July 2005 - openSUSE AMD64 - openSUSE Mailing Lists

RE: [suse-amd64] 9.2 upgrade to 9.3
by Black, Alain 07 Jul '05

07 Jul '05

I've performed the upgrade on a single CPU machine. My only "failure?" if you will is the image for the grub screen is gone so the grub boot loader is in text mode. Other than that everything that I use it for functions normally. There is a section in the release notes (manuals) about what will not be updated doing an upgrade vs fresh install, but none of it was applicable to my system. -Alain. -----Original Message----- From: James D. Parra [mailto:Jamesp@MusicReports.com] Sent: Wednesday, July 06, 2005 3:23 PM To: SuSE AMD64 (E-mail) Subject: [suse-amd64] 9.2 upgrade to 9.3 Hello, I have 9.2 installed on a Dual Opteron and want to upgrade to 9.3. Has anyone had any success with upgrading? Curious to see if anyone has had any good results before I scrub the OS and do a fresh install. Thanks in advance, James -- Check the List-Unsubscribe header to unsubscribe For additional commands, email: suse-amd64-help(a)suse.com

1 0

RE: [suse-amd64] mmap & /dev/shm
by Calkins, Dennis 07 Jul '05

07 Jul '05

Have you looked into HUGEPAGES? These are allocated as 2 MEG pages instead of 4 K pages. You can set aside part of memory for HUGE pages at Boot time. Maybe SUSE doesn't recommend using HUGEPAGES I don't know. /boot/grub/menu.lst and add the following string to the kernel line: hugepages=xx where xx is the number of HUGE pages you want allocated. You can also put a value in /etc/sysctl.conf file: vm/nr_hugepages = xx then on the SHMGET adds the parameter SHM_HUGETLB And the pages will come out of these HUGE pages. Now I have never quite figure out how to calculate a good value for 'xx'. Also I can't find any documentation that states if you can give pages back if you over allocate HUGE pages. or what happens if 'xx' is too small for your requested allocation. Will part of it come out of huge pages and the rest come out of Small pages or will it just fail. --- -----Original Message----- From: Brad Dameron [mailto:brad@seatab.com] Sent: Thursday, July 07, 2005 9:12 AM To: SuSE-Linux-E(a)suse.com; suse-amd64(a)suse.com Subject: Re: [suse-amd64] mmap & /dev/shm Sorry for cross-list messaging. But I use SuSe on both i686 and 64bit. On Thu, 2005-07-07 at 11:48 +0200, Andi Kleen wrote: > On Wed, Jul 06, 2005 at 03:59:32PM -0700, Brad Dameron wrote: > > We have a in-house application that uses mmap to pull a file into > > memory. I can't seem to find any info on where this file is stored > > in memory. I know that it is shared memory. But does the /dev/shm > > have anything to do with it. Isn't shm just a tmpfs? Also can shm be > > Not directly no. Yes. > > > disabled? I have a machine with 16GB and 8GB then goes to the shm. > > Does > > Yes, but you probably don't need to. > > It just means that shm could grow upto 8GB if someone put that much > data into it by creating files in it. Normal mmap doesn't do that. > > > this only allow mmap to use the other 8GB or is all 16GB accessible > > to it? These are the questions I have been trying to figure out for > > over a > > All 16GB accessible for it. Actually a few percent less because there > are some other data structures like page tables that need to fit into > memory too. > > > week now. Not many answers on google and no one has seemed to ask > > this before. > > > > Also is mmap the fastest way to get a file from disk to memory? > > It depends. Normal read() does more aggressive readahead which can be > faster. However mmap() won't copy and you cache the data only once. > > -Andi Andi, First thanks for your response. If I mmap say a 1GB file, then I mmap another 1GB file. It will then have both files in memory correct? And how long will it keep them in memory? Also I noticed it uses PAGE_SIZE as the block size to grab the files in. Is there a way to increase this just for mmap or is the default of 4096kB a good size to use. We mainly deal with 100MB to 5GB file sizes. Nothing really small. And usually everything sequential. Any further answer's appreciated. Thank you, Brad Dameron SeaTab Software www.seatab.com -- Check the List-Unsubscribe header to unsubscribe For additional commands, email: suse-amd64-help(a)suse.com

