maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/polly/test/DeLICM/reduction_overapproximate.ll

159 lines
7.9 KiB
LLVM
Raw Normal View History

[ScopBuilder] Make -polly-stmt-granularity=scalar-indep the default. Splitting basic blocks into multiple statements if there are now additional scalar dependencies gives more freedom to the scheduler, but more statements also means higher compile-time complexity. Switch to finer statement granularity, the additional compile time should be limited by the number of operations quota. The regression tests are written for the -polly-stmt-granularity=bb setting, therefore we add that flag to those tests that break with the new default. Some of the tests only fail because the statements are named differently due to a basic block resulting in multiple statements, but which are removed during simplification of statements without side-effects. Previous commits tried to reduce this effect, but it is not completely avoidable. Differential Revision: https://reviews.llvm.org/D42151 llvm-svn: 324169
2018-02-03 14:59:47 +08:00
; RUN: opt %loadPolly -polly-stmt-granularity=bb -polly-flatten-schedule -polly-delicm-compute-known=true -polly-delicm-overapproximate-writes=true -polly-delicm-partial-writes=false -polly-delicm -analyze < %s | FileCheck %s --check-prefix=APPROX
; RUN: opt %loadPolly -polly-stmt-granularity=bb -polly-flatten-schedule -polly-delicm-compute-known=true -polly-delicm-overapproximate-writes=false -polly-delicm-partial-writes=false -polly-delicm -analyze < %s | FileCheck %s --check-prefix=EXACT
; RUN: opt %loadPolly -polly-stmt-granularity=bb -polly-flatten-schedule -polly-delicm-compute-known=true -polly-delicm-partial-writes=true -polly-delicm -analyze < %s | FileCheck %s --check-prefix=PARTIAL
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
;
; void func(double *A {
; for (int j = -1; j < 3; j += 1) { /* outer */
; double phi = 0.0;
; if (0 < j)
; for (int i = 0; i < j; i += 1) /* reduction */
; phi += 4.2;
; A[j] = phi;
; }
; }
;
define void @func(double* noalias nonnull %A) {
entry:
br label %outer.preheader
outer.preheader:
br label %outer.for
outer.for:
%j = phi i32 [-1, %outer.preheader], [%j.inc, %outer.inc]
%j.cmp = icmp slt i32 %j, 3
br i1 %j.cmp, label %reduction.checkloop, label %outer.exit
reduction.checkloop:
%j2.cmp = icmp slt i32 0, %j
br i1 %j2.cmp, label %reduction.preheader, label %reduction.exit
reduction.preheader:
br label %reduction.for
reduction.for:
%i = phi i32 [0, %reduction.preheader], [%i.inc, %reduction.inc]
%phi = phi double [0.0, %reduction.preheader], [%add, %reduction.inc]
br label %body
body:
%add = fadd double %phi, 4.2
br label %reduction.inc
reduction.inc:
%i.inc = add nuw nsw i32 %i, 1
%i.cmp = icmp slt i32 %i.inc, %j
br i1 %i.cmp, label %reduction.for, label %reduction.exit
reduction.exit:
%val = phi double [%add, %reduction.inc], [0.0, %reduction.checkloop]
%A_idx = getelementptr inbounds double, double* %A, i32 %j
store double %val, double* %A_idx
br label %outer.inc
outer.inc:
%j.inc = add nuw nsw i32 %j, 1
br label %outer.for
outer.exit:
br label %return
return:
ret void
}
; APPROX: After accesses {
; APPROX-NEXT: Stmt_reduction_checkloop
; APPROX-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_reduction_checkloop[i0] -> MemRef_val__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_reduction_checkloop[i0] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: Stmt_reduction_preheader
; APPROX-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_reduction_preheader[i0] -> MemRef_phi__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_reduction_preheader[i0] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: Stmt_reduction_for
; APPROX-NEXT: ReadAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_reduction_for[i0, i1] -> MemRef_phi__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_reduction_for[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_reduction_for[i0, i1] -> MemRef_phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_reduction_for[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: Stmt_body
; APPROX-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_body[i0, i1] -> MemRef_add[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_body[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: ReadAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_body[i0, i1] -> MemRef_phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_body[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: Stmt_reduction_inc
; APPROX-NEXT: ReadAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_reduction_inc[i0, i1] -> MemRef_add[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_reduction_inc[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_reduction_inc[i0, i1] -> MemRef_phi__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_reduction_inc[i0, i1] -> MemRef_A[2] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_reduction_inc[i0, i1] -> MemRef_val__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_reduction_inc[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: Stmt_reduction_exit
; APPROX-NEXT: ReadAccess := [Reduction Type: NONE] [Scalar: 1]
; APPROX-NEXT: { Stmt_reduction_exit[i0] -> MemRef_val__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; APPROX-NEXT: new: { Stmt_reduction_exit[i0] -> MemRef_A[-1 + i0] };
