| View previous topic :: View next topic |
| Author |
Message |
shadowknight Tux's lil' helper


Joined: 03 Aug 2002 Posts: 142
|
Posted: Thu Nov 29, 2007 2:29 pm Post subject: |
|
|
Ok, I got it installed and working, the fix for switching vt's with compiz even works
Something to note though, when running this:
| Code: |
ebuild nvidia-drivers-169.04.ebuild digest
|
it kept wanting to connect to my portage servers to dl the pkg0.run and the freebsd .tar.gz even though the source_uri's in the ebuild were correct. I had to manually dl them to distfiles then everything ran fine after that.
Thanks platojones
I thought it was probably something like that, but I didn't want to add PORTDIR_OVERLAY= to make.conf if it would mess up layman, so I thought I'd ask first  |
|
| Back to top |
|
 |
billydv l33t


Joined: 22 Dec 2006 Posts: 818 Location: Mount Vernon, NY
|
Posted: Fri Nov 30, 2007 11:35 pm Post subject: |
|
|
Okay, Im having problems with this driver, someone else please try this becuase its happening to me on 2 different boxes, x86 and amd64,
With compiz-fusion running (not kwin), open a konqueror window as super user, then switch to a directory such as etc or usr, then click the K icon in the top right hand corner to open a duplicate window
X at that point crashes for me while running the 169 driver. After downgrading to 100.14.23, the problem goes away. Strange thing is that this doesnt happen when running the same driver in Debian. _________________ Billy DeVincentis |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5345 Location: Vienna, Austria; Germany; hello world :)
|
Posted: Fri Nov 30, 2007 11:38 pm Post subject: |
|
|
| billydv wrote: | Okay, Im having problems with this driver, someone else please try this becuase its happening to me on 2 different boxes, x86 and amd64,
With compiz-fusion running (not kwin), open a konqueror window as super user, then switch to a directory such as etc or usr, then click the K icon in the top right hand corner to open a duplicate window
X at that point crashes for me while running the 169 driver. After downgrading to 100.14.23, the problem goes away. Strange thing is that this doesnt happen when running the same driver in Debian. |
works for me - but the thing is - I also don't have the "average box"
gcc-4.2.2 hardened, testing amd64 (non-hardened profile), glibc-2.7;
what toolchain you've got ?
probably a | Code: | | revdep-rebuild -i -p | or updating world might help ...
you might also want to check your kernel-config under kernel-hacking if "error handling" stuff is enabled:
e.g. I configured that way - probably I'll enable more in the future:
| Code: | #
# Kernel hacking
#
CONFIG_TRACE_IRQFLAGS_SUPPORT=y
CONFIG_PRINTK_TIME=y
# CONFIG_ENABLE_WARN_DEPRECATED is not set
# CONFIG_ENABLE_MUST_CHECK is not set
CONFIG_MAGIC_SYSRQ=y
# CONFIG_UNUSED_SYMBOLS is not set
CONFIG_DEBUG_FS=y
# CONFIG_HEADERS_CHECK is not set
CONFIG_DEBUG_KERNEL=y
CONFIG_DEBUG_SHIRQ=y
CONFIG_DETECT_SOFTLOCKUP=y
# CONFIG_SCHED_DEBUG is not set
CONFIG_SCHEDSTATS=y
CONFIG_TIMER_STATS=y
CONFIG_DEBUG_PREEMPT=y
# CONFIG_DEBUG_RT_MUTEXES is not set
# CONFIG_RT_MUTEX_TESTER is not set
CONFIG_DEBUG_SPINLOCK=y
# CONFIG_DEBUG_MUTEXES is not set
# CONFIG_DEBUG_LOCK_ALLOC is not set
# CONFIG_PROVE_LOCKING is not set
# CONFIG_LOCK_STAT is not set
# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
CONFIG_DEBUG_LOCKING_API_SELFTESTS=y
# CONFIG_DEBUG_KOBJECT is not set
# CONFIG_DEBUG_BUGVERBOSE is not set
# CONFIG_DEBUG_INFO is not set
# CONFIG_DEBUG_VM is not set
# CONFIG_DEBUG_LIST is not set
# CONFIG_DEBUG_SG is not set
# CONFIG_FRAME_POINTER is not set
# CONFIG_FORCED_INLINING is not set
# CONFIG_BOOT_PRINTK_DELAY is not set
# CONFIG_RCU_TORTURE_TEST is not set
# CONFIG_RCU_TRACE is not set
# CONFIG_FAULT_INJECTION is not set
# CONFIG_GCC_4003_SUPPORT is not set
# CONFIG_SAMPLES is not set
CONFIG_EARLY_PRINTK=y
CONFIG_DEBUG_STACKOVERFLOW=y
# CONFIG_DEBUG_STACK_USAGE is not set
# CONFIG_DEBUG_RODATA is not set
# CONFIG_IOMMU_DEBUG is not set
|
@others: correct me if anything important isn't enabled / missing - many thanks in advance _________________ Unofficial minimal livecd x86/amd64 w/reiser4+truecrypt (by Neo2)
2.6.37.2_plus_v1: BFS, CFS,THP,compaction, zcache or TOI
Hardcore Linux user since 2004 
Last edited by kernelOfTruth on Fri Nov 30, 2007 11:43 pm; edited 1 time in total |
|
| Back to top |
|
 |
platojones Veteran


