mainline bootstrap failure on sparcv9-sun-solaris2.8

mainline bootstrap failure on sparcv9-sun-solaris2.8

[Brad Lucier]

This has been reported several times before, the following never finishes.

./xgcc -B./ -B/home/c/lucier/local/gcc-test/sparcv9-sun-solaris2.8/bin/ -isystem /home/c/lucier/local/gcc-test/sparcv9-sun-solaris2.8/include -isystem /home/c/lucier/local/gcc-test/sparcv9-sun-solaris2.8/sys-include -O2 -DIN_GCC    -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -isystem ./include  -I. -I. -I../../gcc -I../../gcc/. -I../../gcc/config -I../../gcc/../include  -g0 -finhibit-size-directive -fno-inline-functions -fno-exceptions -fno-zero-initialized-in-bss -fPIC \
   -c ../../gcc/crtstuff.c -DCRT_BEGIN \
  -o crtbegin.o

I tried to see if it was a simple infinite loop using gdb, but no such
luck, so it would be good if someone could investigate this.  It
began failing between

http://gcc.gnu.org/ml/gcc-testresults/2002-07/msg00577.html

and 24 hours later.

Here is the gdb session, but it doesn't seem much help:

gdb ./cc1
 -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -version -finhibit-size-directive -fno-inline-functions -fno-exceptions -fno-zero-initialized-in-bss -fPIC -o crtstuff.s
Starting program: /export/u3/lucier/programs/gcc/gcc-3.2/objdir-sparcv9/gcc/cc1 -fpreprocessed crtstuff.i -quiet -dumpbase crtstuff.c -g0 -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -version -finhibit-size-directive -fno-inline-functions -fno-exceptions -fno-zero-initialized-in-bss -fPIC -o crtstuff.s
GNU C version 3.2 20020722 (experimental) (sparcv9-sun-solaris2.8)
        compiled by GNU C version 3.1 20020509 (prerelease).
^C
Program received signal SIGINT, Interrupt.
0x1002d27a0 in reg_mentioned_p (reg=0x100671a60, in=0x1006b5d90)
    at ../../gcc/rtlanal.c:654
654       switch (code)
(gdb) c
Continuing.
^C
Program received signal SIGINT, Interrupt.
do_local_cprop (x=0x100671a60, insn=0x10079b200, alter_jumps=1)
    at ../../gcc/gcse.c:4303
4303      if (GET_CODE (x) == REG
(gdb) c
Continuing.
^C
Program received signal SIGINT, Interrupt.
0x1002d5700 in find_reg_note (insn=0x10079b200, kind=REG_EQUAL, datum=0x0)
    at ../../gcc/rtlanal.c:1878
1878      return 0;
(gdb) c
Continuing.
^C
Program received signal SIGINT, Interrupt.
0x1002d4ea4 in note_uses (pbody=0x10079b230, 
    fun=0x100291a20 <validate_replace_src_1>, data=0xffffffff7fffe250)
    at ../../gcc/rtlanal.c:1710
1710            if (GET_CODE (dest) == ZERO_EXTRACT)
(gdb) c
Continuing.
^C
Program received signal SIGINT, Interrupt.
0x1002d2864 in reg_mentioned_p (reg=0x100671a60, in=0x100671b00)
    at ../../gcc/rtlanal.c:686
686           if (fmt[i] == 'E')
(gdb) info locals
fmt = 0x10044a870 "ee"
i = 0
code = SET
(gdb) p insn 
No symbol "insn" in current context.
(gdb) c
Continuing.
^C
Program received signal SIGINT, Interrupt.
0x1001cdb48 in local_cprop_pass (alter_jumps=1) at ../../gcc/gcse.c:4380
4380                  for (reg_used = &reg_use_table[0]; reg_use_count > 0;
(gdb) c
Continuing.
^C
Program received signal SIGINT, Interrupt.
reg_mentioned_p (reg=0x100671a60, in=0x100671ac0) at ../../gcc/rtlanal.c:682
682       fmt = GET_RTX_FORMAT (code);

Re: mainline bootstrap failure on sparcv9-sun-solaris2.8

[David S. Miller]

   From: Brad Lucier <lucier@math.purdue.edu>
   Date: Mon, 22 Jul 2002 23:09:26 -0500 (EST)

   This has been reported several times before, the following never finishes.
   
See other postings on this list, including:

	http://gcc.gnu.org/ml/gcc/2002-07/msg01030.html

Which posts a patch which should fix your problem.  Richard
gave the OK for Jan to install those fixes so it should be
in CVS soon if not already there.

Re: mainline bootstrap failure on sparcv9-sun-solaris2.8

[Toon Moene]

"David S. Miller" wrote:
> 
>    From: Brad Lucier <lucier@math.purdue.edu>
>    Date: Mon, 22 Jul 2002 23:09:26 -0500 (EST)
> 
>    This has been reported several times before, the following never finishes.
> 
> See other postings on this list, including:
> 
>         http://gcc.gnu.org/ml/gcc/2002-07/msg01030.html
> 
> Which posts a patch which should fix your problem.  Richard
> gave the OK for Jan to install those fixes so it should be
> in CVS soon if not already there.

Hmmm, Jan said Saturday he would be on holiday soon, and not return before
the 9th (of August, I assume).

See: http://gcc.gnu.org/ml/gcc-patches/2002-07/msg01106.html

-- 
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
Join GNU Fortran 95: http://g95.sourceforge.net/ (under construction)

Re: mainline bootstrap failure on sparcv9-sun-solaris2.8

[Andreas Jaeger]

Toon Moene <toon@moene.indiv.nluug.nl> writes:

> "David S. Miller" wrote:
>> 
>>    From: Brad Lucier <lucier@math.purdue.edu>
>>    Date: Mon, 22 Jul 2002 23:09:26 -0500 (EST)
>> 
>>    This has been reported several times before, the following never finishes.
>> 
>> See other postings on this list, including:
>> 
>>         http://gcc.gnu.org/ml/gcc/2002-07/msg01030.html
>> 
>> Which posts a patch which should fix your problem.  Richard
>> gave the OK for Jan to install those fixes so it should be
>> in CVS soon if not already there.
>
> Hmmm, Jan said Saturday he would be on holiday soon, and not return before
> the 9th (of August, I assume).
>
> See: http://gcc.gnu.org/ml/gcc-patches/2002-07/msg01106.html

He'll leave in a week or so...

Andreas
-- 
 Andreas Jaeger
  SuSE Labs aj@suse.de
   private aj@arthur.inka.de
    http://www.suse.de/~aj