Re: agp stuff
Sound works, and as for nvidia it works as long as I don't run any 3d stuff. And the xf86 supplied driver works like a charm. But I'm working on some 3d code and need the 3d acceleration, the odd thing is that it worked until I upgraded my server from 4.7 to 5.0, I NFS mount everything from the server except the system etc (so I was running on a CURRENT kernel on the machine with the 3d hardware, but with XF86 etc compiled on 4.7). //David On 12 Mar 2003 11:01:10 +0100 CARTER Anthony [EMAIL PROTECTED] wrote: Is your problem with sound or video? If sound, then you need to recompile your Kernel as such: mkdir /root/kernels cp /usr/src/sys/i386/conf/GENERIC /root/kernels/MYKERNEL ln -s /root/kernels/MYKERNEL /usr/src/sys/i386/conf/MYKERNEL now edit the /root/MYKERNEL and at the end add: options pcm options sbc (this one is for some on-Motherboard sound) Now, for your Nvidia problem, there is a quick solution for you, add: options VESA And select VESA for your card in the XFree86 Config setup. This will get you to a stage in which you can at least run X in a decent usable resolution. Now for the compile: cd /usr/src make buildworld (build kernel may need to rely on this) make buildkernel KERNCONF=MYKERNEL make installkernel KERNCONF=MYKERNEL Hope this helps, Anthony Carter On Wed, 2003-03-12 at 00:13, David Holm wrote: Hi, I get the following error (or whatever) in my bootmessage: agp0: VIA Generic host to PCI bridge mem 0xf000-0xf7ff at device 0.0 o n pci0 pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_F OUND I tried playing around in the bios but couldn't find anything that would fix this. Could this be what is causing all my nvidia problems? (and yeah, I'm running on a current kernel, not RELENG_5_0) //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
agp stuff
Hi, I get the following error (or whatever) in my bootmessage: agp0: VIA Generic host to PCI bridge mem 0xf000-0xf7ff at device 0.0 o n pci0 pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_F OUND I tried playing around in the bios but couldn't find anything that would fix this. Could this be what is causing all my nvidia problems? (and yeah, I'm running on a current kernel, not RELENG_5_0) //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Important, agp_via.c missing PCI ID!
http://lxr.linux.no/source/drivers/char/agp/agp.h The linux kernel uses ID 0x0605 as well! //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Important, agp_via.c missing PCI ID!
On Tue, 14 Jan 2003 14:20:21 -0500 (EST) John Baldwin [EMAIL PROTECTED] wrote: On 14-Jan-2003 David Holm wrote: My mistake, I didn't preload the module. I thought I could load it on a booted system. Anyway, with the standard agp module preloaded I get the following: pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND With my modified version I get: Preloaded elf module /boot/kernel/agp.ko at 0xc0551154. agp0: VIA 82C694X (Apollo Pro 133A) host to PCI bridge mem 0xe000-0xefff at device 0.0 I added some debug output to agp_via.c to see what was going on: agp0: id is 0x6051106 (- it prints the device id returned by pci_get_devid(dev)) Could you print out the results of pci_get_vendor()? If it's not 0x1106 then there might be some other bug. It returns 1106, but it seems to work with the generic now for some reason... Did anyone change the code recently? Preloaded elf module /boot/kernel/agp.ko at 0xc0676154. agp0: Vendor id 1106 agp0: VIA Generic host to PCI bridge mem 0xe000-0xefff at device 0.0 on pci0 pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Important, agp_via.c missing PCI ID!
