[Debian GNUstep maintainers] Bug#592751: FTBFS [hppa] - make_services: malloc.c:3097: sYSMALLOc: Assertion failed

[Yavor Doganov]

dann frazier wrote:
> gnustep-gui fails to build on hppa.

Yes, I noticed that it failed to build several times already.  Very
unfortunate.

> make_services: malloc.c:3097: sYSMALLOc: Assertion `(old_top == (((mbinptr) (((char *) &((av)->bins[((1) - 1) * 2])) - __builtin_offsetof (struct malloc_chunk, fd)))) && old_size == 0) || ((unsigned long) (old_size) >= (unsigned long)((((__builtin_offsetof (struct malloc_chunk, fd_nextsize))+((2 * (sizeof(size_t))) - 1)) & ~((2 * (sizeof(size_t))) - 1))) && ((old_top)->size & 0x1) && ((unsigned long)old_end & pagemask) == 0)' failed.

Leaves me completely clueless.  An assertion failure like this one
seems to indicate a toolchain problem.  The `make_services' program
doesn't use malloc directly; so this has to come via some macros like
`CREATE_AUTORELEASE_POOL' or `-new' methods, which allocate memory
under the hood.

Would it be convenient for you to post a backtrace from running this
command (`make_services --test
GSspell.service/Resources/Info-gnustep.plist' or just `make_services
--help') within the build tree?

Does the package build with noopt?  With LDFLAGS in debian/rules
commented out?

If you have gnustep-base-runtime/1.20.1-2 installed, what does `gdnc
--help' output?  (I suspect that there's a possibility most GNUstep
programs to be currently broken on hppa).