maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/Transforms/LoopPredication/invariant_load.ll

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

440 lines
20 KiB
LLVM
Raw Normal View History

Revert "Temporarily Revert "Add basic loop fusion pass."" The reversion apparently deleted the test/Transforms directory. Will be re-reverting again. llvm-svn: 358552
2019-04-17 12:52:47 +08:00
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
[BasicAA] Rename deprecated -basicaa to -basic-aa Follow-up to D82607 Revert an accidental change (empty.ll) of D82683
2020-06-27 11:41:37 +08:00
; RUN: opt -S -basic-aa -loop-predication < %s 2>&1 | FileCheck %s
Revert "Temporarily Revert "Add basic loop fusion pass."" The reversion apparently deleted the test/Transforms directory. Will be re-reverting again. llvm-svn: 358552
2019-04-17 12:52:47 +08:00
; RUN: opt -S -aa-pipeline=basic-aa -passes='require<aa>,require<scalar-evolution>,loop(loop-predication)' < %s 2>&1 | FileCheck %s
declare void @llvm.experimental.guard(i1, ...)
@UNKNOWN = external global i1
define i32 @neg_length_variant(i32* %array, i32* %length, i32 %n) {
; CHECK-LABEL: @neg_length_variant(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[TMP5]], label [[EXIT:%.*]], label [[LOOP_PREHEADER:%.*]]
; CHECK: loop.preheader:
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[LOOP_ACC:%.*]] = phi i32 [ [[LOOP_ACC_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[I_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[UNKNOWN:%.*]] = load volatile i1, i1* @UNKNOWN
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[UNKNOWN]]) [ "deopt"() ]
; CHECK-NEXT: [[LEN:%.*]] = load i32, i32* [[LENGTH:%.*]], align 4
; CHECK-NEXT: [[WITHIN_BOUNDS:%.*]] = icmp ult i32 [[I]], [[LEN]]
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[WITHIN_BOUNDS]], i32 9) [ "deopt"() ]
; CHECK-NEXT: [[I_I64:%.*]] = zext i32 [[I]] to i64
; CHECK-NEXT: [[ARRAY_I_PTR:%.*]] = getelementptr inbounds i32, i32* [[ARRAY:%.*]], i64 [[I_I64]]
; CHECK-NEXT: [[ARRAY_I:%.*]] = load i32, i32* [[ARRAY_I_PTR]], align 4
; CHECK-NEXT: [[LOOP_ACC_NEXT]] = add i32 [[LOOP_ACC]], [[ARRAY_I]]
; CHECK-NEXT: [[I_NEXT]] = add nuw i32 [[I]], 1
; CHECK-NEXT: [[CONTINUE:%.*]] = icmp ult i32 [[I_NEXT]], [[N]]
; CHECK-NEXT: br i1 [[CONTINUE]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
; CHECK: exit.loopexit:
; CHECK-NEXT: [[LOOP_ACC_NEXT_LCSSA:%.*]] = phi i32 [ [[LOOP_ACC_NEXT]], [[LOOP]] ]
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
; CHECK-NEXT: [[RESULT:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[LOOP_ACC_NEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
; CHECK-NEXT: ret i32 [[RESULT]]
;
entry:
%tmp5 = icmp eq i32 %n, 0
br i1 %tmp5, label %exit, label %loop.preheader
loop.preheader:
br label %loop
loop:
%loop.acc = phi i32 [ %loop.acc.next, %loop ], [ 0, %loop.preheader ]
%i = phi i32 [ %i.next, %loop ], [ 0, %loop.preheader ]
%unknown = load volatile i1, i1* @UNKNOWN
call void (i1, ...) @llvm.experimental.guard(i1 %unknown) [ "deopt"() ]
%len = load i32, i32* %length, align 4
%within.bounds = icmp ult i32 %i, %len
call void (i1, ...) @llvm.experimental.guard(i1 %within.bounds, i32 9) [ "deopt"() ]
%i.i64 = zext i32 %i to i64
