Summary: | dev-scheme/plt-scheme-4.1 - make[4]: *** [xsrc/salloc.c] Segmentation fault | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Tiago <tjgq> |
Component: | [OLD] Development | Assignee: | Scheme Project <scheme> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | spatz |
Priority: | High | ||
Version: | 2007.0 | ||
Hardware: | x86 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Tiago
2008-09-12 17:52:34 UTC
I can confirm that `-O3` option is what causes an error; the emerge finishes fine with `-O2`. Probably a GCC issue. (In reply to comment #1) > I can confirm that `-O3` option is what causes an error; the emerge finishes > fine with `-O2`. Probably a GCC issue. Yup, with -O2 it works. Should this bug be closed or marked as a GCC one? (There seem to be lots of -Ox related GCC bugs reported on bugzilla.) make[4]: Entering directory `/var/tmp/portage/dev-scheme/drscheme-4.1/work/plt-4.1/src/mzscheme/gc2' env XFORM_PRECOMP=yes ../mzschemecgc -cqu ./xform.ss --setup . --cpp "i686-pc-linux-gnu-gcc -E -DNEWGC_BTC_ACCOUNT -DMZ_USES_SHARED_LIB -I./.. -I./../include" -o xsrc/precomp.h ./precomp.c make[4]: *** [xsrc/precomp.h] Segmentation fault make[4]: Leaving directory `/var/tmp/portage/dev-scheme/drscheme-4.1/work/plt-4.1/src/mzscheme/gc2' make[3]: *** [all] Error 2 make[3]: Leaving directory `/var/tmp/portage/dev-scheme/drscheme-4.1/work/plt-4.1/src/mzscheme/gc2' make[2]: *** [3m] Error 2 make[2]: Leaving directory `/var/tmp/portage/dev-scheme/drscheme-4.1/work/plt-4.1/src/mzscheme' make[1]: *** [3m] Error 2 make[1]: Leaving directory `/var/tmp/portage/dev-scheme/drscheme-4.1/work/plt-4.1/src' make: *** [all] Error 2 Pretty much the same here. Not necesarily a gcc bug though. The crash happens in scheme_apply which does a lot of error-prone stuff. This could easily be a buffer-overflow that only shows with optimization enabled. *** Bug 256044 has been marked as a duplicate of this bug. *** There's no package named dev-scheme/drscheme, which is why I couldn't find this bug (was it renamed?). Please change the summary :) BTW, the segfault still occurs with the latest version - 4.1.4. yes, it was renamed. The segfault is different, but I'm guessing it's again because of -O3. I'll talk to upstream about it and am open to suggestions. (In reply to comment #1) > I can confirm that `-O3` option is what causes an error; the emerge finishes > fine with `-O2`. Probably a GCC issue. > I confirm this bug and behavior for -march=amdfam10 Still happens in 4.1.5 (fail with -O3, win with -O2). I spoke to upstream and I think they fixed this in 4.2. Can someone confirm that please? (In reply to comment #9) > I spoke to upstream and I think they fixed this in 4.2. > Can someone confirm that please? Fixed in 4.2, I confirm. Thanks. |