Re: Recent sys/vm/ changes and nvidia-driver
On Sat, May 8, 2010 at 4:44 PM, Kip Macy kip.m...@gmail.com wrote: On Sat, May 8, 2010 at 2:39 PM, Brandon Gooch jamesbrandongo...@gmail.com wrote: On Sat, May 8, 2010 at 4:20 PM, b. f. bf1...@googlemail.com wrote: On 05/08/10 13:36, Alan Cox wrote: Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. What performance differences, if any, can we expect on uniprocessors from the vm page lock-related changes? Kip's original post on -arch mentioned some performance improvements and no significant regressions, but on a dual 4-core machine. Alan (or Kip): Is there a document available that describes the work being done on the VM system? I would like to -- or like make and attempt to -- read such a document... Hey b. f., to which post on -arch are you referring? there is no document. The basic idea is straightforward, but there are some unpleasant edges to cope with. http://www.mavetju.org/mail/view_message.php?list=freebsd-archid=3155260thread=no Thanks for the link, surprisingly, I think I get it and it does actually make sense to me :) Is there a final target state regarding the work and/or a time-frame for completion? Perhaps a heads-up e-mail to the freebsd-current mailing list when the first round of commits have settled in? I'm wondering, due to the Nvidia issue stated in this thread, as well as panics I'm seeing using VirtualBox on my computer running HEAD. Thank you so much for taking on this task, all of you, Kip, Alan, and Jeff too -- even though I don't understand all of it, or know about all of the work that goes in to get it done, we all appreciate and benefit from the end result: a faster and better FreeBSD. -Brandon ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On 5/10/10, Brandon Gooch jamesbrandongo...@gmail.com wrote: Is there a final target state regarding the work and/or a time-frame for completion? Perhaps a heads-up e-mail to the freebsd-current mailing list when the first round of commits have settled in? I'm wondering, due to the Nvidia issue stated in this thread, as well as panics I'm seeing using VirtualBox on my computer running HEAD. Kip, could you please add an entry for the __FreeBSD_version 900011 to the Porter's Handbook table in doc/en_US.ISO8859-1/books/porters-handbook/book.sgml , as documented in src/sys/sys/param.h, if no one else beats you to it? In addition to alerting porters to the VM changes, it will remind them that we've also had a version bump after the import of OpenSSL 0.9.8n. Regards, b. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. Is the coast clear yet? :) I have been holding off on updating -current due to the SUJ stuff, but that seems to have mostly settled down now, so I'm hoping that the work on the VM won't prevent me from running nvidia-driver. Thanks, Doug -- ... and that's just a little bit of history repeating. -- Propellerheads Improve the effectiveness of your Internet presence with a domain name makeover!http://SupersetSolutions.com/ ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. Is the coast clear yet? :) I have been holding off on updating -current due to the SUJ stuff, but that seems to have mostly settled down now, so I'm hoping that the work on the VM won't prevent me from running nvidia-driver. No, it's still in a state of flux. Alan ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On 05/08/10 13:36, Alan Cox wrote: Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. Is the coast clear yet? :) I have been holding off on updating -current due to the SUJ stuff, but that seems to have mostly settled down now, so I'm hoping that the work on the VM won't prevent me from running nvidia-driver. No, it's still in a state of flux. Ok, thanks for the info. I'm not going to pretend I understand what you're doing, but it sure _looks_ exciting. :) Doug -- ... and that's just a little bit of history repeating. -- Propellerheads Improve the effectiveness of your Internet presence with a domain name makeover!http://SupersetSolutions.com/ ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On 05/08/10 13:36, Alan Cox wrote: Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. What performance differences, if any, can we expect on uniprocessors from the vm page lock-related changes? Kip's original post on -arch mentioned some performance improvements and no significant regressions, but on a dual 4-core machine. b. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
Doug Barton wrote: On 05/08/10 13:36, Alan Cox wrote: Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. Is the coast clear yet? :) I have been holding off on updating -current due to the SUJ stuff, but that seems to have mostly settled down now, so I'm hoping that the work on the VM won't prevent me from running nvidia-driver. No, it's still in a state of flux. Ok, thanks for the info. I'm not going to pretend I understand what you're doing, but it sure _looks_ exciting. :) I should probably be explicit about one thing. Anyone using nothing but in-tree drivers, shouldn't be worried about updating. Alan ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On Sat, May 8, 2010 at 4:20 PM, b. f. bf1...