Why does core.math exist? It's basically empty, but with a couple of select functions which seem arbitrarily chosen...

On Thursday, 15 March 2018 at 00:16:05 UTC, Manu wrote:
> Why does core.math exist? It's basically empty, but with a couple of select functions which seem arbitrarily chosen...

Isn't core.math compiler intrinsics? The corresponding functions in std.math call the core.math versions.

On Thursday, 15 March 2018 at 00:37:39 UTC, jmh530 wrote:
> On Thursday, 15 March 2018 at 00:16:05 UTC, Manu wrote:
>> Why does core.math exist? It's basically empty, but with a couple of select functions which seem arbitrarily chosen...
>
> Isn't core.math compiler intrinsics? The corresponding functions in std.math call the core.math versions.

They are. I'm in the process of adding a bunch of overloads to core math so that we can stop using `real` unnecessarily (if anyone with a 32-bit machine can build dmd with pr 8002  and run the `std.bitmanip` unit tests and tell me why it crashes that would be much appreciated).

On Thursday, 15 March 2018 at 02:58:12 UTC, Nicholas Wilson wrote:
> On Thursday, 15 March 2018 at 00:37:39 UTC, jmh530 wrote:
>> On Thursday, 15 March 2018 at 00:16:05 UTC, Manu wrote:
>>> Why does core.math exist? It's basically empty, but with a couple of select functions which seem arbitrarily chosen...
>>
>> Isn't core.math compiler intrinsics? The corresponding functions in std.math call the core.math versions.
>
> They are. I'm in the process of adding a bunch of overloads to core math so that we can stop using `real` unnecessarily (if anyone with a 32-bit machine can build dmd with pr 8002  and run the `std.bitmanip` unit tests and tell me why it crashes that would be much appreciated).

dmd -m32?