[llvm] [AMDGPU][MC] Fix printing vcc(_lo) twice for VOPC DPP instrucitons (PR #81158)
Mirko BrkuĊĦanin via llvm-commits
llvm-commits at lists.llvm.org
Thu Feb 8 08:28:54 PST 2024
https://github.com/mbrkusanin created https://github.com/llvm/llvm-project/pull/81158
None
>From b3893dbdae25f374e7348d16e8397e0233542f03 Mon Sep 17 00:00:00 2001
From: Mirko Brkusanin <Mirko.Brkusanin at amd.com>
Date: Thu, 8 Feb 2024 17:11:59 +0100
Subject: [PATCH 1/2] Use full asm with encoding in check lines
---
llvm/test/MC/AMDGPU/gfx11-promotions.s | 156 ++++++++++++-------------
1 file changed, 78 insertions(+), 78 deletions(-)
diff --git a/llvm/test/MC/AMDGPU/gfx11-promotions.s b/llvm/test/MC/AMDGPU/gfx11-promotions.s
index 6be33afb5671b..bf85ba9f8c703 100644
--- a/llvm/test/MC/AMDGPU/gfx11-promotions.s
+++ b/llvm/test/MC/AMDGPU/gfx11-promotions.s
@@ -11,343 +11,343 @@
//===----------------------------------------------------------------------===//
v_mov_b32 v0, v1
-// GFX11: encoding: [0x01,0x03,0x00,0x7e]
+// GFX11: v_mov_b32_e32 v0, v1 ; encoding: [0x01,0x03,0x00,0x7e]
v_mov_b32_e32 v0, v1
-// GFX11: encoding: [0x01,0x03,0x00,0x7e]
+// GFX11: v_mov_b32_e32 v0, v1 ; encoding: [0x01,0x03,0x00,0x7e]
//===----------------------------------------------------------------------===//
// VOP2.
//===----------------------------------------------------------------------===//
v_add_f16 v5, v1, v2
-// GFX11: encoding: [0x01,0x05,0x0a,0x64]
+// GFX11: v_add_f16_e32 v5, v1, v2 ; encoding: [0x01,0x05,0x0a,0x64]
v_add_f16_e32 v5, v1, v2
-// GFX11: encoding: [0x01,0x05,0x0a,0x64]
+// GFX11: v_add_f16_e32 v5, v1, v2 ; encoding: [0x01,0x05,0x0a,0x64]
//===----------------------------------------------------------------------===//
// VOPC.
//===----------------------------------------------------------------------===//
v_cmp_lt_f32 vcc_lo, v1, v2
-// GFX11: encoding: [0x01,0x05,0x22,0x7c]
+// GFX11: v_cmp_lt_f32_e32 vcc_lo, v1, v2 ; encoding: [0x01,0x05,0x22,0x7c]
v_cmp_lt_f32_e32 vcc_lo, v1, v2
-// GFX11: encoding: [0x01,0x05,0x22,0x7c]
+// GFX11: v_cmp_lt_f32_e32 vcc_lo, v1, v2 ; encoding: [0x01,0x05,0x22,0x7c]
//===----------------------------------------------------------------------===//
// VOPCX.
//===----------------------------------------------------------------------===//
v_cmpx_class_f16 v1, v2
-// GFX11: encoding: [0x01,0x05,0xfa,0x7d]
+// GFX11: v_cmpx_class_f16_e32 v1, v2 ; encoding: [0x01,0x05,0xfa,0x7d]
v_cmpx_class_f16_e32 v1, v2
-// GFX11: encoding: [0x01,0x05,0xfa,0x7d]
+// GFX11: v_cmpx_class_f16_e32 v1, v2 ; encoding: [0x01,0x05,0xfa,0x7d]
//===----------------------------------------------------------------------===//
// VOP1.DPP8.
//===----------------------------------------------------------------------===//
v_bfrev_b32 v5, v1 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x70,0x0a,0x7e,0x01,0x77,0x39,0x05]
+// GFX11: v_bfrev_b32_dpp v5, v1 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x70,0x0a,0x7e,0x01,0x77,0x39,0x05]
v_bfrev_b32_dpp v5, v1 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x70,0x0a,0x7e,0x01,0x77,0x39,0x05]
+// GFX11: v_bfrev_b32_dpp v5, v1 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x70,0x0a,0x7e,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOP1.DPP16.
