Re: insn_code -> tree_code in tree-vect-transform.c
On Sat, May 26, 2007 at 12:35:32AM +0200, Thomas Neumann wrote:
> Hi,
>
> as of revision 125076, tree-vect-transform.c contains the following code
> in line 2010:
>
> enum tree_code code, code1 = CODE_FOR_nothing, code2 = CODE_FOR_nothing;
SVN can help you find out who added that piece of code. Use the SVN command
"blame" for bugs/regressions, "praise" for improvements or bug fixes and
"annotate" when you're not sure if the change is good or not. SVN blame
gives:
124784 uros enum tree_code code, code1 = CODE_FOR_nothing, code2 =
CODE_FOR_nothing;
SVN log tells you:
r124784 | uros | 2007-05-17 08:31:05 +0200 (tor, 17 maj 2007) | 75 lines
PR tree-optimization/24659
[...]
And sure enough, the gcc-patches mailing list archive has this:
http://gcc.gnu.org/ml/gcc-patches/2007-05/msg00966.html>. So go ahead
and ask the patch author directly. If you don't have that message in your
mailbox, you can download it from the archive as raw text and demangle it
with this script:
#!/bin/sh
sed -e '1,/^$/ { s/ at /@/g' -e 's/ dot /./g' -e '}'
(Don't ask me why I didn't write it using #!/bin/sed instead.)
--
Rask Ingemann Lambertsen
compiling: gcc4-4.3.0-20070525 failed
On OSX 10.3.9 compiling gcc4-4.3.0-20070525 failed with: ... /bin/sh ./libtool --tag=GCJ --mode=compile /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -o java/lang/Class.lo -fsource-filename=../../../gcc-4.3-20070526/libjava/java/lang/Class.java ../../../gcc-4.3-20070526/libjava/classpath/lib/java/lang/Class.class libtool: compile: /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -fsource-filename=../../../gcc-4.3-20070526/libjava/java/lang/Class.java ../../../gcc-4.3-20070526/libjava/classpath/lib/java/lang/Class.class libtool: compile: mv -f "Class.o" "java/lang/.libs/Class.o" libtool: compile: /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -fsource-filename=../../../gcc-4.3-20070526/libjava/java/lang/Class.java ../../../gcc-4.3-20070526/libjava/classpath/lib/java/lang/Class.class >/dev/null 2>&1 libtool: compile: mv -f "Class.o" "java/lang/Class.o" echo ../../../gcc-4.3-20070526/libjava/classpath/lib/java/lang/PosixProcess*.class > java/process-Posix.list /bin/sh ./libtool --tag=GCJ --mode=compile /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -o java/process-Posix.lo -fsource-filename=/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/classpath/lib/classes -MT java/process-Posix.lo -MD -MP -MF java/process-Posix.deps @java/process-Posix.list libtool: compile: /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -fsource-filename=/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/classpath/lib/classes -MT java/process-Posix.lo -MD -MP -MF java/process-Posix.deps @java/process-Posix.list libtool: compile: mv -f "process-Posix.o" "java/.libs/process-Posix.o" mv: ne peut 'evaluer `process-Posix.o': No such file or directory make[3]: *** [java/process-Posix.lo] Error 1 make[2]: *** [all-recursive] Error 1 make[1]: *** [all-target-libjava] Error 2 make: *** [all] Error 2 TIA Dominique PS I applied the Andrew Pinski' patch #125087 in the hope to be on the safe side.
Re: compiling: gcc4-4.3.0-20070525 failed
Dominique Dhumieres wrote: On OSX 10.3.9 compiling gcc4-4.3.0-20070525 failed with: ... /bin/sh ./libtool --tag=GCJ --mode=compile /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -o java/lang/Class.lo -fsource-filename=../../../gcc-4.3-20070526/libjava/java/lang/Class.java ../../../gcc-4.3-20070526/libjava/classpath/lib/java/lang/Class.class libtool: compile: /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -fsource-filename=../../../gcc-4.3-20070526/libjava/java/lang/Class.java ../../../gcc-4.3-20070526/libjava/classpath/lib/java/lang/Class.class libtool: compile: mv -f "Class.o" "java/lang/.libs/Class.o" libtool: compile: /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -fsource-filename=../../../gcc-4.3-20070526/libjava/java/lang/Class.java ../../../gcc-4.3-20070526/libjava/classpath/lib/java/lang/Class.class >/dev/null 2>&1 libtool: compile: mv -f "Class.o" "java/lang/Class.o" echo ../../../gcc-4.3-20070526/libjava/classpath/lib/java/lang/PosixProcess*.class > java/process-Posix.list /bin/sh ./libtool --tag=GCJ --mode=compile /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -o java/process-Posix.lo -fsource-filename=/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/classpath/lib/classes -MT java/process-Posix.lo -MD -MP -MF java/process-Posix.deps @java/process-Posix.list libtool: compile: /sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/gcj -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/ -B/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/gcc/ -fclasspath= -fbootclasspath=../../../gcc-4.3-20070526/libjava/classpath/lib --encoding=UTF-8 -Wno-deprecated -fbootstrap-classes -g -O2 -c -fsource-filename=/sw/src/fink.build/gcc4-4.3.0-20070526/darwin_objdir/powerpc-apple-darwin7/libjava/classpath/lib/classes -MT java/process-Posix.lo -MD -MP -MF java/process-Posix.deps @java/process-Posix.list libtool: compile: mv -f "process-Posix.o" "java/.libs/process-Posix.o" mv: ne peut 'evaluer `process-Posix.o': No such file or directory make[3]: *** [java/process-Posix.lo] Error 1 make[2]: *** [all-recursive] Error 1 make[1]: *** [all-target-libjava] Error 2 make: *** [all] Error 2 TIA Dominique PS I applied the Andrew Pinski' patch #125087 in the hope to be on the safe side. I also get this exact error under freeBSD regarding compilation of the java frontend. ~/gcc/build/stage1-gcc]$ ./xgcc -v Using built-in specs. Target: i386-unknown-freebsd5.4 Configured with: ../configure --prefix=/home/tebneff/beta --enable-languages=c,c++,fortran,java,objc,obj-c++ Thread model: posix gcc version 4.3.0 20070525 (experimental) Regards, JT
