Thomas H.P. Andersen
d55a69e769
nak: make is_fneg_zero detect -rZ
fold_imm starts with this check:
let SrcRef::Imm32(mut u) = self.src_ref else {
return *self;
};
For -rZ we thus return early and the matching in is_fneg_zero does not
find it. This MR adds a match for SrcRef::Zero + SrcMod::FNeg.
This lets copy propagation fold this:
r3 = shfl.bfly r2 0x2 0x1c03
r3 = fswzadd.ftz r3 r2 [sub, sub, subr, subr]
r3 = fadd.ftz -rZ |r3|
r16 = shfl.bfly r2 0x1 0x1c03
r16 = fswzadd.ftz r16 r2 [sub, subr, sub, subr]
r16 = fadd.ftz -rZ |r16|
r3 = fadd.ftz r16 r3
into:
r3 = shfl.bfly r2 0x2 0x1c03
r3 = fswzadd.ftz r3 r2 [sub, sub, subr, subr]
r16 = shfl.bfly r2 0x1 0x1c03
r16 = fswzadd.ftz r16 r2 [sub, subr, sub, subr]
r3 = fadd.ftz |r16| |r3|
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12480
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33107>
2025-04-17 02:22:02 +00:00
..
2025-04-16 23:13:36 +00:00
2025-04-16 15:24:37 +00:00
2025-04-15 00:17:11 +02:00
2025-01-08 20:37:51 +00:00
2025-04-16 23:13:36 +00:00
2025-04-09 12:15:33 +00:00
2025-04-16 17:53:39 +00:00
2025-04-16 07:40:50 +00:00
2025-04-16 20:19:14 +00:00
2025-04-09 12:15:33 +00:00
2025-04-09 18:29:57 +00:00
2025-03-27 05:46:35 +00:00
2025-02-26 15:19:53 +00:00
2025-04-16 23:13:36 +00:00
2025-03-13 19:45:45 +00:00
2025-04-04 00:18:28 +00:00
2025-04-15 10:37:39 +00:00
2025-04-16 16:02:07 +00:00
2025-04-17 02:22:02 +00:00
2025-04-15 21:30:41 +00:00
2025-04-08 18:29:10 +00:00
2025-04-15 10:37:39 +00:00
2025-04-08 15:38:39 +00:00
2025-04-10 15:28:22 +00:00
2025-02-17 02:50:15 +00:00
2025-04-11 18:01:47 +00:00
2025-04-09 12:15:33 +00:00