-
Notifications
You must be signed in to change notification settings - Fork 15.3k
[AMDGPU] Add GFX12 assembler/disassembler support for v_dual_dot2acc_* #119211
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.
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
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Do for GFX12 what llvm#118984 did for GFX11.
Member
|
@llvm/pr-subscribers-mc @llvm/pr-subscribers-backend-amdgpu Author: Jay Foad (jayfoad) ChangesDo for GFX12 what #118984 did for GFX11. Patch is 228.31 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/119211.diff 3 Files Affected:
diff --git a/llvm/lib/Target/AMDGPU/VOPDInstructions.td b/llvm/lib/Target/AMDGPU/VOPDInstructions.td
index b89cd8b914f8b9..47aa2e8f3cc088 100644
--- a/llvm/lib/Target/AMDGPU/VOPDInstructions.td
+++ b/llvm/lib/Target/AMDGPU/VOPDInstructions.td
@@ -128,13 +128,13 @@ class VOPD_MADK<dag outs, dag ins, string asm, VOP_Pseudo VDX, VOP_Pseudo VDY,
defvar VOPDPseudosCommon = [
"V_FMAC_F32_e32", "V_FMAAK_F32", "V_FMAMK_F32", "V_MUL_F32_e32",
"V_ADD_F32_e32", "V_SUB_F32_e32", "V_SUBREV_F32_e32", "V_MUL_LEGACY_F32_e32",
- "V_MOV_B32_e32", "V_CNDMASK_B32_e32", "V_MAX_F32_e32", "V_MIN_F32_e32"
+ "V_MOV_B32_e32", "V_CNDMASK_B32_e32", "V_MAX_F32_e32", "V_MIN_F32_e32",
+ "V_DOT2C_F32_F16_e32", "V_DOT2C_F32_BF16_e32"
];
-defvar VOPDPseudosGFX11 = ["V_DOT2C_F32_F16_e32", "V_DOT2C_F32_BF16_e32"];
defvar VOPDYOnlyPseudosCommon = ["V_ADD_U32_e32", "V_LSHLREV_B32_e32",
"V_AND_B32_e32"];
-defvar VOPDXPseudosGFX11 = !listconcat(VOPDPseudosCommon, VOPDPseudosGFX11);
+defvar VOPDXPseudosGFX11 = VOPDPseudosCommon;
defvar VOPDXPseudosGFX12 = VOPDPseudosCommon;
defvar VOPDYPseudosGFX11 = !listconcat(VOPDXPseudosGFX11, VOPDYOnlyPseudosCommon);
defvar VOPDYPseudosGFX12 = !listconcat(VOPDXPseudosGFX12, VOPDYOnlyPseudosCommon);
diff --git a/llvm/test/MC/AMDGPU/gfx12_asm_vopd.s b/llvm/test/MC/AMDGPU/gfx12_asm_vopd.s
index 80b107982ae383..e345f24f9a0573 100644
--- a/llvm/test/MC/AMDGPU/gfx12_asm_vopd.s
+++ b/llvm/test/MC/AMDGPU/gfx12_asm_vopd.s
@@ -290,6 +290,78 @@ v_dual_add_f32 v6, null, v5 :: v_dual_cndmask_b32 v255, 0xaf123456, v4
// GFX12: encoding: [0x7c,0x0a,0x12,0xc9,0xff,0x08,0xfe,0x06,0x56,0x34,0x12,0xaf]
// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+v_dual_add_f32 v255, v4, v2 :: v_dual_dot2acc_f32_f16 v6, v1, v3
+// GFX12: encoding: [0x04,0x05,0x18,0xc9,0x01,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, v1, v2 :: v_dual_dot2acc_f32_f16 v6, v255, v3
+// GFX12: encoding: [0x01,0x05,0x18,0xc9,0xff,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, v255, v2 :: v_dual_dot2acc_f32_f16 v6, v2, v3
+// GFX12: encoding: [0xff,0x05,0x18,0xc9,0x02,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, v2, v2 :: v_dual_dot2acc_f32_f16 v6, v3, v3
+// GFX12: encoding: [0x02,0x05,0x18,0xc9,0x03,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, v3, v2 :: v_dual_dot2acc_f32_f16 v6, v4, v3
+// GFX12: encoding: [0x03,0x05,0x18,0xc9,0x04,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, s105, v2 :: v_dual_dot2acc_f32_f16 v6, s1, v3
