[Thumb] Teach ISel how to lower compares of AND bitmasks efficiently
This is essentially a recommit of r285893, but with a correctness fix. The
problem of the original commit was that this:
bic r5, r7, #31
cbz r5, .LBB2_10
got rewritten into:
lsrs r5, r7, #5
beq .LBB2_10
The result in destination register r5 is not the same and this is incorrect
when r5 is not dead. So this fix includes checking the uses of the AND
destination register. And also, compared to the original commit, some regression
tests didn't need changing anymore because of this extra check.
For completeness, this was the original commit message:
For the common pattern (CMPZ (AND x, #bitmask), #0), we can do some more
efficient instruction selection if the bitmask is one consecutive sequence of
set bits (32 - clz(bm) - ctz(bm) == popcount(bm)).
1) If the bitmask touches the LSB, then we can remove all the upper bits and
set the flags by doing one LSLS.
2) If the bitmask touches the MSB, then we can remove all the lower bits and
set the flags with one LSRS.
3) If the bitmask has popcount == 1 (only one set bit), we can shift that bit
into the sign bit with one LSLS and change the condition query from NE/EQ to
MI/PL (we could also implement this by shifting into the carry bit and
branching on BCC/BCS).
4) Otherwise, we can emit a sequence of LSLS+LSRS to remove the upper and lower
zero bits of the mask.
1-3 require only one 16-bit instruction and can elide the CMP. 4 requires two
16-bit instructions but can elide the CMP and doesn't require materializing a
complex immediate, so is also a win.
Differential Revision: https://reviews.llvm.org/D27761
llvm-svn: 289794
diff --git a/llvm/test/CodeGen/ARM/and-cmpz.ll b/llvm/test/CodeGen/ARM/and-cmpz.ll
new file mode 100644
index 0000000..809dc6c
--- /dev/null
+++ b/llvm/test/CodeGen/ARM/and-cmpz.ll
@@ -0,0 +1,71 @@
+; RUN: llc -mtriple=thumbv7m-linux-gnu < %s | FileCheck %s --check-prefix=CHECK --check-prefix=T2
+; RUN: llc -mtriple=thumbv6m-linux-gnu < %s | FileCheck %s --check-prefix=CHECK --check-prefix=T1
+
+; CHECK-LABEL: single_bit:
+; CHECK: lsls r0, r0, #23
+; T2-NEXT: mov
+; T2-NEXT: it
+; T1-NEXT: bmi
+define i32 @single_bit(i32 %p) {
+ %a = and i32 %p, 256
+ %b = icmp eq i32 %a, 0
+ br i1 %b, label %true, label %false
+
+true:
+ ret i32 1
+
+false:
+ ret i32 2
+}
+
+; CHECK-LABEL: multi_bit_lsb_ubfx:
+; CHECK: lsls r0, r0, #24
+; T2-NEXT: mov
+; T2-NEXT: it
+; T1-NEXT: beq
+define i32 @multi_bit_lsb_ubfx(i32 %p) {
+ %a = and i32 %p, 255
+ %b = icmp eq i32 %a, 0
+ br i1 %b, label %true, label %false
+
+true:
+ ret i32 1
+
+false:
+ ret i32 2
+}
+
+; CHECK-LABEL: multi_bit_msb:
+; CHECK: lsrs r0, r0, #24
+; T2-NEXT: mov
+; T2-NEXT: it
+; T1-NEXT: beq
+define i32 @multi_bit_msb(i32 %p) {
+ %a = and i32 %p, 4278190080 ; 0xff000000
+ %b = icmp eq i32 %a, 0
+ br i1 %b, label %true, label %false
+
+true:
+ ret i32 1
+
+false:
+ ret i32 2
+}
+
+; CHECK-LABEL: multi_bit_nosb:
+; T1: lsls r0, r0, #8
+; T1-NEXT: lsrs r0, r0, #24
+; T2: tst.w
+; T2-NEXT: it
+; T1-NEXT: beq
+define i32 @multi_bit_nosb(i32 %p) {
+ %a = and i32 %p, 16711680 ; 0x00ff0000
+ %b = icmp eq i32 %a, 0
+ br i1 %b, label %true, label %false
+
+true:
+ ret i32 1
+
+false:
+ ret i32 2
+}
diff --git a/llvm/test/CodeGen/ARM/arm-and-tst-peephole.ll b/llvm/test/CodeGen/ARM/arm-and-tst-peephole.ll
index 04eae8f..9bd2077 100644
--- a/llvm/test/CodeGen/ARM/arm-and-tst-peephole.ll
+++ b/llvm/test/CodeGen/ARM/arm-and-tst-peephole.ll
@@ -93,7 +93,7 @@
%1 = load i8, i8* %0, align 1
%2 = zext i8 %1 to i32
; ARM: ands
-; THUMB: ands
+; THUMB: ands
; T2: ands
; V8: ands
; V8-NEXT: beq
@@ -150,10 +150,9 @@
%rhs32 = zext i1 %rhs to i32
%diff = sub nsw i32 %lhs32, %rhs32
; ARM: tst r1, #1
-; THUMB: movs [[RTMP:r[0-9]+]], #1
-; THUMB: tst r1, [[RTMP]]
-; T2: tst.w r1, #1
-; V8: tst.w r1, #1
+; THUMB: lsls r1, r1, #31
+; T2: lsls r1, r1, #31
+; V8: lsls r1, r1, #31
ret i32 %diff
}
diff --git a/llvm/test/CodeGen/ARM/arm-shrink-wrapping.ll b/llvm/test/CodeGen/ARM/arm-shrink-wrapping.ll
index be6863d..9cce194 100644
--- a/llvm/test/CodeGen/ARM/arm-shrink-wrapping.ll
+++ b/llvm/test/CodeGen/ARM/arm-shrink-wrapping.ll
@@ -638,14 +638,13 @@
; during PEI with shrink-wrapping enable.
