Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[RV64_DYNAREC] Added more MMX opcodes for vector #2035

Merged
merged 8 commits into from
Nov 14, 2024
Merged

[RV64_DYNAREC] Added more MMX opcodes for vector #2035

Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
4a0e8e9
[RV64_DYNAREC] Added 0F 68 PUNPCKHBW opcode
xctan Nov 12, 2024
9b2c441
[RV64_DYNAREC] Added 0F 69 PUNPCKHWD opcode
xctan Nov 12, 2024
a5674ff
[RV64_DYNAREC] Added 0F 6A PUNPCKHDQ opcode
xctan Nov 12, 2024
a335df2
[RV64_DYNAREC] Updated 0F 68-69 PUNPCKHBW/WD opcodes
xctan Nov 12, 2024
3492910
[RV64_DYNAREC] Added 0F 60 PUNPCKLBW opcode
xctan Nov 12, 2024
833a9cb
[RV64_DYNAREC] Added 0F 61 PUNPCKLWD opcode
xctan Nov 12, 2024
e3697b1
[RV64_DYNAREC] Added 0F 62 PUNPCKLDQ opcode
xctan Nov 12, 2024
5b75a93
[RV64_DYNAREC] Simplified MMX PUNPCK{L,H}{BW,WD,DQ}
xctan Nov 14, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
101 changes: 101 additions & 0 deletions src/dynarec/rv64/dynarec_rv64_0f_vector.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -483,6 +483,52 @@ uintptr_t dynarec64_0F_vector(dynarec_rv64_t* dyn, uintptr_t addr, uintptr_t ip,
VADD_VX(q0, q1, xZR, VECTOR_MASKED);
}
break;
case 0x60:
INST_NAME("PUNPCKLBW Gm, Em");
nextop = F8;
GETGM_vector(q0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW64, 1);
GETEM_vector(q1, 0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW8, 1);
v0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
d0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
VWADDU_VX(d0, q0, xZR, VECTOR_UNMASKED);
VWADDU_VX(v0, q1, xZR, VECTOR_UNMASKED);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW16, 1);
VSLL_VI(v0, v0, 8, VECTOR_UNMASKED);
VOR_VV(q0, d0, v0, VECTOR_UNMASKED);
break;
case 0x61:
INST_NAME("PUNPCKLWD Gm, Em");
nextop = F8;
GETGM_vector(q0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW64, 1);
GETEM_vector(q1, 0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW16, 1);
v0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
d0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
VWADDU_VX(d0, q0, xZR, VECTOR_UNMASKED);
VWADDU_VX(v0, q1, xZR, VECTOR_UNMASKED);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW32, 1);
VSLL_VI(v0, v0, 16, VECTOR_UNMASKED);
VOR_VV(q0, d0, v0, VECTOR_UNMASKED);
break;
case 0x62:
INST_NAME("PUNPCKLDQ Gm, Em");
nextop = F8;
GETGM_vector(q0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW64, 1);
GETEM_vector(q1, 0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW32, 1);
v0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
d0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
MOV32w(x2, 32);
VWADDU_VX(d0, q0, xZR, VECTOR_UNMASKED);
VWADDU_VX(v0, q1, xZR, VECTOR_UNMASKED);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW64, 1);
VSLL_VX(v0, v0, x2, VECTOR_UNMASKED);
VOR_VV(q0, d0, v0, VECTOR_UNMASKED);
break;
case 0x63:
INST_NAME("PACKSSWB Gm, Em");
nextop = F8;
Expand Down Expand Up @@ -515,6 +561,61 @@ uintptr_t dynarec64_0F_vector(dynarec_rv64_t* dyn, uintptr_t addr, uintptr_t ip,
SET_ELEMENT_WIDTH(x1, VECTOR_SEW8, 1);
VNCLIPU_WI(q0, d0, 0, VECTOR_UNMASKED);
break;
case 0x68:
INST_NAME("PUNPCKHBW Gm, Em");
nextop = F8;
GETGM_vector(q0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW64, 1);
GETEM_vector(q1, 0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW8, 1);
v0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
v1 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL1);
d0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
VSLIDEDOWN_VI(v0, q0, 4, VECTOR_UNMASKED);
VSLIDEDOWN_VI(v1, q1, 4, VECTOR_UNMASKED);
VWADDU_VX(d0, v0, xZR, VECTOR_UNMASKED);
VWADDU_VX(v0, v1, xZR, VECTOR_UNMASKED);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW16, 1);
VSLL_VI(v0, v0, 8, VECTOR_UNMASKED);
VOR_VV(q0, d0, v0, VECTOR_UNMASKED);
break;
case 0x69:
INST_NAME("PUNPCKHWD Gm, Em");
nextop = F8;
GETGM_vector(q0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW64, 1);
GETEM_vector(q1, 0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW16, 1);
v0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
v1 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL1);
d0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
VSLIDEDOWN_VI(v0, q0, 2, VECTOR_UNMASKED);
VSLIDEDOWN_VI(v1, q1, 2, VECTOR_UNMASKED);
VWADDU_VX(d0, v0, xZR, VECTOR_UNMASKED);
VWADDU_VX(v0, v1, xZR, VECTOR_UNMASKED);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW32, 1);
VSLL_VI(v0, v0, 16, VECTOR_UNMASKED);
VOR_VV(q0, d0, v0, VECTOR_UNMASKED);
break;
case 0x6A:
INST_NAME("PUNPCKHDQ Gm, Em");
nextop = F8;
GETGM_vector(q0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW64, 1);
GETEM_vector(q1, 0);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW32, 1);
MOV32w(x2, 32);
v0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
v1 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL1);
d0 = fpu_get_scratch_lmul(dyn, VECTOR_LMUL2);
VSLIDEDOWN_VI(v0, q0, 1, VECTOR_UNMASKED);
VSLIDEDOWN_VI(v1, q1, 1, VECTOR_UNMASKED);
VWADDU_VX(d0, v0, xZR, VECTOR_UNMASKED);
VWADDU_VX(v0, v1, xZR, VECTOR_UNMASKED);
SET_ELEMENT_WIDTH(x1, VECTOR_SEW64, 1);
VSLL_VX(v0, v0, x2, VECTOR_UNMASKED);
VOR_VV(q0, d0, v0, VECTOR_UNMASKED);
break;
case 0x6B:
INST_NAME("PACKSSDW Gm, Em");
nextop = F8;
Expand Down
Loading