+// GFX12: encoding: [0x69,0x04,0x18,0xc9,0x01,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, s1, v2 :: v_dual_dot2acc_f32_f16 v6, s105, v3
+// GFX12: encoding: [0x01,0x04,0x18,0xc9,0x69,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, ttmp15, v2 :: v_dual_dot2acc_f32_f16 v6, vcc_lo, v3
+// GFX12: encoding: [0x7b,0x04,0x18,0xc9,0x6a,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, exec_hi, v2 :: v_dual_dot2acc_f32_f16 v6, vcc_hi, v3
+// GFX12: encoding: [0x7f,0x04,0x18,0xc9,0x6b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, exec_lo, v2 :: v_dual_dot2acc_f32_f16 v6, ttmp15, v3
+// GFX12: encoding: [0x7e,0x04,0x18,0xc9,0x7b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, m0, v2 :: v_dual_dot2acc_f32_f16 v6, m0, v3
+// GFX12: encoding: [0x7d,0x04,0x18,0xc9,0x7d,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, vcc_hi, v2 :: v_dual_dot2acc_f32_f16 v6, exec_lo, v3
+// GFX12: encoding: [0x6b,0x04,0x18,0xc9,0x7e,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, vcc_lo, v2 :: v_dual_dot2acc_f32_f16 v6, exec_hi, v3
+// GFX12: encoding: [0x6a,0x04,0x18,0xc9,0x7f,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, 0xaf123456, v2 :: v_dual_dot2acc_f32_f16 v6, null, v3
+// GFX12: encoding: [0xff,0x04,0x18,0xc9,0x7c,0x06,0x06,0xff,0x56,0x34,0x12,0xaf]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, src_scc, v2 :: v_dual_dot2acc_f32_f16 v6, -1, v3
+// GFX12: encoding: [0xfd,0x04,0x18,0xc9,0xc1,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, 0.5, v3 :: v_dual_dot2acc_f32_f16 v6, 0.5, v2
+// GFX12: encoding: [0xf0,0x06,0x18,0xc9,0xf0,0x04,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v255, -1, v4 :: v_dual_dot2acc_f32_f16 v6, src_scc, v5
+// GFX12: encoding: [0xc1,0x08,0x18,0xc9,0xfd,0x0a,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_add_f32 v6, null, v5 :: v_dual_dot2acc_f32_f16 v255, 0xfe0b, v4
+// GFX12: encoding: [0x7c,0x0a,0x18,0xc9,0xff,0x08,0xfe,0x06,0x0b,0xfe,0x00,0x00]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
v_dual_add_f32 v255, v4, v2 :: v_dual_fmaak_f32 v6, v1, v3, 0xaf123456
// GFX12: encoding: [0x04,0x05,0x02,0xc9,0x01,0x07,0x06,0xff,0x56,0x34,0x12,0xaf]
// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
@@ -1370,6 +1442,78 @@ v_dual_cndmask_b32 v6, null, v5 :: v_dual_cndmask_b32 v255, 0xaf123456, v4
// GFX12: encoding: [0x7c,0x0a,0x52,0xca,0xff,0x08,0xfe,0x06,0x56,0x34,0x12,0xaf]
// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+v_dual_cndmask_b32 v255, v4, v2 :: v_dual_dot2acc_f32_f16 v6, v1, v3
+// GFX12: encoding: [0x04,0x05,0x58,0xca,0x01,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v1, v2 :: v_dual_dot2acc_f32_f16 v6, v255, v3
+// GFX12: encoding: [0x01,0x05,0x58,0xca,0xff,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v255, v2 :: v_dual_dot2acc_f32_f16 v6, v2, v3
+// GFX12: encoding: [0xff,0x05,0x58,0xca,0x02,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v2, v2 :: v_dual_dot2acc_f32_f16 v6, v3, v3