@googlemail.com wrote: On 05/08/10 13:36, Alan Cox wrote: Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. What performance differences, if any, can we expect on uniprocessors from the vm page lock-related changes? Kip's original post on -arch mentioned some performance improvements and no significant regressions, but on a dual 4-core machine. Alan (or Kip): Is there a document available that describes the work being done on the VM system? I would like to -- or like make and attempt to -- read such a document... Hey b. f., to which post on -arch are you referring? -Brandon ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On 08.05.2010 22:30 (UTC+1), Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. Is the coast clear yet? :) I have been holding off on updating -current due to the SUJ stuff, but that seems to have mostly settled down now, so I'm hoping that the work on the VM won't prevent me from running nvidia-driver. Hope I do not misunderstand the item. I updated current (amd64) today and it seems all is fine with newest nvidia driver at the moment, Rainer Thanks, Doug ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On Sat, May 8, 2010 at 2:39 PM, Brandon Gooch jamesbrandongo...@gmail.com wrote: On Sat, May 8, 2010 at 4:20 PM, b. f. bf1...@googlemail.com wrote: On 05/08/10 13:36, Alan Cox wrote: Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. What performance differences, if any, can we expect on uniprocessors from the vm page lock-related changes? Kip's original post on -arch mentioned some performance improvements and no significant regressions, but on a dual 4-core machine. Alan (or Kip): Is there a document available that describes the work being done on the VM system? I would like to -- or like make and attempt to -- read such a document... Hey b. f., to which post on -arch are you referring? there is no document. The basic idea is straightforward, but there are some unpleasant edges to cope with. http://www.mavetju.org/mail/view_message.php?list=freebsd-archid=3155260thread=no -Kip ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On 5/8/10, Kip Macy kip.m...@gmail.com wrote: On Sat, May 8, 2010 at 2:39 PM, Brandon Gooch jamesbrandongo...@gmail.com wrote: On Sat, May 8, 2010 at 4:20 PM, b. f. bf1...@googlemail.com wrote: On 05/08/10 13:36, Alan Cox wrote: Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. What performance differences, if any, can we expect on uniprocessors from the vm page lock-related changes? Kip's original post on -arch mentioned some performance improvements and no significant regressions, but on a dual 4-core machine. Alan (or Kip): Is there a document available that describes the work being done on the VM system? I would like to -- or like make and attempt to -- read such a document... Hey b. f., to which post on -arch are you referring? I was referring to the message listed in Kip's link below. there is no document. The basic idea is straightforward, but there are some unpleasant edges to cope with. http://www.mavetju.org/mail/view_message.php?list=freebsd-archid=3155260thread=no -Kip ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On Sat, May 8, 2010 at 2:20 PM, b. f. bf1...@googlemail.com wrote: On 05/08/10 13:36, Alan Cox wrote: Doug Barton wrote: On 05/05/10 11:56, Alan Cox wrote: I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. What performance differences, if any, can we expect on uniprocessors from the vm page lock-related changes? Kip's original post on -arch mentioned some performance improvements and no significant regressions, but on a dual 4-core machine. I wouldn't actually worry about UP since the overhead can largely be disabled by building without SMP. I think we need to be looking at how a dual-core system performs, trading off any regressions there against current processor trends of ever higher core and thread count. Cheers, Kip ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Recent sys/vm/ changes and nvidia-driver
Hi, After recent changes to sys/vm/ by alc@, I'm getting panics as soon as I start xorg-server with nvidia-driver (both 195.22 and 195.36.15): panic: mutex page lock not owned at /home/yuri/src/FreeBSD/head/sys/vm/vm_page.c:1547 (sorry, can't get kernel to dump correctly to provide more info: Cannot access memory at address 0xff007ff8ffe0 (kgdb) #0 0x in ?? () Cannot access memory at address 0x0 ) Commenting out assert on line 1547, sys/vm/vm_page.c gets rid of panic, but even I understand that it isn't correct at all. Could someone please take a look at nvidia-driver kld source for possible fix? TIA, Yuri ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Recent sys/vm/ changes and nvidia-driver
On Wed, May 5, 2010 at 11:19 AM, Yuri Pankov yuri.pan...@gmail.com wrote: Hi, After recent changes to sys/vm/ by alc@, I'm getting panics as soon as I start xorg-server with nvidia-driver (both 195.22 and 195.36.15): panic: mutex page lock not owned at /home/yuri/src/FreeBSD/head/sys/vm/vm_page.c:1547 (sorry, can't get kernel to dump correctly to provide more info: Cannot access memory at address 0xff007ff8ffe0 (kgdb) #0 0x in ?? () Cannot access memory at address 0x0 ) Commenting out assert on line 1547, sys/vm/vm_page.c gets rid of panic, but even I understand that it isn't correct at all. Could someone please take a look at nvidia-driver kld source for possible fix? I'm afraid that I would advise waiting a few days. This round of changes are not yet complete. Alan ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org