On 14 May 2010 06:22, Mike Caron <[email protected]> wrote: > This was meant to go to the whole list, not just James Taylor. > > > On 13/05/2010 2:11 PM, Mike Caron wrote: >> >> No, it's not part of the distribution package, it's an internal testing >> suite. >> >> TMC, it wasn't crashing before. However, NOW, in revision 3535, it >> crashes. :\ > > > Anyway, it's not actually crashing outright. However, there is some funny > stuff going on, because it's triggering the Windows Application Verifier (a > tool vaguely similar to valgrind, but for win32) which is causing a crash > (if not attached to a debugger). > > I don't have time right at the moment to debug it, but I'll check it out > later...
What version of FB are you using? I've noticed that reloadtest.bat doesn't use "-lang deprecated", which is a mistake, because lang fb changes the default argument passing behaviour to byval. reloadtest doesn't compile with lang fb for me, but now that I look closely, the exact error message is: /mnt/d/OHRRPGCE/SOURCE/CURRENT/wip/util.bi(274) error 136: Default types or suffixes are only valid in -lang deprecated or fblite or qb, found ',' in 'declare sub flusharray (array(), byval size as integer=-1, byval value as integer=0)' Which is obviously a compiler bug, and probably a version-specific one. I'm using FreeBASIC Compiler - Version 0.20.0 (05-29-2009) for linux (target:linux) >> On 13/05/2010 2:08 PM, james taylor wrote: >>> >>> nope, i have the latest version already and still done see where its >>> broken >>> >>> On Thu, May 13, 2010 at 2:04 PM, james taylor <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> reloadtest? What is that? ill redownload ohrrpgce and see if its >>> been updated since i last downloaded it >>> >>> >>> On Thu, May 13, 2010 at 10:33 AM, James Paige >>> <[email protected] <mailto:[email protected]>> wrote: >>> >>> On Thu, May 13, 2010 at 05:29:12AM -0700, >>> [email protected] wrote: >>> > teeemcee >>> > 2010-05-13 05:29:12 -0700 (Thu, 13 May 2010) >>> > 363 >>> > reloadtest always crashed in comparenodes! Shock horror, this >>> is a BYREF problem AGAIN! I modified all functions in reload.bas >>> to use BYVAL... funny, I've already done this before, but Mike's >>> been merrily writing new code (I did spot two function modifying >>> BYREF arguments which aren't used yet) >>> > >>> > So how come nobody noticed that reloadtest was completely broken? >>> >>> For me it was not crashing... I am not sure why. >>> >>> --- >>> James >>> _______________________________________________ >>> Ohrrpgce mailing list >>> [email protected] >>> <mailto:[email protected]> >>> http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org >>> >>> >>> >> > > -- > Mike > _______________________________________________ > Ohrrpgce mailing list > [email protected] > http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org > _______________________________________________ Ohrrpgce mailing list [email protected] http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