+// GFX12: encoding: [0x02,0x05,0x58,0xca,0x03,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v3, v2 :: v_dual_dot2acc_f32_f16 v6, v4, v3
+// GFX12: encoding: [0x03,0x05,0x58,0xca,0x04,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, s105, v2 :: v_dual_dot2acc_f32_f16 v6, s105, v3
+// GFX12: encoding: [0x69,0x04,0x58,0xca,0x69,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, s1, v2 :: v_dual_dot2acc_f32_f16 v6, s1, v3
+// GFX12: encoding: [0x01,0x04,0x58,0xca,0x01,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, ttmp15, v2 :: v_dual_dot2acc_f32_f16 v6, ttmp15, v3
+// GFX12: encoding: [0x7b,0x04,0x58,0xca,0x7b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, exec_hi, v2 :: v_dual_dot2acc_f32_f16 v6, exec_hi, v3
+// GFX12: encoding: [0x7f,0x04,0x58,0xca,0x7f,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, exec_lo, v2 :: v_dual_dot2acc_f32_f16 v6, exec_lo, v3
+// GFX12: encoding: [0x7e,0x04,0x58,0xca,0x7e,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, m0, v2 :: v_dual_dot2acc_f32_f16 v6, m0, v3
+// GFX12: encoding: [0x7d,0x04,0x58,0xca,0x7d,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, vcc_hi, v2 :: v_dual_dot2acc_f32_f16 v6, vcc_hi, v3
+// GFX12: encoding: [0x6b,0x04,0x58,0xca,0x6b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, vcc_lo, v2 :: v_dual_dot2acc_f32_f16 v6, vcc_lo, v3
+// GFX12: encoding: [0x6a,0x04,0x58,0xca,0x6a,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, 0xaf123456, v2 :: v_dual_dot2acc_f32_f16 v6, null, v3
+// GFX12: encoding: [0xff,0x04,0x58,0xca,0x7c,0x06,0x06,0xff,0x56,0x34,0x12,0xaf]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, src_scc, v2 :: v_dual_dot2acc_f32_f16 v6, -1, v3
+// GFX12: encoding: [0xfd,0x04,0x58,0xca,0xc1,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, 0.5, v3 :: v_dual_dot2acc_f32_f16 v6, 0.5, v2
+// GFX12: encoding: [0xf0,0x06,0x58,0xca,0xf0,0x04,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, -1, v4 :: v_dual_dot2acc_f32_f16 v6, src_scc, v5
+// GFX12: encoding: [0xc1,0x08,0x58,0xca,0xfd,0x0a,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v6, null, v5 :: v_dual_dot2acc_f32_f16 v255, 0xfe0b, v4
+// GFX12: encoding: [0x7c,0x0a,0x58,0xca,0xff,0x08,0xfe,0x06,0x0b,0xfe,0x00,0x00]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
v_dual_cndmask_b32 v255, v4, v2 :: v_dual_fmaak_f32 v6, v1, v3, 0xaf123456
// GFX12: encoding: [0x04,0x05,0x42,0xca,0x01,0x07,0x06,0xff,0x56,0x34,0x12,0xaf]
// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
@@ -2010,84 +2154,1228 @@ v_dual_cndmask_b32 v255, 0.5, v3 :: v_dual_sub_f32 v6, 0.5, v2
// GFX12: encoding: [0xf0,0x06,0x4a,0xca,0xf0,0x04,0x06,0xff]
// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
-v_dual_cndmask_b32 v255, -1, v4 :: v_dual_sub_f32 v6, src_scc, v5
-// GFX12: encoding: [0xc1,0x08,0x4a,0xca,0xfd,0x0a,0x06,0xff]
+v_dual_cndmask_b32 v255, -1, v4 :: v_dual_sub_f32 v6, src_scc, v5
+// GFX12: encoding: [0xc1,0x08,0x4a,0xca,0xfd,0x0a,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v6, null, v5 :: v_dual_sub_f32 v255, 0xaf123456, v4
