llvm-project/polly/test/DependenceInfo/reduction_privatization_dep...

83 lines
3.2 KiB
LLVM
Raw Normal View History

; RUN: opt %loadPolly -polly-dependences -analyze < %s | FileCheck %s
;
; CHECK: RAW dependences:
2016-01-15 23:54:45 +08:00
; CHECK-NEXT: { Stmt_S2[i0, i1] -> Stmt_S3[o0] : i1 <= 1 - i0 and -i1 < o0 <= 1 and o0 <= 1 + i0 - i1; Stmt_S3[i0] -> Stmt_S2[o0, 1 - i0] : 0 <= i0 <= 1 and i0 < o0 <= 98; Stmt_S1[i0] -> Stmt_S3[2 + i0] : 0 <= i0 <= 96 }
; CHECK-NEXT: WAR dependences:
[Polly] [DependenceInfo] change WAR, WAW generation to correct semantics = Change of WAR, WAW generation: = - `buildFlow(Sink, MustSource, MaySource, Sink)` treates any flow of the form `sink <- may source <- must source` as a *may* dependence. - we used to call: ```lang=cpp, name=old-flow-call.cpp Flow = buildFlow(MustWrite, MustWrite, Read, Schedule); WAW = isl_union_flow_get_must_dependence(Flow); WAR = isl_union_flow_get_may_dependence(Flow); ``` - This caused some WAW dependences to be treated as WAR dependences. - Incorrect semantics. - Now, we call WAR and WAW correctly. == Correct WAW: == ```lang=cpp, name=new-waw-call.cpp Flow = buildFlow(Write, MustWrite, MayWrite, Schedule); WAW = isl_union_flow_get_may_dependence(Flow); isl_union_flow_free(Flow); ``` == Correct WAR: == ```lang=cpp, name=new-war-call.cpp Flow = buildFlow(Write, Read, MustaWrite, Schedule); WAR = isl_union_flow_get_must_dependence(Flow); isl_union_flow_free(Flow); ``` - We want the "shortest" WAR possible (exact dependences). - We mark all the *must-writes* as may-source, reads as must-souce. - Then, we ask for *must* dependence. - This removes all the reads that flow through a *must-write* before reaching a sink. - Note that we only block ealier writes with *must-writes*. This is intuitively correct, as we do not want may-writes to block must-writes. - Leaves us with direct (R -> W). - This affects reduction generation since RED is built using WAW and WAR. = New StrictWAW for Reductions: = - We used to call: ```lang=cpp,name=old-waw-war-call.cpp Flow = buildFlow(MustWrite, MustWrite, Read, Schedule); WAW = isl_union_flow_get_must_dependence(Flow); WAR = isl_union_flow_get_may_dependence(Flow); ``` - This *is* the right model of WAW we need for reductions, just not in general. - Reductions need to track only *strict* WAW, without any interfering reductions. = Explanation: Why the new WAR dependences in tests are correct: = - We no longer set WAR = WAR - WAW - Hence, we will have WAR dependences that were originally removed. - These may look incorrect, but in fact make sense. == Code: == ```lang=llvm, name=new-war-dependence.ll ; void manyreductions(long *A) { ; for (long i = 0; i < 1024; i++) ; for (long j = 0; j < 1024; j++) ; S0: *A += 42; ; ; for (long i = 0; i < 1024; i++) ; for (long j = 0; j < 1024; j++) ; S1: *A += 42; ; ``` === WAR dependence: === { S0[1023, 1023] -> S1[0, 0] } - Between `S0[1023, 1023]` and `S1[0, 0]`, we will have the dependences: ```lang=cpp, name=dependence-incorrect, counterexample S0[1023, 1023]: *-- tmp = *A (load0)--* WAR 2 add = tmp + 42 | *-> *A = add (store0) | WAR 1 S1[0, 0]: | tmp = *A (load1) | add = tmp + 42 | A = add (store1)<-* ``` - One may assume that WAR2 *hides* WAR1 (since store0 happens before store1). However, within a statement, Polly has no idea about the ordering of loads and stores. - Hence, according to Polly, the code may have looked like this: ```lang=cpp, name=dependence-correct S0[1023, 1023]: A = add (store0) tmp = A (load0) ---* add = A + 42 | WAR 1 S1[0, 0]: | tmp = A (load1) | add = A + 42 | A = add (store1) <-* ``` - So, Polly generates (correct) WAR dependences. It does not make sense to remove these dependences, since they are correct with respect to Polly's model. Reviewers: grosser, Meinersbur tags: #polly Differential revision: https://reviews.llvm.org/D31386 llvm-svn: 299429
2017-04-04 21:08:23 +08:00
; CHECK-NEXT: { Stmt_S2[i0, i1] -> Stmt_S3[o0] : i1 <= 1 - i0 and -i1 < o0 <= 1 and o0 <= 1 + i0 - i1; Stmt_S3[i0] -> Stmt_S2[o0, 1 - i0] : 0 <= i0 <= 1 and i0 < o0 <= 98; Stmt_S1[i0] -> Stmt_S3[2 + i0] : 0 <= i0 <= 96 }
; CHECK-NEXT: WAW dependences:
2016-01-15 23:54:45 +08:00
; CHECK-NEXT: { Stmt_S2[i0, i1] -> Stmt_S3[o0] : i1 <= 1 - i0 and -i1 < o0 <= 1 and o0 <= 1 + i0 - i1; Stmt_S3[i0] -> Stmt_S2[o0, 1 - i0] : 0 <= i0 <= 1 and i0 < o0 <= 98; Stmt_S1[i0] -> Stmt_S3[2 + i0] : 0 <= i0 <= 96 }
; CHECK-NEXT: Reduction dependences:
2016-01-15 23:54:45 +08:00
; CHECK-NEXT: { Stmt_S2[i0, i1] -> Stmt_S2[1 + i0, i1] : 0 <= i0 <= 97 and i1 >= 0 and 2 - i0 <= i1 <= 98 - i0; Stmt_S2[0, 0] -> Stmt_S2[1, 0] }
;
; void f(int *sum) {
; int i, j;
; for (i = 0; i < 99; i++) {
; S1: sum[i + 1] += 42;
; for (j = i; j < 100; j++)
; S2: sum[i - j] += i * j;
; S3: sum[i - 1] += 7;
; }
; }
;
target datalayout = "e-m:e-p:32:32-i64:64-v128:64:128-n32-S64"
define void @f(i32* %sum) {
entry:
br label %for.cond
for.cond: ; preds = %for.inc10, %entry
%i.0 = phi i32 [ 0, %entry ], [ %inc11, %for.inc10 ]
%exitcond1 = icmp ne i32 %i.0, 99
br i1 %exitcond1, label %for.body, label %for.end12
for.body: ; preds = %for.cond
br label %S1
S1: ; preds = %for.body
%add = add nsw i32 %i.0, 1
%arrayidx = getelementptr inbounds i32, i32* %sum, i32 %add
%tmp = load i32, i32* %arrayidx, align 4
%add1 = add nsw i32 %tmp, 42
store i32 %add1, i32* %arrayidx, align 4
br label %for.cond2
for.cond2: ; preds = %for.inc, %S1
%j.0 = phi i32 [ %i.0, %S1 ], [ %inc, %for.inc ]
%exitcond = icmp ne i32 %j.0, 100
br i1 %exitcond, label %for.body4, label %for.end
for.body4: ; preds = %for.cond2
br label %S2
S2: ; preds = %for.body4
%mul = mul nsw i32 %i.0, %j.0
%sub = sub nsw i32 %i.0, %j.0
%arrayidx5 = getelementptr inbounds i32, i32* %sum, i32 %sub
%tmp2 = load i32, i32* %arrayidx5, align 4
%add6 = add nsw i32 %tmp2, %mul
store i32 %add6, i32* %arrayidx5, align 4
br label %for.inc
for.inc: ; preds = %S2
%inc = add nsw i32 %j.0, 1
br label %for.cond2
for.end: ; preds = %for.cond2
br label %S3
S3: ; preds = %for.end
%sub7 = add nsw i32 %i.0, -1
%arrayidx8 = getelementptr inbounds i32, i32* %sum, i32 %sub7
%tmp3 = load i32, i32* %arrayidx8, align 4
%add9 = add nsw i32 %tmp3, 7
store i32 %add9, i32* %arrayidx8, align 4
br label %for.inc10
for.inc10: ; preds = %S3
%inc11 = add nsw i32 %i.0, 1
br label %for.cond
for.end12: ; preds = %for.cond
ret void
}