[PATCH] D30265: [APInt] Add APInt::setBits() method to set all bits in range

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon created this revision.

The current pattern for setting bits in range is typically:

  Mask |= APInt::getBitsSet(MaskSizeInBits, LoPos, HiPos);

Which can be particularly slow for large APInts (MaskSizeInBits > 64) as they require the allocation memory for the temporary variable.

This is one of the key compile time issues identified in PR32037.

This patch adds the APInt::setBits() helper method which avoids the temporary memory allocation completely, this first implementation uses setBit() internally instead but already significantly reduces the regression in PR32037 (~10% drop). Additional optimization may be possible.

I investigated whether there is need for APInt::clearBits() and APInt::flipBits() equivalents but haven't seen these patterns to be particularly common.


Repository:
  rL LLVM

https://reviews.llvm.org/D30265

Files:
  include/llvm/ADT/APInt.h
  lib/Support/APInt.cpp
  lib/Target/X86/X86ISelLowering.cpp
  lib/Target/X86/X86ShuffleDecodeConstantPool.cpp
  unittests/ADT/APIntTest.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D30265.89395.patch
Type: text/x-patch
Size: 5880 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170222/727d4063/attachment.bin>