+// GFX12: encoding: [0x7c,0x0a,0x4a,0xca,0xff,0x08,0xfe,0x06,0x56,0x34,0x12,0xaf]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v4, v2 :: v_dual_subrev_f32 v6, v1, v3
+// GFX12: encoding: [0x04,0x05,0x4c,0xca,0x01,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v1, v2 :: v_dual_subrev_f32 v6, v255, v3
+// GFX12: encoding: [0x01,0x05,0x4c,0xca,0xff,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v255, v2 :: v_dual_subrev_f32 v6, v2, v3
+// GFX12: encoding: [0xff,0x05,0x4c,0xca,0x02,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v2, v2 :: v_dual_subrev_f32 v6, v3, v3
+// GFX12: encoding: [0x02,0x05,0x4c,0xca,0x03,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, v3, v2 :: v_dual_subrev_f32 v6, v4, v3
+// GFX12: encoding: [0x03,0x05,0x4c,0xca,0x04,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, s105, v2 :: v_dual_subrev_f32 v6, s105, v3
+// GFX12: encoding: [0x69,0x04,0x4c,0xca,0x69,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, s1, v2 :: v_dual_subrev_f32 v6, s1, v3
+// GFX12: encoding: [0x01,0x04,0x4c,0xca,0x01,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, ttmp15, v2 :: v_dual_subrev_f32 v6, ttmp15, v3
+// GFX12: encoding: [0x7b,0x04,0x4c,0xca,0x7b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, exec_hi, v2 :: v_dual_subrev_f32 v6, exec_hi, v3
+// GFX12: encoding: [0x7f,0x04,0x4c,0xca,0x7f,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, exec_lo, v2 :: v_dual_subrev_f32 v6, exec_lo, v3
+// GFX12: encoding: [0x7e,0x04,0x4c,0xca,0x7e,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, m0, v2 :: v_dual_subrev_f32 v6, m0, v3
+// GFX12: encoding: [0x7d,0x04,0x4c,0xca,0x7d,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, vcc_hi, v2 :: v_dual_subrev_f32 v6, vcc_hi, v3
+// GFX12: encoding: [0x6b,0x04,0x4c,0xca,0x6b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, vcc_lo, v2 :: v_dual_subrev_f32 v6, vcc_lo, v3
+// GFX12: encoding: [0x6a,0x04,0x4c,0xca,0x6a,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, 0xaf123456, v2 :: v_dual_subrev_f32 v6, null, v3
+// GFX12: encoding: [0xff,0x04,0x4c,0xca,0x7c,0x06,0x06,0xff,0x56,0x34,0x12,0xaf]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, src_scc, v2 :: v_dual_subrev_f32 v6, -1, v3
+// GFX12: encoding: [0xfd,0x04,0x4c,0xca,0xc1,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, 0.5, v3 :: v_dual_subrev_f32 v6, 0.5, v2
+// GFX12: encoding: [0xf0,0x06,0x4c,0xca,0xf0,0x04,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v255, -1, v4 :: v_dual_subrev_f32 v6, src_scc, v5
+// GFX12: encoding: [0xc1,0x08,0x4c,0xca,0xfd,0x0a,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_cndmask_b32 v6, null, v5 :: v_dual_subrev_f32 v255, 0xaf123456, v4
+// GFX12: encoding: [0x7c,0x0a,0x4c,0xca,0xff,0x08,0xfe,0x06,0x56,0x34,0x12,0xaf]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v4, v2 :: v_dual_add_f32 v6, v1, v3
+// GFX12: encoding: [0x04,0x05,0x08,0xcb,0x01,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v1, v2 :: v_dual_add_f32 v6, v255, v3