1 0

mmap & /dev/shm
by Brad Dameron 07 Jul '05

07 Jul '05

We have a in-house application that uses mmap to pull a file into memory. I can't seem to find any info on where this file is stored in memory. I know that it is shared memory. But does the /dev/shm have anything to do with it. Isn't shm just a tmpfs? Also can shm be disabled? I have a machine with 16GB and 8GB then goes to the shm. Does this only allow mmap to use the other 8GB or is all 16GB accessible to it? These are the questions I have been trying to figure out for over a week now. Not many answers on google and no one has seemed to ask this before. Also is mmap the fastest way to get a file from disk to memory? Any help appreciated. Thank you, Brad Dameron SeaTab Software www.seatab.com

2 2

compiling gimp plugin for amd64
by Dave Howorth 06 Jul '05

06 Jul '05

Hello, I'm trying to use a gimp plugin called wideangle, but it's crashing with a segfault. I've talked to the developer, who doesn't have an amd64. It seems possible that the problem is some amd64 'gotcha' like a compiler flag or something (I was once a C programmer but am no longer current). I haven't found anything in the FAQ or by searching the archives. So I'm wondering if there's anywhere specific I should be looking or if the following details suggest any solution. I'm running Suse 9.3 on an athlon64 box. I installed gimp from the release with Yast and built the plugin from source using the command: /opt/gnome/bin/gimptool-2.0 --install wideangle.c It logged the build: /usr/bin/install -c -d /root/.gimp-2.2/plug-ins gcc -O2 -fmessage-length=0 -Wall -g -fno-strict-aliasing -I/opt/gnome/include/gimp-2.0 -DXTHREADS -D_REENTRANT -DXUSE_MTSAFE_API -I/opt/gnome/include/gtk-2.0 -I/opt/gnome/lib64/gtk-2.0/include -I/usr/X11R6/include -I/opt/gnome/include/atk-1.0 -I/opt/gnome/include/pango-1.0 -I/usr/include/freetype2 -I/usr/include/freetype2/config -I/opt/gnome/include/glib-2.0 -I/opt/gnome/lib64/glib-2.0/include -o /root/.gimp-2.2/plug-ins/wideangle wideangle.c -L/opt/gnome/lib64 -lgimpui-2.0 -lgimpwidgets-2.0 -lgimp-2.0 -lgimpcolor-2.0 -lgimpmath-2.0 -lgimpbase-2.0 -lgtk-x11-2.0 -lgdk-x11-2.0 -latk-1.0 -lgdk_pixbuf-2.0 -lm -lpangoxft-1.0 -lpangox-1.0 -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -ldl -lglib-2.0 When I try to use the plugin, I see the following error message: /home/dhoworth/.gimp-2.2/plug-ins/wideangle: fatal error: Segmentation fault and when I turn on gimp's stack trace I see the following: #0 0x00002aaaac4c1e4b in fork () from /lib64/tls/libc.so.6 #1 0x00002aaaac2bcbfd in g_on_error_stack_trace () #2 0x00002aaaaaecccd4 in gimp_main () from /opt/gnome/lib64/libgimp-2.0.so.0 #3 <signal handler called> #4 0x00002aaaac4a26a0 in strlen () from /lib64/tls/libc.so.6 #5 0x00002aaaac2e7d63 in g_strdup () from /opt/gnome/lib64/libglib-2.0.so.0 #6 0x00002aaaabf96995 in g_value_dup_string () #7 0x00002aaaabf7bc93 in g_object_new_valist () #8 0x00002aaaabf7bf47 in g_object_new () #9 0x00002aaaab3712f5 in gtk_button_new_from_stock () #10 0x00002aaaab3b4c57 in gtk_dialog_add_button () #11 0x00002aaaaacf8a03 in gimp_dialog_add_buttons_valist () #12 0x00002aaaaacf8bf1 in gimp_dialog_new_valist () #13 0x00002aaaaacf8da7 in gimp_dialog_new () #14 0x0000000000404228 in run (name=<value optimized out>, nparams=3, #15 0x00002aaaaaecc98f in gimp_main () from /opt/gnome/lib64/libgimp-2.0.so.0 #16 0x00002aaaac45054a in __libc_start_main () from /lib64/tls/libc.so.6 #17 0x00000000004028ea in _start () at start.S:113 #18 0x00007fffffffde38 in ?? () #19 0x0000000000000000 in ?? () <snip> Thanks, Dave -- No virus found in this outgoing message. Checked by AVG Anti-Virus. Version: 7.0.323 / Virus Database: 267.8.9/42 - Release Date: 06/07/05

