyed at Linus in that scenario would be like me criticising X.Org in this one. That would be silly. It's a good job I'm not doing that, really.
| slycordinator wrote: |
~x86 = testing branch for x86
Perhaps they put it in the testing branch to get more people testing it?
|
Lots of problems there. For one thing, take a quick trawl through these forums and see how problems are "solved" by upgrading to unstable. Similarly for bugzilla. Then take a look at the wiki and see how many HOWTOs start with "install the unstable version..." So, regardless of the original intent, I don't think ~x86 can be said to be just for testing any longer.
Then there is the question of what they were testing for. They knew before they released 7.1 that it was going to break so it's not like they'd get a lot of useful test data from that part of the userbase.
The quality of the test data they got is also questionable. Gentoo has a tradition of debugging new software using private overlays and small, dedicated test groups, and the approach seems to work very well. GCC 3.4 and 4.0 both went live as a result of such groups. So did bootsplash, and later gensplash. Look through the forums - I'm sure you'll find others
But the biggest problem with that line of thought is that it places a greater importance on test data than it does on users having working desktops. I don't think that's a sensible ordering of priorities myself.
| slycordinator wrote: |
Furthermore, what exactly made you lose 2 weeks of work? Unless I'm mistaken, when they put xorg-7.1 into the testing branch they also put in the blocks for ati-drivers and nvidia-kernel. |
heh. This is a support thread for people who had their desktops broken by X.Org 7.1. There were enough people affected that they made the thread sticky to help folks find it. It's quite a long thread.
How well do you think those blocks might have worked?
You obviously haven't been affected personally by this issue, so maybe you should go and read through this thread from the start. There's quite a bit more too it than just not being able to run emerge -avuD world
_________________
Don't let THEM immanentize the Eschaton! |
|
| Back to top |
|
|
sternklang
Veteran
Joined: 10 Sep 2005
Posts: 1641
Location: Somewhere in time and space
|
| Posted: Sat Jul 22, 2006 4:56 pm Post subject: |
|
|
| Could I suggest that as this is a support forum, the discussion on what might or might not have been done regarding xorg 7.1 masking/blocking/etc. should be continued in Gentoo Chat? It's not really relevant to people trying to work around problems with the ABI change. |
|
| Back to top |
|
|
boroshan
l33t
Joined: 16 Apr 2003
Posts: 730
Location: upside down
|
| Posted: Sat Jul 22, 2006 6:05 pm Post subject: |
|
|
| sternklang wrote: | Could I suggest that as this is a support forum, the discussion on what might or might not have been done regarding xorg 7.1 masking/blocking/etc. should be continued in Gentoo Chat? It's not really relevant to people trying to work around problems with the ABI change.
|
And I did say I didn't want to clutter up a support forum with a flame war. Therefore...
Anyone who wants to discuss the whys and wherefores behind this issue, please comment on the new thread.
_________________
Don't let THEM immanentize the Eschaton! |
|
| Back to top |
|
|
STEDevil
Apprentice
Joined: 24 Apr 2003
Posts: 156
|
| Posted: Sun Jul 23, 2006 9:03 pm Post subject: |
|
|
| sedorox wrote: | | Xithix wrote: | Why did this go ~arch? 7.0->7.1 caused massive breakage of binary video drivers. It should still be masked, as a *LOT* of users, probably the majority, use these drivers.
Now I have to manually mask it and re-downgrade... |
Mmmm, I think most users running ~arch will check before they upgrade...
|
What about users running arch that just followed the modular-x-howto that STILL specifies that you should add the following list of 296(sic!) packages to your package.keywords without fixed version numbers.
Of course since I read it there has been added a (quite fuzzy) warning about 7.1, but its not like one rereads every single howto you have ever read in your life before you make an emerge --sync and update world.
Then again this is by far not the first time arch users of Gentoo get fubared systems because specific instructions on official Gentoo pages mixed with later updates become highly volatile/unstable since "the right hand is not aware of what the left hand is doing". |
|
| Back to top |
|
|
drohr
n00b
Joined: 29 Sep 2003
Posts: 26
Location: Stockholm, Sweden
|
| Posted: Sat Jul 29, 2006 4:30 am Post subject: |
|
|
Using the new ATI drivers released yesterday (8.27.10), and they actually work great with Xorg 7.1 with DRI.
No ebuild yet, but works fine to download manually and install. |
|
| Back to top |
|
|
sternklang
Veteran
Joined: 10 Sep 2005
Posts: 1641
Location: Somewhere in time and space
|
| Posted: Sat Jul 29, 2006 4:40 am Post subject: |
|
|
| drohr wrote: | | Using the new ATI drivers released yesterday (8.27.10), and they actually work great with Xorg 7.1 with DRI. |
Thanks for the info -- I will update the initial post right now. |
|
| Back to top |
|
|
CosminG
Apprentice
Joined: 19 Jan 2005
Posts: 166
|
| Posted: Sat Jul 29, 2006 8:45 am Post subject: |
|
|
Ati was boght by AMD maby that's the reason  |
|
| Back to top |
|
|
slycordinator
Advocate
Joined: 31 Jan 2004
Posts: 3065
Location: Korea
|
| Posted: Sun Jul 30, 2006 12:16 am Post subject: |
|
|
| drohr wrote: | Using the new ATI drivers released yesterday (8.27.10), and they actually work great with Xorg 7.1 with DRI.