//===----------------------------------------------------------------------===//
v_bfrev_b32 v5, v1 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x70,0x0a,0x7e,0x01,0x1b,0x00,0xff]
+// GFX11: v_bfrev_b32_dpp v5, v1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x70,0x0a,0x7e,0x01,0x1b,0x00,0xff]
v_bfrev_b32_dpp v5, v1 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x70,0x0a,0x7e,0x01,0x1b,0x00,0xff]
+// GFX11: v_bfrev_b32_dpp v5, v1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x70,0x0a,0x7e,0x01,0x1b,0x00,0xff]
//===----------------------------------------------------------------------===//
// VOP2.DPP8.
//===----------------------------------------------------------------------===//
v_add_f16 v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x04,0x0a,0x64,0x01,0x77,0x39,0x05]
+// GFX11: v_add_f16_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0x0a,0x64,0x01,0x77,0x39,0x05]
v_add_f16_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x04,0x0a,0x64,0x01,0x77,0x39,0x05]
+// GFX11: v_add_f16_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0x0a,0x64,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOP2.DPP16.
//===----------------------------------------------------------------------===//
v_add_f16 v5, v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x04,0x0a,0x64,0x01,0x1b,0x00,0xff]
+// GFX11: v_add_f16_dpp v5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0x0a,0x64,0x01,0x1b,0x00,0xff]
v_add_f16_dpp v5, v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x04,0x0a,0x64,0x01,0x1b,0x00,0xff]
+// GFX11: v_add_f16_dpp v5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0x0a,0x64,0x01,0x1b,0x00,0xff]
//===----------------------------------------------------------------------===//
// VOPC.DPP8.
//===----------------------------------------------------------------------===//
v_cmp_le_u16 v1, v2 dpp8:[7,7,7,3,4,4,6,7] fi:1
-// GFX11: encoding: [0xea,0x04,0x76,0x7c,0x01,0xff,0x47,0xfa]
+// GFX11: v_cmp_le_u16 vcc_lo, v1, v2 dpp8:[7,7,7,3,4,4,6,7] fi:1 ; encoding: [0xea,0x04,0x76,0x7c,0x01,0xff,0x47,0xfa]
v_cmp_le_u16_dpp v1, v2 dpp8:[7,7,7,3,4,4,6,7] fi:1
-// GFX11: encoding: [0xea,0x04,0x76,0x7c,0x01,0xff,0x47,0xfa]
+// GFX11: v_cmp_le_u16 vcc_lo, v1, v2 dpp8:[7,7,7,3,4,4,6,7] fi:1 ; encoding: [0xea,0x04,0x76,0x7c,0x01,0xff,0x47,0xfa]
//===----------------------------------------------------------------------===//
// VOPC.DPP16.
//===----------------------------------------------------------------------===//
v_cmp_gt_u16 v1, v2 row_shl:0x7 row_mask:0x0 bank_mask:0x0 fi:1
-// GFX11: encoding: [0xfa,0x04,0x78,0x7c,0x01,0x07,0x05,0x00]
+// GFX11: v_cmp_gt_u16 vcc_lo, v1, v2 row_shl:7 row_mask:0x0 bank_mask:0x0 fi:1 ; encoding: [0xfa,0x04,0x78,0x7c,0x01,0x07,0x05,0x00]
v_cmp_gt_u16_dpp v1, v2 row_shl:0x7 row_mask:0x0 bank_mask:0x0 fi:1
-// GFX11: encoding: [0xfa,0x04,0x78,0x7c,0x01,0x07,0x05,0x00]
+// GFX11: v_cmp_gt_u16 vcc_lo, v1, v2 row_shl:7 row_mask:0x0 bank_mask:0x0 fi:1 ; encoding: [0xfa,0x04,0x78,0x7c,0x01,0x07,0x05,0x00]
//===----------------------------------------------------------------------===//
// VOPCX.DPP8.
//===----------------------------------------------------------------------===//
v_cmpx_class_f16 v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x04,0xfa,0x7d,0x01,0x77,0x39,0x05]
+// GFX11: v_cmpx_class_f16 v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0xfa,0x7d,0x01,0x77,0x39,0x05]
v_cmpx_class_f16_dpp v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x04,0xfa,0x7d,0x01,0x77,0x39,0x05]
+// GFX11: v_cmpx_class_f16 v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0xfa,0x7d,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOPCX.DPP16.
//===----------------------------------------------------------------------===//
v_cmpx_class_f16 v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x04,0xfa,0x7d,0x01,0x1b,0x00,0xff]
+// GFX11: v_cmpx_class_f16 v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0xfa,0x7d,0x01,0x1b,0x00,0xff]
v_cmpx_class_f16_dpp v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x04,0xfa,0x7d,0x01,0x1b,0x00,0xff]
+// GFX11: v_cmpx_class_f16 v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0xfa,0x7d,0x01,0x1b,0x00,0xff]
//===----------------------------------------------------------------------===//
// VOP1 -> VOP3.