%array.i.ptr = getelementptr inbounds i32, i32* %array, i64 %i.i64
%array.i = load i32, i32* %array.i.ptr, align 4
%loop.acc.next = add i32 %loop.acc, %array.i
%i.next = add nuw i32 %i, 1
%continue = icmp ult i32 %i.next, %n
br i1 %continue, label %loop, label %exit
exit:
%result = phi i32 [ 0, %entry ], [ %loop.acc.next, %loop ]
ret i32 %result
}
define i32 @invariant_load_guard_limit(i32* %array, i32* %length, i32 %n) {
; CHECK-LABEL: @invariant_load_guard_limit(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[TMP5]], label [[EXIT:%.*]], label [[LOOP_PREHEADER:%.*]]
; CHECK: loop.preheader:
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[LOOP_ACC:%.*]] = phi i32 [ [[LOOP_ACC_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[I_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[UNKNOWN:%.*]] = load volatile i1, i1* @UNKNOWN
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[UNKNOWN]]) [ "deopt"() ]
; CHECK-NEXT: [[LEN:%.*]] = load i32, i32* [[LENGTH:%.*]], align 4, !invariant.load !0
[LoopPredication] Allow predication of loop invariant computations (within the loop) The purpose of this patch is to eliminate a pass ordering dependence between LoopPredication and LICM. To understand the purpose, consider the following snippet of code inside some loop 'L' with IV 'i' A = _a.length; guard (i < A) a = _a[i] B = _b.length; guard (i < B); b = _b[i]; ... Z = _z.length; guard (i < Z) z = _z[i] accum += a + b + ... + z; Today, we need LICM to hoist the length loads, LoopPredication to make the guards loop invariant, and TrivialUnswitch to eliminate the loop invariant guard to establish must execute for the next length load. Today, if we can't prove speculation safety, we'd have to iterate these three passes 26 times to reduce this example down to the minimal form. Using the fact that the array lengths are known to be invariant, we can short circuit this iteration. By forming the loop invariant form of all the guards at once, we remove the need for LoopPredication from the iterative cycle. At the moment, we'd still have to iterate LICM and TrivialUnswitch; we'll leave that part for later. As a secondary benefit, this allows LoopPred to expose peeling oppurtunities in a much more obvious manner. See the udiv test changes as an example. If the udiv was not hoistable (i.e. we couldn't prove speculation safety) this would be an example where peeling becomes obviously profitable whereas it wasn't before. A couple of subtleties in the implementation: - SCEV's isSafeToExpand guarantees speculation safety (i.e. let's us expand at a new point). It is not a precondition for expansion if we know the SCEV corresponds to a Value which dominates the requested expansion point. - SCEV's isLoopInvariant returns true for expressions which compute the same value across all iterations executed, regardless of where the original Value is located. (i.e. it can be in the loop) This implies we have a speculation burden to prove before expanding them outside loops. - invariant_loads and AA->pointsToConstantMemory are two cases that SCEV currently does not handle, but meets the SCEV definition of invariance. I plan to sink this part into SCEV once this has baked for a bit. Differential Revision: https://reviews.llvm.org/D60093 llvm-svn: 358684