Joined: 23 Oct 2002 Posts: 1430 Location: Just over the horizon
|
Posted: Fri Nov 30, 2007 11:42 pm Post subject: |
|
|
| Nope, no problems here...running the new beta driver, kde 3.5.7 and compiz-fuzion 0.6.0. Navigated from my home directory to /etc and /user then hit the new screen window and I'm still alive. I am using emerald as the windows-decorator here. This is an x86 machine, btw. Are you using any unstable packages for kde or X? Could it be your window decorator? |
|
| Back to top |
|
 |
shadowknight Tux's lil' helper


Joined: 03 Aug 2002 Posts: 142
|
Posted: Sat Dec 01, 2007 1:58 am Post subject: |
|
|
It works here as well using gnome as desktop, compiz 0.6.0. I do seem to have a couple of issues but nothing major. For me the only thing I've noticed so far is that Extreme Tux Racer (my 4 yr old niece loves Tux )runs a good bit slower than it used to, Doom3 settings had to be cut back a bit, but WoW runs perfectly and that's the only game I really play anyway so I'm not complaining. I was going to eventually post this somewhere anyway, and although it's not related to the new driver directly, it is graphic related, so I'll ask here first and then post somewhere else if I need to. I know glxgears is not really for benchmarking but before my last HD crash, my fps were topping 11,000 with all the same settings I have now. Now I'm not even hitting 100 and even compiz-fusion hovers in the 59-60 range. It's not a issue really since the WoW runs perfectly well, but I'm curious what could be causing this. When my HD went down I was without a pc for several months, and when I got back up and running, I used a fresh install and just copied over my old config files from backup, (make.conf, xorg.conf, etc). Again, it's not a problem really, but I would like to get some opinions on what could be the cause. This was happening with the stable drivers as well, so I know it's not an issue related to the new drivers. |
|
| Back to top |
|
 |
billydv l33t


Joined: 22 Dec 2006 Posts: 818 Location: Mount Vernon, NY
|
Posted: Sat Dec 01, 2007 3:39 am Post subject: |
|
|
as for kernel hacking heres my config
#
# Kernel hacking
#
CONFIG_TRACE_IRQFLAGS_SUPPORT=y
# CONFIG_PRINTK_TIME is not set
CONFIG_ENABLE_MUST_CHECK=y
CONFIG_MAGIC_SYSRQ=y
CONFIG_UNUSED_SYMBOLS=y
# CONFIG_DEBUG_FS is not set
# CONFIG_HEADERS_CHECK is not set
# CONFIG_DEBUG_KERNEL is not set
# CONFIG_SLUB_DEBUG_ON is not set
# CONFIG_DEBUG_BUGVERBOSE is not set
stable only toolchain
gnome 2.20
kde 3.5.8
x 7.2
my compiz is from the live ebuilds, revdep shows nothing
Think I should rememerge compiz and then try it. _________________ Billy DeVincentis |
|
| Back to top |
|
 |
billydv l33t


