/tmp/cc4koWpS.s: Assembler messages: /tmp/cc4koWpS.s:1214: Error: operands given don't match any known 386 instruction /tmp/cc4koWpS.s:1221: Error: operands given don't match any known 386 instruction make[1]: *** [signals.o] Error 1 make[1]: Leaving directory `/6/backup/glibc-2.2.5/linuxthreads' make: *** [linuxthreads/others] Error 2
Help :-(
gcc version 2.95.3 20010315 (release) (3.0.x doesn't compile on my system)
glibc-2.2.5
linuxthreads: glibc-linuxthreads-2.2.5.tar.gz
Kernel Linux 2.4.8.
It's an AMD Duron or something.
On Sunday 14 July 2002 14:44, you wrote:
/tmp/cc4koWpS.s: Assembler messages: /tmp/cc4koWpS.s:1214: Error: operands given don't match any known 386 instruction gcc version 2.95.3 20010315 (release) (3.0.x doesn't compile on my system)
perhaps you need some specfic configure options... which release of 3.0x ... I had a problem with one of the earlier ones ( specifically the one from RHL 7.1 ).. thats well know now... & yeah.. are you using RHL ? your problems seems to stem from the instruction set of your chip somehow giving a prob.
which is why this statement "disturbs" me ;-) ( below ) " > It's an AMD Duron or something. "... or something ?????? what does your kernel report ...
.... then again it could be another error ....
C'ya Erle
On Jul 15, 2002 at 10:10, Erle Pereira wrote:
perhaps you need some specfic configure options... which release of 3.0x ...
I had a problem with one of the earlier ones ( specifically the one from RHL
gcc 3.0.4
7.1 ).. thats well know now... & yeah.. are you using RHL ?
Yes, it's RHL. Usually I'd say that's irrelevant, but I think my current libraries are still RHL 6.x.
your problems seems to stem from the instruction set of your chip somehow giving a prob.
Yes, it thinks I'm using an i386?
which is why this statement "disturbs" me ;-) ( below )
" > It's an AMD Duron or something. "... or something ?????? what does your kernel report ...
No, I configured the kernel correctly, it's an i686 chip.
The complete error:
[satyap]/6/backup/glibc-2.2.5/linuxthreads$ make gcc signals.c -c -O2 -Wall -Winline -Wstrict-prototypes -Wwrite-strings -g -fno-omit-frame-pointer -I../include -I. -I.. -I../libio -I../sysdeps/i386/elf -I../linuxthreads/sysdeps/unix/sysv/linux/i386 -I../linuxthreads/sysdeps/unix/sysv/linux -I../linuxthreads/sysdeps/pthread -I../sysdeps/pthread -I../linuxthreads/sysdeps/unix/sysv -I../linuxthreads/sysdeps/unix -I../linuxthreads/sysdeps/i386/i686 -I../linuxthreads/sysdeps/i386 -I../sysdeps/unix/sysv/linux/i386/i686 -I../sysdeps/unix/sysv/linux/i386 -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet -I../sysdeps/unix/sysv/i386 -I../sysdeps/unix/sysv -I../sysdeps/unix/i386 -I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/i386/i686/fpu -I../sysdeps/i386/i686 -I../sysdeps/i386/i486 -I../sysdeps/i386/fpu -I../sysdeps/i386 -I../sysdeps/wordsize-32 -I../sysdeps/ieee754/ldbl-96 -I../sysdeps/ieee754/dbl-64 -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754 -I../sysdeps/generic/elf -I../sysdeps/generic -D_LIBC_REENTRANT -include ../include/libc-symbols.h -o signals.o /tmp/ccOXB4Wg.s: Assembler messages: /tmp/ccOXB4Wg.s:1214: Error: operands given don't match any known 386 instruction /tmp/ccOXB4Wg.s:1221: Error: operands given don't match any known 386 instruction make: *** [signals.o] Error 1 [satyap]/6/backup/glibc-2.2.5/linuxthreads$
(Wow, that's nicely formatted.)
[satyap]/6/backup/glibc-2.2.5/linuxthreads$ ls sysdeps/i386/ Makefile i586/ i686/ pspinlock.c pt-machine.h useldt.h
[satyap]/6/backup/glibc-2.2.5/linuxthreads$ ls sysdeps/unix/sysv/linux/ Implies bits/ configure hppa/ i386/ ia64/ s390/ sh/ sparc/
[satyap]/6/backup/glibc-2.2.5/linuxthreads$ ls sysdeps/unix/sysv/linux/i386/ bits/
[satyap]/6/backup/glibc-2.2.5/linuxthreads$ ls sysdeps/unix/sysv/linux/i386/bits/ posix_opt.h
uname is: Linux gort.cjb.net 2.4.18 #1 Sat Jul 13 13:20:07 PDT 2002 i686 unknown
(Note the i686)
(PS: What's the #1 in the uname?)
gcc version 2.95.3 20010315 (release) (3.0.x doesn't compile on my system)
glibc-2.2.5
linuxthreads: glibc-linuxthreads-2.2.5.tar.gz
It's an AMD Duron or something.
Almost certainly a Duron 700 or 900, something like that. Does it matter?