2019-04-19 00:33:17 +08:00
; CHECK-NEXT: [[TMP0:%.*]] = icmp ule i32 [[N]], [[LEN]]
; CHECK-NEXT: [[TMP1:%.*]] = icmp ult i32 0, [[LEN]]
; CHECK-NEXT: [[TMP2:%.*]] = and i1 [[TMP1]], [[TMP0]]
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[TMP2]], i32 9) [ "deopt"() ]
Revert "Temporarily Revert "Add basic loop fusion pass."" The reversion apparently deleted the test/Transforms directory. Will be re-reverting again. llvm-svn: 358552
2019-04-17 12:52:47 +08:00
; CHECK-NEXT: [[I_I64:%.*]] = zext i32 [[I]] to i64
; CHECK-NEXT: [[ARRAY_I_PTR:%.*]] = getelementptr inbounds i32, i32* [[ARRAY:%.*]], i64 [[I_I64]]
; CHECK-NEXT: [[ARRAY_I:%.*]] = load i32, i32* [[ARRAY_I_PTR]], align 4
; CHECK-NEXT: [[LOOP_ACC_NEXT]] = add i32 [[LOOP_ACC]], [[ARRAY_I]]
; CHECK-NEXT: [[I_NEXT]] = add nuw i32 [[I]], 1
; CHECK-NEXT: [[CONTINUE:%.*]] = icmp ult i32 [[I_NEXT]], [[N]]
; CHECK-NEXT: br i1 [[CONTINUE]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
; CHECK: exit.loopexit:
; CHECK-NEXT: [[LOOP_ACC_NEXT_LCSSA:%.*]] = phi i32 [ [[LOOP_ACC_NEXT]], [[LOOP]] ]
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
; CHECK-NEXT: [[RESULT:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[LOOP_ACC_NEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
; CHECK-NEXT: ret i32 [[RESULT]]
;
entry:
%tmp5 = icmp eq i32 %n, 0
br i1 %tmp5, label %exit, label %loop.preheader
loop.preheader:
br label %loop
loop:
%loop.acc = phi i32 [ %loop.acc.next, %loop ], [ 0, %loop.preheader ]
%i = phi i32 [ %i.next, %loop ], [ 0, %loop.preheader ]
%unknown = load volatile i1, i1* @UNKNOWN
call void (i1, ...) @llvm.experimental.guard(i1 %unknown) [ "deopt"() ]
%len = load i32, i32* %length, align 4, !invariant.load !{}
%within.bounds = icmp ult i32 %i, %len
call void (i1, ...) @llvm.experimental.guard(i1 %within.bounds, i32 9) [ "deopt"() ]
%i.i64 = zext i32 %i to i64
%array.i.ptr = getelementptr inbounds i32, i32* %array, i64 %i.i64
%array.i = load i32, i32* %array.i.ptr, align 4
%loop.acc.next = add i32 %loop.acc, %array.i
%i.next = add nuw i32 %i, 1
%continue = icmp ult i32 %i.next, %n
br i1 %continue, label %loop, label %exit
exit:
%result = phi i32 [ 0, %entry ], [ %loop.acc.next, %loop ]
ret i32 %result
}
[LoopPred] Fix a blatantly obvious bug in r358684 The bug is that I didn't check whether the operand of the invariant_loads were themselves invariant. I don't know how this got missed in the patch and review. I even had an unreduced test case locally, and I remember handling this case, but I must have lost it in one of the rebases. Oops. llvm-svn: 358688
2019-04-19 01:01:19 +08:00
; Case where we have an invariant load, but it's not loading from a loop
; invariant location.
define i32 @neg_varying_invariant_load_op(i32* %array, i32* %lengths, i32 %n) {
; CHECK-LABEL: @neg_varying_invariant_load_op(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[TMP5]], label [[EXIT:%.*]], label [[LOOP_PREHEADER:%.*]]
; CHECK: loop.preheader:
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[LOOP_ACC:%.*]] = phi i32 [ [[LOOP_ACC_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[I_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[UNKNOWN:%.*]] = load volatile i1, i1* @UNKNOWN
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[UNKNOWN]]) [ "deopt"() ]