Joined: 22 Dec 2006 Posts: 818 Location: Mount Vernon, NY
|
Posted: Sat Dec 01, 2007 4:12 am Post subject: |
|
|
No question, removed and reemerged all compiz and returned the 169 driver, did it right away, must put 2 or 3 open root windows in folders other than root and then try to open duplicate window, crashes x every time. _________________ Billy DeVincentis |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5345 Location: Vienna, Austria; Germany; hello world :)
|
Posted: Sat Dec 01, 2007 11:26 am Post subject: |
|
|
| billydv wrote: | | No question, removed and reemerged all compiz and returned the 169 driver, did it right away, must put 2 or 3 open root windows in folders other than root and then try to open duplicate window, crashes x every time. |
perhaps trying latest xorg-server + drivers (keyboard, mouse) does the trick ?
running amd64 & unstable / testing here _________________ Unofficial minimal livecd x86/amd64 w/reiser4+truecrypt (by Neo2)
2.6.37.2_plus_v1: BFS, CFS,THP,compaction, zcache or TOI
Hardcore Linux user since 2004  |
|
| Back to top |
|
 |
billydv l33t


Joined: 22 Dec 2006 Posts: 818 Location: Mount Vernon, NY
|
Posted: Sat Dec 01, 2007 1:57 pm Post subject: |
|
|
My problem with running xorg 7.3 and server 1.4 is the keyboard problems, when I had 7.3 the caps led and numlock lights were reacting strangely, in fact there were no caps leds at all. I checked the bug status just a couple of days ago and its still broken, but they claim that xorg server 1.4.1 will be out very soon and that will include the fixes. Thats when I'm going to upgrade. _________________ Billy DeVincentis |
|
| Back to top |
|
 |
platojones Veteran


Joined: 23 Oct 2002 Posts: 1430 Location: Just over the horizon
|
Posted: Sat Dec 01, 2007 1:58 pm Post subject: |
|
|
Well, you are using newer versions of both kde and X than I am (for X I am on stable 7.1), so I'm wondering if that could be difference. I tried again with several root windows open and still could not duplicate the problem. Of course, since it only happens with the new driver for you, it sounds like it's interacting poorly with something on your system.
I happen to have an issue that nobody else seems to be having, though it's pretty minor. My 'konsole' sessions don't refresh correctly when I rotate the compiz-fusion cube. When I run 'top' in one of the windows then rotate to another screen and back, it looks pretty garbled. But like I said, pretty minor stuff and I'm betting if I dig and tinker a bit, I'll find a way to clear that up. That too is a driver difference since it only showed up after I merged the new driver in. |
|
| Back to top |
|
 |
billydv l33t


Joined: 22 Dec 2006 Posts: 818 Location: Mount Vernon, NY
|
Posted: Sat Dec 01, 2007 8:39 pm Post subject: |
|
|
I now have updated xorg on my amd64 box and have once again tried to open several root konqueror windows, same exact thing, it may open 1 okay especially outside of the root folder but the 2nd or 3rd now lock up the computer, reboot necessary. Dropped again to the 23 driver and it works fine. Its definitely the 169 driver but it doesnt happen if compiz isnt running. _________________ Billy DeVincentis |
|
| Back to top |
|
 |
billydv l33t


Joined: 22 Dec 2006 Posts: 818 Location: Mount Vernon, NY
|
Posted: Sun Dec 09, 2007 10:25 pm Post subject: |
|
|
Okay, I've got a better handle on this, it doesn't matter if you are su or not, if you open enough windows, especially ones with lots of files and folders you get a crash. I want to upgrade my system to a fully ~amd64 and then I am going to give it another try. BTW, the regular nvidia installer done on a debian box works perfectly, therefore it may be an issue with the ebuild or gentoo implementation. _________________ Billy DeVincentis |
|
| Back to top |
|
 |
slycordinator Advocate