; CHECK-LABEL: debug_info:
;
-; ENABLE: tst{{(\.w)?}} r2, #1
+; ENABLE: {{tst r2, #1|lsls r1, r2, #31}}
; ENABLE-NEXT: beq [[BB13:LBB[0-9_]+]]
;
; CHECK: push
;
-; DISABLE: tst{{(\.w)?}} r2, #1
-; DISABLE-NEXT: vst1.64
-; DISABLE-NEXT: beq [[BB13:LBB[0-9_]+]]
+; DISABLE: {{tst r2, #1|lsls r1, r2, #31}}
+; DISABLE: beq [[BB13:LBB[0-9_]+]]
;
; CHECK: bl{{x?}} _pow
;
diff --git a/llvm/test/CodeGen/ARM/call-tc.ll b/llvm/test/CodeGen/ARM/call-tc.ll
index 2277a58..c5cfb9d 100644
--- a/llvm/test/CodeGen/ARM/call-tc.ll
+++ b/llvm/test/CodeGen/ARM/call-tc.ll
@@ -120,7 +120,7 @@
br i1 %tobool2, label %if.end5, label %if.then3
if.then3: ; preds = %if.end
-; CHECKT2D: bne.w _b
+; CHECKT2D: bmi.w _b
%call4 = tail call i32 @b(i32 %x) nounwind
br label %return
diff --git a/llvm/test/CodeGen/ARM/debug-info-branch-folding.ll b/llvm/test/CodeGen/ARM/debug-info-branch-folding.ll
index 7a4483e..e80e5e1 100644
--- a/llvm/test/CodeGen/ARM/debug-info-branch-folding.ll
+++ b/llvm/test/CodeGen/ARM/debug-info-branch-folding.ll
@@ -3,8 +3,7 @@
target triple = "thumbv7-apple-macosx10.6.7"
;CHECK: vadd.f32 q4, q8, q8
-;CHECK: LBB0_1:
-;CHECK-NOT: beq LBB0_1
+;CHECK-NEXT: LBB0_1
;CHECK: @DEBUG_VALUE: x <- %Q4{{$}}
;CHECK-NEXT: @DEBUG_VALUE: y <- %Q4{{$}}
diff --git a/llvm/test/CodeGen/Thumb/thumb-shrink-wrapping.ll b/llvm/test/CodeGen/Thumb/thumb-shrink-wrapping.ll
index 0fa790c..6114b72 100644
--- a/llvm/test/CodeGen/Thumb/thumb-shrink-wrapping.ll
+++ b/llvm/test/CodeGen/Thumb/thumb-shrink-wrapping.ll
@@ -650,11 +650,14 @@
; CHECK: tst r3, r4
; ENABLE-NEXT: pop {r4}
-; ENABLE-NEXT: pop {r3}
-; ENABLE-NEXT: mov lr, r3
+; ENABLE-NEXT: mov r12, r{{.*}}
+; ENABLE-NEXT: pop {r0}
+; ENABLE-NEXT: mov lr, r0
+; ENABLE-NEXT: mov r0, r12
; CHECK-NEXT: beq [[EXIT_LABEL]]
; CHECK: str r1, [r2]
+; CHECK: str r3, [r2]
; CHECK-NEXT: movs r0, #0
; CHECK-NEXT: [[EXIT_LABEL]]: @ %cleanup
; ENABLE-NEXT: bx lr
@@ -675,6 +678,7 @@
if.end4:
store i32 %head, i32* %y, align 4
+ store volatile i32 %z, i32* %y, align 4
br label %cleanup
cleanup:
diff --git a/llvm/test/CodeGen/Thumb2/float-ops.ll b/llvm/test/CodeGen/Thumb2/float-ops.ll
index c9f93f2..3c29b4d 100644
--- a/llvm/test/CodeGen/Thumb2/float-ops.ll
+++ b/llvm/test/CodeGen/Thumb2/float-ops.ll
@@ -259,9 +259,9 @@
define float @select_f(float %a, float %b, i1 %c) {
; CHECK-LABEL: select_f:
-; NONE: tst.w r2, #1
+; NONE: lsls r2, r2, #31
; NONE: moveq r0, r1
-; HARD: tst.w r0, #1
+; HARD: lsls r0, r0, #31
; VFP4-ALL: vmovne.f32 s1, s0
; VFP4-ALL: vmov.f32 s0, s1
; FP-ARMv8: vseleq.f32 s0, s1, s0
@@ -271,8 +271,8 @@
define double @select_d(double %a, double %b, i1 %c) {
; CHECK-LABEL: select_d:
-; NONE: ldr.w [[REG:r[0-9]+]], [sp]
-; NONE: ands [[REG]], [[REG]], #1
+; NONE: ldr{{(.w)?}} [[REG:r[0-9]+]], [sp]
+; NONE ands [[REG]], [[REG]], #1
; NONE: moveq r0, r2
; NONE: moveq r1, r3
; SP: ands r0, r0, #1
@@ -282,7 +282,7 @@
; SP-DAG: movne [[BLO]], [[ALO]]
; SP-DAG: movne [[BHI]], [[AHI]]
; SP: vmov d0, [[BLO]], [[BHI]]
-; DP: tst.w r0, #1
+; DP: lsls r0, r0, #31
; VFP4-DP: vmovne.f64 d1, d0
; VFP4-DP: vmov.f64 d0, d1
; FP-ARMV8: vseleq.f64 d0, d1, d0