//===----------------------------------------------------------------------===//
v_sin_f32 v5, 0.5 mul:2
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0xf0,0x00,0x00,0x08]
+// GFX11: v_sin_f32_e64 v5, 0.5 mul:2 ; encoding: [0x05,0x00,0xb5,0xd5,0xf0,0x00,0x00,0x08]
v_sin_f32_e64 v5, 0.5 mul:2
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0xf0,0x00,0x00,0x08]
+// GFX11: v_sin_f32_e64 v5, 0.5 mul:2 ; encoding: [0x05,0x00,0xb5,0xd5,0xf0,0x00,0x00,0x08]
v_sin_f32_e64 v5, v1
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0x01,0x01,0x00,0x00]
+// GFX11: v_sin_f32_e64 v5, v1 ; encoding: [0x05,0x00,0xb5,0xd5,0x01,0x01,0x00,0x00]
v_sin_f32 v5, v1
-// GFX11: encoding: [0x01,0x6b,0x0a,0x7e]
+// GFX11: v_sin_f32_e32 v5, v1 ; encoding: [0x01,0x6b,0x0a,0x7e]
//===----------------------------------------------------------------------===//
// VOP2 -> VOP3.
//===----------------------------------------------------------------------===//
v_add_f32 v5, v1, -v2
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0x01,0x05,0x02,0x40]
+// GFX11: v_add_f32_e64 v5, v1, -v2 ; encoding: [0x05,0x00,0x03,0xd5,0x01,0x05,0x02,0x40]
v_add_f32_e64 v5, v1, -v2
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0x01,0x05,0x02,0x40]
+// GFX11: v_add_f32_e64 v5, v1, -v2 ; encoding: [0x05,0x00,0x03,0xd5,0x01,0x05,0x02,0x40]
v_add_f32_e64 v5, v1, v2
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0x01,0x05,0x02,0x00]
+// GFX11: v_add_f32_e64 v5, v1, v2 ; encoding: [0x05,0x00,0x03,0xd5,0x01,0x05,0x02,0x00]
v_add_f32 v5, v1, v2
-// GFX11: encoding: [0x01,0x05,0x0a,0x06]
+// GFX11: v_add_f32_e32 v5, v1, v2 ; encoding: [0x01,0x05,0x0a,0x06]
//===----------------------------------------------------------------------===//
// VOPC -> VOP3.
//===----------------------------------------------------------------------===//
v_cmp_f_f32 s10, -v1, v2
-// GFX11: encoding: [0x0a,0x00,0x10,0xd4,0x01,0x05,0x02,0x20]
+// GFX11: v_cmp_f_f32_e64 s10, -v1, v2 ; encoding: [0x0a,0x00,0x10,0xd4,0x01,0x05,0x02,0x20]
v_cmp_f_f32_e64 s10, -v1, v2
-// GFX11: encoding: [0x0a,0x00,0x10,0xd4,0x01,0x05,0x02,0x20]
+// GFX11: v_cmp_f_f32_e64 s10, -v1, v2 ; encoding: [0x0a,0x00,0x10,0xd4,0x01,0x05,0x02,0x20]
v_cmp_f_f32_e64 vcc_lo, v1, v2
-// GFX11: encoding: [0x6a,0x00,0x10,0xd4,0x01,0x05,0x02,0x00]
+// GFX11: v_cmp_f_f32_e64 vcc_lo, v1, v2 ; encoding: [0x6a,0x00,0x10,0xd4,0x01,0x05,0x02,0x00]
v_cmp_f_f32 vcc_lo, v1, v2
-// GFX11: encoding: [0x01,0x05,0x20,0x7c]
+// GFX11: v_cmp_f_f32_e32 vcc_lo, v1, v2 ; encoding: [0x01,0x05,0x20,0x7c]
//===----------------------------------------------------------------------===//
// VOPCX -> VOP3.
