mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-18 18:42:46 +02:00
[AMDGPU][MC][GFX8+][DISASSEMBLER] Corrected 1/2pi value for 64-bit operands
See bug 39332: https://bugs.llvm.org/show_bug.cgi?id=39332 Reviewers: artem.tamazov, arsenm Differential Revision: https://reviews.llvm.org/D56794 llvm-svn: 351555
This commit is contained in:
parent
4b6e08d50e
commit
97d150850a
@ -523,22 +523,22 @@ Floating-point *inline constants* are converted to
|
||||
:ref:`expected operand type<amdgpu_syn_instruction_type>`
|
||||
as described :ref:`here<amdgpu_synid_fp_const_conv>`.
|
||||
|
||||
================================== ===================================================== ==================
|
||||
Value Note Availability
|
||||
================================== ===================================================== ==================
|
||||
0.0 The same as integer constant 0. All GPUs
|
||||
0.5 Floating-point constant 0.5 All GPUs
|
||||
1.0 Floating-point constant 1.0 All GPUs
|
||||
2.0 Floating-point constant 2.0 All GPUs
|
||||
4.0 Floating-point constant 4.0 All GPUs
|
||||
-0.5 Floating-point constant -0.5 All GPUs
|
||||
-1.0 Floating-point constant -1.0 All GPUs
|
||||
-2.0 Floating-point constant -2.0 All GPUs
|
||||
-4.0 Floating-point constant -4.0 All GPUs
|
||||
0.1592 1.0/(2.0*pi). Use only for 16-bit operands. GFX8, GFX9
|
||||
0.15915494 1.0/(2.0*pi). Use only for 16- and 32-bit operands. GFX8, GFX9
|
||||
0.159154943091895317852646485335 1.0/(2.0*pi). GFX8, GFX9
|
||||
================================== ===================================================== ==================
|
||||
===================== ===================================================== ==================
|
||||
Value Note Availability
|
||||
===================== ===================================================== ==================
|
||||
0.0 The same as integer constant 0. All GPUs
|
||||
0.5 Floating-point constant 0.5 All GPUs
|
||||
1.0 Floating-point constant 1.0 All GPUs
|
||||
2.0 Floating-point constant 2.0 All GPUs
|
||||
4.0 Floating-point constant 4.0 All GPUs
|
||||
-0.5 Floating-point constant -0.5 All GPUs
|
||||
-1.0 Floating-point constant -1.0 All GPUs
|
||||
-2.0 Floating-point constant -2.0 All GPUs
|
||||
-4.0 Floating-point constant -4.0 All GPUs
|
||||
0.1592 1.0/(2.0*pi). Use only for 16-bit operands. GFX8, GFX9
|
||||
0.15915494 1.0/(2.0*pi). Use only for 16- and 32-bit operands. GFX8, GFX9
|
||||
0.15915494309189532 1.0/(2.0*pi). GFX8, GFX9
|
||||
===================== ===================================================== ==================
|
||||
|
||||
.. WARNING:: GFX7 does not support inline constants for *f16* operands.
|
||||
|
||||
|
@ -491,7 +491,7 @@ void AMDGPUInstPrinter::printImmediate64(uint64_t Imm,
|
||||
O << "-4.0";
|
||||
else if (Imm == 0x3fc45f306dc9c882 &&
|
||||
STI.getFeatureBits()[AMDGPU::FeatureInv2PiInlineImm])
|
||||
O << "0.15915494";
|
||||
O << "0.15915494309189532";
|
||||
else {
|
||||
assert(isUInt<32>(Imm) || Imm == 0x3fc45f306dc9c882);
|
||||
|
||||
|
@ -449,7 +449,7 @@ define amdgpu_kernel void @add_inline_imm_neg_4.0_f64(double addrspace(1)* %out,
|
||||
