On Tue, 8 Jan 2008, James Keenan via RT wrote:
> Patch posted by Jarkko Hietaniemi on parrot-porters. I'll be looking at
> this patch in conjunction with other fixes to this class and its tests.
His patch makes perfect sense to me. (But then his nearly always do.)
You can't know what passing random -Woptions to random compilers might do.
Traditional SysV compilers, for example, take -W as a prefix to hand off
the next arguments to a later stage in the build, so -Wa,-O would pass
'-O' to the assembler. Other compilers could do other random things.
That said, I wonder if icc ought to be included in those tests. It tries
to pretend it's gcc in other ways; I don't know about warnings.
--
Andy Dougherty [EMAIL PROTECTED]