; CHECK-NEXT: [[LENGTH_ADDR:%.*]] = getelementptr i32, i32* [[LENGTHS:%.*]], i32 [[I]]
; CHECK-NEXT: [[LEN:%.*]] = load i32, i32* [[LENGTH_ADDR]], align 4, !invariant.load !0
; CHECK-NEXT: [[WITHIN_BOUNDS:%.*]] = icmp ult i32 [[I]], [[LEN]]
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[WITHIN_BOUNDS]], i32 9) [ "deopt"() ]
; CHECK-NEXT: [[I_I64:%.*]] = zext i32 [[I]] to i64
; CHECK-NEXT: [[ARRAY_I_PTR:%.*]] = getelementptr inbounds i32, i32* [[ARRAY:%.*]], i64 [[I_I64]]
; CHECK-NEXT: [[ARRAY_I:%.*]] = load i32, i32* [[ARRAY_I_PTR]], align 4
; CHECK-NEXT: [[LOOP_ACC_NEXT]] = add i32 [[LOOP_ACC]], [[ARRAY_I]]
; CHECK-NEXT: [[I_NEXT]] = add nuw i32 [[I]], 1
; CHECK-NEXT: [[CONTINUE:%.*]] = icmp ult i32 [[I_NEXT]], [[N]]
; CHECK-NEXT: br i1 [[CONTINUE]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
; CHECK: exit.loopexit:
; CHECK-NEXT: [[LOOP_ACC_NEXT_LCSSA:%.*]] = phi i32 [ [[LOOP_ACC_NEXT]], [[LOOP]] ]
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
; CHECK-NEXT: [[RESULT:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[LOOP_ACC_NEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
; CHECK-NEXT: ret i32 [[RESULT]]
;
entry:
%tmp5 = icmp eq i32 %n, 0
br i1 %tmp5, label %exit, label %loop.preheader
loop.preheader:
br label %loop
loop:
%loop.acc = phi i32 [ %loop.acc.next, %loop ], [ 0, %loop.preheader ]
%i = phi i32 [ %i.next, %loop ], [ 0, %loop.preheader ]
%unknown = load volatile i1, i1* @UNKNOWN
call void (i1, ...) @llvm.experimental.guard(i1 %unknown) [ "deopt"() ]
%length.addr = getelementptr i32, i32* %lengths, i32 %i
%len = load i32, i32* %length.addr, align 4, !invariant.load !{}
%within.bounds = icmp ult i32 %i, %len
call void (i1, ...) @llvm.experimental.guard(i1 %within.bounds, i32 9) [ "deopt"() ]
%i.i64 = zext i32 %i to i64
%array.i.ptr = getelementptr inbounds i32, i32* %array, i64 %i.i64
%array.i = load i32, i32* %array.i.ptr, align 4
%loop.acc.next = add i32 %loop.acc, %array.i
%i.next = add nuw i32 %i, 1
%continue = icmp ult i32 %i.next, %n
br i1 %continue, label %loop, label %exit
exit:
%result = phi i32 [ 0, %entry ], [ %loop.acc.next, %loop ]
ret i32 %result
}
Revert "Temporarily Revert "Add basic loop fusion pass."" The reversion apparently deleted the test/Transforms directory. Will be re-reverting again. llvm-svn: 358552
2019-04-17 12:52:47 +08:00
; This is a case where moving the load which provides the limit for the latch
; would be invalid, so we can't preform the tempting transform. Loading the
; latch limit may fault since we could always fail the guard.
define i32 @neg_invariant_load_latch_limit(i32* %array, i32* %length, i32 %n) {
; CHECK-LABEL: @neg_invariant_load_latch_limit(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[TMP5]], label [[EXIT:%.*]], label [[LOOP_PREHEADER:%.*]]
; CHECK: loop.preheader:
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[LOOP_ACC:%.*]] = phi i32 [ [[LOOP_ACC_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[I_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[UNKNOWN:%.*]] = load volatile i1, i1* @UNKNOWN
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[UNKNOWN]]) [ "deopt"() ]
; CHECK-NEXT: [[WITHIN_BOUNDS:%.*]] = icmp ult i32 [[I]], [[N]]