1 0

Problem with VMware and 2.6.8-24.16 kernel
by donald.leslie＠comcast.net 05 Jul '05

05 Jul '05

When I run vmware-config.pl I get the following: make: Leaving directory `/tmp/vmware-config8/vmmon-only' Unable to make a vmmon module that can be loaded in the running kernel: insmod: error inserting '/tmp/vmware-config8/vmmon.o': -1 Invalid parameters There is probably a slight difference in the kernel configuration between the set of C header files you specified and your running kernel. You may want to rebuild a kernel based on that directory, or specify another directory. Any suggestions. I got the kernel-source with apt-get from ftp://ftp.gwdg.de SuSE/9.2-x86_64 Don

4 4

Matlab 7.0 SP2 process
by Karaoulis Marios 04 Jul '05

04 Jul '05

When i start matlab (64 bit version) on my suse 9.3 x86_64 it appears on the process table about 10 times consuming each process some memory (10 times more memory). This behavior doesn't appear on my other pc (running debian 3.1 , kernel 2.6.8 i386, same version of matlab but 32 bit). Does anyone have noticed this behavior? __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

1 0

Re: [suse-amd64] Puzzling wild variation of execution time
by Andi Kleen 03 Jul '05

03 Jul '05

[Readding mailing list] On Sun, Jul 03, 2005 at 12:42:48AM -0700, Sheo Shanker Prasad wrote: > (1) How do I boot with that option. Just enter it at the grub boot prompt. > > (2) I have 4 GB of PC3200 (DDR) memory in 4 dimms that fill two dimms on cpu0 > and two dimms on cpu1 installed in Tyan Thunder K8W (S2885 ANRF) system > board. In this case what should be the optimal number to put in mem= xxxxM? Some smaller number with which your application still runs. Let's say mem=3000M This would be just to test if the bad memory is at the end. > > (3) Does my present problem have any connection with bug that you and Mark > Langsdorf (of AMD, Inc)? [To refresh your memory about this correspondence > (which I picked up in the course of Internet Search) this was aTyan BIOS MTRR > initialization bug (i.e., Some Tyan AMD BIOS don't initialize the first fixed > range MTRR, which caused it to contain random bogus values. When the MTRR > tries to duplicate the MTRR state to other CPUs at startup it oopses because > of this). I also gathered that you had a patch to overcome this problem. ] That one caused a crash a boot, so unlikely. -Andi

1 0

Re: Re: [suse-amd64] Marvell GigE chip under SuSE 9.x
by Arne 01 Jul '05

01 Jul '05

I have a Shuttle XPC SN95G5 (v1) with motherboard FN95 that includes a Marvell Yukon 8001 GB network device and have similar problems as with the two A makers. That is I get VPD keys software errors from kernel, it allways reports zero bytes send/recive and small temporar lockups (~0.1s) when monitoring the device. The ethernet device works though, so it's not a that big problem, just wanted to report the problem. Problem is with SuSE64 9.2 and 9.3 latest BIOS update and OS update. /Arne On 19.6 2005 17:06, Andi Kleen wrote: > >On Sun, 19 Jun 2005 01:32:30 +0200 (CEST) >Harald Milz <hm(a)seneca.muc.de> wrote: > >> Bernd Paysan <bernd.paysan(a)gmx.de> wrote: >> >> > I haven't seen any problems with the Marvell GigE chip for 1.5 years, >> > using all SuSE distributions along. >> >> There was a confirmed bug in the SUSE 9.2 kernels. > >No actually there was a hardware bug on some motherboards by >a particular big Taiwanese vendor starting with A (they had >incorrect NIC eeprom checksums) We later added a workaround for that >to ignore the wrong checksum. But the Linux code was correct. > > >> See the January thread. >> If you didn't have any problems, you were lucky. I had. > >You should have complained to your motherboard vendor. > >> BTW has this problem been solved in the latest 9.2 update kernels? Andi? > >Yes, the patch made it into the update kernel. > >-Andi > >-- >Check the List-Unsubscribe header to unsubscribe >For additional commands, email: suse-amd64-help(a)suse.com > > ------------------------------------------------------------------ Gratis svenskpråkig e-post och förmånliga textmeddelanden http://webmail.surfnet.fi ------------------------------------------------------------------

