October 2004 - openSUSE AMD64 - openSUSE Mailing Lists

Mellanox Infiniband on SuSE 9.0 - xp0 module, etc
by Mikhail Kuzminsky 18 Oct '04

18 Oct '04

Dear colleagues ! I've some problem w/Mellanox IB software installation (in particular, the absence of xp0 kernel module). I've installed Infiniband HCA (PCI-X Infinihost MT23108 low profile) to upgarde my interconnect on Tyan S2880 under SuSE Linux 9.0 (2.4.21-243). It's official software platform supported by whole Mellanox IBHPC-0.5.0 software collection. 1)After minor source modification (in mosal.c) the installation (INSTALL script) was finished successfully. But after reboot I see that a) mst tools started successfully b) and I see then following boot messages: Setting up network interfaces : eth0 eth1 - both done ib0: modprobe: modprobe: can't locate module xp0 and ib0 interface is down (it's not connected through IB cable really). But I may do ib0 up manually; in particular, /etc/init.d/network start put ib0 in "up" state. I didn't find xb0.o in /lib/modules, and in any Mellanox software rpm's also ! What should I do now - and what/where is xp0 module ?? 2) I configured IPoIB at the installation. (To try IBsNice) I issued vapi start after boot, and see the message Loading mod_ib_mgt FAILED ("manual" modprobe mod_ib_mgt leads to init_module: device or reesource busy) If I run IBsNice, then I receive the same message about mod_ib_mgt but IBsNice create virtual eth2 , and ping to the IP of eth2 works normally. I'll be very appreciate if somebody clarify me this situation w/mod_ib_mgt ! 3) TO BE MORE CORRECT: pls take into account, that my host w/installed software& Mellanox hardware *isn't connected* currently with IB switch (i.e. is "standalone" server !) Yours Mikhail Kuzminsky Zelinsky Institute of Organic Chemistry Moscow

2 1

Re: [suse-amd64] MPlayer 1.0pre5 - Compile Error `CONFIG_X86_L1_CACHE_SHIFT'
by Scott Leighton 18 Oct '04

18 Oct '04

On Thursday 29 July 2004 4:33 am, you wrote: > I had no problems... Try easing back on the functions your compiling > in... Humm, that zings right over my head. I have to admit my ignorance here. As far as I know, I'm not 'compiling in' anything other than what the package itself calls for. I did a... ./configure --enable-libsuffix=64 make Is there something I should be doing to 'ease back on the functions'? Scott > > On Thu, 2004-07-29 at 16:47, Scott Leighton wrote: > > Having a problem compiling the latest version of MPlayer (I did get > > an earlier version compiled with no problems). > > > > Has anyone gotten the 64 bit version of this to compile, or does > > anyone have a suggestion on how to resolve the error: > > `CONFIG_X86_L1_CACHE_SHIFT' undeclared here (not in a function) > > > > Scott > > > > > > helphand@helphand:~/MPlayer-1.0pre5> make > > make -C libvo > > make[1]: Entering directory `/home/helphand/MPlayer-1.0pre5/libvo' > > cc -c -I../libvo -I../../libvo -I/usr/X11/include -O4 -pipe -ffast-math > > -fomit-frame-pointer -D_REENTRANT -D_LARGEFILE_SOURCE > > -D_FILE_OFFSET_BITS=64 -I. -I.. -I../osdep -I/usr/include/freetype2 > > -I/usr/X11/include -I/usr/include/directfb -DMPG12PLAY -o vo_fbdev.o > > vo_fbdev.c > > In file included from /usr/include/asm/pda.h:4, > > from /usr/include/asm-x86_64/thread_info.h:14, > > from /usr/include/asm/thread_info.h:4, > > from /usr/include/linux/thread_info.h:21, > > from ../osdep/kerneltwosix.h:4, > > from vo_fbdev.c:21: > > /usr/include/asm-x86_64/pda.h:26: error: `CONFIG_X86_L1_CACHE_SHIFT' > > undeclared here (not in a function) > > /usr/include/asm-x86_64/pda.h:26: error: requested alignment is not a > > constant In file included from /usr/include/asm/processor.h:4, > > from /usr/include/linux/prefetch.h:13, > > from /usr/include/linux/list.h:7, > > from ../osdep/kerneltwosix.h:5, > > from vo_fbdev.c:21: > > /usr/include/asm-x86_64/processor.h:229: error: > > `CONFIG_X86_L1_CACHE_SHIFT' undeclared here (not in a function) > > /usr/include/asm-x86_64/processor.h:229: error: requested alignment is > > not a constant > > make[1]: *** [vo_fbdev.o] Error 1 > > make[1]: Leaving directory `/home/helphand/MPlayer-1.0pre5/libvo' > > make: *** [libvo/libvo.a] Error 2 > > helphand@helphand:~/MPlayer-1.0pre5> > > -- > > POPFile, the OpenSource EMail Classifier > > http://popfile.sourceforge.net/ > > Linux 2.6.5-7.95-default x86_64 -- POPFile, the OpenSource EMail Classifier http://popfile.sourceforge.net/ Linux 2.6.5-7.95-default x86_64