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[WITHIN_BOUNDS]], i32 9) [ "deopt"() ]
; CHECK-NEXT: [[I_I64:%.*]] = zext i32 [[I]] to i64
; CHECK-NEXT: [[ARRAY_I_PTR:%.*]] = getelementptr inbounds i32, i32* [[ARRAY:%.*]], i64 [[I_I64]]
; CHECK-NEXT: [[ARRAY_I:%.*]] = load i32, i32* [[ARRAY_I_PTR]], align 4
; CHECK-NEXT: [[LOOP_ACC_NEXT]] = add i32 [[LOOP_ACC]], [[ARRAY_I]]
; CHECK-NEXT: [[I_NEXT]] = add nuw i32 [[I]], 1
; CHECK-NEXT: [[LEN:%.*]] = load i32, i32* [[LENGTH:%.*]], align 4, !invariant.load !0
; CHECK-NEXT: [[CONTINUE:%.*]] = icmp ult i32 [[I_NEXT]], [[LEN]]
; CHECK-NEXT: br i1 [[CONTINUE]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
; CHECK: exit.loopexit:
; CHECK-NEXT: [[LOOP_ACC_NEXT_LCSSA:%.*]] = phi i32 [ [[LOOP_ACC_NEXT]], [[LOOP]] ]
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
; CHECK-NEXT: [[RESULT:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[LOOP_ACC_NEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
; CHECK-NEXT: ret i32 [[RESULT]]
;
entry:
%tmp5 = icmp eq i32 %n, 0
br i1 %tmp5, label %exit, label %loop.preheader
loop.preheader:
br label %loop
loop:
%loop.acc = phi i32 [ %loop.acc.next, %loop ], [ 0, %loop.preheader ]
%i = phi i32 [ %i.next, %loop ], [ 0, %loop.preheader ]
%unknown = load volatile i1, i1* @UNKNOWN
call void (i1, ...) @llvm.experimental.guard(i1 %unknown) [ "deopt"() ]
%within.bounds = icmp ult i32 %i, %n
call void (i1, ...) @llvm.experimental.guard(i1 %within.bounds, i32 9) [ "deopt"() ]
%i.i64 = zext i32 %i to i64
%array.i.ptr = getelementptr inbounds i32, i32* %array, i64 %i.i64
%array.i = load i32, i32* %array.i.ptr, align 4
%loop.acc.next = add i32 %loop.acc, %array.i
%i.next = add nuw i32 %i, 1
%len = load i32, i32* %length, align 4, !invariant.load !{}
%continue = icmp ult i32 %i.next, %len
br i1 %continue, label %loop, label %exit
exit:
%result = phi i32 [ 0, %entry ], [ %loop.acc.next, %loop ]
ret i32 %result
}
define i32 @invariant_load_latch_limit(i32* %array,
; CHECK-LABEL: @invariant_load_latch_limit(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[TMP5]], label [[EXIT:%.*]], label [[LOOP_PREHEADER:%.*]]
; CHECK: loop.preheader:
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[LOOP_ACC:%.*]] = phi i32 [ [[LOOP_ACC_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[I_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[UNKNOWN:%.*]] = load volatile i1, i1* @UNKNOWN
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[UNKNOWN]]) [ "deopt"() ]
; CHECK-NEXT: [[WITHIN_BOUNDS:%.*]] = icmp ult i32 [[I]], [[N]]
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[WITHIN_BOUNDS]], i32 9) [ "deopt"() ]
; CHECK-NEXT: [[I_I64:%.*]] = zext i32 [[I]] to i64
; CHECK-NEXT: [[ARRAY_I_PTR:%.*]] = getelementptr inbounds i32, i32* [[ARRAY:%.*]], i64 [[I_I64]]
; CHECK-NEXT: [[ARRAY_I:%.*]] = load i32, i32* [[ARRAY_I_PTR]], align 4
; CHECK-NEXT: [[LOOP_ACC_NEXT]] = add i32 [[LOOP_ACC]], [[ARRAY_I]]
; CHECK-NEXT: [[I_NEXT]] = add nuw i32 [[I]], 1
; CHECK-NEXT: [[LEN:%.*]] = load i32, i32* [[LENGTH:%.*]], align 4, !invariant.load !0
; CHECK-NEXT: [[CONTINUE:%.*]] = icmp ult i32 [[I_NEXT]], [[LEN]]
; CHECK-NEXT: br i1 [[CONTINUE]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
; CHECK: exit.loopexit:
; CHECK-NEXT: [[LOOP_ACC_NEXT_LCSSA:%.*]] = phi i32 [ [[LOOP_ACC_NEXT]], [[LOOP]] ]