//===----------------------------------------------------------------------===//
v_cmpx_f_f32 -v1, v2
-// GFX11: encoding: [0x7e,0x00,0x90,0xd4,0x01,0x05,0x02,0x20]
+// GFX11: v_cmpx_f_f32_e64 -v1, v2 ; encoding: [0x7e,0x00,0x90,0xd4,0x01,0x05,0x02,0x20]
v_cmpx_f_f32_e64 -v1, v2
-// GFX11: encoding: [0x7e,0x00,0x90,0xd4,0x01,0x05,0x02,0x20]
+// GFX11: v_cmpx_f_f32_e64 -v1, v2 ; encoding: [0x7e,0x00,0x90,0xd4,0x01,0x05,0x02,0x20]
v_cmpx_f_f32_e64 v1, v2
-// GFX11: encoding: [0x7e,0x00,0x90,0xd4,0x01,0x05,0x02,0x00]
+// GFX11: v_cmpx_f_f32_e64 v1, v2 ; encoding: [0x7e,0x00,0x90,0xd4,0x01,0x05,0x02,0x00]
v_cmpx_f_f32 v1, v2
-// GFX11: encoding: [0x01,0x05,0x20,0x7d]
+// GFX11: v_cmpx_f_f32_e32 v1, v2 ; encoding: [0x01,0x05,0x20,0x7d]
//===----------------------------------------------------------------------===//
// VOP3.
//===----------------------------------------------------------------------===//
v_add3_u32 v5, v1, v2, s3
-// GFX11: encoding: [0x05,0x00,0x55,0xd6,0x01,0x05,0x0e,0x00]
+// GFX11: v_add3_u32 v5, v1, v2, s3 ; encoding: [0x05,0x00,0x55,0xd6,0x01,0x05,0x0e,0x00]
v_add3_u32_e64 v5, v1, v2, s3
-// GFX11: encoding: [0x05,0x00,0x55,0xd6,0x01,0x05,0x0e,0x00]
+// GFX11: v_add3_u32 v5, v1, v2, s3 ; encoding: [0x05,0x00,0x55,0xd6,0x01,0x05,0x0e,0x00]
//===----------------------------------------------------------------------===//
// VOP1 -> VOP3.DPP8.
//===----------------------------------------------------------------------===//
v_sin_f32 v5, v1 div:2 dpp8:[0,0,0,0,0,0,0,0]
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0xe9,0x00,0x00,0x18,0x01,0x00,0x00,0x00]
+// GFX11: v_sin_f32_e64_dpp v5, v1 div:2 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0x05,0x00,0xb5,0xd5,0xe9,0x00,0x00,0x18,0x01,0x00,0x00,0x00]
v_sin_f32_e64_dpp v5, v1 div:2 dpp8:[0,0,0,0,0,0,0,0]
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0xe9,0x00,0x00,0x18,0x01,0x00,0x00,0x00]
+// GFX11: v_sin_f32_e64_dpp v5, v1 div:2 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0x05,0x00,0xb5,0xd5,0xe9,0x00,0x00,0x18,0x01,0x00,0x00,0x00]
v_sin_f32_e64_dpp v5, v1 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0xe9,0x00,0x00,0x00,0x01,0x77,0x39,0x05]
+// GFX11: v_sin_f32_e64_dpp v5, v1 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0xb5,0xd5,0xe9,0x00,0x00,0x00,0x01,0x77,0x39,0x05]
v_sin_f32 v5, v1 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x6a,0x0a,0x7e,0x01,0x77,0x39,0x05]
+// GFX11: v_sin_f32_dpp v5, v1 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x6a,0x0a,0x7e,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOP2 -> VOP3.DPP8.
//===----------------------------------------------------------------------===//
v_add_f32 v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0xe9,0x04,0x02,0x10,0x01,0x77,0x39,0x05]
+// GFX11: v_add_f32_e64_dpp v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x03,0xd5,0xe9,0x04,0x02,0x10,0x01,0x77,0x39,0x05]
v_add_f32_e64_dpp v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0xe9,0x04,0x02,0x10,0x01,0x77,0x39,0x05]
+// GFX11: v_add_f32_e64_dpp v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x03,0xd5,0xe9,0x04,0x02,0x10,0x01,0x77,0x39,0x05]
v_add_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
+// GFX11: v_add_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x03,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
v_add_f32 v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x04,0x0a,0x06,0x01,0x77,0x39,0x05]
+// GFX11: v_add_f32_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0x0a,0x06,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOPC -> VOP3.DPP8.
//===----------------------------------------------------------------------===//
v_cmp_class_f32 s5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x7e,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
+// GFX11: v_cmp_class_f32_e64_dpp s5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x7e,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
v_cmp_class_f32_e64_dpp s5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x7e,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
+// GFX11: v_cmp_class_f32_e64_dpp s5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x7e,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
v_cmp_class_f32_e64_dpp vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x6a,0x00,0x7e,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
+// GFX11: v_cmp_class_f32_e64_dpp vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x6a,0x00,0x7e,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
v_cmp_class_f32 vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x04,0xfc,0x7c,0x01,0x77,0x39,0x05]
+// GFX11: v_cmp_class_f32 vcc_lo, vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0xfc,0x7c,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOPCX -> VOP3.DPP8.
