* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
@ 2012-05-07 17:43 ` pinskia at gcc dot gnu.org
2012-05-07 17:45 ` skunk at iskunk dot org
` (9 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: pinskia at gcc dot gnu.org @ 2012-05-07 17:43 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> 2012-05-07 17:27:53 UTC ---
mulhwu is a powerpc but not rs6000 instruction.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
2012-05-07 17:43 ` [Bug bootstrap/53266] " pinskia at gcc dot gnu.org
@ 2012-05-07 17:45 ` skunk at iskunk dot org
2012-09-25 7:31 ` segher at gcc dot gnu.org
` (8 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: skunk at iskunk dot org @ 2012-05-07 17:45 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #2 from Daniel Richard G. <skunk at iskunk dot org> 2012-05-07 17:42:43 UTC ---
(In reply to comment #1)
> mulhwu is a powerpc but not rs6000 instruction.
When a file failed to compile, I noticed that specifying -mcpu=powerpc got
things going again. I'm not clear on how GCC distinguishes powerpc versus
rs6000, but as far as -mcpu is concerned, the former value seems to exclude
mulhwu.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
2012-05-07 17:43 ` [Bug bootstrap/53266] " pinskia at gcc dot gnu.org
2012-05-07 17:45 ` skunk at iskunk dot org
@ 2012-09-25 7:31 ` segher at gcc dot gnu.org
2012-09-25 10:57 ` skunk at iskunk dot org
` (7 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: segher at gcc dot gnu.org @ 2012-09-25 7:31 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
Segher Boessenkool <segher at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |segher at gcc dot gnu.org
--- Comment #3 from Segher Boessenkool <segher at gcc dot gnu.org> 2012-09-25 07:31:33 UTC ---
Hi Daniel,
Does this still fail?
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
` (2 preceding siblings ...)
2012-09-25 7:31 ` segher at gcc dot gnu.org
@ 2012-09-25 10:57 ` skunk at iskunk dot org
2012-09-25 11:45 ` segher at gcc dot gnu.org
` (6 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: skunk at iskunk dot org @ 2012-09-25 10:57 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #4 from Daniel Richard G. <skunk at iskunk dot org> 2012-09-25 10:56:41 UTC ---
(In reply to comment #3)
> Does this still fail?
I haven't built GCC on the AIX 4.3 (PowerPC 604) system lately, but my scripts
are set up to do so using
--with-cpu=powerpc --disable-aix64 --disable-powercpu
in order to avoid this issue. Should I be able to drop any of these, at this
point?
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
` (3 preceding siblings ...)
2012-09-25 10:57 ` skunk at iskunk dot org
@ 2012-09-25 11:45 ` segher at gcc dot gnu.org
2012-09-25 21:38 ` skunk at iskunk dot org
` (5 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: segher at gcc dot gnu.org @ 2012-09-25 11:45 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #5 from Segher Boessenkool <segher at gcc dot gnu.org> 2012-09-25 11:44:42 UTC ---
> I haven't built GCC on the AIX 4.3 (PowerPC 604) system lately, but my scripts
> are set up to do so using
>
> --with-cpu=powerpc --disable-aix64 --disable-powercpu
>
> in order to avoid this issue. Should I be able to drop any of these, at this
> point?
You can remove the --with-cpu; the default on AIX4.3 is 604e,
which should work fine. I don't think you can yet remove the
--disable-powercpu since I didn't touch the multilib code (I
completely missed it, and wow it's a nice little maze all of
itself). I have no idea about --disable-aix64. Try and see :-)
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
` (4 preceding siblings ...)
2012-09-25 11:45 ` segher at gcc dot gnu.org
@ 2012-09-25 21:38 ` skunk at iskunk dot org
2012-09-26 5:49 ` segher at gcc dot gnu.org
` (4 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: skunk at iskunk dot org @ 2012-09-25 21:38 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #6 from Daniel Richard G. <skunk at iskunk dot org> 2012-09-25 21:38:32 UTC ---
Okay, I'm bootstrapping an unmodified 4.7.2, using a previously-built 4.7.0 as
CC and the following configuration options:
--prefix=/opt/tg
--disable-dependency-tracking
--disable-maintainer-mode
--program-prefix=tg-
--disable-shared
--disable-nls
--enable-languages=c,c++
--enable-version-specific-runtime-libs
--with-pic
--with-mpc=/tg/freeport/arch/aix32
--with-mpfr=/tg/freeport/arch/aix32
--with-gmp=/tg/freeport/arch/aix32
--disable-powercpu
Last time I tried this with 4.7.0/4.7.1, there were numerous problems (none
having to do with bad PPC instructions, since I just used --with-cpu= et al. to
work around that). I actually have patches for most of the issues, and have
previously posted them at
http://gcc.gnu.org/PR53348#c7
but David Edelsohn took no notice of them. If you're interested, I'll be happy
to update the patch for current SVN.
At any rate, I'll let you know how the build goes. This machine isn't exactly a
speed demon, so give it a day or three....
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
` (5 preceding siblings ...)