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
; CHECK-NEXT: [[RESULT:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[LOOP_ACC_NEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
; CHECK-NEXT: ret i32 [[RESULT]]
;
i32* dereferenceable(4) %length,
i32 %n) {
entry:
%tmp5 = icmp eq i32 %n, 0
br i1 %tmp5, label %exit, label %loop.preheader
loop.preheader:
br label %loop
loop:
%loop.acc = phi i32 [ %loop.acc.next, %loop ], [ 0, %loop.preheader ]
%i = phi i32 [ %i.next, %loop ], [ 0, %loop.preheader ]
%unknown = load volatile i1, i1* @UNKNOWN
call void (i1, ...) @llvm.experimental.guard(i1 %unknown) [ "deopt"() ]
%within.bounds = icmp ult i32 %i, %n
call void (i1, ...) @llvm.experimental.guard(i1 %within.bounds, i32 9) [ "deopt"() ]
%i.i64 = zext i32 %i to i64
%array.i.ptr = getelementptr inbounds i32, i32* %array, i64 %i.i64
%array.i = load i32, i32* %array.i.ptr, align 4
%loop.acc.next = add i32 %loop.acc, %array.i
%i.next = add nuw i32 %i, 1
%len = load i32, i32* %length, align 4, !invariant.load !{}
%continue = icmp ult i32 %i.next, %len
br i1 %continue, label %loop, label %exit
exit:
%result = phi i32 [ 0, %entry ], [ %loop.acc.next, %loop ]
ret i32 %result
}
@Length = external constant i32
define i32 @constant_memory(i32* %array, i32 %n) {
; CHECK-LABEL: @constant_memory(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[TMP5]], label [[EXIT:%.*]], label [[LOOP_PREHEADER:%.*]]
; CHECK: loop.preheader:
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[LOOP_ACC:%.*]] = phi i32 [ [[LOOP_ACC_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[I_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[UNKNOWN:%.*]] = load volatile i1, i1* @UNKNOWN
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[UNKNOWN]]) [ "deopt"() ]
; CHECK-NEXT: [[LEN:%.*]] = load i32, i32* @Length, align 4
[LoopPredication] Allow predication of loop invariant computations (within the loop) The purpose of this patch is to eliminate a pass ordering dependence between LoopPredication and LICM. To understand the purpose, consider the following snippet of code inside some loop 'L' with IV 'i' A = _a.length; guard (i < A) a = _a[i] B = _b.length; guard (i < B); b = _b[i]; ... Z = _z.length; guard (i < Z) z = _z[i] accum += a + b + ... + z; Today, we need LICM to hoist the length loads, LoopPredication to make the guards loop invariant, and TrivialUnswitch to eliminate the loop invariant guard to establish must execute for the next length load. Today, if we can't prove speculation safety, we'd have to iterate these three passes 26 times to reduce this example down to the minimal form. Using the fact that the array lengths are known to be invariant, we can short circuit this iteration. By forming the loop invariant form of all the guards at once, we remove the need for LoopPredication from the iterative cycle. At the moment, we'd still have to iterate LICM and TrivialUnswitch; we'll leave that part for later. As a secondary benefit, this allows LoopPred to expose peeling oppurtunities in a much more obvious manner. See the udiv test changes as an example. If the udiv was not hoistable (i.e. we couldn't prove speculation safety) this would be an example where peeling becomes obviously