This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
| Other format: | [Raw text] | |
Please try this patch.In recent CVS checkouts of GCC mainline, I have noticed a weird bootstrapping issue. The steps are:
1. make bootstrap 2. touch $GCC_SRC_DIR/gcc/java/check-init.c 3. make bubblestrap 4. touch $GCC_SRC_DIR/gcc/java/check-init.c 5. make bubblestrap
In step #3, I see "genattrab" and other "gen*" programs running once again, while in #5 (and subsequent such iterations) they don't. I have seen this on two separate x86-Linux boxen.
It seems that you've discovered a regression from 3.4. Unfortunately, it likely is related to my moving the GCC build executables into a build/ directory.
Please create a new PR and feel free to assign it to me <kcook@gcc.gnu.org>.
diff -prcd gcc-orig/gcc/Makefile.in gcc-snapshot/gcc/Makefile.in *** gcc-orig/gcc/Makefile.in Wed Jan 5 12:31:37 2005 --- gcc-snapshot/gcc/Makefile.in Wed Jan 12 16:30:33 2005 *************** STAGECOPYSTUFF = insn-flags.h insn-confi *** 950,959 **** genrtl.c genrtl.h gt-*.h gtype-*.h gtype-desc.c gtyp-gen.h
# Files to be moved away after each stage in building. ! STAGEMOVESTUFF = *$(objext) s-* \ xgcc$(exeext) cpp$(exeext) cc1$(exeext) $(EXTRA_PASSES) \ $(EXTRA_PARTS) $(EXTRA_PROGRAMS) gcc-cross$(exeext) \ protoize$(exeext) unprotoize$(exeext) \ Only in gcc-snapshot/gcc/config/i386: i386.md.patch
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |