On Sat, Dec 29, 2001 at 03:40:31PM +0100, Michael Schmitz wrote: > > > Does anyone have time to dig into bug #123015 a little? I've tried the > > > obvious fix (make sure the built-in CRASHES_ON_STARTUP workaround is > > > defined :)), but Mark Brown says that doesn't help. The only unstable > > > > Another occurrence of reusing a va_list variable. Although in this > > case, it looks like it was just an oversight. In addition to fixing > > That's what it looked like from the gdb trace. I'll keep your patch > solution around - even where __va_copy is defined for other archs it > should not hurt to use it and work on a copy of va_list, right?
Absolutely. Someday va_copy should be available everywhere... as with all changes to C, it may be a while. Actually, in GCC 3: #if !defined(__STRICT_ANSI__) || __STDC_VERSION__ + 0 >= 199900L #define va_copy(d,s) __builtin_va_copy((d),(s)) #endif #define __va_copy(d,s) __builtin_va_copy((d),(s)) So C99 in non-ANSI mode has va_copy, and otherwise everyone has __va_copy. -- Daniel Jacobowitz Carnegie Mellon University MontaVista Software Debian GNU/Linux Developer
