3
0
Fork 0
mirror of https://github.com/YosysHQ/yosys synced 2026-04-26 22:03:36 +00:00
Commit graph

18 commits

Author SHA1 Message Date
Krystine Sherwin
31679685a3
symfpu: Missed a space 2026-03-11 15:38:59 +13:00
Krystine Sherwin
ad32d66553
symfpu_convert: Handle signed ints
Use input wire `is_signed` to select between signed and unsigned handling.
2026-03-11 12:58:59 +13:00
Krystine Sherwin
f95a5ac7df
symfpu: Use ubv for convert flags 2026-03-11 12:58:59 +13:00
Krystine Sherwin
e6e8075b3a
symfpu: Convert with flags 2026-03-11 12:58:59 +13:00
Krystine Sherwin
43c1b82e64
symfpu: Add symfpu_convert
Convert one input to three outputs (int -> float, float -> int, float -> float).
No rounding mode, no flags (yet).
2026-03-11 12:58:58 +13:00
Krystine Sherwin
53de8b6abe
symfpu: Add -compare mode
Also `min` and `max` ops.  RISC-V uses IEEE 754-2019 semantics where `min(+0,-0) == -0` and `max(+0,-0) == +0` so we do the same here.  We could make it optional, but as I understand it the newer behavior is still backwards compatible (since previously it was valid to have selected either).
2026-03-11 12:58:58 +13:00
Krystine Sherwin
529df986b0
Add symfpu -classify
Add description text for standard `symfpu` signature.
2026-03-11 12:58:58 +13:00
Krystine Sherwin
5372cf5047
symfpu: Add altsqrt
No denormalization here.  That can be a problem for later (or not at all).
2026-03-11 12:58:57 +13:00
Krystine Sherwin
8a764230af
symfpu: Add alt2div
`altdiv` but without denormalization, because as it turns out HardFloat unpacks subnorms in the same way, so lets just support both styles.
2026-03-11 12:58:56 +13:00
Krystine Sherwin
1312668092
symfpu: Add altdiv 2026-03-11 12:58:56 +13:00
Krystine Sherwin
7b270a79c0
symfpu: Dynamic rounding mode 2026-03-11 12:58:55 +13:00
Krystine Sherwin
9cb347a1a6
symfpu: Tidying output
Also switching to cleaner library branch
2026-03-11 12:58:54 +13:00
Krystine Sherwin
cc4e9e52a8
symfpu: floatWithStatusFlags
Now with verified muladd exceptions.
2026-03-11 12:58:54 +13:00
Krystine Sherwin
cd28709b81
symfpu: Configurable rounding modes
Including tests, but currently only testing rounding modes on multiply.
Also missing the ...01 case.
2026-03-11 12:58:54 +13:00
Krystine Sherwin
c1bc302174
symfpu: Add flags
Use symfpu fork.
Add tests for symfpu properties and extra edge case checking for flags.
2026-03-11 12:58:53 +13:00
Krystine Sherwin
77f3248a11
symfpu: Configurable op 2026-03-11 12:58:53 +13:00
Krystine Sherwin
e7588492a0
symfpu: Configurable eb and sb 2026-03-11 12:58:52 +13:00
Jannis Harder
25848020cc
wip: symfpu pass 2026-03-11 12:58:52 +13:00