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 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 3
; Test high-part i64->i128 multiplications on z13.
;
; RUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=z13 | FileCheck %s
; Check zero-extended multiplication in which only the high part is used.
define i64 @f1(i64 %dummy, i64 %a, i64 %b) {
; CHECK-LABEL: f1:
; CHECK: # %bb.0:
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlgr %r2, %r4
; CHECK-NEXT: # kill: def $r2d killed $r2d killed $r2q
; CHECK-NEXT: br %r14
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check sign-extended multiplication in which only the high part is used.
; This needs a rather convoluted sequence.
define i64 @f2(i64 %dummy, i64 %a, i64 %b) {
; CHECK-LABEL: f2:
; CHECK: # %bb.0:
; CHECK-NEXT: srag %r1, %r4, 63
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: srag %r0, %r3, 63
; CHECK-NEXT: ngr %r1, %r3
; CHECK-NEXT: mlgr %r2, %r4
; CHECK-NEXT: ngr %r0, %r4
; CHECK-NEXT: agr %r0, %r1
; CHECK-NEXT: sgr %r2, %r0
; CHECK-NEXT: br %r14
%ax = sext i64 %a to i128
%bx = sext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check zero-extended multiplication in which only part of the high half
; is used.
define i64 @f3(i64 %dummy, i64 %a, i64 %b) {
; CHECK-LABEL: f3:
; CHECK: # %bb.0:
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlgr %r2, %r4
; CHECK-NEXT: srlg %r2, %r2, 3
; CHECK-NEXT: br %r14
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 67
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check zero-extended multiplication in which the result is split into
; high and low halves.
define i64 @f4(i64 %dummy, i64 %a, i64 %b) {
; CHECK-LABEL: f4:
; CHECK: # %bb.0:
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlgr %r2, %r4
; CHECK-NEXT: ogr %r2, %r3
; CHECK-NEXT: br %r14
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
%low = trunc i128 %mulx to i64
%or = or i64 %high, %low
ret i64 %or
}
; Check division by a constant, which should use multiplication instead.
define i64 @f5(i64 %dummy, i64 %a) {
; CHECK-LABEL: f5:
; CHECK: # %bb.0:
; CHECK-NEXT: llihf %r0, 1782028570
; CHECK-NEXT: oilf %r0, 598650223
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlgr %r2, %r0
; CHECK-NEXT: srlg %r2, %r2, 9
; CHECK-NEXT: br %r14
%res = udiv i64 %a, 1234
ret i64 %res
}
; Check MLG with no displacement.
define i64 @f6(i64 %dummy, i64 %a, ptr %src) {
; CHECK-LABEL: f6:
; CHECK: # %bb.0:
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlg %r2, 0(%r4)
; CHECK-NEXT: # kill: def $r2d killed $r2d killed $r2q
; CHECK-NEXT: br %r14
%b = load i64, ptr %src
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check the high end of the aligned MLG range.
define i64 @f7(i64 %dummy, i64 %a, ptr %src) {
; CHECK-LABEL: f7:
; CHECK: # %bb.0:
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlg %r2, 524280(%r4)
; CHECK-NEXT: # kill: def $r2d killed $r2d killed $r2q
; CHECK-NEXT: br %r14
%ptr = getelementptr i64, ptr %src, i64 65535
%b = load i64, ptr %ptr
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check the next doubleword up, which requires separate address logic.
; Other sequences besides this one would be OK.
define i64 @f8(i64 %dummy, i64 %a, ptr %src) {
; CHECK-LABEL: f8:
; CHECK: # %bb.0:
; CHECK-NEXT: agfi %r4, 524288
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlg %r2, 0(%r4)
; CHECK-NEXT: # kill: def $r2d killed $r2d killed $r2q
; CHECK-NEXT: br %r14
%ptr = getelementptr i64, ptr %src, i64 65536
%b = load i64, ptr %ptr
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check the high end of the negative aligned MLG range.
define i64 @f9(i64 %dummy, i64 %a, ptr %src) {
; CHECK-LABEL: f9:
; CHECK: # %bb.0:
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlg %r2, -8(%r4)
; CHECK-NEXT: # kill: def $r2d killed $r2d killed $r2q
; CHECK-NEXT: br %r14
%ptr = getelementptr i64, ptr %src, i64 -1
%b = load i64, ptr %ptr
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check the low end of the MLG range.
define i64 @f10(i64 %dummy, i64 %a, ptr %src) {
; CHECK-LABEL: f10:
; CHECK: # %bb.0:
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlg %r2, -524288(%r4)
; CHECK-NEXT: # kill: def $r2d killed $r2d killed $r2q
; CHECK-NEXT: br %r14
%ptr = getelementptr i64, ptr %src, i64 -65536
%b = load i64, ptr %ptr
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check the next doubleword down, which needs separate address logic.
; Other sequences besides this one would be OK.
define i64 @f11(ptr %dest, i64 %a, ptr %src) {
; CHECK-LABEL: f11:
; CHECK: # %bb.0:
; CHECK-NEXT: agfi %r4, -524296
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlg %r2, 0(%r4)
; CHECK-NEXT: # kill: def $r2d killed $r2d killed $r2q
; CHECK-NEXT: br %r14
%ptr = getelementptr i64, ptr %src, i64 -65537
%b = load i64, ptr %ptr
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
; Check that MLG allows an index.
define i64 @f12(ptr %dest, i64 %a, i64 %src, i64 %index) {
; CHECK-LABEL: f12:
; CHECK: # %bb.0:
; CHECK-NEXT: # kill: def $r3d killed $r3d def $r2q
; CHECK-NEXT: mlg %r2, 524287(%r5,%r4)
; CHECK-NEXT: # kill: def $r2d killed $r2d killed $r2q
; CHECK-NEXT: br %r14
%add1 = add i64 %src, %index
%add2 = add i64 %add1, 524287
%ptr = inttoptr i64 %add2 to ptr
%b = load i64, ptr %ptr
%ax = zext i64 %a to i128
%bx = zext i64 %b to i128
%mulx = mul i128 %ax, %bx
%highx = lshr i128 %mulx, 64
%high = trunc i128 %highx to i64
ret i64 %high
}
|