On Tue, Feb 26, 2008 at 6:53 AM, Neil Booth <[EMAIL PROTECTED]> wrote:
> Chris Lattner wrote:-
>
>
>  > On Feb 25, 2008, at 3:13 PM, Eli Friedman wrote:
>  > > (clang unfortunately has to be
>  > > extremely loose about pointer conversions in general.)
>  >
>  >
>  > I think we can tighten this up in C++ mode, right?
>
>  Shouldn't we warn even for C without -pedantic?  I know I as
>  a user would like that.

Yeah, we probably should... gcc warns, so it shouldn't be an issue.
We should be careful not to warn by default about implicit
signed/unsigned conversion, though; people are likely to get annoyed
at a compiler warning that they're passing an int16_t* to a function
expecting a uint16_t*.

-Eli
_______________________________________________
cfe-dev mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev

Reply via email to