aboutsummaryrefslogtreecommitdiffstats
path: root/include/llvm-c
diff options
context:
space:
mode:
authorHal Finkel <hfinkel@anl.gov>2013-11-16 23:59:05 +0000
committerHal Finkel <hfinkel@anl.gov>2013-11-16 23:59:05 +0000
commitbebe48dbfe00078329341945bfb11f778ace6d12 (patch)
tree5efe3810ad3db10985cd43351ed4a7bd83727908 /include/llvm-c
parent0ccb37a7339883e1fd090beadc2deb1ce40ea7d4 (diff)
downloadexternal_llvm-bebe48dbfe00078329341945bfb11f778ace6d12.zip
external_llvm-bebe48dbfe00078329341945bfb11f778ace6d12.tar.gz
external_llvm-bebe48dbfe00078329341945bfb11f778ace6d12.tar.bz2
Add a loop rerolling pass
This adds a loop rerolling pass: the opposite of (partial) loop unrolling. The transformation aims to take loops like this: for (int i = 0; i < 3200; i += 5) { a[i] += alpha * b[i]; a[i + 1] += alpha * b[i + 1]; a[i + 2] += alpha * b[i + 2]; a[i + 3] += alpha * b[i + 3]; a[i + 4] += alpha * b[i + 4]; } and turn them into this: for (int i = 0; i < 3200; ++i) { a[i] += alpha * b[i]; } and loops like this: for (int i = 0; i < 500; ++i) { x[3*i] = foo(0); x[3*i+1] = foo(0); x[3*i+2] = foo(0); } and turn them into this: for (int i = 0; i < 1500; ++i) { x[i] = foo(0); } There are two motivations for this transformation: 1. Code-size reduction (especially relevant, obviously, when compiling for code size). 2. Providing greater choice to the loop vectorizer (and generic unroller) to choose the unrolling factor (and a better ability to vectorize). The loop vectorizer can take vector lengths and register pressure into account when choosing an unrolling factor, for example, and a pre-unrolled loop limits that choice. This is especially problematic if the manual unrolling was optimized for a machine different from the current target. The current implementation is limited to single basic-block loops only. The rerolling recognition should work regardless of how the loop iterations are intermixed within the loop body (subject to dependency and side-effect constraints), but the significant restriction is that the order of the instructions in each iteration must be identical. This seems sufficient to capture all current use cases. This pass is not currently enabled by default at any optimization level. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194939 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'include/llvm-c')
-rw-r--r--include/llvm-c/Transforms/Scalar.h3
1 files changed, 3 insertions, 0 deletions
diff --git a/include/llvm-c/Transforms/Scalar.h b/include/llvm-c/Transforms/Scalar.h
index 2456c6c..355e8dc 100644
--- a/include/llvm-c/Transforms/Scalar.h
+++ b/include/llvm-c/Transforms/Scalar.h
@@ -65,6 +65,9 @@ void LLVMAddLoopIdiomPass(LLVMPassManagerRef PM);
/** See llvm::createLoopRotatePass function. */
void LLVMAddLoopRotatePass(LLVMPassManagerRef PM);
+/** See llvm::createLoopRerollPass function. */
+void LLVMAddLoopRerollPass(LLVMPassManagerRef PM);
+
/** See llvm::createLoopUnrollPass function. */
void LLVMAddLoopUnrollPass(LLVMPassManagerRef PM);