//===----------------------------------------------------------------------===//
v_cmpx_class_f32 -v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x7e,0x00,0xfe,0xd4,0xe9,0x04,0x02,0x20,0x01,0x77,0x39,0x05]
+// GFX11: v_cmpx_class_f32_e64_dpp -v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x7e,0x00,0xfe,0xd4,0xe9,0x04,0x02,0x20,0x01,0x77,0x39,0x05]
v_cmpx_class_f32_e64_dpp -v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x7e,0x00,0xfe,0xd4,0xe9,0x04,0x02,0x20,0x01,0x77,0x39,0x05]
+// GFX11: v_cmpx_class_f32_e64_dpp -v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x7e,0x00,0xfe,0xd4,0xe9,0x04,0x02,0x20,0x01,0x77,0x39,0x05]
v_cmpx_class_f32_e64_dpp v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x7e,0x00,0xfe,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
+// GFX11: v_cmpx_class_f32_e64_dpp v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x7e,0x00,0xfe,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
v_cmpx_class_f32 v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0xe9,0x04,0xfc,0x7d,0x01,0x77,0x39,0x05]
+// GFX11: v_cmpx_class_f32 v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0xfc,0x7d,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOP1 -> VOP3.DPP16.
//===----------------------------------------------------------------------===//
v_sin_f32 v5, v1 div:2 row_xmask:15
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0xfa,0x00,0x00,0x18,0x01,0x6f,0x01,0xff]
+// GFX11: v_sin_f32_e64_dpp v5, v1 div:2 row_xmask:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0xb5,0xd5,0xfa,0x00,0x00,0x18,0x01,0x6f,0x01,0xff]
v_sin_f32_e64_dpp v5, v1 div:2 row_xmask:15
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0xfa,0x00,0x00,0x18,0x01,0x6f,0x01,0xff]
+// GFX11: v_sin_f32_e64_dpp v5, v1 div:2 row_xmask:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0xb5,0xd5,0xfa,0x00,0x00,0x18,0x01,0x6f,0x01,0xff]
v_sin_f32_e64_dpp v5, v1 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0xb5,0xd5,0xfa,0x00,0x00,0x00,0x01,0x1b,0x00,0xff]
+// GFX11: v_sin_f32_e64_dpp v5, v1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0xb5,0xd5,0xfa,0x00,0x00,0x00,0x01,0x1b,0x00,0xff]
v_sin_f32 v5, v1 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x6a,0x0a,0x7e,0x01,0x1b,0x00,0xff]
+// GFX11: v_sin_f32_dpp v5, v1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x6a,0x0a,0x7e,0x01,0x1b,0x00,0xff]
//===----------------------------------------------------------------------===//
// VOP2 -> VOP3.DPP16.
//===----------------------------------------------------------------------===//
v_add_f32 v5, v1, v2 div:2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0xfa,0x04,0x02,0x18,0x01,0x1b,0x00,0xff]
+// GFX11: v_add_f32_e64_dpp v5, v1, v2 div:2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x03,0xd5,0xfa,0x04,0x02,0x18,0x01,0x1b,0x00,0xff]
v_add_f32_e64_dpp v5, v1, v2 div:2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0xfa,0x04,0x02,0x18,0x01,0x1b,0x00,0xff]
+// GFX11: v_add_f32_e64_dpp v5, v1, v2 div:2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x03,0xd5,0xfa,0x04,0x02,0x18,0x01,0x1b,0x00,0xff]
v_add_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x03,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
+// GFX11: v_add_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x03,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
v_add_f32 v5, v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x04,0x0a,0x06,0x01,0x1b,0x00,0xff]
+// GFX11: v_add_f32_dpp v5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0x0a,0x06,0x01,0x1b,0x00,0xff]
//===----------------------------------------------------------------------===//
// VOPC -> VOP3.DPP16.