profitable whereas it wasn't before. A couple of subtleties in the implementation: - SCEV's isSafeToExpand guarantees speculation safety (i.e. let's us expand at a new point). It is not a precondition for expansion if we know the SCEV corresponds to a Value which dominates the requested expansion point. - SCEV's isLoopInvariant returns true for expressions which compute the same value across all iterations executed, regardless of where the original Value is located. (i.e. it can be in the loop) This implies we have a speculation burden to prove before expanding them outside loops. - invariant_loads and AA->pointsToConstantMemory are two cases that SCEV currently does not handle, but meets the SCEV definition of invariance. I plan to sink this part into SCEV once this has baked for a bit. Differential Revision: https://reviews.llvm.org/D60093 llvm-svn: 358684
2019-04-19 00:33:17 +08:00
; CHECK-NEXT: [[TMP0:%.*]] = icmp ule i32 [[N]], [[LEN]]
; CHECK-NEXT: [[TMP1:%.*]] = icmp ult i32 0, [[LEN]]
; CHECK-NEXT: [[TMP2:%.*]] = and i1 [[TMP1]], [[TMP0]]
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[TMP2]], i32 9) [ "deopt"() ]
Revert "Temporarily Revert "Add basic loop fusion pass."" The reversion apparently deleted the test/Transforms directory. Will be re-reverting again. llvm-svn: 358552
2019-04-17 12:52:47 +08:00
; CHECK-NEXT: [[I_I64:%.*]] = zext i32 [[I]] to i64
; CHECK-NEXT: [[ARRAY_I_PTR:%.*]] = getelementptr inbounds i32, i32* [[ARRAY:%.*]], i64 [[I_I64]]
; CHECK-NEXT: [[ARRAY_I:%.*]] = load i32, i32* [[ARRAY_I_PTR]], align 4
; CHECK-NEXT: [[LOOP_ACC_NEXT]] = add i32 [[LOOP_ACC]], [[ARRAY_I]]
; CHECK-NEXT: [[I_NEXT]] = add nuw i32 [[I]], 1
; CHECK-NEXT: [[CONTINUE:%.*]] = icmp ult i32 [[I_NEXT]], [[N]]
; CHECK-NEXT: br i1 [[CONTINUE]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
; CHECK: exit.loopexit:
; CHECK-NEXT: [[LOOP_ACC_NEXT_LCSSA:%.*]] = phi i32 [ [[LOOP_ACC_NEXT]], [[LOOP]] ]
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
; CHECK-NEXT: [[RESULT:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[LOOP_ACC_NEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
; CHECK-NEXT: ret i32 [[RESULT]]
;
entry:
%tmp5 = icmp eq i32 %n, 0
br i1 %tmp5, label %exit, label %loop.preheader
loop.preheader:
br label %loop
loop:
%loop.acc = phi i32 [ %loop.acc.next, %loop ], [ 0, %loop.preheader ]
%i = phi i32 [ %i.next, %loop ], [ 0, %loop.preheader ]
%unknown = load volatile i1, i1* @UNKNOWN
call void (i1, ...) @llvm.experimental.guard(i1 %unknown) [ "deopt"() ]
%len = load i32, i32* @Length, align 4
%within.bounds = icmp ult i32 %i, %len
call void (i1, ...) @llvm.experimental.guard(i1 %within.bounds, i32 9) [ "deopt"() ]
%i.i64 = zext i32 %i to i64
%array.i.ptr = getelementptr inbounds i32, i32* %array, i64 %i.i64
%array.i = load i32, i32* %array.i.ptr, align 4
%loop.acc.next = add i32 %loop.acc, %array.i
%i.next = add nuw i32 %i, 1
%continue = icmp ult i32 %i.next, %n
br i1 %continue, label %loop, label %exit
exit:
%result = phi i32 [ 0, %entry ], [ %loop.acc.next, %loop ]
ret i32 %result
}
define i32 @constant_length(i32* %array, i32 %n) {
; CHECK-LABEL: @constant_length(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[TMP5]], label [[EXIT:%.*]], label [[LOOP_PREHEADER:%.*]]