5 13

memtest from sles9?
by mmarion＠qualcomm.com 18 Oct '04

18 Oct '04

Does this work for anyone? I've got some hosts that I wanted to run it on, tried via pxe, and now from the SLES9 cd1.. all 3 hosts panic immediately, even the one host that I know is fine. -- Mike Marion-Unix SysAdmin/Staff Engineer-http://www.qualcomm.com My pid is Inigo Montoya. You kill -9 my parent process. Prepare to vi.

7 7

Experience with Acer Aspire 1501LCe
by Sid Boyce 18 Oct '04

18 Oct '04

About a month ago during starup, the CDRW/DVD just put out gibberish instead of the model name etc., this was before it even got to booting up, so yesterday I called Acer Support in the UK with a view of sending it back to have it put right. Luckily as I was talking to Support, I noticed it was OK again. When asked what my operating system was I said it was Linux, no Windows! I was asked to hang on and when he came back he told me his manager said that Linux writes to the BIOS, which may have caused a corruption and that he couldn't help me. He said he had RedHat and Mandrake installed on a few machines there, but he'd never experienced the problem. Seeing as I'm usually polite and tolerant to begin with, I didn't tell him his manager was talking out of a strange piece of his anatomy. The fact that all seemed OK and I was able again to boot knoppix and access the unit, we ended the conversation. Now, Acer supplies laptops with Linux loaded, but XP Home Ed. on this one lasted as long as it took to boot once, then it was gone, so their reaction was a little strange, though somewhat expected. We are still seeing this sort of sick reaction to Linux from the manufacturers. Sometimes you need to get ratty, "Look pal, I'm quite capable of supporting Linux myself, if I needed Linux support, you'd be the last person on earth I'd ask." My friend Pete also had the folks at blueyonder.co.uk tell him that the customer's cablemodem problem was down to Linux even after he was told that the same problem was there using W2K, then he put the phone down on him. The customer called them back and gave them a rocket, they sent someone out to fix the cablemodem and gave him a free link upgrade to 1M. The guy they sent out was not Linux savvy, but he said he could test the modem out using Linux, which he did and passed on a useful tip, so we now know to point a browser at 192.168.100.1 and admin the cablemodem ourselves. May be we need to go on the offensive straight away, letting them know we don't want to hear "We don't support LINE-UX" when we have a genuine hardware problem. Happily running 9.1 x86_64 on this beast, fully functional again. Regards Sid. -- Sid Boyce .... Hamradio G3VBV and keen Flyer =====LINUX ONLY USED HERE=====

