Build fails in src_compile with a segmentation fault when first executing temacs: if test "no" = "yes"; then \ rm -f bootstrap-emacs; \ ln temacs bootstrap-emacs; \ else \ ./temacs --batch --load loadup bootstrap || exit 1; \ test "X" = X || -zex emacs; \ mv -f emacs bootstrap-emacs; \ fi /bin/sh: line 7: 30823 Segmentation fault ./temacs --batch --load loadup bootstrap make[1]: *** [Makefile:816: bootstrap-emacs] Error 1 make[1]: Leaving directory '/var/tmp/portage/app-editors/emacs-24.5-r3/work/emacs-24.5/src' make: *** [Makefile:387: src] Error 2 * ERROR: app-editors/emacs-24.5-r3::gentoo failed (compile phase): * emake failed (gdb) run Starting program: /var/tmp/portage/app-editors/emacs-24.5-r3/work/emacs-24.5/src/temacs [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". Program received signal SIGSEGV, Segmentation fault. 0x00000000005ce6e5 in calloc ( nmemb=<error reading variable: DWARF-2 expression error: Loop detected (257).>, size=size@entry=1) at gmalloc.c:1529 1529 result = malloc (bytes); Not sure what has triggered this. Possibly due to glibc update from 2.23 to 2.24.
Created attachment 464170 [details] build.log
Created attachment 464172 [details] emerge --info
Same problem here (on hardened) with sys-libs/glibc-2.23-r4
Created attachment 478020 [details] build.log
Created attachment 478022 [details] emerge --info
Looks like duplicate of #622146 and caused by improper pax flags : paxctl-ng -v temacs temacs: PT_PAX : -e--- XATTR_PAX : -e-r- temacs does not segfault after resetting pax flags using: paxctl-ng -z temacs
(In reply to Aurélien Francillon from comment #3) Please don't re-use an existing bug report for an unrelated issue. *This* bug is about a build failure of emacs-24.5-r3 on non-hardened amd64. Your emacs-25.2 issue is most likely the same problem as bug 607990 (in particular, see bug 607990 comment 8).
OK sorry to disturb here, crash was on the same step of compilation and this was unclear initially if that was hardened related or not (or if original post was hardened too...). Thanks for the pointers though.
I cannot reproduce this any more. Closing.