On Mon, 13 Jan 2003 14:35:43 -0500 (EST) John Baldwin [EMAIL PROTECTED] wrote: According to /usr/share/misc/pci_vendors, that device is not for the Pro 133A but for the PM133: 0605VT8605 ProSavage PM133 System Controller 0691VT82C691/693A/694X Apollo Pro/133/133A System Controller I promise you it's an Apollo Pro 133A and it identifies with 0605, not with 0691. This is the mobo I have: http://www.motherboard.cz/mb/asus/CUV4X.htm The patch should make no functional difference since the generic probe should work (it should be listed as a VIA Generic host to PCI bridge). It doesn't, without adding that line my computer reboots when running OpenGL apps. I have tried maybe 5 times to recompile the kernel without it, same result every time. Is there anything I can do to confirm, check why I need this or something? I have tried checking the syslogs after the computer reboots but I never found anything related to it. I'd prefer not to run too many tests though, since sometimes my drives get corrupted. I just switched back to CURRENT to check how it would behave when I added these lines. When I load agp.ko now I get this extra output: on pci0 pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND and I had to set the AGP rate to 0, otherwise the machine will hang randomly (not OpenGL related this time since it will hang even though I haven't run any OpenGL apps). I have tried not loading agp.ko and using nvidias agp (by enabling in XF86Config), but that yields the same result as agp_via.c without the added ID. Now this is very strange, I did a search and indeed, the id is 0691. I was unable to find the page that said that it was 0605 =(. Now I'm positive I have a CUV4X mobo which has a 694X AGP chipset. Do you have any hints to how I can debug this, because it works with id 0605 and it doesn't work without it, which does not make any sense I guess. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Important, agp_via.c missing PCI ID!
On Mon, 13 Jan 2003 16:16:51 -0500 (EST) John Baldwin [EMAIL PROTECTED] wrote: On 13-Jan-2003 David Holm wrote: What is the output of 'dmesg | grep agp' both with and without the patch? Now this is strange. I cvsupped before recompiling the kernel. Now all I get when loading the module (with or without the patch): pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND I don't understand why it doesn't fall back to the generic. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Important, agp_via.c missing PCI ID!
On Tue, 14 Jan 2003 03:14:59 +0100 David Holm [EMAIL PROTECTED] wrote: On Mon, 13 Jan 2003 16:16:51 -0500 (EST) John Baldwin [EMAIL PROTECTED] wrote: On 13-Jan-2003 David Holm wrote: What is the output of 'dmesg | grep agp' both with and without the patch? Now this is strange. I cvsupped before recompiling the kernel. Now all I get when loading the module (with or without the patch): pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND I don't understand why it doesn't fall back to the generic. My mistake, I didn't preload the module. I thought I could load it on a booted system. Anyway, with the standard agp module preloaded I get the following: pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND With my modified version I get: Preloaded elf module /boot/kernel/agp.ko at 0xc0551154. agp0: VIA 82C694X (Apollo Pro 133A) host to PCI bridge mem 0xe000-0xefff at device 0.0 on pci0 pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND I added some debug output to agp_via.c to see what was going on: agp0: id is 0x6051106 (- it prints the device id returned by pci_get_devid(dev)) pcib1: ACPI PCI-PCI bridge at device 1.0 on pci0 pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND pci1: ACPI PCI bus on pcib1 pci1: display, VGA at device 0.0 (no driver attached) It seems it never enters the generic but goes past it and returns NULL. (I even modified the line to and the result of pci_get_vendor(dev) with 0x) Also, acpi returns this which confirms my mobo model: acpi0: ASUS CUV4Xon motherboard //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Possible PCI/AGP error
Hi, I get the following error when loading agp.ko in CURRENT (I did not get this in STABLE): Preloaded elf module /boot/kernel/agp.ko at 0xc0700154. agp0: VIA 82C694X (Apollo Pro 133A) host to PCI bridge mem 0xe000-0xefff at device 0.0 on pci0 pcib1: could not get PCI interrupt routing table for \\_SB_.PCI0.AGP_ - AE_NOT_FOUND //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Important, agp_via.c missing PCI ID!
