[parisc-linux] Binutils update

[parisc-linux] Binutils update

[Alan Modra]

I've just checked in the following:

Rewrite much of the stub handling code (still using r8)
Fix expression parsing for `!' line separator. 
Fix sizing of branches & other miscellaneous small fixes. 

Feedback appreciated.
Paul, did you get tar to run?

Alan Modra
-- 
Linuxcare.  Support for the Revolution.

Re: [parisc-linux] Binutils update

[Paul Bame]

= Paul, did you get tar to run?

Yes!  After it linked it worked fine.

	-P

Re: [parisc-linux] Binutils update

[Paul Bame]

Oops, it would probably be helpful to see a little more context:

gcc -W -Wall -g -O2 -o as-new app.o as.o atof-generic.o bignum-copy.o cond.o depend.o dwarf2dbg.o ecoff.o ehopt.o expr.o flonum-copy.o flonum-konst.o flonum-mult.o frags.o hash.o input-file.o input-scrub.o listing.o literal.o macro.o messages.o output-file.o read.o sb.o stabs.o subsegs.o symbols.o write.o tc-hppa.o obj-elf.o atof-ieee.o ../opcodes/.libs/libopcodes.a ../bfd/.libs/libbfd.a ../libiberty/libiberty.a
read.o: In function `read_a_source_file':
/users/bame/palinux/binutils-2.10/gas/read.c:1107: undefined reference to `found_comment'
collect2: ld returned 1 exit status

Re: [parisc-linux] Binutils update

[Grant Grundler]

Alan Modra wrote:
> I've just checked in the following:
> 
> Rewrite much of the stub handling code (still using r8)
> Fix expression parsing for `!' line separator. 
> Fix sizing of branches & other miscellaneous small fixes. 
> 
> Feedback appreciated.

Ok.  I get an error when building:
...
gcc -W -Wall -g -O2 -o as-new app.o as.o atof-generic.o bignum-copy.o cond.o depend.o dwarf2dbg.o ecoff.o ehopt.o expr.o flonum-copy.o flonum-konst.o flonum-mult.o frags.o hash.o input-file.o input-scrub.o listing.o literal.o macro.o messages.o output-file.o read.o sb.o stabs.o subsegs.o symbols.o write.o tc-hppa.o obj-elf.o atof-ieee.o ../opcodes/.libs/libopcodes.a ../bfd/.libs/libbfd.a ../libiberty/libiberty.a
read.o: In function `read_a_source_file':
/linux/grundler/build-2.10-32/gas/../../binutils-2.10/gas/read.c:1107: undefined reference to `found_comment'
collect2: ld returned 1 exit status
make[3]: *** [as-new] Error 1
make[3]: Leaving directory `/linux/grundler/build-2.10-32/gas'
make[2]: *** [all-recursive] Error 1
...

Any idea what's missing?

config.status contains:
../binutils-2.10/configure --host=i686-pc-linux-gnu --target=hppa1.1-linux --prefix=/opt/puffin --disable-nls --norecursion 

I assume that's the original config line plus some defaults.

thanks,
grant

Grant Grundler
Unix Development Lab
+1.408.447.7253

Re: [parisc-linux] Binutils update

[Alan Modra]

On Thu, 1 Jun 2000, Paul Bame wrote:

> gcc -W -Wall -g -O2 -o as-new app.o as.o atof-generic.o bignum-copy.o cond.o depend.o dwarf2dbg.o ecoff.o ehopt.o expr.o flonum-copy.o flonum-konst.o flonum-mult.o frags.o hash.o input-file.o input-scrub.o listing.o literal.o macro.o messages.o output-file.o read.o sb.o stabs.o subsegs.o symbols.o write.o tc-hppa.o obj-elf.o atof-ieee.o ../opcodes/.libs/libopcodes.a ../bfd/.libs/libbfd.a ../libiberty/libiberty.a
> read.o: In function `read_a_source_file':
> /users/bame/palinux/binutils-2.10/gas/read.c:1107: undefined reference to `found_comment'
> collect2: ld returned 1 exit status

Oops.  That leaked in from a patch I made to help find semicolons lurking
in C asm statements.  I've reverted the patch to gas/read.c

-- 
Linuxcare.  Support for the Revolution.

[parisc-linux] Binutils update.

[Ryan Bradetich]

Hello all,

I have finished working through all the compiler warnings and errors in
the binutils-2.10 build for the HP ansi compiler in ANSI and non-ANSI
mode.  Unfortunately I am with out a i386 system atm :( (Will be
remedied
soon!) so I can not test the cross compiler.  If someone with access to
an
i386 would test this for me I would appreciate it. :)

Just wanted to give an update to the list, so people could try out my
changes
and make sure I didn't break anything! *fingers crossed*

Well off to step 3 in the recipe now....

Later,
- Ryan

[parisc-linux] Recipe

[Brian S. Julin]

A couple questions about the recipe and the status of the 
cvs sources/snapshots -- for i386, my guess is you actually
don't want to specify --host=hppa-hpux, right?  It looks like
someone just cut and pasted the directions for compiling on hpux
without modifiying them at all.  

Secondly, either files are missing from the binutils cvs tree that
prevent libtool/automake from constructing bfd.h and a libbfd.la,
or I'm missing tools or one of my tools is a lower version than
needed.

I've been trying to get myself off the ground for a while now,
off and on (Ran out of disk on my HPUX box, which was 9.x anyway
so it was a pain, then found out cross-compile from ALPHA just wasn't
going to happen no matter how much I tried to fix binutils, and now I
finally have an i386 in here and I can't cross-compile there either.)  
I want to work on some of the input and video code, but I pretty much
can't make any contributions unless I can get the kernel to compile, 
a precompiled kernel doesn't have any development value.  (Although 
it would be nice to try a newer precompiled kernel just for the fun of
actually getting a working sash, so whoever was going to post one, please 
do.)

I think if you guys want more people participating in the project,
it's pretty important to make sure that the web pages are kept up
to date and the recipe works.  Including a list of last known
compiling/working snapshots and minimum versions of all tools required
would be a big plus.

--
Brian S. Julin

[parisc-linux] binutils update

[Alan Modra]

I've just committed a change to binutils that cures an ELF spec violation.
We now no longer create plabels for DT_INIT and DT_FINI and instead rely
on the dynamic linker to create them for us.  This means you need an
updated glibc, that I committed a little earlier today, to create the
plabel function pointers if you update your binutils.

You have been warned...

Alan Modra
-- 
Linuxcare.  Support for the Revolution.