On Thu, Dec 18, 2008 at 16:05, Derek M Jones <derek@xxxxxxxxxxxx> wrote: > Alexey, > >>> "This is a GCC extension. sizeof(void) is invalid C and should error out >>> and does with -pedantic-errors" >>> >>> Yuk, who ever thought to use -pedantic-errors! >>> >>> -ansi or -std=c89 or -std=c99 don't elicit any diagnostics. >> >> I don't really see why you take this as a tragedy. People seem >> to like this extension: > > I appreciate the reason for wanting to make this extension to > be friendly to people wanting to migrate pre-C standard code > and start using void. > > There is plenty of opportunity for all sorts of subtle bugs here > and I would have hoped that gcc was more proactive in warning > users. If people writing new code want a pointer to behave like > a char * then why not declare the fact. > > Of course sparse has to support this usage, but I would expect it > to flag any usage as an error. I don't agree on this, but even if I would, it is generaly accepted that sparse should not duplicate gcc warnings. -- To unsubscribe from this list: send the line "unsubscribe linux-sparse" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