1 0

Puzzling wild variation of execution time
by Sheo Shanker Prasad 01 Jul '05

01 Jul '05

I will greatly appreciate help in understanding the puzzling wild variation in the execution time of small test/benchmark program, as explained below. (1) The nature of the benchmark/test program: A floating point arithmetic intensive atmospheric chemistry modeling task. It used to take, in a consistent manner, around 2m 37s on my machine when it was running under SuSe 9.1 Pro. When I run the benchmark program, I shut-off all other programs. Thus, during its run, it is the only active program. THe rest are belonging to root are sleeping. There is no other user. (2) The machine: Dual Opteron 250 with 4 GB of PC3200 (DDR) memory in 4 dimms that fill two dimms on cpu0 and two dimms on cpu1 installed in Tyan Thunder K8 system board that has American MegaTrend AMIBIOS V2-04. It now runs under SuSE 9.3 Pro and the Linux version 2.6.11.4-21.7-smp (geeko@buildhost) (gcc version 3.3.5 20050117 (prerelease) (SUSE Linux)) #1 SMP Thu Jun 2 14:23:14 UTC 2005 (3) Background of the problem: After only 4 months of the delivery of the machine by the vendor, the mother board failed. The machine was sent to the vendor. He then replaced the mother board with a new one and installed SuSE 9.3 Pro to bring the machine up-to-date with the Linux version 2.6.11.4-21.7-smp (geeko@buildhost) (gcc version 3.3.5 20050117 (prerelease) (SUSE Linux)) #1 SMP Thu Jun 2 14:23:14 UTC 2005 (4) The problem: After the repair and upgrade to 9.3 Pro, the execution times of the same test/benchmark program varies widely between run to run. It is never consistent. It varies from 2m 37s to more than 6m. A pattern is seen in the variation. Usually, the longest execution time is when the machine is started after remaining shut-off for a few hours. The shortest execution time is obtained after the machine has been constantly on for abot 7 to 10 hours. (5) My puzzle: While a computer is well known to be reproducible, the wild variation by a factor of almost 3 in the execution is very disturbing. What could be the cause? Does it mean that some vital component involved in execution of the numerical modeling (RAM memory or the memory on the CPUs) are loose or erratic and may be about to go bad? I am just puzzled. Any clue from you all will be really very very much appreciated. -- Best regards. Sheo (Sheo S. Prasad) Creative Research Enterprises 6354 Camino del Lago Pleasanton, CA 94566, USA Voice Phone: (+1) 925 426-9341 Fax Phone: (+1) 925 426-9417 e-mail: ssp(a)CreativeResearch.org

5 5

smbfs: unsupported module, tainting kernel.
by detlinux＠gmx.ch 01 Jul '05

01 Jul '05

Hi, I don't suceed with smbmount. The smbfs module seems not to be ok: With "dmesg | grep smb" I get " smbfs: unsupported module, tainting kernel. " I have the 2.6.11.4-21.7-smp kernel. What can I do? Christian Geschenkt: 3 Monate GMX ProMail gratis + 3 Ausgaben stern gratis ++ Jetzt anmelden & testen ++ http://www.gmx.net/de/go/promail ++ -- Weitersagen: GMX DSL-Flatrates mit Tempo-Garantie! Ab 4,99 Euro/Monat: http://www.gmx.net/de/go/dsl

5 10