Joined: 31 Jan 2004 Posts: 3050 Location: Houston, TX
|
Posted: Mon Dec 10, 2007 8:49 am Post subject: |
|
|
| shadowknight wrote: | | it kept wanting to connect to my portage servers to dl the pkg0.run and the freebsd .tar.gz even though the source_uri's in the ebuild were correct. I had to manually dl them to distfiles then everything ran fine after that. |
That's because portage ALWAYS DOES THAT. It first tries your portage servers then it goes to the one specified in the ebuild after those fail.
In fact, all ebuilds in the tree specify an upstream URI which is only used if the source file is not in the gentoo servers you've specified. This is a feature of portage because most upstream servers are slower than the ones used as gentoo file mirrors as the gentoo servers usually give you a faster download than the upstream one (especially if you've chosen a server that's close to you). |
|
| Back to top |
|
 |
shadowknight Tux's lil' helper


Joined: 03 Aug 2002 Posts: 142
|
Posted: Mon Dec 10, 2007 2:08 pm Post subject: |
|
|
| slycordinator wrote: | | shadowknight wrote: | | it kept wanting to connect to my portage servers to dl the pkg0.run and the freebsd .tar.gz even though the source_uri's in the ebuild were correct. I had to manually dl them to distfiles then everything ran fine after that. |
That's because portage ALWAYS DOES THAT. It first tries your portage servers then it goes to the one specified in the ebuild after those fail.
|
Ahh, wasn't aware of that. I just assumed that the uri's specified were the ones it would use. Guess that's what I get for assuming again One of these days I'll learn to stop making assumptions Thanks for the tip off. |
|
| Back to top |
|
 |
cvweiss n00b


Joined: 25 Apr 2005 Posts: 67 Location: Detroit!
|
Posted: Fri Dec 14, 2007 1:15 pm Post subject: |
|
|
Has anyone filed a bug report against this? I'm quite surprised to not see 169.04 in portage yet, even if it were masked.
My X will randomly lock up for up to 2 minutes with the current versions, and 169.04 fixes the problem on the same hardware on another machine running Ubuntu. _________________ If at first you don't suceed, try again does NOT apply to skydiving. |
|
| Back to top |
|
 |
slycordinator Advocate


Joined: 31 Jan 2004 Posts: 3050 Location: Houston, TX
|
|
| Back to top |
|
 |
shadowknight Tux's lil' helper


Joined: 03 Aug 2002 Posts: 142
|
Posted: Sun Dec 16, 2007 1:41 pm Post subject: |
|
|
The last post on bugzilla says:
| Quote: | Everyone can agree until they're blue in the face. I do not add beta drivers to
the tree. Cut and dry. |
Although I know I've seen beta drivers in the tree before, so IMHO I don't see what the problem would be to add it and have it masked with missing keyword or something. But, then again, we always have the option of using an overlay which is what we're doing now anyway. So I guess it really makes no real difference in the end. Just thought I'd show my support for making it official though. |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5345 Location: Vienna, Austria; Germany; hello world :)
|
Posted: Sun Dec 16, 2007 1:47 pm Post subject: |
|
|
| shadowknight wrote: | The last post on bugzilla says:
| Quote: | Everyone can agree until they're blue in the face. I do not add beta drivers to
the tree. Cut and dry. |
Although I know I've seen beta drivers in the tree before, so IMHO I don't see what the problem would be to add it and have it masked with missing keyword or something. But, then again, we always have the option of using an overlay which is what we're doing now anyway. So I guess it really makes no real difference in the end. Just thought I'd show my support for making it official though. |
well well - so Doug is the current nvidia drivers maintainer ? anyone willing to take his place ? (just kidding)
that drivers fixed a lot problems for me & in addition they add support for unsupported hardware - I'm curious what's his point ... _________________ Unofficial minimal livecd x86/amd64 w/reiser4+truecrypt (by Neo2)
2.6.37.2_plus_v1: BFS, CFS,THP,compaction, zcache or TOI
Hardcore Linux user since 2004  |
|
| Back to top |
|
 |
dodo1122 Guru


Joined: 02 Sep 2006 Posts: 347 Location: York, England
|
Posted: Sun Dec 16, 2007 2:04 pm Post subject: |
|
|
i started using those drivers since 2 days ago, and i have no problems with them. Seriously, the previous beta drivers are in the tree, so why not this one?
dodo _________________ #zen-sources on irc.rizon.net |
|
| Back to top |
|
 |
platojones Veteran