+// GFX12: encoding: [0x01,0x05,0x08,0xcb,0xff,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v255, v2 :: v_dual_add_f32 v6, v2, v3
+// GFX12: encoding: [0xff,0x05,0x08,0xcb,0x02,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v2, v2 :: v_dual_add_f32 v6, v3, v3
+// GFX12: encoding: [0x02,0x05,0x08,0xcb,0x03,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v3, v2 :: v_dual_add_f32 v6, v4, v3
+// GFX12: encoding: [0x03,0x05,0x08,0xcb,0x04,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, s105, v2 :: v_dual_add_f32 v6, s1, v3
+// GFX12: encoding: [0x69,0x04,0x08,0xcb,0x01,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, s1, v2 :: v_dual_add_f32 v6, s105, v3
+// GFX12: encoding: [0x01,0x04,0x08,0xcb,0x69,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, ttmp15, v2 :: v_dual_add_f32 v6, vcc_lo, v3
+// GFX12: encoding: [0x7b,0x04,0x08,0xcb,0x6a,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, exec_hi, v2 :: v_dual_add_f32 v6, vcc_hi, v3
+// GFX12: encoding: [0x7f,0x04,0x08,0xcb,0x6b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, exec_lo, v2 :: v_dual_add_f32 v6, ttmp15, v3
+// GFX12: encoding: [0x7e,0x04,0x08,0xcb,0x7b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, m0, v2 :: v_dual_add_f32 v6, m0, v3
+// GFX12: encoding: [0x7d,0x04,0x08,0xcb,0x7d,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, vcc_hi, v2 :: v_dual_add_f32 v6, exec_lo, v3
+// GFX12: encoding: [0x6b,0x04,0x08,0xcb,0x7e,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, vcc_lo, v2 :: v_dual_add_f32 v6, exec_hi, v3
+// GFX12: encoding: [0x6a,0x04,0x08,0xcb,0x7f,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, 0xfe0b, v2 :: v_dual_add_f32 v6, null, v3
+// GFX12: encoding: [0xff,0x04,0x08,0xcb,0x7c,0x06,0x06,0xff,0x0b,0xfe,0x00,0x00]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, src_scc, v2 :: v_dual_add_f32 v6, -1, v3
+// GFX12: encoding: [0xfd,0x04,0x08,0xcb,0xc1,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, 0.5, v3 :: v_dual_add_f32 v6, 0.5, v2
+// GFX12: encoding: [0xf0,0x06,0x08,0xcb,0xf0,0x04,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, -1, v4 :: v_dual_add_f32 v6, src_scc, v5
+// GFX12: encoding: [0xc1,0x08,0x08,0xcb,0xfd,0x0a,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v6, null, v5 :: v_dual_add_f32 v255, 0xaf123456, v4
+// GFX12: encoding: [0x7c,0x0a,0x08,0xcb,0xff,0x08,0xfe,0x06,0x56,0x34,0x12,0xaf]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v4, v2 :: v_dual_add_nc_u32 v6, v1, v3
+// GFX12: encoding: [0x04,0x05,0x20,0xcb,0x01,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v1, v2 :: v_dual_add_nc_u32 v6, v255, v3
+// GFX12: encoding: [0x01,0x05,0x20,0xcb,0xff,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v255, v2 :: v_dual_add_nc_u32 v6, v2, v3
+// GFX12: encoding: [0xff,0x05,0x20,0xcb,0x02,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v2, v2 :: v_dual_add_nc_u32 v6, v3, v3
+// GFX12: encoding: [0x02,0x05,0x20,0xcb,0x03,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v3, v2 :: v_dual_add_nc_u32 v6, v4, v3
