Hi,
While I was reading the documentation, I noticed a divergence between the earlier and later sections.
:
This is the link I referenced: CUDA Math API Documentation - BFLOAT16 Arithmetic
Hi,
While I was reading the documentation, I noticed a divergence between the earlier and later sections.
:
No.
That appears to be a typo or HTML rendering issue.
The actual function documentation as you show indicates the double - marking, which is correct. The list of prototypes at the top should also reflect the double - marking.
Some text tools (e.g. microsoft word) will convert a typed double dash --
to a long dash: -- -> –
like so:
If you look at the list of prototypes, that appears to be what happened:
As you have already indicated, the actual function documentation shows the correct syntax. If this is concerning to you, you can file a bug.
OK I got it, thanks.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.