Hi, the FreeBSD kernel is missing the PCI ID for the Apollo Pro 133A PCI bridge even though it is supported by the agp_via code. Please see this pr: http://www.freebsd.org/cgi/query-pr.cgi?pr=46983 //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: upgrading to CURRENT from STABLE
On Thu, 9 Jan 2003 20:34:19 +0100 David Holm [EMAIL PROTECTED] wrote: You can find info on how to set maximum optimizations with gcc 3.2 here: http://www.freehackers.org/gentoo/gccflags/flag_gcc3opt.html BTW, some of these flags will only work with gcc 3.1 and higher, so you won't be able to set them until after you have upgraded to 5.0. But as I said, you don't want to set the when compiling world and kernel anyway ;). I usually have two sets of CFLAGS and CXXFLAGS in my make.conf, one which I use when compiling world/kernel and one which I use for everything else. It's just a matter of moving two '#' when switching, the lazy could even write a script =). On Thu, 09 Jan 2003 11:22:39 -0800 Joe Laughlin [EMAIL PROTECTED] wrote: In the UPDATING file, the following is noted: FreeBSD 5.0-CURRENT has many debugging features turned on, in both the kernel and userland. These features attempt to detect incorrect use of system primitives, and encourage loud failure through extra sanity checking and fail stop semantics. They also substantially impact system performance. If you want to do performance measurement, benchmarking, and optimization, you'll want to turn them off. How do I turn them off? Thanks, Joe To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: upgrading to CURRENT from STABLE
On Thu, 9 Jan 2003 20:34:19 +0100 David Holm [EMAIL PROTECTED] wrote: You can find info on how to set maximum optimizations with gcc 3.2 here: http://www.freehackers.org/gentoo/gccflags/flag_gcc3opt.html I'm the master of forgetting to tell people important stuff =(. You should not use -march=somearch, instead use this on a separate line: CPUTYPE?=somearch On Thu, 09 Jan 2003 11:22:39 -0800 Joe Laughlin [EMAIL PROTECTED] wrote: In the UPDATING file, the following is noted: FreeBSD 5.0-CURRENT has many debugging features turned on, in both the kernel and userland. These features attempt to detect incorrect use of system primitives, and encourage loud failure through extra sanity checking and fail stop semantics. They also substantially impact system performance. If you want to do performance measurement, benchmarking, and optimization, you'll want to turn them off. How do I turn them off? Thanks, Joe To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: troubles with realplay-er
MPlayer plays these streams perfectly here. On Sunday 05 January 2003 01:29, Mikhail Teterin wrote: This Linux program used to work well, but now it crashes (with SIGABRT) some URLs, such as pnm://rm.content.loudeye.com/~iii-600111/0255135_0101_00_0002.ra pnm://rm.content.loudeye.com/~a-600111/0631342_0103_00_0002.ra or hangs... The crashes are persistent -- the same URL will alway cause the SIGABRT. The hangs are intermittent -- after ``killall -9 realplay'' the new instance will play the same URL. Then -- eventually hang on another... -mi P.S. Teaching libfetch about pnm:// together with mplayer would eliminate the need for RealPlayer :-) Ducks... To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
nVidia opengl works as root but not as user
Hi, it's getting boring seeing all the nvidia posts but I'm so darn close to having it working here. I have disabled INVARIANTS in the kernel and AGP and running with nvAgp (although I get the same results using kernel agp as well). The thing is that OpenGL apps run perfectly as root, but whenever I use my normal user I can run 1-2 OpenGL apps without any problems but when I run them the next time the machine locks instantly and then reboots, this never happens when I'm running as root. Since it works for root I must be close to having it fully working. Do anyone have any ideas? (No, I don't want to set the suid bit on my opengl apps ;). I've never seen any useful output in my system logs after these reboots, is there any way I can debug the driver without setting up a serial console? //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: nVidia opengl works as root but not as user