No ebuild yet, but works fine to download manually and install. |
Ati release notes confirm.
| Code: | Minimum System Requirements
Before attempting to install the ATI Proprietary Linux driver, the following software must be installed:
* XOrg 6.7, 6.8, 6.9,7.0 or 7.1; XFree86 version 4.3 |
And in "New features" section
| Code: | X.org 7.1 Support
This release of the ATI Proprietary Linux driver introduces support for X.org 7.1 X Server via the ATI Driver Installer as well as via packages for supported distributions that utilize X.org 7.1. |
https://a248.e.akamai.net/f/674/9206/0/www2.ati.com/drivers/linux/linux_8.27.10.html |
|
| Back to top |
|
|
Xamindar
Veteran
Joined: 03 Oct 2004
Posts: 1155
Location: California
|
| Posted: Mon Jul 31, 2006 5:48 am Post subject: |
|
|
| hey |
|
| Back to top |
|
|
allex87
Apprentice
Joined: 02 Aug 2003
Posts: 154
Location: Ontario, Canada
|
| Posted: Tue Aug 01, 2006 2:26 pm Post subject: |
|
|
| And nothing from nVIDIA? It's actually funny that ATi released this before them, as it usually was the other way around |
|
| Back to top |
|
|
slycordinator
Advocate
Joined: 31 Jan 2004
Posts: 3065
Location: Korea
|
| Posted: Tue Aug 01, 2006 8:10 pm Post subject: |
|
|
| allex87 wrote: | | And nothing from nVIDIA? It's actually funny that ATi released this before them, as it usually was the other way around |
nvidia's waiting to provide REAL support. ati's "support" for 7.1 means you can use it w/ 7.1 but it gives no 3d acceleration or aiglx support. |
|
| Back to top |
|
|
El_Presidente_Pufferfish
Veteran
Joined: 11 Jul 2002
Posts: 1179
Location: Seattle
|
| Posted: Wed Aug 02, 2006 1:46 am Post subject: |
|
|
| slycordinator wrote: | | allex87 wrote: | | And nothing from nVIDIA? It's actually funny that ATi released this before them, as it usually was the other way around |
nvidia's waiting to provide REAL support. ati's "support" for 7.1 means you can use it w/ 7.1 but it gives no 3d acceleration or aiglx support. |
3d works for me and at least several other people in 7.1... :/ |
|
| Back to top |
|
|
slycordinator
Advocate
Joined: 31 Jan 2004
Posts: 3065
Location: Korea
|
| Posted: Wed Aug 02, 2006 7:06 am Post subject: |
|
|
| El_Presidente_Pufferfish wrote: |
3d works for me and at least several other people in 7.1... :/ |
That's weird because ati themselves say that they have no 3d acceleration.
Perhaps they "support" 3d output but not accelerated? |
|
| Back to top |
|
|
zsitvaij
n00b
Joined: 11 Jan 2006
Posts: 74
|
| Posted: Wed Aug 02, 2006 11:11 pm Post subject: |
|
|
| slycordinator wrote: | | El_Presidente_Pufferfish wrote: |
3d works for me and at least several other people in 7.1... :/ |
That's weird because ati themselves say that they have no 3d acceleration.
Perhaps they "support" 3d output but not accelerated? |
How about actually reading the release notes? |
|
| Back to top |
|
|
boroshan
l33t
Joined: 16 Apr 2003
Posts: 730
Location: upside down
|
| Posted: Thu Aug 03, 2006 7:23 am Post subject: Deja Vu, anyone? |
|
|
Had the oddest experience last night. Installed the new ati-drivers at work, got home, turned the laptop back on and X didn't work. ABI incompatibility, it said.
No big deal, I just reverted to the previous version which I knew worked.
That said, the release notes say this lot are supposed to work with 7.0. Anyone got any ideas what I'm missing here?
_________________
Don't let THEM immanentize the Eschaton! |
|
| Back to top |
|
|
Phlogiston
Veteran
Joined: 27 Jan 2004
Posts: 1925
Location: Europe, Swizerland
|
| Posted: Thu Aug 03, 2006 12:34 pm Post subject: |
|
|
Hmm why is this still blocking? I searched the xorg-server ebuild and the version is correct...
| Code: |
emerge -up xorg-server
These are the packages that would be merged, in order:
Calculating dependencies... done!
[blocks B ] x11-drivers/ati-drivers (is blocking x11-base/xorg-server-1.1.1-r1)
[ebuild U ] x11-base/xorg-server-1.1.1-r1 [1.0.2-r7] USE="dri nptl% sdl% xorg% xprint -3dfx% -debug -dmx% -ipv6 -kdrive% -minimal" INPUT_DEVICES="evdev% keyboard% mouse% synaptics% -acecad% -aiptek% -calcomp% -citron% -digitaledge% -dmc% -dynapro% -elo2300% -elographics% -fpit% -hyperpen% -jamstudio% -joystick% -magellan% -microtouch% -mutouch% -palmax% -penmount% -spaceorb% -summa% -tek4957% -ur98% -vmmouse% -void% -wacom%" VIDEO_CARDS="radeon% -apm% -ark% -chips% -cirrus% -cyrix% -dummy% -epson% -fbdev% -glint% -i128% -i740% -i810% -imstt% -mach64% -mga% -neomagic% -nsc% -nv% -r128% -rendition% -s3% -s3virge% -savage% -siliconmotion% -sis% -sisusb% -tdfx% -tga% -trident% -tseng% -v4l% -vesa% -vga% -via% -vmware% -voodoo%" 6,105 kB [2]
[ebuild U ] x11-drivers/xf86-input-mouse-1.1.1 [1.0.4] USE="-debug" 261 kB
[ebuild U ] x11-drivers/xf86-video-ati-6.6.1_p20060625 [6.5.8.0] USE="dri -debug" 0 kB |
_________________
Workstation: 5.1 SurroundSound, LIRC remote control; Laptop [IBM-T43]: patched sources, s2disk/ram, fingerprint sensor |
|
| Back to top |
|
|
|
Desktop Environments