[DeLICM] Add -polly-delicm-overapproximate-writes option. One of the current limitations of DeLICM is that it only creates PHI WRITEs that it knows are read by some PHI. Such writes may not span all instances of a statement. Polly's code generator currently does not support MemoryAccesses that are not executed in all instances ('partial accesses') and so has to give up on a possible mapping. This workaround has once been suggested by Tobias Grosser: Try to interpolate an arbitrary expansion to all instances. It will be checked for possible conflicts with the existing Knowledge and can be applied if the conflict checking result is that no semantics are changed. Expansion is done by simplifying the mapping by coalescing with the hope that coalescing will find a polyhedral 'rule' of the relevant map. It is then 'gist'-ed using the domain of the relevant instances such that the rule is expanded to the universe and finally intersected with the domain of all statement instances. The expansion makes conflicts become more likely, the found rule may still not encompass all statement instances and the found rule exposes internals of isl's implementation of coalesce and gist. The latter means that the result depends on how much effort the implementation invests into finding a rule which may change between versions of isl. Trivial implementations of gist and coalesce just return the input arguments. A patch that makes codegen support partial accesses is in preparation as well. Differential Revision: https://reviews.llvm.org/D30763 llvm-svn: 297373
2017-03-09 19:23:22 +08:00
; APPROX-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 0]
; APPROX-NEXT: { Stmt_reduction_exit[i0] -> MemRef_A[-1 + i0] };
; APPROX-NEXT: }
; EXACT: No modification has been made
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL: After accesses {
; PARTIAL-NEXT: Stmt_reduction_checkloop
; PARTIAL-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_reduction_checkloop[i0] -> MemRef_val__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_reduction_checkloop[i0] -> MemRef_A[-1 + i0] : i0 <= 1 };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: Stmt_reduction_preheader
; PARTIAL-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_reduction_preheader[i0] -> MemRef_phi__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_reduction_preheader[i0] -> MemRef_A[-1 + i0] };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: Stmt_reduction_for
; PARTIAL-NEXT: ReadAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_reduction_for[i0, i1] -> MemRef_phi__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_reduction_for[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_reduction_for[i0, i1] -> MemRef_phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_reduction_for[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: Stmt_body
; PARTIAL-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_body[i0, i1] -> MemRef_add[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_body[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: ReadAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_body[i0, i1] -> MemRef_phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_body[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: Stmt_reduction_inc
; PARTIAL-NEXT: ReadAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_reduction_inc[i0, i1] -> MemRef_add[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_reduction_inc[i0, i1] -> MemRef_A[-1 + i0] };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_reduction_inc[i0, i1] -> MemRef_phi__phi[] };
; PARTIAL-NEXT: new: { Stmt_reduction_inc[3, 0] -> MemRef_A[2] };
; PARTIAL-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_reduction_inc[i0, i1] -> MemRef_val__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_reduction_inc[i0, -2 + i0] -> MemRef_A[-1 + i0] };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: Stmt_reduction_exit
; PARTIAL-NEXT: ReadAccess := [Reduction Type: NONE] [Scalar: 1]
; PARTIAL-NEXT: { Stmt_reduction_exit[i0] -> MemRef_val__phi[] };
[ScopInfo] Simplify new access functions under domain context Summary: We do not keep domain constraints on access functions when building the scop. Hence, for consistency reasons, it makes also sense to not include them when storing a new access function. This change results in simpler access functions that make output easier to read. This patch also helps to make DeLICMed memory accesses to be understood by our matrix multiplication pattern matching pass. Further changes to the matrix multiplication pattern matching are needed for this to work, so the corresponding test case will be added in a future commit. Reviewers: Meinersbur, bollu, gareevroman, efriedma, huihuiz, sebpop, simbuerg Subscribers: pollydev, llvm-commits Tags: #polly Differential Revision: https://reviews.llvm.org/D35237 llvm-svn: 308215
2017-07-18 04:47:10 +08:00
; PARTIAL-NEXT: new: { Stmt_reduction_exit[i0] -> MemRef_A[-1 + i0] };
[DeLICM] Partial writes for PHIs. Enable the use for partial writes for PHI write accesses with a switch. This simply skips the test for whether a PHI write would be partial. The analog test for partial value writes also protects for partial reads which we do not support (yet). It is possible to test for partial reads separately such that we could skip the partial write check as well. In case this shows up to be useful, I can implement it as well. Differential Revision: https://reviews.llvm.org/D33487 llvm-svn: 303762
2017-05-24 23:23:06 +08:00
; PARTIAL-NEXT: MustWriteAccess := [Reduction Type: NONE] [Scalar: 0]
; PARTIAL-NEXT: { Stmt_reduction_exit[i0] -> MemRef_A[-1 + i0] };
; PARTIAL-NEXT: }