Skip to content

[X86] Correct the cdisp8 encoding for VSCATTER/VGATHER prefetch #122051

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

Merged
merged 1 commit into from
Jan 11, 2025
Merged

[X86] Correct the cdisp8 encoding for VSCATTER/VGATHER prefetch #122051

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
16 changes: 8 additions & 8 deletions llvm/lib/Target/X86/X86InstrAVX512.td
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10388,10 +10388,10 @@ defm VGATHERPF0DPS: avx512_gather_scatter_prefetch<0xC6, MRM1m, "vgatherpf0dps",
VK16WM, vz512mem>, EVEX_V512, EVEX_CD8<32, CD8VT1>;

defm VGATHERPF0QPS: avx512_gather_scatter_prefetch<0xC7, MRM1m, "vgatherpf0qps",
VK8WM, vz256mem>, EVEX_V512, EVEX_CD8<64, CD8VT1>;
VK8WM, vz256mem>, EVEX_V512, EVEX_CD8<32, CD8VT1>;

defm VGATHERPF0DPD: avx512_gather_scatter_prefetch<0xC6, MRM1m, "vgatherpf0dpd",
VK8WM, vy512xmem>, EVEX_V512, REX_W, EVEX_CD8<32, CD8VT1>;
VK8WM, vy512xmem>, EVEX_V512, REX_W, EVEX_CD8<64, CD8VT1>;

defm VGATHERPF0QPD: avx512_gather_scatter_prefetch<0xC7, MRM1m, "vgatherpf0qpd",
VK8WM, vz512mem>, EVEX_V512, REX_W, EVEX_CD8<64, CD8VT1>;
Expand All @@ -10400,10 +10400,10 @@ defm VGATHERPF1DPS: avx512_gather_scatter_prefetch<0xC6, MRM2m, "vgatherpf1dps",
VK16WM, vz512mem>, EVEX_V512, EVEX_CD8<32, CD8VT1>;

defm VGATHERPF1QPS: avx512_gather_scatter_prefetch<0xC7, MRM2m, "vgatherpf1qps",
VK8WM, vz256mem>, EVEX_V512, EVEX_CD8<64, CD8VT1>;
VK8WM, vz256mem>, EVEX_V512, EVEX_CD8<32, CD8VT1>;

defm VGATHERPF1DPD: avx512_gather_scatter_prefetch<0xC6, MRM2m, "vgatherpf1dpd",
VK8WM, vy512xmem>, EVEX_V512, REX_W, EVEX_CD8<32, CD8VT1>;
VK8WM, vy512xmem>, EVEX_V512, REX_W, EVEX_CD8<64, CD8VT1>;

defm VGATHERPF1QPD: avx512_gather_scatter_prefetch<0xC7, MRM2m, "vgatherpf1qpd",
VK8WM, vz512mem>, EVEX_V512, REX_W, EVEX_CD8<64, CD8VT1>;
Expand All @@ -10412,10 +10412,10 @@ defm VSCATTERPF0DPS: avx512_gather_scatter_prefetch<0xC6, MRM5m, "vscatterpf0dps
VK16WM, vz512mem>, EVEX_V512, EVEX_CD8<32, CD8VT1>;

defm VSCATTERPF0QPS: avx512_gather_scatter_prefetch<0xC7, MRM5m, "vscatterpf0qps",
VK8WM, vz256mem>, EVEX_V512, EVEX_CD8<64, CD8VT1>;
VK8WM, vz256mem>, EVEX_V512, EVEX_CD8<32, CD8VT1>;

defm VSCATTERPF0DPD: avx512_gather_scatter_prefetch<0xC6, MRM5m, "vscatterpf0dpd",
VK8WM, vy512xmem>, EVEX_V512, REX_W, EVEX_CD8<32, CD8VT1>;
VK8WM, vy512xmem>, EVEX_V512, REX_W, EVEX_CD8<64, CD8VT1>;

defm VSCATTERPF0QPD: avx512_gather_scatter_prefetch<0xC7, MRM5m, "vscatterpf0qpd",
VK8WM, vz512mem>, EVEX_V512, REX_W, EVEX_CD8<64, CD8VT1>;
Expand All @@ -10424,10 +10424,10 @@ defm VSCATTERPF1DPS: avx512_gather_scatter_prefetch<0xC6, MRM6m, "vscatterpf1dps
VK16WM, vz512mem>, EVEX_V512, EVEX_CD8<32, CD8VT1>;

