maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/polly/test/Isl/CodeGen/non-affine-phi-node-expansi...

58 lines
1.8 KiB
LLVM
Raw Normal View History

tests: Drop -polly-detect-unprofitable and -polly-no-early-exit These flags are now always passed to all tests and need to be disabled if not needed. Disabling these flags, rather than passing them to almost all tests, significantly simplfies our RUN: lines. llvm-svn: 249422
2015-10-06 23:36:44 +08:00
; RUN: opt %loadPolly -polly-codegen \
Do not detect Scops with only one loop. If a region does not have more than one loop, we do not identify it as a Scop in ScopDetection. The main optimizations Polly is currently performing (tiling, preparation for outer-loop vectorization and loop fusion) are unlikely to have a positive impact on individual loops. In some cases, Polly's run-time alias checks or conditional hoisting may still have a positive impact, but those are mostly enabling transformations which LLVM already performs for individual loops. As we do not focus on individual loops, we leave them untouched to not introduce compile time regressions and execution time noise. This results in good compile time reduction (oourafft: -73.99%, smg2000: -56.25%). Contributed-by: Pratik Bhatu <cs12b1010@iith.ac.in> Reviewers: grosser Differential Revision: http://reviews.llvm.org/D12268 llvm-svn: 246161
2015-08-28 00:55:18 +08:00
; RUN: -S < %s | FileCheck %s
Only use instructions as insert locations for SCEVExpander SCEVExpander, which we are using during code generation, only allows instructions as insert locations, but breaks in case BasicBlock->end() iterators are passed to it due to it trying to obtain the basic block in which code should be generated by calling Instruction->getParent(), which is not defined for ->end() iterators. This change adds an assert to Polly that ensures we only pass valid instructions to SCEVExpander and it fixes one case, where we used IRBuilder->SetInsertBlock() to set an ->end() insert location which was later passed to SCEVExpander. In general, Polly is always trying to build up the CFG first, before we actually insert instructions into the CFG sceleton. As a result, each basic block should already have at least one branch instruction before we start adding code. Hence, always requiring the IRBuilder insert location to be set to a real instruction should always be possible. Thanks Utpal Bora <cs14mtech11017@iith.ac.in> for his help with test case reduction. llvm-svn: 243830
2015-08-01 17:07:57 +08:00
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
%struct.wombat = type {[4 x i32]}
[FIX] Do not hoist from inside a non-affine subregion We have to skip accesses in non-affine subregions during hoisting as they might not be executed under the same condition as the entry of the non-affine subregion. llvm-svn: 249139
2015-10-02 22:51:00 +08:00
; CHECK-NOT: polly.preload.begin:
[FIX] Repair broken commit The last invariant load fix was based on a later patch not polly/master, thus needs to be adjusted. llvm-svn: 249145
2015-10-02 23:35:03 +08:00
; CHECK-NOT: %polly.access.B
[FIX] Do not hoist from inside a non-affine subregion We have to skip accesses in non-affine subregions during hoisting as they might not be executed under the same condition as the entry of the non-affine subregion. llvm-svn: 249139
2015-10-02 22:51:00 +08:00
; CHECK-NOT: %polly.access.B.load
Identify and hoist definitively invariant loads As a first step in the direction of assumed invariant loads (loads that are not written in some context) we now detect and hoist definitively invariant loads. These invariant loads will be preloaded in the code generation and used in the optimized version of the SCoP. If the load is only conditionally executed the preloaded version will also only be executed under the same condition, hence we will never access memory that wouldn't have been accessed otherwise. This is also the most distinguishing feature to licm. As hoisting can make statements empty we will simplify the SCoP and remove empty statements that would otherwise cause artifacts in the code generation. Differential Revision: http://reviews.llvm.org/D13194 llvm-svn: 248861
2015-09-30 07:47:21 +08:00
[FIX] Repair broken commit The last invariant load fix was based on a later patch not polly/master, thus needs to be adjusted. llvm-svn: 249145
2015-10-02 23:35:03 +08:00
; CHECK: polly.split_new_and_old
Only use instructions as insert locations for SCEVExpander SCEVExpander, which we are using during code generation, only allows instructions as insert locations, but breaks in case BasicBlock->end() iterators are passed to it due to it trying to obtain the basic block in which code should be generated by calling Instruction->getParent(), which is not defined for ->end() iterators. This change adds an assert to Polly that ensures we only pass valid instructions to SCEVExpander and it fixes one case, where we used IRBuilder->SetInsertBlock() to set an ->end() insert location which was later passed to SCEVExpander. In general, Polly is always trying to build up the CFG first, before we actually insert instructions into the CFG sceleton. As a result, each basic block should already have at least one branch instruction before we start adding code. Hence, always requiring the IRBuilder insert location to be set to a real instruction should always be possible. Thanks Utpal Bora <cs14mtech11017@iith.ac.in> for his help with test case reduction. llvm-svn: 243830
2015-08-01 17:07:57 +08:00
; CHECK: polly.stmt.bb3.entry: ; preds = %polly.start
; CHECK: br label %polly.stmt.bb3
; CHECK: polly.stmt.bb3: ; preds = %polly.stmt.bb3.entry
; CHECK: br i1 true, label %polly.stmt.bb4, label %polly.stmt.bb5
; CHECK: polly.stmt.bb4: ; preds = %polly.stmt.bb3
; CHECK: br label %polly.stmt.bb13.exit
; CHECK: polly.stmt.bb5: ; preds = %polly.stmt.bb3
[FIX] Repair broken commit The last invariant load fix was based on a later patch not polly/master, thus needs to be adjusted. llvm-svn: 249145
2015-10-02 23:35:03 +08:00
; CHECK: load i32, i32* %B
Only use instructions as insert locations for SCEVExpander SCEVExpander, which we are using during code generation, only allows instructions as insert locations, but breaks in case BasicBlock->end() iterators are passed to it due to it trying to obtain the basic block in which code should be generated by calling Instruction->getParent(), which is not defined for ->end() iterators. This change adds an assert to Polly that ensures we only pass valid instructions to SCEVExpander and it fixes one case, where we used IRBuilder->SetInsertBlock() to set an ->end() insert location which was later passed to SCEVExpander. In general, Polly is always trying to build up the CFG first, before we actually insert instructions into the CFG sceleton. As a result, each basic block should already have at least one branch instruction before we start adding code. Hence, always requiring the IRBuilder insert location to be set to a real instruction should always be possible. Thanks Utpal Bora <cs14mtech11017@iith.ac.in> for his help with test case reduction. llvm-svn: 243830
2015-08-01 17:07:57 +08:00
; Function Attrs: nounwind uwtable
define void @quux(%struct.wombat* %arg, i32* %B) {
bb:
br i1 undef, label %bb2, label %bb1
bb1: ; preds = %bb
br label %bb2
bb2: ; preds = %bb1, %bb
%tmp = phi i1 [ true, %bb ], [ undef, %bb1 ]
br label %bb3
bb3: ; preds = %bb13, %bb2
br i1 %tmp, label %bb4, label %bb5
bb4: ; preds = %bb3
br label %bb13
bb5: ; preds = %bb3
%tmp7 = load i32, i32* %B
%tmp12 = getelementptr inbounds %struct.wombat, %struct.wombat* %arg, i64 0, i32 0, i64 0
store i32 %tmp7, i32* %tmp12
br label %bb13
bb13: ; preds = %bb5, %bb4
br i1 false, label %bb3, label %bb14
bb14: ; preds = %bb13
br label %bb15
bb15: ; preds = %bb14
unreachable
}