On Monday 15 July 2002 17:02, Satya wrote:
Yes, it thinks I'm using an i386?
:-) what I meant was .. few of the operands in the assembler file ( *.s ) do not match or are commented, hence the compiler is giving you an error. Nothing necessarily wrong with it detecting an i386, its probably recognising it as an i386 class ( as against sparc, arm etc ....)
[satyap]/6/backup/glibc-2.2.5/linuxthreads$ make gcc signals.c -c -O2 -Wall -Winline -Wstrict-prototypes -Wwrite-strings -g -fno-omit-frame-pointer -I../include
< snip >
/tmp/ccOXB4Wg.s: Assembler messages: /tmp/ccOXB4Wg.s:1214: Error: operands given don't match any known 386 instruction /tmp/ccOXB4Wg.s:1221: Error: operands given don't match any known 386 instruction make: *** [signals.o] Error 1
{ it could be just a random bug ... but... I have a tendency to go off the deep end... so ............................ }
Most likely to be some kind of syntax / code error, either due to oversight, or somehow that file is corrupted --- start looking at the code from "signals.c " and the includes etc... :-) ... yeah i know ... the "FUN" part of working with code. 1214 , 1221 will be the line # at which you will find it.... it probably be a "*.s" file... _or_ What is probably happening is that it is detecting some instructions related to your chip which it can't / won't compile or trying to compile but doesn't know how...
Check in the configure options.. if there is anything to disable special detection of anyting related to your hardware ( like 3d processing... I do not know why / if this could be an issue... but its worth a check )
~~~ you will get a similar error if you compile mpg123 ( 0.59r ) with 3d-now instructions.. reason being that the one of the files containing the instructions used when 3dnow is enabled... has got some comments, after uncomenting, it works... ..hopefully this case is somewhat similar ( if you could ... try compiling it on a different distro... ( non RHL )) I could try it... but really tied up at present.. If you could wait up a bit.. we could get to the bottom of this..
(Wow, that's nicely formatted.)
yep... good work ...!!!! :-)
uname is: Linux gort.cjb.net 2.4.18 #1 Sat Jul 13 13:20:07 PDT 2002 i686 unknown
(PS: What's the #1 in the uname?)
'#1' is one of the fields reported when you ask for the operating system version... mine reports #24 ( slackware 8.0, kernel 2.4.18 , gcc 2.95.3 )... I do not think its related to the kernel proper... ( my kernel version is the same.. ) , my best guess some kind of internal versioning system... not documented in the man , info pages
Almost certainly a Duron 700 or 900, something like that. Does it matter?
not a bit... why i had asked this was... in case you were using a one of the later (new ) Athlons, thinking of going for a Athlon 18+ myself... was just checking.
C'ya Erle.
On Tue, 2002-07-16 at 04:37, Erle Pereira wrote:
[snip]
uname is: Linux gort.cjb.net 2.4.18 #1 Sat Jul 13 13:20:07 PDT 2002 i686 unknown
(PS: What's the #1 in the uname?)
'#1' is one of the fields reported when you ask for the operating system version... mine reports #24 ( slackware 8.0, kernel 2.4.18 , gcc 2.95.3 )... I do not think its related to the kernel proper... ( my kernel version is the same.. ) , my best guess some kind of internal versioning system... not documented in the man , info pages
[snip]
uname info page says its part of OSVERSION,
OSVERSION includes , #1 Sat Jul 13 13:20:07 PDT 2002
is this the number of successful builds from a given kernel source ? perhaps not.
-Rajesh
On Wednesday 17 July 2002 09:54, Rajesh Deo wrote:
OSVERSION includes , #1 Sat Jul 13 13:20:07 PDT 2002
is this the number of successful builds from a given kernel source ? perhaps not.
Yes... thats what it seems to be... anybody knows why would this be stored ... ??
C'ya Erle.
On Jul 16, 2002 at 14:07, Erle Pereira wrote:
/tmp/ccOXB4Wg.s: Assembler messages: /tmp/ccOXB4Wg.s:1214: Error: operands given don't match any known 386 instruction /tmp/ccOXB4Wg.s:1221: Error: operands given don't match any known 386 instruction make: *** [signals.o] Error 1
start looking at the code from "signals.c " and the includes etc... :-) ... yeah i know ... the "FUN" part of working with code. 1214 , 1221 will be the line # at which you will find it.... it probably be a "*.s" file...
It's in a .s file, which is deleted as soon as the make stops. Those files are in /tmp.
Check in the configure options.. if there is anything to disable special
The configure script consists of:
exit 0
detection of anyting related to your hardware ( like 3d processing... I do not know why / if this could be an issue... but its worth a check )
Possible. I've had trouble with 3D games and AMD processors before.
Googling suggests that the binutils may be at fault, so I've just downloaded a new binutils (2.12). I'll try it out.
Gah, that's a big package! It's got a very long configure process, too, it's still running. Okay, seems to be compiling fine. I'll drop to single user to install it.
Alright, new binutils in place and.... whaddya know! linuxthreads compiles! Yippee.
-
Erle Pereira -
Rajesh Deo -
Satya