sin vs floor : clang bug

Eli Friedman eli.friedman at
Tue Jul 30 10:18:41 PDT 2013

On Tue, Jul 30, 2013 at 3:55 AM, Joerg Sonnenberger
<joerg at> wrote:
> On Mon, Jul 29, 2013 at 06:12:30PM -0700, Eli Friedman wrote:
>> The clang "bug" isn't actually a bug; we do this intentionally so
>> -fmath-errno works.
> Isn't there also the difference that sin is affected by the rounding
> mode and floor not?

We don't currently support programs which change the default rounding
mode anyway.  See .


More information about the cfe-commits mailing list