Float: Introduce floating point api eq #567
Conversation
Incarnation-p-lee
commented
Jun 3, 2024
•
Incarnation-p-lee
commented
- The eq(equal) api(s) introduced.
- Add new type fp_bits_x2.
- Add new type fp_bool_and_flags.
- Overload operator == for test to compare fp_bool_and_flags tuple.
- Add test cases for half, single and double floating point.
- Add eq(equal) to interface.
* The eq(equal) api(s) introduced. * Add new type fp_bits_x2. * Add new type fp_bool_and_flags. * Add new operator ==_t to compare fp_bool_and_flags tuple. * Add test cases for half, single and double floating point. * Add eq(equal) to interface. Signed-off-by: Pan Li <[email protected]>
|
I'm away today, but it shouldn't be a problem for me to do a rebase before merging when I'm back. |
No hurry and take your time. You can review #565 first and then I can do the rebase anyway. I am afraid that you may review these 2 PR(s) and then merge. Then you may need to resolve the conflict, which is my duty. |
lib/float/common.sail
Outdated
| @@ -74,4 +77,13 @@ function float_decompose(op) = { | |||
| } | |||
| } | |||
|
|
|||
| /* Floating point operator implementations */ | |||
| infix 2 ==_t /* The suffix _t indicates tuple */ | |||
There was a problem hiding this comment.
I think these are bad for readability and it would be better to just spell it out.
There was a problem hiding this comment.
Or just scoped to the test file.
There was a problem hiding this comment.
Within a test file you could do $include <generic_equality.sail> and use ==, but it shouldn't be included in ordinary library code.
There was a problem hiding this comment.
Thanks for comments, that make sense to me, will update it after #565 .
There was a problem hiding this comment.
Rebase and updated.
Signed-off-by: Pan Li <[email protected]>
|
Kindly ping for this PR, thanks! |