Hmm, when I mounted my home dir as read-only I was able to use OpenGL as a normal user. But I've looked all through my home dir and I can't find anything there that would affect OpenGL if the user had write access to the partition =(. //David Holm On Sunday 05 January 2003 11:38, David Holm wrote: Hi, it's getting boring seeing all the nvidia posts but I'm so darn close to having it working here. I have disabled INVARIANTS in the kernel and AGP and running with nvAgp (although I get the same results using kernel agp as well). The thing is that OpenGL apps run perfectly as root, but whenever I use my normal user I can run 1-2 OpenGL apps without any problems but when I run them the next time the machine locks instantly and then reboots, this never happens when I'm running as root. Since it works for root I must be close to having it fully working. Do anyone have any ideas? (No, I don't want to set the suid bit on my opengl apps ;). I've never seen any useful output in my system logs after these reboots, is there any way I can debug the driver without setting up a serial console? //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Compilation failure on x86
Hi,
I tried upgrading this morning but compilation fails with the following error
(I waited a while and cvsup'd again in case someone was commiting, didn't
help)
=== lib/libkvm
Warning: Object directory not changed from original /usr/src/lib/libkvm
gcc -O2 -pipe -march=pentiumpro -DLIBC_SCCS -I/usr/src/lib/libkvm -c
kvm_getswapinfo.c -o kvm_getswapinfo.o
In file included from kvm_getswapinfo.c:20:
/usr/include/vm/swap_pager.h:75: syntax error before vm_object_t
kvm_getswapinfo.c: In function `kvm_getswapinfo_kvm':
kvm_getswapinfo.c:191: storage size of `swinfo' isn't known
kvm_getswapinfo.c:194: invalid use of undefined type `struct swdevt'
kvm_getswapinfo.c:194: dereferencing pointer to incomplete type
kvm_getswapinfo.c: In function `nlist_init':
kvm_getswapinfo.c:559: storage size of `swinfo' isn't known
kvm_getswapinfo.c:561: invalid use of undefined type `struct swdevt'
kvm_getswapinfo.c:561: dereferencing pointer to incomplete type
*** Error code 1
/usr/include/vm/swap_pager.h:
* from: @(#)swap_pager.h 7.1 (Berkeley) 12/5/90
* $FreeBSD: src/sys/vm/swap_pager.h,v 1.35 2002/12/15 19:17:57 dillon Exp $
struct swblock {
struct swblock *swb_hnext;
vm_object_t swb_object; - line 75
vm_pindex_t swb_index;
int swb_count;
daddr_t swb_pages[SWAP_META_PAGES];
};
//David Holm
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: Compilation failure on x86
On Saturday 04 January 2003 15:28, [EMAIL PROTECTED] wrote: In message [EMAIL PROTECTED], David Holm writes: It seems like your #includes are not in sync with your userland, did you use make buildworld or did you simple make all in /usr/src ? I got up too early it seems. Ran make instead of make buildworld. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
yet another nvidia post
Hi, I have the nvidia drivers up and running on my current box and 2d-wise they seem to be working great. But whenever I activate glx I can run glxgears two times and it runs accelerated (I even got a higher framerate than in linux), but the third time I ran it it ran for about 5s then the computer froze. 5-10s later it just reboots. This has happened every time I tried it (the last two or three times I updated world, I update approximately after 1-2 weeks). Has anyone else had this problem and/or know a solution to it? I'm not loading dri, and I've tried disabling agp. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: yet another nvidia post
On Thursday 26 December 2002 00:51, Cliff L. Biffle wrote: I've managed to reproduce a very similar situation using RC1. Does your kernel have various debugging features in place? If so...try turning them /off/. I kid you not. On the RC1 box here the nVidia drivers were very unstable doing 3D until the kernel was recompiled without debugging. (Chris, the owner of the machine, wanted to do it for 'performance' reasons -- getting the graphics working was an unexpected boon.) I can get you the kernel config, or make it available to the list if anyone's interested in this fleeting heisenbug. -Cliff L. Biffle Yes, I've turned off all the debugging stuff. I've also removed Load dri from XF86Config etc. Please post your kernel conf, maybe I can see where they differ (if they do). //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
netinet/tcp.h compile error
