Problems with v2.1.30-v2.1.34

Problems with v2.1.30-v2.1.34

Post by Mike Dowlin » Fri, 18 Apr 1997 04:00:00



I realise that these are development versions, but I nevertheless wonder if
anybody else has tried these kernels out.

At home, version 2.1.32 would not compile.   The system complained that the
kernel was too big.  I had no problems compiling and running it at work, where
I have fewer system requirements like no PPP, DIP, soundcard, etc.

Version 2.1.33 compiled at work, but the the control-arrow key combinations and
home, end etc. were not being interpreted correctly.

version 2.1.34 both at home and at work would not compile.  I get

e.o ioport.o ldt.o setup.o time.o sys_i386.o smp.o *oline.o
smp.o: In function `x86_do_profile':
smp.o(.text+0x108c): multiple definition of `x86_do_profile'
time.o(.text+0x2ec): first defined here
make[1]: *** [kernel.o] Error 1
make[1]: Leaving directory `/usr/src/linux/arch/i386/kernel'
make: *** [linuxsubdirs] Error 2

Normally, I don't have problems with new kernels, and, when problems do occur,
they are quickly rectified.  Surely there cannot be a spate of bad bugs, or has
something changed drastically for recent kernels?  (I am mostly concerned
about the keyboard support; changes there would cost a lot of work!)

Any ideas anybody?

Cheers,
        Mike

 
 
 

Problems with v2.1.30-v2.1.34

Post by Dragon Fl » Fri, 18 Apr 1997 04:00:00




Quote:

> I realise that these are development versions, but I nevertheless wonder
if
> anybody else has tried these kernels out.

> At home, version 2.1.32 would not compile.   The system complained that
the
> kernel was too big.  I had no problems compiling and running it at work,
where
> I have fewer system requirements like no PPP, DIP, soundcard, etc.

2.1.32 did not compile for either.
But both 2.1.30, 2.1.33, 2.1.34  compiled without a glitch.

Sergei

 
 
 

Problems with v2.1.30-v2.1.34

Post by Andries Brouw » Fri, 18 Apr 1997 04:00:00


: Version 2.1.33 compiled at work, but the the control-arrow key
: combinations and home, end etc. were not being interpreted correctly.

: version 2.1.34 both at home and at work would not compile.

Yes, someone polished vt.c and lost a variable initialisation.
I already posted a fix here (and submitted it to Linus -
maybe it'll be fixed in 2.1.35 again).

It is true that for recent kernels one needs a few attempts
before hitting a configuration that compiles, but such flaws
are quickly rectified.

So far, 2.0.29 and 2.1.29 were good kernels (for me).