mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-30 15:32:52 +01:00
c7495a0fca
Add a mapping from register-based <INSN>R instructions to the corresponding memory-based <INSN>. Use it to cut down on the number of spill loads. Some instructions extend their operands from smaller fields, so this required a new TSFlags field to say how big the unextended operand is. This optimisation doesn't trigger for C(G)R and CL(G)R because in practice we always combine those instructions with a branch. Adding a test for every other case probably seems excessive, but it did catch a missed optimisation for DSGF (fixed in r185435). llvm-svn: 185529
265 lines
6.7 KiB
LLVM
265 lines
6.7 KiB
LLVM
; Test high-part i64->i128 multiplications.
|
|
;
|
|
; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck %s
|
|
|
|
declare i64 @foo()
|
|
|
|
; Check zero-extended multiplication in which only the high part is used.
|
|
define i64 @f1(i64 %dummy, i64 %a, i64 %b) {
|
|
; CHECK: f1:
|
|
; CHECK-NOT: {{%r[234]}}
|
|
; CHECK: mlgr %r2, %r4
|
|
; CHECK: 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: f2:
|
|
; CHECK: mlgr
|
|
; CHECK: agr
|
|
; CHECK: agr
|
|
; CHECK: 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: f3:
|
|
; CHECK-NOT: {{%r[234]}}
|
|
; CHECK: mlgr %r2, %r4
|
|
; CHECK: srlg %r2, %r2, 3
|
|
; CHECK: 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: f4:
|
|
; CHECK-NOT: {{%r[234]}}
|
|
; CHECK: mlgr %r2, %r4
|
|
; CHECK: ogr %r2, %r3
|
|
; CHECK: 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: f5:
|
|
; CHECK: mlgr %r2,
|
|
; CHECK: srlg %r2, %r2,
|
|
; CHECK: br %r14
|
|
%res = udiv i64 %a, 1234
|
|
ret i64 %res
|
|
}
|
|
|
|
; Check MLG with no displacement.
|
|
define i64 @f6(i64 %dummy, i64 %a, i64 *%src) {
|
|
; CHECK: f6:
|
|
; CHECK-NOT: {{%r[234]}}
|
|
; CHECK: mlg %r2, 0(%r4)
|
|
; CHECK: br %r14
|
|
%b = load i64 *%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, i64 *%src) {
|
|
; CHECK: f7:
|
|
; CHECK: mlg %r2, 524280(%r4)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i64 *%src, i64 65535
|
|
%b = load i64 *%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, i64 *%src) {
|
|
; CHECK: f8:
|
|
; CHECK: agfi %r4, 524288
|
|
; CHECK: mlg %r2, 0(%r4)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i64 *%src, i64 65536
|
|
%b = load i64 *%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, i64 *%src) {
|
|
; CHECK: f9:
|
|
; CHECK: mlg %r2, -8(%r4)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i64 *%src, i64 -1
|
|
%b = load i64 *%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, i64 *%src) {
|
|
; CHECK: f10:
|
|
; CHECK: mlg %r2, -524288(%r4)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i64 *%src, i64 -65536
|
|
%b = load i64 *%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(i64 *%dest, i64 %a, i64 *%src) {
|
|
; CHECK: f11:
|
|
; CHECK: agfi %r4, -524296
|
|
; CHECK: mlg %r2, 0(%r4)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i64 *%src, i64 -65537
|
|
%b = load i64 *%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(i64 *%dest, i64 %a, i64 %src, i64 %index) {
|
|
; CHECK: f12:
|
|
; CHECK: mlg %r2, 524287(%r5,%r4)
|
|
; CHECK: br %r14
|
|
%add1 = add i64 %src, %index
|
|
%add2 = add i64 %add1, 524287
|
|
%ptr = inttoptr i64 %add2 to i64 *
|
|
%b = load i64 *%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 multiplications of spilled values can use MLG rather than MLGR.
|
|
define i64 @f13(i64 *%ptr0) {
|
|
; CHECK: f13:
|
|
; CHECK: brasl %r14, foo@PLT
|
|
; CHECK: mlg {{%r[0-9]+}}, 160(%r15)
|
|
; CHECK: br %r14
|
|
%ptr1 = getelementptr i64 *%ptr0, i64 2
|
|
%ptr2 = getelementptr i64 *%ptr0, i64 4
|
|
%ptr3 = getelementptr i64 *%ptr0, i64 6
|
|
%ptr4 = getelementptr i64 *%ptr0, i64 8
|
|
%ptr5 = getelementptr i64 *%ptr0, i64 10
|
|
%ptr6 = getelementptr i64 *%ptr0, i64 12
|
|
%ptr7 = getelementptr i64 *%ptr0, i64 14
|
|
%ptr8 = getelementptr i64 *%ptr0, i64 16
|
|
%ptr9 = getelementptr i64 *%ptr0, i64 18
|
|
|
|
%val0 = load i64 *%ptr0
|
|
%val1 = load i64 *%ptr1
|
|
%val2 = load i64 *%ptr2
|
|
%val3 = load i64 *%ptr3
|
|
%val4 = load i64 *%ptr4
|
|
%val5 = load i64 *%ptr5
|
|
%val6 = load i64 *%ptr6
|
|
%val7 = load i64 *%ptr7
|
|
%val8 = load i64 *%ptr8
|
|
%val9 = load i64 *%ptr9
|
|
|
|
%ret = call i64 @foo()
|
|
|
|
%retx = zext i64 %ret to i128
|
|
%val0x = zext i64 %val0 to i128
|
|
%mul0d = mul i128 %retx, %val0x
|
|
%mul0x = lshr i128 %mul0d, 64
|
|
|
|
%val1x = zext i64 %val1 to i128
|
|
%mul1d = mul i128 %mul0x, %val1x
|
|
%mul1x = lshr i128 %mul1d, 64
|
|
|
|
%val2x = zext i64 %val2 to i128
|
|
%mul2d = mul i128 %mul1x, %val2x
|
|
%mul2x = lshr i128 %mul2d, 64
|
|
|
|
%val3x = zext i64 %val3 to i128
|
|
%mul3d = mul i128 %mul2x, %val3x
|
|
%mul3x = lshr i128 %mul3d, 64
|
|
|
|
%val4x = zext i64 %val4 to i128
|
|
%mul4d = mul i128 %mul3x, %val4x
|
|
%mul4x = lshr i128 %mul4d, 64
|
|
|
|
%val5x = zext i64 %val5 to i128
|
|
%mul5d = mul i128 %mul4x, %val5x
|
|
%mul5x = lshr i128 %mul5d, 64
|
|
|
|
%val6x = zext i64 %val6 to i128
|
|
%mul6d = mul i128 %mul5x, %val6x
|
|
%mul6x = lshr i128 %mul6d, 64
|
|
|
|
%val7x = zext i64 %val7 to i128
|
|
%mul7d = mul i128 %mul6x, %val7x
|
|
%mul7x = lshr i128 %mul7d, 64
|
|
|
|
%val8x = zext i64 %val8 to i128
|
|
%mul8d = mul i128 %mul7x, %val8x
|
|
%mul8x = lshr i128 %mul8d, 64
|
|
|
|
%val9x = zext i64 %val9 to i128
|
|
%mul9d = mul i128 %mul8x, %val9x
|
|
%mul9x = lshr i128 %mul9d, 64
|
|
|
|
%mul9 = trunc i128 %mul9x to i64
|
|
ret i64 %mul9
|
|
}
|