3 5

Intel 64-bit compatible compiler for AMD64?
by Jyh-Shyong Ho 18 Oct '04

18 Oct '04

Hi, Recently Intel released 64-bit compatible compiler 8.1 which supports both Intel 64-bit x86 and AMD x86-64 chips. I wonder who has actually tried this new version of compiler on and AMD x86-64 system? I downloaded the non-commercial version of the compiler and installed on my Opteron system, however, only the 32-bit version can be installed. Best regards Jyh-Shyong Ho, Ph.D. Research Scientist National Center for High Performance Computing Hsinchu, Taiwan, ROC

3 2

map_user_kiobuf on 2.4.21-241-numa kernel running suse 8.1 amd64
by Satish Kumar 18 Oct '04

18 Oct '04

Hi, I have a filesystem driver using the map_user_kiobuf error = map_user_kiobuf(WRITE, user_mem, (unsigned long)buf, length_to_be_written); On this new kernel all 'dd' with bs>16k hangs the process and 'top' indicates it's spinning. cheetah:/home# dd if=/dev/zero of=test bs=16834 count=10 <---hangs--> I see that this function has changed a bit in 2.4.21-241 and i am wondering if the new changes are causing this. The thread is looping in map_user_kiobuf and it can not be killed. PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 4281 root 19 0 564 548 444 R 99.6 0.0 108:21 dd Any help is appreciated Thanks, Skumar