; CHECK: loop.preheader:
; CHECK-NEXT: [[TMP0:%.*]] = icmp ule i32 [[N]], 20
Revert "[IRBuilder] Fold consistently for or/and whether constant is LHS or RHS" This reverts commit r365260 which broke the following tests: Clang :: CodeGenCXX/cfi-mfcall.cpp Clang :: CodeGenObjC/ubsan-nullability.m LLVM :: Transforms/LoopVectorize/AArch64/pr36032.ll llvm-svn: 365284
2019-07-08 06:12:01 +08:00
; CHECK-NEXT: [[TMP1:%.*]] = and i1 true, [[TMP0]]
Revert "Temporarily Revert "Add basic loop fusion pass."" The reversion apparently deleted the test/Transforms directory. Will be re-reverting again. llvm-svn: 358552
2019-04-17 12:52:47 +08:00
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[LOOP_ACC:%.*]] = phi i32 [ [[LOOP_ACC_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[I_NEXT:%.*]], [[LOOP]] ], [ 0, [[LOOP_PREHEADER]] ]
; CHECK-NEXT: [[UNKNOWN:%.*]] = load volatile i1, i1* @UNKNOWN
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[UNKNOWN]]) [ "deopt"() ]
Revert "[IRBuilder] Fold consistently for or/and whether constant is LHS or RHS" This reverts commit r365260 which broke the following tests: Clang :: CodeGenCXX/cfi-mfcall.cpp Clang :: CodeGenObjC/ubsan-nullability.m LLVM :: Transforms/LoopVectorize/AArch64/pr36032.ll llvm-svn: 365284
2019-07-08 06:12:01 +08:00
; CHECK-NEXT: call void (i1, ...) @llvm.experimental.guard(i1 [[TMP1]], i32 9) [ "deopt"() ]
Revert "Temporarily Revert "Add basic loop fusion pass."" The reversion apparently deleted the test/Transforms directory. Will be re-reverting again. llvm-svn: 358552
2019-04-17 12:52:47 +08:00
; CHECK-NEXT: [[I_I64:%.*]] = zext i32 [[I]] to i64
; CHECK-NEXT: [[ARRAY_I_PTR:%.*]] = getelementptr inbounds i32, i32* [[ARRAY:%.*]], i64 [[I_I64]]
; CHECK-NEXT: [[ARRAY_I:%.*]] = load i32, i32* [[ARRAY_I_PTR]], align 4
; CHECK-NEXT: [[LOOP_ACC_NEXT]] = add i32 [[LOOP_ACC]], [[ARRAY_I]]
; CHECK-NEXT: [[I_NEXT]] = add nuw i32 [[I]], 1
; CHECK-NEXT: [[CONTINUE:%.*]] = icmp ult i32 [[I_NEXT]], [[N]]
; CHECK-NEXT: br i1 [[CONTINUE]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
; CHECK: exit.loopexit:
; CHECK-NEXT: [[LOOP_ACC_NEXT_LCSSA:%.*]] = phi i32 [ [[LOOP_ACC_NEXT]], [[LOOP]] ]
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
; CHECK-NEXT: [[RESULT:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[LOOP_ACC_NEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
; CHECK-NEXT: ret i32 [[RESULT]]
;
entry:
%tmp5 = icmp eq i32 %n, 0
br i1 %tmp5, label %exit, label %loop.preheader
loop.preheader:
br label %loop
loop:
%loop.acc = phi i32 [ %loop.acc.next, %loop ], [ 0, %loop.preheader ]
%i = phi i32 [ %i.next, %loop ], [ 0, %loop.preheader ]
%unknown = load volatile i1, i1* @UNKNOWN
call void (i1, ...) @llvm.experimental.guard(i1 %unknown) [ "deopt"() ]
%within.bounds = icmp ult i32 %i, 20
call void (i1, ...) @llvm.experimental.guard(i1 %within.bounds, i32 9) [ "deopt"() ]
%i.i64 = zext i32 %i to i64
%array.i.ptr = getelementptr inbounds i32, i32* %array, i64 %i.i64
%array.i = load i32, i32* %array.i.ptr, align 4
%loop.acc.next = add i32 %loop.acc, %array.i
%i.next = add nuw i32 %i, 1
%continue = icmp ult i32 %i.next, %n
br i1 %continue, label %loop, label %exit
exit:
%result = phi i32 [ 0, %entry ], [ %loop.acc.next, %loop ]
ret i32 %result
}