//===----------------------------------------------------------------------===//
v_cmp_class_f32 s5, v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x7e,0xd4,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
+// GFX11: v_cmp_class_f32_e64_dpp s5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x7e,0xd4,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
v_cmp_class_f32_e64_dpp s5, v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0x05,0x00,0x7e,0xd4,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
+// GFX11: v_cmp_class_f32_e64_dpp s5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x7e,0xd4,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
v_cmp_class_f32_e64_dpp vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf
-// GFX11: encoding: [0x6a,0x00,0x7e,0xd4,0xfa,0x04,0x02,0x00,0x01,0x50,0x01,0xff]
+// GFX11: v_cmp_class_f32_e64_dpp vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0x6a,0x00,0x7e,0xd4,0xfa,0x04,0x02,0x00,0x01,0x50,0x01,0xff]
v_cmp_class_f32 vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf
-// GFX11: encoding: [0xfa,0x04,0xfc,0x7c,0x01,0x50,0x01,0xff]
+// GFX11: v_cmp_class_f32 vcc_lo, vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0xfc,0x7c,0x01,0x50,0x01,0xff]
//===----------------------------------------------------------------------===//
// VOPCX -> VOP3.DPP16.
//===----------------------------------------------------------------------===//
v_cmpx_class_f32_e64_dpp v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0x7e,0x00,0xfe,0xd4,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
+// GFX11: v_cmpx_class_f32_e64_dpp v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x7e,0x00,0xfe,0xd4,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
v_cmpx_class_f32 v1, v2 quad_perm:[3,2,1,0]
-// GFX11: encoding: [0xfa,0x04,0xfc,0x7d,0x01,0x1b,0x00,0xff]
+// GFX11: v_cmpx_class_f32 v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0xfc,0x7d,0x01,0x1b,0x00,0xff]
//===----------------------------------------------------------------------===//
// VOP3P.
//===----------------------------------------------------------------------===//
v_dot2_f32_f16 v0, v1, v2, v3
-// GFX11: encoding: [0x00,0x40,0x13,0xcc,0x01,0x05,0x0e,0x1c]
+// GFX11: v_dot2_f32_f16 v0, v1, v2, v3 ; encoding: [0x00,0x40,0x13,0xcc,0x01,0x05,0x0e,0x1c]
v_dot2_f32_f16_e64 v0, v1, v2, v3
-// GFX11: encoding: [0x00,0x40,0x13,0xcc,0x01,0x05,0x0e,0x1c]
+// GFX11: v_dot2_f32_f16 v0, v1, v2, v3 ; encoding: [0x00,0x40,0x13,0xcc,0x01,0x05,0x0e,0x1c]
//===----------------------------------------------------------------------===//
// VOP3P.DPP8.
//===----------------------------------------------------------------------===//
v_dot2_f32_f16 v0, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x00,0x00,0x13,0xcc,0xe9,0x04,0x0e,0x04,0x01,0x77,0x39,0x05]
+// GFX11: v_dot2_f32_f16_e64_dpp v0, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x00,0x00,0x13,0xcc,0xe9,0x04,0x0e,0x04,0x01,0x77,0x39,0x05]
v_dot2_f32_f16_e64_dpp v0, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: encoding: [0x00,0x00,0x13,0xcc,0xe9,0x04,0x0e,0x04,0x01,0x77,0x39,0x05]
+// GFX11: v_dot2_f32_f16_e64_dpp v0, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x00,0x00,0x13,0xcc,0xe9,0x04,0x0e,0x04,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOP3P.DPP16.
//===----------------------------------------------------------------------===//
v_dot2_f32_f16 v0, v1, v2, v3 quad_perm:[1,2,3,0]
-// GFX11: encoding: [0x00,0x00,0x13,0xcc,0xfa,0x04,0x0e,0x04,0x01,0x39,0x00,0xff]
+// GFX11: v_dot2_f32_f16_e64_dpp v0, v1, v2, v3 quad_perm:[1,2,3,0] row_mask:0xf bank_mask:0xf ; encoding: [0x00,0x00,0x13,0xcc,0xfa,0x04,0x0e,0x04,0x01,0x39,0x00,0xff]
