3 Nov
2023
3 Nov
'23
1:51 a.m.
On Fri Nov 3 01:49:39 2023 +0000, Conor McCarthy wrote:
In a similar vein, I didn't add a `DREM` to go with `FREM`. Separate opcodes for `double` are redundant because we have the register data types. I'm not planning on adding `half` variants for native 16-bit support either. To keep things consistent I think we should normalise all TPF `double` variant opcodes to `float` opcodes, and validate the register data types. To keep things moving, I support doing all of that later.
-- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/409#note_50752