[4.3][doc patch] document new tree-codes for multipletypes vectorization project, and fortran build problem

Dorit Nuzman DORIT@il.ibm.com
Sat Nov 4 13:16:00 GMT 2006


While preparing the multiple-datatypes project to be committed I realized I
never got this bit reviewed - Diego pointed out that documentation of new
tree-codes was missing from doc/c-tree.texi. Attached is the missing doc
patch. Is it OK?

Other than that the multiple-datatypes-vectorization project is ready to be
committed, except I haven't been able to build gfortran in the last few
days - getting this error:

/Develop/mainline_svn/build1/./gcc/xgcc
-B/Develop/mainline_svn/build1/./gcc/
-B/Develop/mainline_svn1/i386-redhat-linux/bin/
-B/Develop/mainline_svn1/i386-redhat-linux/lib/ -isystem
/Develop/mainline_svn1/i386-redhat-linux/include -isystem
/Develop/mainline_svn1/i386-redhat-linux/sys-include -DHAVE_CONFIG_H -I.
-I../../../gcc/libgfortran -I. -iquote../../../gcc/libgfortran/io
-I../../../gcc/libgfortran/../gcc -I../../../gcc/libgfortran/../gcc/config
-I../.././gcc -D_GNU_SOURCE -std=gnu99 -Wall -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -Wextra -Wwrite-strings -O2 -g
-O2 -c ../../../gcc/libgfortran/io/unix.c  -fPIC -DPIC -o .libs/unix.o
/tmp/cccbx6w4.s: Assembler messages:
/tmp/cccbx6w4.s:4398: Error: symbol `fstat64' is already defined
/tmp/cccbx6w4.s:4429: Error: symbol `lstat64' is already defined
/tmp/cccbx6w4.s:4571: Error: symbol `stat64' is already defined
make[3]: *** [unix.lo] Error 1
make[3]: Leaving directory
`/Develop/mainline_svn/build1/i386-redhat-linux/libgfortran'
make[2]: *** [all] Error 2
make[2]: Leaving directory
`/Develop/mainline_svn/build1/i386-redhat-linux/libgfortran'
make[1]: *** [all-target-libgfortran] Error 2
make[1]: Leaving directory `/Develop/mainline_svn/build1'
make: *** [all] Error 2

I assumed that this is related to the c99 issue (and
http://gcc.gnu.org/ml/gcc/2006-11/msg00054.html ?).
So my most recent bootstrapping/testing on ppc and x86 is without fortran
(I have tested it with fortran in the past).
Is it ok to commit the project this way,
or is there a way to build fortran with current mainline?

:ADDPATCH docs (vectorizer):

thanks,
dorit

(See attached file: texipatch.txt)
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: texipatch.txt
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20061104/debc84f6/attachment.txt>


More information about the Gcc-patches mailing list