defm VSCATTERPF1QPS: avx512_gather_scatter_prefetch<0xC7, MRM6m, "vscatterpf1qps",
VK8WM, vz256mem>, EVEX_V512, EVEX_CD8<64, CD8VT1>;
VK8WM, vz256mem>, EVEX_V512, EVEX_CD8<32, CD8VT1>;

defm VSCATTERPF1DPD: avx512_gather_scatter_prefetch<0xC6, MRM6m, "vscatterpf1dpd",
VK8WM, vy512xmem>, EVEX_V512, REX_W, EVEX_CD8<32, CD8VT1>;
VK8WM, vy512xmem>, EVEX_V512, REX_W, EVEX_CD8<64, CD8VT1>;

defm VSCATTERPF1QPD: avx512_gather_scatter_prefetch<0xC7, MRM6m, "vscatterpf1qpd",
VK8WM, vz512mem>, EVEX_V512, REX_W, EVEX_CD8<64, CD8VT1>;
Expand Down
64 changes: 64 additions & 0 deletions llvm/test/MC/X86/avx512pf-64-att.s
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,3 +63,67 @@ vscatterpf0qpd (%r14,%zmm14){%k7}
// CHECK: vscatterpf1qpd (%r15,%zmm13) {%k1}
// CHECK: encoding: [0x62,0x92,0xfd,0x49,0xc7,0x34,0x2f]
vscatterpf1qpd (%r15,%zmm13){%k1}

// CHECK: vgatherpf0dpd 8(%rcx,%ymm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0xfd,0x49,0xc6,0x4c,0x51,0x01]
vgatherpf0dpd 8(%rcx,%ymm2,2){%k1}

// CHECK: vgatherpf0dps 4(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0x7d,0x49,0xc6,0x4c,0x51,0x01]
vgatherpf0dps 4(%rcx,%zmm2,2){%k1}

// CHECK: vgatherpf0qpd 8(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0xfd,0x49,0xc7,0x4c,0x51,0x01]
vgatherpf0qpd 8(%rcx,%zmm2,2){%k1}

// CHECK: vgatherpf0qps 4(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0x7d,0x49,0xc7,0x4c,0x51,0x01]
vgatherpf0qps 4(%rcx,%zmm2,2){%k1}

// CHECK: vgatherpf1dpd 8(%rcx,%ymm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0xfd,0x49,0xc6,0x54,0x51,0x01]
vgatherpf1dpd 8(%rcx,%ymm2,2){%k1}

// CHECK: vgatherpf1dps 4(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0x7d,0x49,0xc6,0x54,0x51,0x01]
vgatherpf1dps 4(%rcx,%zmm2,2){%k1}

// CHECK: vgatherpf1qpd 8(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0xfd,0x49,0xc7,0x54,0x51,0x01]
vgatherpf1qpd 8(%rcx,%zmm2,2){%k1}

// CHECK: vgatherpf1qps 4(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0x7d,0x49,0xc7,0x54,0x51,0x01]
vgatherpf1qps 4(%rcx,%zmm2,2){%k1}

// CHECK: vscatterpf0dpd 8(%rcx,%ymm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0xfd,0x49,0xc6,0x6c,0x51,0x01]
vscatterpf0dpd 8(%rcx,%ymm2,2){%k1}

// CHECK: vscatterpf0dps 4(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0x7d,0x49,0xc6,0x6c,0x51,0x01]
vscatterpf0dps 4(%rcx,%zmm2,2){%k1}

// CHECK: vscatterpf0qpd 8(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0xfd,0x49,0xc7,0x6c,0x51,0x01]
vscatterpf0qpd 8(%rcx,%zmm2,2){%k1}

// CHECK: vscatterpf0qps 4(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0x7d,0x49,0xc7,0x6c,0x51,0x01]
vscatterpf0qps 4(%rcx,%zmm2,2){%k1}

// CHECK: vscatterpf1dpd 8(%rcx,%ymm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0xfd,0x49,0xc6,0x74,0x51,0x01]
vscatterpf1dpd 8(%rcx,%ymm2,2){%k1}

// CHECK: vscatterpf1dps 4(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0x7d,0x49,0xc6,0x74,0x51,0x01]
vscatterpf1dps 4(%rcx,%zmm2,2){%k1}

// CHECK: vscatterpf1qpd 8(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0xfd,0x49,0xc7,0x74,0x51,0x01]
vscatterpf1qpd 8(%rcx,%zmm2,2){%k1}

// CHECK: vscatterpf1qps 4(%rcx,%zmm2,2) {%k1}
// CHECK: encoding: [0x62,0xf2,0x7d,0x49,0xc7,0x74,0x51,0x01]
vscatterpf1qps 4(%rcx,%zmm2,2){%k1}
Loading