File: inlineasm-packed.ll

package info (click to toggle)
llvm-toolchain-17 1%3A17.0.6-22
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 1,799,624 kB
  • sloc: cpp: 6,428,607; ansic: 1,383,196; asm: 793,408; python: 223,504; objc: 75,364; f90: 60,502; lisp: 33,869; pascal: 15,282; sh: 9,684; perl: 7,453; ml: 4,937; awk: 3,523; makefile: 2,889; javascript: 2,149; xml: 888; fortran: 619; cs: 573
file content (57 lines) | stat: -rw-r--r-- 2,109 bytes parent folder | download | duplicates (4)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s

; GCN-LABEL: {{^}}inline_asm_input_v2i16:
; GCN: s_mov_b32 s{{[0-9]+}}, s{{[0-9]+}}
define amdgpu_kernel void @inline_asm_input_v2i16(ptr addrspace(1) %out, <2 x i16> %in) #0 {
entry:
  %val = call i32 asm "s_mov_b32 $0, $1", "=r,r"(<2 x i16> %in) #0
  store i32 %val, ptr addrspace(1) %out
  ret void
}

; GCN-LABEL: {{^}}inline_asm_input_v2f16:
; GCN: s_mov_b32 s0, s{{[0-9]+}}
define amdgpu_kernel void @inline_asm_input_v2f16(ptr addrspace(1) %out, <2 x half> %in) #0 {
entry:
  %val = call i32 asm "s_mov_b32 $0, $1", "=r,r"(<2 x half> %in) #0
  store i32 %val, ptr addrspace(1) %out
  ret void
}

; GCN-LABEL: {{^}}inline_asm_output_v2i16:
; GCN: s_mov_b32 s{{[0-9]+}}, s{{[0-9]+}}
define amdgpu_kernel void @inline_asm_output_v2i16(ptr addrspace(1) %out, i32 %in) #0 {
entry:
  %val = call <2 x i16> asm "s_mov_b32 $0, $1", "=r,r"(i32 %in) #0
  store <2 x i16> %val, ptr addrspace(1) %out
  ret void
}

; GCN-LABEL: {{^}}inline_asm_output_v2f16:
; GCN: v_mov_b32 v{{[0-9]+}}, s{{[0-9]+}}
define amdgpu_kernel void @inline_asm_output_v2f16(ptr addrspace(1) %out, i32 %in) #0 {
entry:
  %val = call <2 x half> asm "v_mov_b32 $0, $1", "=v,r"(i32 %in) #0
  store <2 x half> %val, ptr addrspace(1) %out
  ret void
}

; GCN-LABEL: {{^}}inline_asm_packed_v2i16:
; GCN: v_pk_add_u16 v{{[0-9]+}}, s{{[0-9]+}}, v{{[0-9]+}}
define amdgpu_kernel void @inline_asm_packed_v2i16(ptr addrspace(1) %out, <2 x i16> %in0, <2 x i16> %in1) #0 {
entry:
  %val = call <2 x i16> asm "v_pk_add_u16 $0, $1, $2", "=v,r,v"(<2 x i16> %in0, <2 x i16> %in1) #0
  store <2 x i16> %val, ptr addrspace(1) %out
  ret void
}

; GCN-LABEL: {{^}}inline_asm_packed_v2f16:
; GCN: v_pk_add_f16 v{{[0-9]+}}, s{{[0-9]+}}, v{{[0-9]+}}
define amdgpu_kernel void @inline_asm_packed_v2f16(ptr addrspace(1) %out, <2 x half> %in0, <2 x half> %in1) #0 {
entry:
  %val = call <2 x half> asm "v_pk_add_f16 $0, $1, $2", "=v,r,v"(<2 x half> %in0, <2 x half> %in1) #0
  store <2 x half> %val, ptr addrspace(1) %out
  ret void
}

attributes #0 = { nounwind }