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/MemAccess/simple_analyze.ll

77 lines
4.2 KiB
LLVM
Raw Normal View History

ScopDetection: Only detect scops that have at least one read and one write Scops that only read seem generally uninteresting and scops that only write are most likely initializations where there is also little to optimize. To not waste compile time we bail early. Differential Revision: http://reviews.llvm.org/D7735 llvm-svn: 229820
2015-02-19 13:31:07 +08:00
;RUN: opt %loadPolly -polly-detect-unprofitable -polly-import-jscop -analyze -polly-import-jscop-dir=%S -polly-import-jscop-postfix=transformed < %s | FileCheck %s
;RUN: opt %loadPolly -polly-detect-unprofitable -polly-import-jscop -polly-import-jscop-dir=%S -polly-import-jscop-postfix=transformed -polly-codegen-isl -polly-vectorizer=polly -S < %s | FileCheck %s --check-prefix=JSCOPVEC
Memaccess: Display Changed Access Relation The changed access relations imported from JSCOP file is shown as output of -analyze pass. llvm-svn: 136774
2011-08-03 21:47:59 +08:00
target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:32:32-n8:16:32"
@A = common global [100 x i32] zeroinitializer, align 4
@B = common global [100 x i32] zeroinitializer, align 4
test: Remove memaccess prefix The prefix is not needed, as all test cases are already in a separate folder. llvm-svn: 153320
2012-03-23 16:24:04 +08:00
define i32 @simple() nounwind {
Memaccess: Display Changed Access Relation The changed access relations imported from JSCOP file is shown as output of -analyze pass. llvm-svn: 136774
2011-08-03 21:47:59 +08:00
entry:
br label %for.cond
for.cond: ; preds = %for.inc, %entry
%0 = phi i32 [ 0, %entry ], [ %inc, %for.inc ]
%arrayidx = getelementptr [100 x i32]* @A, i32 0, i32 %0
%exitcond1 = icmp ne i32 %0, 12
br i1 %exitcond1, label %for.body, label %for.end
for.body: ; preds = %for.cond
store i32 %0, i32* %arrayidx
br label %for.inc
for.inc: ; preds = %for.body
%inc = add nsw i32 %0, 1
br label %for.cond
for.end: ; preds = %for.cond
Switch default code generation backend to isl The isl based backend has been tested since a long time and with the recently commited OpenMP support the last missing piece of functionality was ported from the CLooG backend. The isl based backend gives us interesting new functionality: - Run-time alias checks (enabled by default) Optimize scops that contain possibly aliasing pointers. This feature has largely increased the number of loop nests we consider for optimization. Thanks Johannes! - Delinearization (not yet enabled by default) Model accesses to multi-dimensional arrays precisely. This will allow us to understand kernels with multi-dimensional VLAs written in Julia, boost::ublas, coremark or C99. Thanks Sebastian! - Generation of higher quality code Sven and me spent a long time to optimize the quality of the generated code. A major focus were expressions as they result from modulos/divisions or piecewise affine expressions (a ? b : c). - Full/Partial tile separation, polyhedral unrolling The isl code generation provides functionality to generate specialized code for core and cleanup loops and to specialize code using polyhedral context information while unrolling statements. (not yet exploited in Polly) - Modifieable access functions We can now use standard isl functionality to remap memory accesses to new data locations. A standard use case is the use of shared memory, where accesses to a larger region in global memory need to be mapped to a smaller shared memory region using a modulo mapping. (not yet exploited in Polly) The cloog based code generation is still available for comparision, but is scheduled for removal. llvm-svn: 222101
2014-11-17 01:02:11 +08:00
fence seq_cst
Memaccess: Display Changed Access Relation The changed access relations imported from JSCOP file is shown as output of -analyze pass. llvm-svn: 136774
2011-08-03 21:47:59 +08:00
br label %for.cond4
for.cond4: ; preds = %for.inc11, %for.end
%1 = phi i32 [ 0, %for.end ], [ %inc13, %for.inc11 ]
%arrayidx10 = getelementptr [100 x i32]* @B, i32 0, i32 %1
%exitcond = icmp ne i32 %1, 12
br i1 %exitcond, label %for.body7, label %for.end14
for.body7: ; preds = %for.cond4
store i32 %1, i32* %arrayidx10
br label %for.inc11
for.inc11: ; preds = %for.body7
%inc13 = add nsw i32 %1, 1
br label %for.cond4
for.end14: ; preds = %for.cond4
ret i32 0
}
[RTC] Runtime Alias Checks for the ISL backend This change will build all alias groups (minimal/maximal accesses to possible aliasing base pointers) we have to check before we can assume an alias free environment. It will also use these to create Runtime Alias Checks (RTC) in the ISL code generation backend, thus allow us to optimize SCoPs despite possibly aliasing pointers when this backend is used. This feature will be enabled for the isl code generator, e.g., --polly-code-generator=isl, but disabled for: - The cloog code generator (still the default). - The case delinearization is enabled. - The case non-affine accesses are allowed. llvm-svn: 218046
2014-09-18 19:17:17 +08:00
; CHECK-DAG: New access function '{ Stmt_for_body7[i0] -> MemRef_B[0] }'detected in JSCOP file
; CHECK-DAG: New access function '{ Stmt_for_body[i0] -> MemRef_A[0] }'detected in JSCOP file
Allow the VectorBlockGenerator to use the IslExprBuilder. This also enables the VectorBlockGenerator to build load store accesses according to the newAccessRelation of a MemoryAccess. llvm-svn: 219321
2014-10-09 01:25:30 +08:00
; Verify that the new access function (see above) is actually used during vector code generation.
; JSCOPVEC: store i32 0, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 1, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 2, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 3, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 4, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 5, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 6, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 7, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 8, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 9, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 10, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 11, i32* getelementptr inbounds ([100 x i32]* @B, i32 0, i32 0)
; JSCOPVEC: store i32 0, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 1, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 2, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 3, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 4, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 5, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 6, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 7, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 8, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 9, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 10, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)
; JSCOPVEC: store i32 11, i32* getelementptr inbounds ([100 x i32]* @A, i32 0, i32 0)