Hi, when trying to install adasockets I got the following error message: /usr/include/netinet/tcp.h:44: syntax error before tcp_seq /usr/include/netinet/tcp.h:44: warning: data definition has no type or storage class /usr/include/netinet/tcp.h:45: syntax error before tcp_cc /usr/include/netinet/tcp.h:45: warning: data definition has no type or storage class /usr/include/netinet/tcp.h:55: syntax error before u_short I replaced #include sys/cdefs.h with #include sys/types.h to make it work. Is this a bug? //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Problems connecting locally
Hi, after updating my -CURRENT today (sun dec 15) I can no longer connect to my local cvs and sshd. cvs returns: Connection closed by 217.208.105.23 cvs [commit aborted]: end of file from server (consult above messages if any) and ssh returns: Connection closed by 217.208.105.23 Both these worked flawlessly before the update, the previous version I was running on was about 2 weeks old. The thing is, if I try to connect from remote I have no problems accessing both cvs and ssh. And apache seems unaffected, I can access my webserver both locally and remotely. I have not modified my fiewall since the upgrade, but just in case, it looks like this: 00050 divert 8668 ip from any to any via rl0 00100 allow ip from any to any via lo0 00200 deny ip from any to 127.0.0.0/8 00300 deny ip from 127.0.0.0/8 to any 01050 deny ip from any to any dst-port 139 via rl0 01100 deny ip from any to any dst-port 587 via rl0 01150 deny ip from any to any dst-port 783 via rl0 01200 deny ip from any to any dst-port 901 via rl0 01250 deny ip from any to any dst-port 1024 via rl0 65000 allow ip from any to any 65535 deny ip from any to any //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Problems connecting locally
I ran truss on both cvs and ssh and they stop at the following points: cvs: open(/home/avatar/.cvsignore,0x0,0666) ERR#2 'No such file or directory' access(/home/avatar/.cvswrappers,0)ERR#2 'No such file or directory' pipe() = 3 (0x3) pipe() = 5 (0x5) vfork() = 14645 (0x3935) close(3) = 0 (0x0) close(6) = 0 (0x0) fcntl(0x4,0x2,0x1) = 0 (0x0) fcntl(0x5,0x2,0x1) = 0 (0x0) getdtablesize() = 7322 (0x1c9a) fcntl(0x4,0x3,0x0) = 2 (0x2) fcntl(0x5,0x3,0x0) = 2 (0x2) __sysctl(0xbfbffa08,0x2,0x281cb29c,0xbfbffa04,0x0,0x0) = 0 (0x0) break(0x80d5000) = 0 (0x0) fstat(4,0xbfbff850) = 0 (0x0) break(0x80d9000) = 0 (0x0) write(4,0x80d5000,372) = 372 (0x174) fstat(5,0xbfbff8a0) = 0 (0x0) break(0x80dd000) = 0 (0x0) ssh: socket(0x2,0x1,0x0) = 3 (0x3) . . . open(/home/avatar/.ssh/known_hosts,0x0,0666) = 4 (0x4) fstat(4,0xbfbfca00) = 0 (0x0) read(0x4,0x8076000,0x1000) = 4096 (0x1000) read(0x4,0x8076000,0x1000) = 4096 (0x1000) read(0x4,0x8076000,0x1000) = 1212 (0x4bc) close(4) = 0 (0x0) write(3,0x807,16)= 16 (0x10) write(3,0x807,48)= 48 (0x30) select(0x4,0x806c360,0x0,0x0,0x0)= 1 (0x1) read(0x3,0xbfbfd900,0x2000) = 48 (0x30) write(3,0x807,64)= 64 (0x40) select(0x4,0x806c360,0x0,0x0,0x0)= 1 (0x1) read(0x3,0xbfbfd8f0,0x2000) = 80 (0x50) stat(/home/avatar/.ssh/identity,0xbfbff6d0)ERR#2 'No such file or directory' stat(/home/avatar/.ssh/id_rsa,0xbfbff6d0) ERR#2 'No such file or directory' stat(/home/avatar/.ssh/id_dsa,0xbfbff6d0) ERR#2 'No such file or directory' write(3,0x807,96)= 96 (0x60) //David On Sunday 15 December 2002 22:44, David Holm wrote: Hi, after updating my -CURRENT today (sun dec 15) I can no longer connect to my local cvs and sshd. cvs returns: Connection closed by 217.208.105.23 cvs [commit aborted]: end of file from server (consult above messages if any) and ssh returns: Connection closed by 217.208.105.23 Both these worked flawlessly before the update, the previous version I was running on was about 2 weeks old. The thing is, if I try to connect from remote I have no problems accessing both cvs and ssh. And apache seems unaffected, I can access my webserver both locally and remotely. I have not modified my fiewall since the upgrade, but just in case, it looks like this: 00050 divert 8668 ip from any to any via rl0 00100 allow ip from any to any via lo0 00200 deny ip from any to 127.0.0.0/8 00300 deny ip from 127.0.0.0/8 to any 01050 deny ip from any to any dst-port 139 via rl0 01100 deny ip from any to any dst-port 587 via rl0 01150 deny ip from any to any dst-port 783 via rl0 01200 deny ip from any to any dst-port 901 via rl0 01250 deny ip from any to any dst-port 1024 via rl0 65000 allow ip from any to any 65535 deny ip from any to any //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: jdk 1.3.1_6 linux compilation.
