-> Gerhard Schneider (gs@ilfb03.tuwien.ac.at) wrote:
-> >Kernel: 1.2.8 and 1.2.11
-> >Distribution: Slackware 2.3
-> >gcc: 2.6.3 (a.out)
-> >Hardware: Pentium P90 Premiere/PCI II motherboard (BIOS 1.0.10 up to
-> 1.0.13)
-> > NCR 53c815 SCSI (3.0.6), SMC EtherPower PCI (SMC 8432BT, Tulip)
-> >For 1.2.8 the board works well with the above hardware and software
-> >(old and new tulip driver).
-> >
-> >For 1.2.11 there is no bootup problem, no kernel panic, no anything, only
-> >gcc and various other programs are dying w/ signal 11 while starting.
-> >It's not the typical one of signal 11 while compiling the kernel
-> (motherboard,
-> >memory, cache) - it's while starting. I'm confused.
-> >The same software (1.2.8 AND 1.2.11) (and hardware) is doing well on a P60
-> >with a cheap 586SP4 motherboard.
->
->
-> First off, sorry this is being replied to so late. At home, I have a
-> Gateway P90 (Overclocked
-> to 100MHz), 40MB RAM, 1.08GB SCSI-2 Funitsu HD, Adaptech 2940 SCSI
-> controller. When
-> I was in the process of upgrading from Linux 1.2.11 to Linux 1.3.4 when I
-> got this error:
-> gcc: Internal compiler error: program cc1 got fatal signal 11.
-> This was running gcc i2.6.3 (Pentium compiler, -m 486 flag), Linux 1.3.4.
-> I started worrying that
-> my gcc was broken so I changed back to gcc 2.5.8 (from a Slackware
-> release). Again, I got
-> the same error. I rebooted the machine back to the 1.2.11 kernel and this
-> time the 2.5.8 gcc
-> AND the i2.6.3 gcc worked. To make sure it was not a fluke, when I
-> rebooted back in to
-> the 1.3.4 kernel, I still got the same error.
-> Each time, I did a "make clean ; make zImage" in the 1.3.4 source tree
-> and each time it
-> stopped on the sched.c source code. Unfortunatly, I don't have this exact
-> setup anymore
-> since when I tried to upgrade to a later Slackware release of GCC I
-> accidentally deleted my
-> i2.5.8 (Pentium) compiler and a whole chunk of my /usr directory.
->
-> Has anyone else seen any weird behavior with the compiler under the 1.3.4
-> kernel, or does
-> the 1.3.5 or greater kernel fix things?
->
-> Dan
->
-> ****************************************************************
-> * Dan Linder - Inacom NSSC
-> * SCO ACE, Compaq SCO UNIX ASE
-> ****************************************************************
Ok, Gerhard Schneider and Dan Linder have nearly the same Problem as i
have on one of my systems at work (my location at the moment!). I
installed an old DLD-Linux V1.3 on an AMD486-DX40, 16MB RAM,
AHA1542CF with a Conner 510MB, a Segate 210MB, a Toshiba and a NEC CD-ROM.
On this machine a compiled Kernel-Versions 1.1xx up to 1.3.3 _without_ any
problems. Then suddenly my Conner HD crashed (it isnt initializing
anymore! Adaptec says it is there, but thats all! No FDISK possible 8-( ),
so i changed to my second system (normaly my Novell-Netware-Administration
PC) which is a AMD486-DX/2-80 with two IDE-Drives (Conner 407MB and Segate
408MB) and 16MB RAM. This machine drives me nuts, `cause i'm not able to
compile a kernel (independent of it's version). On both systems i am using
GCC 2.5.8 (a.out) but only on the faster machine i get:
gcc: Internal compiler error: program cc1 got fatal signal 11 (or 6)
mostly in sched.c.
Everything works fine up to "make zImage"!
I checked my BIOS and set my system from a fast one to an incredible slow
creeper but that wasn't the reason (same errors!).
So i think this is a problem, independent of the processor-type and also
independent of the HD-type! But what is the problem? Are we doing
something wrong?
So what: ...happy LINUX'ing!
N N W W OO Dirk Traenapp, NWO, Zum Oelhafen 207, 26384 Wilhelmshaven
NN N W W O O Germany
N NN W W W O O E-Mail: d.traenapp@nwofs2.nwowhv.de
N N W W OO Phone: +49(0)4421 62-364
Intel outside! Fax: +49(0)4421 62-381 -382 -342