3 5

Compiling Perl Modules (Frequencies.pm) and IVTV
by Hilary L Hertzoff 17 Oct '04

17 Oct '04

I have a WinTV PVR 250 that I am trying to get working. After several false starts I was able to compile IVTV. I then tried to run the included utility to set the channel (ptune.pl) and came up with a message that I was missing the Frequencies.pm module, so I located the file it was included in - Video-Capture-V4l-0.224, unpacked it and attempted to compile it. The command I used was perl Makefile.pl This seemed to work fine, but when I try to run make, I get the following output (below). The relevant bit seemed to be about 6 rows from the bottom - a message that I should recompile with -fPIC I tried running perl Makefile.pl -fPIC make and got the same error. So what should I try next? Is there another way to set the channel on the card. I've tried xawtv (which crashes) and Kwintv (which realizes that there's a video card available, but doesn't recognize any input). I have a running vcr attached to the card (the card was intended for transferring old video tapes to dvds). Thanks, Hilary This is the output of the make command: make[1]: Entering directory `/home/hhertzof/Video-Capture-V4l-0.224/V4l' cc -c -D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS -fno-strict-aliasing -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -fmessage-length=0 -Wall -Wall -pipe -DVERSION=\"0.224\" -DXS_VERSION=\"0.224\" -fPIC "-I/usr/lib/perl5/5.8.3/x86_64-linux-thread-multi/CORE" V4l.c V4l.c: In function `XS_Video__Capture__V4l_reduce2': V4l.c:1766: warning: unused variable `RETVAL' V4l.c: In function `XS_Video__Capture__V4l_findmin': V4l.xs:652: warning: `mindata' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Tuner_mode': V4l.c:1679: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Picture_palette': V4l.c:1511: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Picture_depth': V4l.c:1483: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Picture_whiteness': V4l.c:1455: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Picture_contrast': V4l.c:1427: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Picture_colour': V4l.c:1399: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Picture_hue': V4l.c:1371: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Picture_brightness': V4l.c:1343: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Audio_step': V4l.c:1315: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Audio_balance': V4l.c:1287: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Audio_mode': V4l.c:1259: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Audio_flags': V4l.c:1203: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Audio_treble': V4l.c:1175: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Audio_bass': V4l.c:1147: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Audio_volume': V4l.c:1119: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Channel_norm': V4l.c:1063: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Tuner_tuner': V4l.c:1539: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Audio_audio': V4l.c:1091: warning: `RETVAL' might be used uninitialized in this function V4l.c: In function `XS_Video__Capture__V4l__Channel_channel': V4l.c:923: warning: `RETVAL' might be used uninitialized in this function Running Mkbootstrap for Video::Capture::V4l () chmod 644 V4l.bs rm -f ../blib/arch/auto/Video/Capture/V4l/V4l.so LD_RUN_PATH="" cc -shared -L/usr/local/lib64 V4l.o -o ../blib/arch/auto/Video/Capture/V4l/V4l.so -lpthread chmod 755 ../blib/arch/auto/Video/Capture/V4l/V4l.so cp V4l.bs ../blib/arch/auto/Video/Capture/V4l/V4l.bs chmod 644 ../blib/arch/auto/Video/Capture/V4l/V4l.bs Manifying ../blib/man3/Video::Capture::V4l.3pm make[1]: Leaving directory `/home/hhertzof/Video-Capture-V4l-0.224/V4l' make[1]: Entering directory `/home/hhertzof/Video-Capture-V4l-0.224/VBI' cc -c -D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS -fno-strict-aliasing -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -fmessage-length=0 -Wall -Wall -pipe -DVERSION=\"0.05\" -DXS_VERSION=\"0.05\" -fPIC "-I/usr/lib/perl5/5.8.3/x86_64-linux-thread-multi/CORE" VBI.c VBI.xs: In function `unham16': VBI.xs:258: warning: unused variable `bit' VBI.xs: In function `decoder_decode_vps': VBI.xs:296: warning: suggest parentheses around arithmetic in operand of | VBI.xs:296: warning: suggest parentheses around arithmetic in operand of | VBI.xs:296: warning: suggest parentheses around arithmetic in operand of | VBI.xs:296: warning: suggest parentheses around arithmetic in operand of | VBI.xs:302: warning: suggest parentheses around arithmetic in operand of | VBI.xs: In function `decoder_decode_vt': VBI.xs:380: warning: suggest parentheses around comparison in operand of & VBI.xs: In function `decoder_decode_other': VBI.xs:397: warning: unused variable `pos' VBI.xs: In function `decode_stream': VBI.xs:780: warning: suggest parentheses around assignment used as truth value VBI.xs: In function `decode_block': VBI.xs:1175: warning: suggest parentheses around arithmetic in operand of | VBI.c: In function `XS_Video__Capture__VBI__EPG_decode_block': VBI.c:1425: warning: unused variable `RETVAL' VBI.xs: In function `get_byte_SE': VBI.xs:206: warning: `byte' might be used uninitialized in this function VBI.xs: In function `decode_block': VBI.xs:1198: warning: control reaches end of non-void function VBI.xs:1098: warning: `msg_len' might be used uninitialized in this function VBI.c: In function `XS_Video__Capture__VBI_decode_ansi': VBI.xs:692: warning: `o' might be used uninitialized in this function VBI.c: At top level: VBI.xs:32: warning: `invtab' defined but not used Running Mkbootstrap for Video::Capture::VBI () chmod 644 VBI.bs rm -f ../blib/arch/auto/Video/Capture/VBI/VBI.so LD_RUN_PATH="" cc -shared -L/usr/local/lib64 VBI.o -o ../blib/arch/auto/Video/Capture/VBI/VBI.so chmod 755 ../blib/arch/auto/Video/Capture/VBI/VBI.so cp VBI.bs ../blib/arch/auto/Video/Capture/VBI/VBI.bs chmod 644 ../blib/arch/auto/Video/Capture/VBI/VBI.bs Manifying ../blib/man3/Video::Capture::VBI.3pm make[1]: Leaving directory `/home/hhertzof/Video-Capture-V4l-0.224/VBI' make[1]: Entering directory `/home/hhertzof/Video-Capture-V4l-0.224/RTjpeg' cc -c -D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS -fno-strict-aliasing -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -fmessage-length=0 -Wall -Wall -pipe -DVERSION=\"0.012\" -DXS_VERSION=\"0.012\" -fPIC "-I/usr/lib/perl5/5.8.3/x86_64-linux-thread-multi/CORE" RTjpeg.c RTjpeg.xs: In function `boot_Video__RTjpeg': RTjpeg.xs:112: warning: unused variable `stash' cc -c -D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS -fno-strict-aliasing -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -fmessage-length=0 -Wall -Wall -pipe -DVERSION=\"0.012\" -DXS_VERSION=\"0.012\" -o codec/RTjpeg.o codec/RTjpeg.c codec/RTjpeg.c: In function `RTjpeg_b2s': codec/RTjpeg.c:99: warning: use of cast expressions as lvalues is deprecated Running Mkbootstrap for Video::RTjpeg () chmod 644 RTjpeg.bs rm -f ../blib/arch/auto/Video/RTjpeg/RTjpeg.so LD_RUN_PATH="" cc -shared -L/usr/local/lib64 RTjpeg.o codec/RTjpeg.o -o ../blib/arch/auto/Video/RTjpeg/RTjpeg.so /usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld: codec/RTjpeg.o: relocation R_X86_64_32S can not be used when making a shared object; recompile with -fPIC codec/RTjpeg.o: could not read symbols: Bad value collect2: ld returned 1 exit status make[1]: *** [../blib/arch/auto/Video/RTjpeg/RTjpeg.so] Error 1 make[1]: Leaving directory `/home/hhertzof/Video-Capture-V4l-0.224/RTjpeg' make: *** [subdirs] Error 2