On Tuesday 26 November 2002 18:19, Marcin Dalecki wrote: During a compile of the openoffice-pl package I have just noticed, that 1. Well it depends on java... 2. java depends on the Linux emulation stuff. 3. The automatic fetching of the linux stuff is still looking for RedHat 7.1 files at the ftp.redhat.com FTP site. Well I could live with using OpenOffice without *any* java crap. I never noticed any functionality provided by it. You can add java applets to your documents. I guess it's a portable replacement for microsofts macro crap without the security holes. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: GTK problem after updating system
On Saturday 23 November 2002 07:19, Joe Marcus Clarke wrote: On Fri, 2002-11-22 at 19:26, David Holm wrote: These locales are not valid. Set your locale to: en_US.ISO_8859-1 This gives the exact same error, locale not supported by C library etc. And you should be set. The real problem seems to be your DISPLAY variable. What is it set to? Where are you running xmms from? DISPLAY is set to :0.0 and I'm running locally. xdpyinfo returns name of display::0.0. Besides, so far xmms is the only app that dies with this message, most other apps works fine, but some apps compiled against kdelibs/qt seems to die on certain things that worked before my upgrade. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: GTK problem after updating system
On Saturday 23 November 2002 13:25, David Holm wrote: I found the problem. I had accidentally run make install in the pthread dir of the fbsd source tree before following the recent libpthread question thread on the list which explains that it should not be used. After removing libpthreads and recompiling xmms it works again. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
buildworld fails on x86
Hi, I've tried to build world since yesterday, doing cvsup's inbetween hoping for a fix. But it always fails on bin/cat with a bunch of unresolvable pthread symbols at linking. I have cleared out my old /usr/obj. === bin/cat cc -march=pentium3 -O3 -pipe-Werror -Wall -Wno-format-y2k -Wno-uninitialized -Wformat=2 -Wno-format-extra-args -Werror -static -o cat cat.o /usr/obj/usr/src/i386/usr/lib/libc.a(atexit.o): In function `atexit': atexit.o(.text+0xc7): undefined reference to `_pthread_mutex_unlock' atexit.o(.text+0xd8): undefined reference to `_pthread_mutex_lock' atexit.o(.text+0xe8): undefined reference to `_pthread_mutex_unlock' atexit.o(.text+0x109): undefined reference to `_pthread_mutex_lock' atexit.o(.text+0x11a): undefined reference to `_pthread_mutex_unlock' atexit.o(.text+0x141): undefined reference to `_pthread_mutex_lock' /usr/obj/usr/src/i386/usr/lib/libc.a(_flock_stub.o): In function `flockfile': _flock_stub.o(.text+0x10): undefined reference to `_pthread_self' _flock_stub.o(.text+0x25): undefined reference to `_pthread_mutex_lock' /usr/obj/usr/src/i386/usr/lib/libc.a(_flock_stub.o): In function `_flockfile_debug': _flock_stub.o(.text+0x60): undefined reference to `_pthread_self' _flock_stub.o(.text+0x75): undefined reference to `_pthread_mutex_lock' /usr/obj/usr/src/i386/usr/lib/libc.a(_flock_stub.o): In function `ftrylockfile': _flock_stub.o(.text+0xb5): undefined reference to `_pthread_self' _flock_stub.o(.text+0xca): undefined reference to `_pthread_mutex_trylock' /usr/obj/usr/src/i386/usr/lib/libc.a(_flock_stub.o): In function `funlockfile': _flock_stub.o(.text+0x10d): undefined reference to `_pthread_self' _flock_stub.o(.text+0x149): undefined reference to `_pthread_mutex_unlock' //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
GTK problem after updating system