; SI: v_add_f64 v{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}}, v{{\[}}[[LO_VREG]]:[[HI_VREG]]{{\]}}
|
||||
|
||||
; VI: s_load_dwordx2 [[VAL:s\[[0-9]+:[0-9]+\]]], {{s\[[0-9]+:[0-9]+\]}}, 0x4c
|
||||
; VI: v_add_f64 [[REG:v\[[0-9]+:[0-9]+\]]], [[VAL]], 0.15915494{{$}}
|
||||
; VI: v_add_f64 [[REG:v\[[0-9]+:[0-9]+\]]], [[VAL]], 0.15915494309189532{{$}}
|
||||
; VI: buffer_store_dwordx2 [[REG]]
|
||||
define amdgpu_kernel void @add_inline_imm_inv_2pi_f64(double addrspace(1)* %out, [8 x i32], double %x) {
|
||||
%y = fadd double %x, 0x3fc45f306dc9c882
|
||||
|
@ -429,7 +429,7 @@ v_and_b32_e32 v0, 0xffffffffffffffff, v1
|
||||
v_trunc_f32_e32 v0, 0x3fc45f306dc9c882
|
||||
|
||||
// NOSICI: error: invalid operand for instruction
|
||||
// VI: v_fract_f64_e32 v[0:1], 0.15915494 ; encoding: [0xf8,0x64,0x00,0x7e]
|
||||
// VI: v_fract_f64_e32 v[0:1], 0.15915494309189532 ; encoding: [0xf8,0x64,0x00,0x7e]
|
||||
v_fract_f64_e32 v[0:1], 0x3fc45f306dc9c882
|
||||
|
||||
// SICI: v_trunc_f32_e32 v0, 0x3e22f983 ; encoding: [0xff,0x42,0x00,0x7e,0x83,0xf9,0x22,0x3e]
|
||||
@ -445,7 +445,7 @@ v_fract_f64_e32 v[0:1], 0x3e22f983
|
||||
v_trunc_f32_e64 v0, 0x3fc45f306dc9c882
|
||||
|
||||
// NOSICI: error: invalid operand for instruction
|
||||
// VI: v_fract_f64_e64 v[0:1], 0.15915494 ; encoding: [0x00,0x00,0x72,0xd1,0xf8,0x00,0x00,0x00]
|
||||
// VI: v_fract_f64_e64 v[0:1], 0.15915494309189532 ; encoding: [0x00,0x00,0x72,0xd1,0xf8,0x00,0x00,0x00]
|
||||
v_fract_f64_e64 v[0:1], 0x3fc45f306dc9c882
|
||||
|
||||
// NOSICI: error: invalid operand for instruction
|
||||
@ -457,7 +457,7 @@ v_trunc_f32_e64 v0, 0x3e22f983
|
||||
v_fract_f64_e64 v[0:1], 0x3e22f983
|
||||
|
||||
// NOSICI: error: invalid operand for instruction
|
||||
// VI: s_mov_b64 s[0:1], 0.15915494 ; encoding: [0xf8,0x01,0x80,0xbe]
|
||||
// VI: s_mov_b64 s[0:1], 0.15915494309189532 ; encoding: [0xf8,0x01,0x80,0xbe]
|
||||
s_mov_b64_e32 s[0:1], 0.159154943091895317852646485335
|
||||
|
||||
// SICI: v_and_b32_e32 v0, 0x3e22f983, v1 ; encoding: [0xff,0x02,0x00,0x36,0x83,0xf9,0x22,0x3e]
|
||||
@ -469,7 +469,7 @@ v_and_b32_e32 v0, 0.159154943091895317852646485335, v1
|
||||
v_and_b32_e64 v0, 0.159154943091895317852646485335, v1
|
||||
|
||||
// SICI: v_fract_f64_e32 v[0:1], 0x3fc45f30 ; encoding: [0xff,0x7c,0x00,0x7e,0x30,0x5f,0xc4,0x3f]
|
||||
// VI: v_fract_f64_e32 v[0:1], 0.15915494 ; encoding: [0xf8,0x64,0x00,0x7e]
|
||||
// VI: v_fract_f64_e32 v[0:1], 0.15915494309189532 ; encoding: [0xf8,0x64,0x00,0x7e]
|
||||
v_fract_f64 v[0:1], 0.159154943091895317852646485335
|
||||
|
||||
// SICI: v_trunc_f32_e32 v0, 0x3e22f983 ; encoding: [0xff,0x42,0x00,0x7e,0x83,0xf9,0x22,0x3e]
|
||||
|
7
test/MC/Disassembler/AMDGPU/literal_vi.txt
Normal file
7
test/MC/Disassembler/AMDGPU/literal_vi.txt
Normal file
@ -0,0 +1,7 @@
|
||||
# RUN: llvm-mc -arch=amdgcn -mcpu=tonga -disassemble -show-encoding %s | FileCheck -check-prefix=VI %s
|
||||
|
||||
# VI: v_fract_f64_e32 v[0:1], 0.15915494309189532 ; encoding: [0xf8,0x64,0x00,0x7e]
|
||||
0xf8,0x64,0x00,0x7e
|
||||
|
||||
# VI: v_trunc_f32_e32 v0, 0.15915494 ; encoding: [0xf8,0x38,0x00,0x7e]
|
||||
0xf8,0x38,0x00,0x7e
|
Loading…
Reference in New Issue
Block a user