PR25477 and PR24685
Mike and Eric, The gfortran developers are wondering if Darwin PPC will ever have PR24685 and PR25477 fixed. Looking at this again I noticed a proposed patch which I seemed to have missed when I last looked into this issue... http://gcc.gnu.org/ml/gcc-patches/2005-12/msg01545.html http://gcc.gnu.org/ml/gcc-patches/2005-12/txt00089.txt This patch looks pretty complete and only seemed to have minor issues with it according to you and Geoff... http://gcc.gnu.org/ml/gcc-patches/2005-12/msg01549.html http://gcc.gnu.org/ml/gcc-patches/2005-12/msg01550.html Does that original patch look like a reasonable starting place for the gfortran developers to attempt to fix this? Or might you take a stab at touching up this patch so we could finally get this issue resolved? Jack
LLVM 2.0 Released
Hi Everyone, As an FYI, LLVM 2.0 was released on Wednesday. This is a critical release for us, with many new features, optimizer improvements, and compile-time speedups. Among other things, this release add support for many missing GCC extensions, so we can now build things like mozilla, qt, koffice and broad range of other apps straight out of the box. If you're interested, information about the delta from LLVM 1.9 is at: http://lists.cs.uiuc.edu/pipermail/llvm-announce/2007-May/23.html http://lists.cs.uiuc.edu/pipermail/llvm-announce/2007-February/ 21.html and/or: http://llvm.org/releases/2.0/docs/ReleaseNotes.html If you have thoughts or comments, please respond to the llvmdev list - http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev - to avoid polluting this list. -Chris
Re: Dynamically linking against GMP and MPFR
On Fri, 25 May 2007, Tim Prince wrote: > I spent quite a while getting out of the tangle I got into when I built > mpfr and gmp with --disable-shared, leaving older incompatible shared > libraries in the path. No doubt, it can be made to work, but with > plenty of ways to go wrong. The above isn't enough description to take any action. It would help if you filed a PR with the specific things your did and the symptoms you were experiencing so we can determine if there's a bug or where in the process things could be clarified through better documentation. Would you please do that? Thanks, --Kaveh -- Kaveh R. Ghazi [EMAIL PROTECTED]
libgfortran generated and long double calls
While looking at Andrew's patch for fixing the long double issue on Darwin PPC, I noticed something interesting in the symbols in the object files created for gcc/libgfortran/generated. Some of these object files have symbols with the proper $LDBL128 appended on Darwin PPC. For example... nm rrspacing_r16.o | grep expl U _frexpl$LDBL128 U _ldexpl$LDBL128 This contrasts with other object files created from gcc/libgfortran/generated... nm _exp_r16.o | grep expl U _expl The difference seems to be that rrspacing_r16.c explicitly calls frexpl() whereas _exp_r16.c calls exp() rather than expl(). This leads me to wonder if this problem with large_real_kind_2 failing on Darwin PPC isn't due to various *_r10.c and *_r16.c files making calls to foo() rather than fool(). Is it possible that this issue (and the necessary fix) could really be that simple? Jack
RE: libgfortran generated and long double calls
A quick stab at changing... specific__exp_r16 = exp (parm) to specific__exp_r16 = expl (parm) ...in gcc/libgfortran/generated/_exp_r16.F90 produced the followimg compilation error on Darwin PPC... ../../../gcc-4.2-20070526/libgfortran/generated/_exp_r16.F90:47.23: specific__exp_r16 = expl (parm) 1 Error: Function reference to 'expl' at (1) is to a non-PURE procedure within a PURE procedure make[3]: *** [_exp_r16.lo] Error 1 make[2]: *** [all] Error 2 make[1]: *** [all-target-libgfortran] Error 2 make: *** [all] Error 2 Which led me to take a look at gcc/libgfortran/c99_protos.h. There I noticed we have no mention of expl() despite it being listed as a valid C99 function... http://en.wikibooks.org/wiki/C_Programming/Further_math Again, is it possible that Darwin PPC simply is exposing the fact that we are missing valid C99 functions that should be used in the generated *.F90 routines for r10 and r16? Jack
Re: libgfortran generated and long double calls
On Sun, May 27, 2007 at 01:50:53AM -0400, Jack Howarth wrote: >A quick stab at changing... > >specific__exp_r16 = exp (parm) > > to > >specific__exp_r16 = expl (parm) > > ...in gcc/libgfortran/generated/_exp_r16.F90 > produced the followimg compilation error on > Darwin PPC... > > ../../../gcc-4.2-20070526/libgfortran/generated/_exp_r16.F90:47.23: > >specific__exp_r16 = expl (parm) > 1 > Error: Function reference to 'expl' at (1) is to a non-PURE procedure within > a PURE procedure Well, that was a stupid thing to do. _exp_r16.F90 is compiled by a Fortran compiler and expl() is not an intrinsic function in Fortran. -- Steve