2012-09-25 21:38 ` skunk at iskunk dot org
@ 2012-09-26 5:49 ` segher at gcc dot gnu.org
2012-09-26 6:06 ` skunk at iskunk dot org
` (3 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: segher at gcc dot gnu.org @ 2012-09-26 5:49 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #7 from Segher Boessenkool <segher at gcc dot gnu.org> 2012-09-26 05:49:09 UTC ---
Sorry, I didn't notice this was for 4.7 -- the removal of all
suport for the POWER ISA is only on mainline, and won't be
backported. Could you test mainline instead? All fixes happen
there (or happen there first).
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
` (6 preceding siblings ...)
2012-09-26 5:49 ` segher at gcc dot gnu.org
@ 2012-09-26 6:06 ` skunk at iskunk dot org
2012-09-26 6:44 ` segher at gcc dot gnu.org
` (2 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: skunk at iskunk dot org @ 2012-09-26 6:06 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #8 from Daniel Richard G. <skunk at iskunk dot org> 2012-09-26 06:06:11 UTC ---
Sure, can do. Is mainline a synonym for SVN trunk? Is there somewhere I can
download a ready-made tarball, just to make sure everything is autogen'ed
correctly?
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
` (7 preceding siblings ...)
2012-09-26 6:06 ` skunk at iskunk dot org
@ 2012-09-26 6:44 ` segher at gcc dot gnu.org
2012-09-27 14:02 ` skunk at iskunk dot org
2021-07-19 7:13 ` pinskia at gcc dot gnu.org
10 siblings, 0 replies; 12+ messages in thread
From: segher at gcc dot gnu.org @ 2012-09-26 6:44 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #9 from Segher Boessenkool <segher at gcc dot gnu.org> 2012-09-26 06:43:41 UTC ---
Yes, mainline is SVN trunk.
There are weekly snapshots on http://gcc.gnu.org/snapshots.html (just
get the newest one, it should be more than new enough for this).
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
` (8 preceding siblings ...)
2012-09-26 6:44 ` segher at gcc dot gnu.org
@ 2012-09-27 14:02 ` skunk at iskunk dot org
2021-07-19 7:13 ` pinskia at gcc dot gnu.org
10 siblings, 0 replies; 12+ messages in thread
From: skunk at iskunk dot org @ 2012-09-27 14:02 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
--- Comment #10 from Daniel Richard G. <skunk at iskunk dot org> 2012-09-27 14:02:03 UTC ---
Okay, I grabbed a copy of gcc-4.8-20120923, and started bootstrapping using an
older 4.5.2 (since the 4.7.0 C++ compiler isn't working correctly; am I correct
in observing that GCC can no longer be bootstrapped using just a C compiler?).
Things went well up until this point:
tg-g++ -c -g -DIN_GCC -fno-exceptions -fno-rtti -W -Wall -Wwrite-strings
-Wcast-qual -Wmissing-format-attribute -pedantic -Wno-long-long
-Wno-variadic-macros -Wno-overlength-strings -fno-common -DHAVE_CONFIG_H -I.
-I. -I/home/src/gcc-4.8-20120923/gcc -I/home/src/gcc-4.8-20120923/gcc/.
-I/home/src/gcc-4.8-20120923/gcc/../include
-I/home/src/gcc-4.8-20120923/gcc/../libcpp/include
-I/tg/freeport/arch/aix32/include -I/tg/freeport/arch/aix32/include
-I/tg/freeport/arch/aix32/include
-I/home/src/gcc-4.8-20120923/gcc/../libdecnumber
-I/home/src/gcc-4.8-20120923/gcc/../libdecnumber/dpd -I../libdecnumber \
/home/src/gcc-4.8-20120923/gcc/config/rs6000/rs6000.c -o rs6000.o
/home/src/gcc-4.8-20120923/gcc/config/rs6000/rs6000.c: In function 'void
rs6000_code_end()':
/home/src/gcc-4.8-20120923/gcc/config/rs6000/rs6000.c:28292:51: error:
'ASM_WEAKEN_DECL' was not declared in this scope
gmake[3]: *** [rs6000.o] Error 1
gmake[3]: Leaving directory `/tmp/gcc-4.8-test/gcc'
gmake[2]: *** [all-stage1-gcc] Error 2
gmake[2]: Leaving directory `/tmp/gcc-4.8-test'
gmake[1]: *** [stage1-bubble] Error 2
gmake[1]: Leaving directory `/tmp/gcc-4.8-test'
gmake: *** [bootstrap-lean] Error 2
This is one of the issues addressed by my aforementioned patch.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug bootstrap/53266] Error: Unrecognized opcode: `mulhwu'
2012-05-07 14:31 [Bug bootstrap/53266] New: Error: Unrecognized opcode: `mulhwu' skunk at iskunk dot org
` (9 preceding siblings ...)
2012-09-27 14:02 ` skunk at iskunk dot org
@ 2021-07-19 7:13 ` pinskia at gcc dot gnu.org
10 siblings, 0 replies; 12+ messages in thread
From: pinskia at gcc dot gnu.org @ 2021-07-19 7:13 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53266
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |WONTFIX
Status|UNCONFIRMED |RESOLVED
Target Milestone|--- |7.0
--- Comment #11 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
AIX 4.3 is no longer supported since GCC 7.
^ permalink raw reply [flat|nested] 12+ messages in thread