Hi, I updated my CURRENT system today, previously I was running on a one week old build. After I updated I can no longer run xmms, it returns with the following messages: Gdk-WARNING **: locale not supported by C library Gdk-WARNING **: locale not supported by Xlib, locale set to C Gdk-WARNING **: can not set locale modifiers ** CRITICAL **: Unable to open display I have searched for a solution and found some pages saying I should set $LANG or $LANGUAGE or $LC_LANG but changing these do not help, even if I set one of them to say ISO8859-1 (or even the path to iso8859-1 (/usr/X11R6/lib/X11/locale/iso8859-1)) I still get the same message complaining about locale C not supported. And yeah, the first message, local not supported by C library only appears if I set one of the shell variables. Otherwise I only get the 3 lines below it. I also found one post where a user had recompiled X and got it working so I tried recompiling and reinstalling XFree86-4-Libraries, but it did not help. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: New NVidia drivers on -current
On Monday 18 November 2002 07.29, Mikhail Teterin wrote: Has anyone tried them yet? After removing the #error triggered by __FreeBSD_version being over 50, I got the thing nvidia.ko to build, but: 00:50:30 aldan shutdown: reboot by root: New world, kernel. Nvidia drivers 00:56:12 aldan kernel: Preloaded elf module /boot/kernel/nvidia.ko at 0xc06470b0. 00:56:12 aldan kernel: VESA: NVidia 00:56:12 aldan kernel: VESA: NVidia Corporation NV15 Reference Board Chip Rev A0 00:56:14 aldan kernel: nvidia0: GeForce2 GTS mem 0xf000-0xf7ff,0xe900-0xe9ff irq 11 at device 0.0 on pci1 00:56:14 aldan kernel: pcib1: device nvidia0 requested decoded memory range 0xe900-0xe9ff 00:56:14 aldan kernel: pcib1: device nvidia0 requested unsupported memory range 0x0-0xe9ff (decoding 0xe900-0xe9ff, 0xf000-0xf7ff) 00:56:14 aldan kernel: nvidia0: Unable to allocate NVIDIA memory resource. 00:56:14 aldan kernel: device_probe_and_attach: nvidia0 attach returned 6 Any clues, or is, indeed, a major porting effort required to get it to work on -current? Thanks! -mi Hi, I suggest you read the following post to this list: http://www.freebsd.org/cgi/getmsg.cgi?fetch=1510354+0+/usr/local/www/db/text/2002/freebsd-current/20021110.freebsd-current It talks about the (now deprecated) kernel config option PCI_ALLOW_UNSUPPORTED_IO_RANGE. This was moved to be a loader tunable instead. If you enable this at boot time you should be able to use nvidias driver without any problems on -CURRENT. I haven't had time to try this myself yet so I cannot promise you that it will work. You can find the setting if you try sysctl -a|grep pci, hw.pci.allow_unsupported_io_range should be set to one. Note that this cannot be set during runtime! Hope this helps. //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: New NVidia drivers on -current
On Monday 18 November 2002 09.03, David Holm wrote: Hi, I suggest you read the following post to this list: http://www.freebsd.org/cgi/getmsg.cgi?fetch=1510354+0+/usr/local/www/db/tex t/2002/freebsd-current/20021110.freebsd-current It talks about the (now deprecated) kernel config option PCI_ALLOW_UNSUPPORTED_IO_RANGE. This was moved to be a loader tunable instead. If you enable this at boot time you should be able to use nvidias driver without any problems on -CURRENT. I haven't had time to try this myself yet so I cannot promise you that it will work. You can find the setting if you try sysctl -a|grep pci, hw.pci.allow_unsupported_io_range should be set to one. Note that this cannot be set during runtime! Hope this helps. //David Holm Note that the patch has already been applied so no need to patch your kernel! BTW, why hasn't anyone set the mailing list to automatically set the reply-to address to [EMAIL PROTECTED]? //David To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: alpha tinderbox failure
On Wednesday 13 November 2002 22.57, Dag-Erling Smorgrav wrote: === lib/libc /h/des/src/lib/libc/gen/_pthread_stubs.c:59: conflicting types for `__thr_jtable' /h/des/src/lib/libc/include/libc_private.h:106: previous declaration of `__thr_jtable' *** Error code 1 Stop in /h/des/src/lib/libc. *** Error code 1 Hi, I have the exact same problem on an i386. Checked out the source yesterday (20021113). //David Holm To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