+// GFX12: encoding: [0x03,0x05,0x20,0xcb,0x04,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, s105, v2 :: v_dual_add_nc_u32 v6, s1, v3
+// GFX12: encoding: [0x69,0x04,0x20,0xcb,0x01,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, s1, v2 :: v_dual_add_nc_u32 v6, s105, v3
+// GFX12: encoding: [0x01,0x04,0x20,0xcb,0x69,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, ttmp15, v2 :: v_dual_add_nc_u32 v6, vcc_lo, v3
+// GFX12: encoding: [0x7b,0x04,0x20,0xcb,0x6a,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, exec_hi, v2 :: v_dual_add_nc_u32 v6, vcc_hi, v3
+// GFX12: encoding: [0x7f,0x04,0x20,0xcb,0x6b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, exec_lo, v2 :: v_dual_add_nc_u32 v6, ttmp15, v3
+// GFX12: encoding: [0x7e,0x04,0x20,0xcb,0x7b,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, m0, v2 :: v_dual_add_nc_u32 v6, m0, v3
+// GFX12: encoding: [0x7d,0x04,0x20,0xcb,0x7d,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, vcc_hi, v2 :: v_dual_add_nc_u32 v6, exec_lo, v3
+// GFX12: encoding: [0x6b,0x04,0x20,0xcb,0x7e,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, vcc_lo, v2 :: v_dual_add_nc_u32 v6, exec_hi, v3
+// GFX12: encoding: [0x6a,0x04,0x20,0xcb,0x7f,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, 0xfe0b, v2 :: v_dual_add_nc_u32 v6, null, v3
+// GFX12: encoding: [0xff,0x04,0x20,0xcb,0x7c,0x06,0x06,0xff,0x0b,0xfe,0x00,0x00]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, src_scc, v2 :: v_dual_add_nc_u32 v6, -1, v3
+// GFX12: encoding: [0xfd,0x04,0x20,0xcb,0xc1,0x06,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, 0.5, v3 :: v_dual_add_nc_u32 v6, 0.5, v2
+// GFX12: encoding: [0xf0,0x06,0x20,0xcb,0xf0,0x04,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, -1, v4 :: v_dual_add_nc_u32 v6, src_scc, v5
+// GFX12: encoding: [0xc1,0x08,0x20,0xcb,0xfd,0x0a,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v6, null, v5 :: v_dual_add_nc_u32 v255, 0xaf123456, v4
+// GFX12: encoding: [0x7c,0x0a,0x20,0xcb,0xff,0x08,0xfe,0x06,0x56,0x34,0x12,0xaf]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v4, v2 :: v_dual_and_b32 v6, v1, v3
+// GFX12: encoding: [0x04,0x05,0x24,0xcb,0x01,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v1, v2 :: v_dual_and_b32 v6, v255, v3
+// GFX12: encoding: [0x01,0x05,0x24,0xcb,0xff,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v255, v2 :: v_dual_and_b32 v6, v2, v3
+// GFX12: encoding: [0xff,0x05,0x24,0xcb,0x02,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v2, v2 :: v_dual_and_b32 v6, v3, v3
+// GFX12: encoding: [0x02,0x05,0x24,0xcb,0x03,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, v3, v2 :: v_dual_and_b32 v6, v4, v3
+// GFX12: encoding: [0x03,0x05,0x24,0xcb,0x04,0x07,0x06,0xff]
+// W64-ERR: :[[@LINE-2]]:{{[0-9]+}}: error
+
+v_dual_dot2acc_f32_f16 v255, s105, v2 :: v_dual_and_b32 v6, s1, v3
+// GFX12: encoding: [0x69,0x0...
[truncated]
|
mbrkusanin
approved these changes
Dec 9, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Do for GFX12 what #118984 did for GFX11.