v_dot2_f32_f16_e64_dpp v0, v1, v2, v3 quad_perm:[1,2,3,0]
-// GFX11: encoding: [0x00,0x00,0x13,0xcc,0xfa,0x04,0x0e,0x04,0x01,0x39,0x00,0xff]
+// GFX11: v_dot2_f32_f16_e64_dpp v0, v1, v2, v3 quad_perm:[1,2,3,0] row_mask:0xf bank_mask:0xf ; encoding: [0x00,0x00,0x13,0xcc,0xfa,0x04,0x0e,0x04,0x01,0x39,0x00,0xff]
>From 3cc965006e751e20181ad0b2f4208af03a7a49d7 Mon Sep 17 00:00:00 2001
From: Mirko Brkusanin <Mirko.Brkusanin at amd.com>
Date: Thu, 8 Feb 2024 17:12:35 +0100
Subject: [PATCH 2/2] Fix printing vcc(_lo) twice for VOPC DPP instrucitons
---
.../AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp | 3 ++-
llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp | 16 ++++++++++++++++
llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h | 3 +++
llvm/lib/Target/AMDGPU/VOPCInstructions.td | 4 ++++
llvm/lib/Target/AMDGPU/VOPInstructions.td | 13 +++++++++++++
llvm/test/MC/AMDGPU/gfx11-promotions.s | 4 ++--
6 files changed, 40 insertions(+), 3 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp
index abfa4a3531e8e..4ab3aa5a0240a 100644
--- a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp
+++ b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp
@@ -709,6 +709,7 @@ bool AMDGPUInstPrinter::needsImpliedVcc(const MCInstrDesc &Desc,
unsigned OpNo) const {
return OpNo == 0 && (Desc.TSFlags & SIInstrFlags::DPP) &&
(Desc.TSFlags & SIInstrFlags::VOPC) &&
+ !isVOPCAsmOnly(Desc.getOpcode()) &&
(Desc.hasImplicitDefOfPhysReg(AMDGPU::VCC) ||
Desc.hasImplicitDefOfPhysReg(AMDGPU::VCC_LO));
}
@@ -725,7 +726,7 @@ void AMDGPUInstPrinter::printOperand(const MCInst *MI, unsigned OpNo,
// printOperandAndIntInputMods will be called instead
if ((OpNo == 0 ||
(OpNo == 1 && (Desc.TSFlags & SIInstrFlags::DPP) && ModIdx != -1)) &&
- (Desc.TSFlags & SIInstrFlags::VOPC) &&
+ (Desc.TSFlags & SIInstrFlags::VOPC) && !isVOPCAsmOnly(Desc.getOpcode()) &&
(Desc.hasImplicitDefOfPhysReg(AMDGPU::VCC) ||
Desc.hasImplicitDefOfPhysReg(AMDGPU::VCC_LO)))
printDefaultVccOperand(true, STI, O);
diff --git a/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp b/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
index 33335ac75df76..800dfcf3076dd 100644
--- a/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
@@ -342,6 +342,14 @@ struct VOPC64DPPInfo {
uint16_t Opcode;
};
+struct VOPCDPPAsmOnlyInfo {
+ uint16_t Opcode;
+};
+
+struct VOP3CDPPAsmOnlyInfo {
+ uint16_t Opcode;
+};
+
struct VOPDComponentInfo {
uint16_t BaseVOP;
uint16_t VOPDOp;
@@ -376,6 +384,10 @@ struct VOPTrue16Info {
#define GET_VOPC64DPPTable_IMPL
#define GET_VOPC64DPP8Table_DECL
#define GET_VOPC64DPP8Table_IMPL
+#define GET_VOPCAsmOnlyInfoTable_DECL
+#define GET_VOPCAsmOnlyInfoTable_IMPL
+#define GET_VOP3CAsmOnlyInfoTable_DECL
+#define GET_VOP3CAsmOnlyInfoTable_IMPL
#define GET_VOPDComponentTable_DECL
#define GET_VOPDComponentTable_IMPL
#define GET_VOPDPairs_DECL
@@ -477,6 +489,10 @@ bool isVOPC64DPP(unsigned Opc) {
return isVOPC64DPPOpcodeHelper(Opc) || isVOPC64DPP8OpcodeHelper(Opc);
}
+bool isVOPCAsmOnly(unsigned Opc) {
+ return isVOPCAsmOnlyOpcodeHelper(Opc) || isVOP3CAsmOnlyOpcodeHelper(Opc);
+}
+
bool getMAIIsDGEMM(unsigned Opc) {
const MAIInstInfo *Info = getMAIInstInfoHelper(Opc);
return Info ? Info->is_dgemm : false;
diff --git a/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h b/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h
index f24b9f0e3615d..f9eeede0fd8a1 100644
--- a/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h
+++ b/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h
@@ -489,6 +489,9 @@ bool getVOP3IsSingle(unsigned Opc);
LLVM_READONLY
bool isVOPC64DPP(unsigned Opc);
+LLVM_READONLY
+bool isVOPCAsmOnly(unsigned Opc);
+
/// Returns true if MAI operation is a double precision GEMM.
