[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Chicken-users] Re: silex.exe dies
From: |
Brandon J. Van Every |
Subject: |
[Chicken-users] Re: silex.exe dies |
Date: |
Wed, 31 May 2006 18:57:43 -0700 |
User-agent: |
Thunderbird 1.5.0.2 (Windows/20060308) |
Brandon J. Van Every wrote:
Brandon J. Van Every wrote:
I tried adding typedef __int64 int64_t to chicken.h. I also tried
adding __int64 to easyffi.l. Neither approach worked. I would
appreciate advice or a resolution to this problem as it's holding up
my builds.
Hrm. I unrecorded a bunch of stuff in my own Darcs repository. The
failure of silex.exe predates integer64 support. I don't know what
causes the failure. I will try unrecording even more stuff until I
find it. If that doesn't work, I'll have to actually think about
what's going on.
Ok I looked back in Darcs and I don't know why silex.exe dies.
easyffi.scm last changed on May 4, 2006. silex.scm changed on May 13,
2006, that was the "separate input and output file" capability. It
worked fine. easyffi.l hasn't changed since Feb. 13, 2006. I guess
it's not silex itself, but something silex uses from libchicken that
causes the death. The death is: "The exception unknown software
exception (oxc00000fd) occurred in the application at location 0x00476407."
Haven't really gone up a debugger learning curve for this stuff. If
anyone has any ideas, I'd love to hear 'em before embarking on this.
I will point out that only the CMake build compiles a silex.exe. It was
either that or search for an extant csi.exe. Since I already had the
search code to find chicken.exe, I thought it would be easier to just
compile silex.exe. Maybe I'm revealing a bug or inconsistency between
interpreted and compiled versions that's always been present, but the
interpreter is more forgiving?
Cheers,
Brandon Van Every