Joined: 23 Oct 2002 Posts: 1430 Location: Just over the horizon
|
Posted: Sun Dec 16, 2007 3:28 pm Post subject: |
|
|
Well, there have been some problems reported with these drivers over in nvNEWS, but like many have said, they are fairly stable for a beta release. In my case, they cause a refresh issue in KDE Konsole when using compiz-fusion. Also, switching virtual terminals while running compiz-fuzion doesn't crash X, but it kills compiz-fusion for me. Not biggies, but again, not stable. Since the creating an overlay ebuild is so easy to do, I don't find it a big issue that they aren't in the official tree and am more that willing to leave that kind of policy decision to the dev who is managing this driver. I don't think criticizing him for this reasonable decision will help anything. The true solution is, as always, that Nvidia fix the few problems in this driver and release an RC driver. Given their track record, I think they will before long. I am getting tired of this Konsole refresh bug  |
|
| Back to top |
|
 |
slycordinator Advocate


Joined: 31 Jan 2004 Posts: 3050 Location: Houston, TX
|
Posted: Sun Dec 16, 2007 4:28 pm Post subject: |
|
|
| shadowknight wrote: | The last post on bugzilla says:
| Quote: | Everyone can agree until they're blue in the face. I do not add beta drivers to
the tree. Cut and dry. |
Although I know I've seen beta drivers in the tree before, so IMHO I don't see what the problem would be to add it and have it masked with missing keyword or something. But, then again, we always have the option of using an overlay which is what we're doing now anyway. So I guess it really makes no real difference in the end. Just thought I'd show my support for making it official though. |
You're assuming that the beta drivers you saw before were added by the current maintainer. |
|
| Back to top |
|
 |
shadowknight Tux's lil' helper


Joined: 03 Aug 2002 Posts: 142
|
Posted: Sun Dec 16, 2007 7:12 pm Post subject: |
|
|
| slycordinator wrote: | | shadowknight wrote: | The last post on bugzilla says:
| Quote: | Everyone can agree until they're blue in the face. I do not add beta drivers to
the tree. Cut and dry. |
Although I know I've seen beta drivers in the tree before, so IMHO I don't see what the problem would be to add it and have it masked with missing keyword or something. But, then again, we always have the option of using an overlay which is what we're doing now anyway. So I guess it really makes no real difference in the end. Just thought I'd show my support for making it official though. |
You're assuming that the beta drivers you saw before were added by the current maintainer. |
Actually I wasn't assuming anything, just making an observation, that they have been in there before regardless of who put them there  |
|
| Back to top |
|
 |
slycordinator Advocate


Joined: 31 Jan 2004 Posts: 3050 Location: Houston, TX
|
Posted: Sun Dec 16, 2007 7:56 pm Post subject: |
|
|
| Nonsequitur. The dev says that he never adds beta drivers (and he's said it before with other beta drivers) then you point out that other drivers that he isn't in control of have had beta drivers added. |
|
| Back to top |
|
 |
shadowknight Tux's lil' helper


Joined: 03 Aug 2002 Posts: 142
|
Posted: Sun Dec 16, 2007 8:34 pm Post subject: |
|
|
| I think you misunderstood my intent. I wasn't trying to start an argument. I merely pointing out that previous drivers had beta's in the tree. I have no knowledge of who put them there or why. Also, he's the maintainer so he can do what he wants, doesn't matter to me. I can and have put it in an overlay, so the point it's really a moot point. I was simply voicing my opinion that I PERSONALLY agreed that I thought it should be added, but that's my opinion, to which I am entitled. There's no point in debating it any further. |
|
| Back to top |
|
 |
slycordinator Advocate


Joined: 31 Jan 2004 Posts: 3050 Location: Houston, TX
|
Posted: Mon Dec 17, 2007 9:21 am Post subject: |
|
|
Now I'm confused
| cardoe@gentoo.org wrote: | Everyone can agree until they're blue in the face. I do not add beta drivers to
the tree. Cut and dry. |
| cardoe@gentoo.org wrote: | Added to the tree but masked and with a note that they're beta drivers and
unsupported. |
The first is obviously the current bug report and the second is the one for 100.14.23. Seems it's not so "cut and dry."
I assumed when he said he doesn't do it that someone would've been the one who did. |
|
| Back to top |
|
 |
|