LLVM_READONLY
bool getMAIIsDGEMM(unsigned Opc);
diff --git a/llvm/lib/Target/AMDGPU/VOPCInstructions.td b/llvm/lib/Target/AMDGPU/VOPCInstructions.td
index 861c9f75e866d..fe52a0e39e4f1 100644
--- a/llvm/lib/Target/AMDGPU/VOPCInstructions.td
+++ b/llvm/lib/Target/AMDGPU/VOPCInstructions.td
@@ -1165,6 +1165,8 @@ class VOPC_DPPe_Common<bits<8> op> : Enc64 {
class VOPC_DPP_Base<bits<8> op, string OpName, VOPProfile P>
: VOP_DPP_Base<OpName, P, P.InsDPP16, " " #P.AsmDPP16>,
VOPC_DPPe_Common<op> {
+ Instruction Opcode = !cast<Instruction>(NAME);
+
bits<2> src0_modifiers;
bits<8> src0;
bits<2> src1_modifiers;
@@ -1194,6 +1196,8 @@ class VOPC_DPP_Base<bits<8> op, string OpName, VOPProfile P>
class VOPC_DPP8_Base<bits<8> op, string OpName, VOPProfile P>
: VOP_DPP8_Base<OpName, P, P.InsDPP8, " " #P.AsmDPP8>,
VOPC_DPPe_Common<op> {
+ Instruction Opcode = !cast<Instruction>(NAME);
+
bits<8> src0;
bits<24> dpp8;
bits<9> fi;
diff --git a/llvm/lib/Target/AMDGPU/VOPInstructions.td b/llvm/lib/Target/AMDGPU/VOPInstructions.td
index 20d7c88fb7e59..e8f85cb7d5f79 100644
--- a/llvm/lib/Target/AMDGPU/VOPInstructions.td
+++ b/llvm/lib/Target/AMDGPU/VOPInstructions.td
@@ -1669,6 +1669,19 @@ class VOPC64Table <string Format> : GenericTable {
def VOPC64DPPTable : VOPC64Table<"DPP">;
def VOPC64DPP8Table : VOPC64Table<"DPP8">;
+class AsmOnlyInfoTable <string Format, string Class>: GenericTable {
+ let FilterClass = Class;
+ let FilterClassField = "isAsmParserOnly";
+ let CppTypeName = Format # "DPPAsmOnlyInfo";
+ let Fields = ["Opcode"];
+
+ let PrimaryKey = ["Opcode"];
+ let PrimaryKeyName = "is" # Format # "AsmOnlyOpcodeHelper";
+}
+
+def VOPCAsmOnlyInfoTable : AsmOnlyInfoTable <"VOPC", "VOPC_DPPe_Common">;
+def VOP3CAsmOnlyInfoTable : AsmOnlyInfoTable <"VOP3C", "VOP3_DPPe_Common_Base">;
+
def VOPTrue16Table : GenericTable {
let FilterClass = "VOP_Pseudo";
let CppTypeName = "VOPTrue16Info";
diff --git a/llvm/test/MC/AMDGPU/gfx11-promotions.s b/llvm/test/MC/AMDGPU/gfx11-promotions.s
index bf85ba9f8c703..0bd90266457ee 100644
--- a/llvm/test/MC/AMDGPU/gfx11-promotions.s
+++ b/llvm/test/MC/AMDGPU/gfx11-promotions.s
@@ -246,7 +246,7 @@ v_cmp_class_f32_e64_dpp vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
// GFX11: v_cmp_class_f32_e64_dpp vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x6a,0x00,0x7e,0xd4,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
v_cmp_class_f32 vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
-// GFX11: v_cmp_class_f32 vcc_lo, vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0xfc,0x7c,0x01,0x77,0x39,0x05]
+// GFX11: v_cmp_class_f32 vcc_lo, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0xfc,0x7c,0x01,0x77,0x39,0x05]
//===----------------------------------------------------------------------===//
// VOPCX -> VOP3.DPP8.
@@ -310,7 +310,7 @@ v_cmp_class_f32_e64_dpp vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf
// GFX11: v_cmp_class_f32_e64_dpp vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0x6a,0x00,0x7e,0xd4,0xfa,0x04,0x02,0x00,0x01,0x50,0x01,0xff]
v_cmp_class_f32 vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf
-// GFX11: v_cmp_class_f32 vcc_lo, vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0xfc,0x7c,0x01,0x50,0x01,0xff]
+// GFX11: v_cmp_class_f32 vcc_lo, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x04,0xfc,0x7c,0x01,0x50,0x01,0xff]
//===----------------------------------------------------------------------===//
// VOPCX -> VOP3.DPP16.
More information about the llvm-commits
mailing list