digitalmars.D.bugs - lazy in main
- Carlos Santander <csantander619 gmail.com> Sep 05 2006
- Kirk McDonald <kirklin.mcdonald gmail.com> Sep 05 2006
- Thomas Kuehne <thomas-dloop kuehne.cn> Sep 07 2006
This compiles successfully:
int main (lazy char [][] args)
{
return args.length;
}
But at runtime I get "Illegal instruction". Tested with gdc from SVN repos
(uses
dmd 0.166) on Mac OS X.
--
Carlos Santander Bernal
Sep 05 2006
Carlos Santander wrote:This compiles successfully: int main (lazy char [][] args) { return args.length; } But at runtime I get "Illegal instruction". Tested with gdc from SVN repos (uses dmd 0.166) on Mac OS X.
I would call this a bug in the compiler. The spec states there are only four valid forms of main: void main() { ... } void main(char[][] args) { ... } int main() { ... } int main(char[][] args) { ... } http://digitalmars.com/d/function.html Since that isn't one of them, I would think the compiler should issue an error. -- Kirk McDonald Pyd: Wrapping Python with D http://pyd.dsource.org
Sep 05 2006
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Kirk McDonald schrieb am 2006-09-05:Carlos Santander wrote:This compiles successfully: int main (lazy char [][] args) { return args.length; } But at runtime I get "Illegal instruction". Tested with gdc from SVN repos (uses dmd 0.166) on Mac OS X.
I would call this a bug in the compiler. The spec states there are only four valid forms of main: void main() { ... } void main(char[][] args) { ... } int main() { ... } int main(char[][] args) { ... } http://digitalmars.com/d/function.html Since that isn't one of them, I would think the compiler should issue an error.
Added to DStress as http://dstress.kuehne.cn/nocompile/m/main_07_A.d http://dstress.kuehne.cn/nocompile/m/main_07_B.d Thomas -----BEGIN PGP SIGNATURE----- iD8DBQFE/9ZHLK5blCcjpWoRAtB/AJ9RwF/tVkSjJber3vkS8mzZYHAxOwCfSRAm qwpddA9mBw+tJ5Xg2gEDdVQ= =yM84 -----END PGP SIGNATURE-----
Sep 07 2006








Thomas Kuehne <thomas-dloop kuehne.cn>