1 0

Small RPM gotcha
by Peter Dalgaard 17 Oct '04

17 Oct '04

Only a minor point, and I think I can wiggle my way around it, but as it might break scripts for others, I thought I should report it. The following is abstracted from a script by Detlef Steuer (Bcc'ed so that he doesn't get every followup...) to compute a list of required packages for a binary. Doesn't quite work on AMD64 as you will see: pd@linux:~/> ldd /usr/lib/R/modules/R_X11.so | grep jpeg libjpeg.so.62 => /usr/lib64/libjpeg.so.62 (0x0000002a95ace000) pd@linux:~/> rpm -q --whatprovides libjpeg.so.62 libjpeg-32bit-9.1-200404070910 I.e. 32 bit compatibility packages are found and not the correct 64 bit ones. One workaround seems to be based on pd@linux:~/> rpm -q --whatprovides 'libjpeg.so.62()(64bit)' libjpeg-6.2.0-320 which is a bit of a pain because the script is really like rpm -q --whatprovides `ldd ... | gawk '{print $1}' |....` and you have to deal with the fact that some elements come out with explicit paths, and RPM doesn't cope; only the latter of the following two will work: rpm -q --whatprovides '/lib64/ld-linux-x86-64.so.2()(64bit)' rpm -q --whatprovides 'ld-linux-x86-64.so.2()(64bit)' Another, simpler, workaround is to use the RHS of the "=>" in the ldd output (gawk '{print $3}'). However, this is sometimes "not found" since the application jiggles the library path. -- O__ ---- Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~~~~~~~~~ - (p.dalgaard(a)biostat.ku.dk) FAX: (+45) 35327907

2 2

32bit or 64 bit installation?
by Raimund Dold 16 Oct '04

16 Oct '04

Hi everybody! I am planing to buy a new desktop which will most probably be an amd64 system (e.g. www.deutschlandpc.de) I am listening to the list since quite a time to get an impression what the status of the 64bit Suse is for a "normal" user. Questions: 1. Most systems I saw do have an ATI graphic driver, which are not supported in 64 bit, right? What does that mean? No X? 2. I have a AVM Fritz! DSL card, is this one supported in 9.2? 3. Would it be better for a desktop to install the 32 bit version, because hw support is more complete? Thanx Raimund

5 4

Flash for 64-bit
by Tom Nielsen 15 Oct '04

15 Oct '04

Is anyone using Flash on their browser? I normally use Firebird. I can't seem to find a version of Flash that will work. Is there one? If so, where can I get it? Thanks! -- Tom Nielsen Neuro Logic Systems 1.805.389.5435 x18 www.neuro-logic.com

4 3