Hi, On Tue, 17 Oct 2006, Paolo Bonzini wrote:
> > The other problem is that the configure script doesn't configure the fault > > handler correctly, the m68k check is too late and CFG_FAULT already has been > > set fault-posix.h. > > > This means that, in theory, the fault-posix.h handler was found to work. > > In fact, from > http://buildd.debian.org/fetch.cgi?pkg=gnu-smalltalk;ver=2.1.8-2.1;arch=m68k;stamp=1156726548 > I see this: > > checking whether a fault handler according to POSIX works... yes > ... > checking whether a fault handler according to Linux/m68k works... yes > ... > checking if the system supports catching SIGSEGV... yes > > > I have applied your patch to my copy of libsigsegv, but I hesitate a bit in > applying it upstream. Indeed, it seems the posix handler works better than I expected, I reverted that patch and rerun the checks: Test passed. PASS: sigsegv1 Test passed. PASS: sigsegv2 SKIP: stackoverflow1 SKIP: stackoverflow2 ====================== All 2 tests passed (2 tests were not run) ====================== This is the same result as with the m68k fault handler, so you can drop this part of the patch. Probably what had me confused here is that I've seen upstream complete all tests: http://buildd.debian.org/fetch.cgi?pkg=libsigsegv;ver=2.4-3;arch=m68k;stamp=1159196519 but that's not an immediate issue, I'll look into that later... Anyway, thanks for reacting so quickly. bye, Roman -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
