[PATCH] Re: [PATCH] was Re: perldoc segfaulting in XS_UNIVERSAL_VERSION

2005-07-15 Thread Michael G Schwern
On Wed, Jul 13, 2005 at 04:43:15PM -0700, Michael G Schwern wrote: $ bleadperl -wle 'package Foo; Foo-VERSION' Bus error Segfaults are handled with fresh_perl_is/like(). Unfortunately t/op/universal.t uses ad hoc testing so I'll have to convert it to t/test.pl first. Back in a bit.

Re: [PATCH] Re: [PATCH] was Re: perldoc segfaulting in XS_UNIVERSAL_VERSION

2005-07-15 Thread H.Merijn Brand
On Fri, 15 Jul 2005 00:49:12 -0700, Michael G Schwern [EMAIL PROTECTED] wrote: On Wed, Jul 13, 2005 at 04:43:15PM -0700, Michael G Schwern wrote: $ bleadperl -wle 'package Foo; Foo-VERSION' Bus error Segfaults are handled with fresh_perl_is/like(). Unfortunately t/op/universal.t

[PATCH] was Re: perldoc segfaulting in XS_UNIVERSAL_VERSION

2005-07-13 Thread John Peacock
John Peacock wrote: I'm backreving to before those patches to see if the cores go away. That was it and this fixes it for good: === universal.c == --- universal.c (revision 17957) +++ universal.c (local) @@ -338,7 +338,7 @@

Re: [PATCH] was Re: perldoc segfaulting in XS_UNIVERSAL_VERSION

2005-07-13 Thread Nicholas Clark
On Wed, Jul 13, 2005 at 02:24:06PM -0400, John Peacock wrote: John Peacock wrote: I'm backreving to before those patches to see if the cores go away. That was it and this fixes it for good: === universal.c == --- universal.c

Re: [PATCH] was Re: perldoc segfaulting in XS_UNIVERSAL_VERSION

2005-07-13 Thread Nicholas Clark
On Wed, Jul 13, 2005 at 09:10:34PM +0100, Nicholas Clark wrote: Strangely the patch didn't want to apply automatically, so I did it by hand. Aha: Content-Type: text/plain; charset=ISO-8859-1; format=flowed No surprise there then. Nicholas Clark

Re: [PATCH] was Re: perldoc segfaulting in XS_UNIVERSAL_VERSION

2005-07-13 Thread John Peacock
Nicholas Clark wrote: Aha: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sorry, that's a typo, it should be format=flawed... :-( As for a regression test, I was thinking about that. I don't how to write a test for SEGV, but Schwern's example is as good as any: BEFORE

Re: [PATCH] was Re: perldoc segfaulting in XS_UNIVERSAL_VERSION

2005-07-13 Thread Michael G Schwern
On Wed, Jul 13, 2005 at 04:35:30PM -0400, John Peacock wrote: As for a regression test, I was thinking about that. I don't how to write a test for SEGV, but Schwern's example is as good as any: BEFORE PATCH: $ LD_LIBRARY_PATH=$PWD ./perl -Ilib utils/perldoc perldoc Segmentation fault $