[PATCH] D86295: [analyzer] Reorder the layout of MemRegion and cache by hand for optimal size

Fri Aug 21 08:23:50 PDT 2020

steakhal added a comment.

In D86295#2229712 <https://reviews.llvm.org/D86295#2229712>, @NoQ wrote:

> Heh, nice! Did you try to measure the actual impact of this change on memory and/or performance?

Eh, I don't really know how to bench this.
Here is how I did it in nutshell:
Added STATISTIC counters to MemRegion ctor, dtor, getAsOffset begining and the path of just returning the cached value.
F12722457: add-memregion-statistics.patch <https://reviews.llvm.org/F12722457>

I have analyzed a faily big (50k+ LOC) TU of the llvm repository (`X86ISelLowering.cpp`).
I was using the following command:

  /home/myuser/git/llvm-project/build/bin/clang --analyze -Qunused-arguments -Xclang -analyzer-opt-analyze-headers -Xclang -analyzer-output=text -Xclang -analyzer-config -Xclang expand-macros=true -Xclang -analyzer-checker=core -Xclang -analyzer-config -Xclang aggressive-binary-operation-simplification=true  -x c++ --target=x86_64-linux-gnu -std=gnu++14 -DGTEST_HAS_RTTI=0 -D_DEBUG -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -I/home/myuser/git/llvm-project/build/lib/Target/X86 -I/home/myuser/git/llvm-project/llvm/lib/Target/X86 -I/home/myuser/git/llvm-project/build/include -I/home/myuser/git/llvm-project/llvm/include -fPIC -fvisibility-inlines-hidden -Wno-unused-parameter -Wno-missing-field-initializers -pedantic -Wno-long-long -Wno-maybe-uninitialized -Wno-noexcept-type -Wno-comment -fdiagnostics-color -fPIC -fno-exceptions -fno-rtti -std=c++14 -isystem /usr/include/c++/7 -isystem /usr/include/x86_64-linux-gnu/c++/7 -isystem /usr/include/c++/7/backward -isystem /usr/local/include -isystem /usr/include/x86_64-linux-gnu -isystem /usr/include -Xclang -analyzer-stats /home/myuser/git/llvm-project/llvm/lib/Target/X86/X86ISelLowering.cpp

The relevant part of the log is:

   1051489 MemRegion        - The # of MemRegion objects were alive at a time during analysis
  48603499 MemRegion        - The # of times the MemRegion::getAsOffset was called and the result was already cached
  49258796 MemRegion        - The # of times the MemRegion::getAsOffset was called
   1051489 MemRegion        - The # of MemRegion objects alive at the given moment
   1051489 MemRegion        - The # of MemRegion objects created during the analysis

---

> I think it'd make perfect sense to keep the offset in a side map. We don't compute it for all regions, and for most regions it doesn't need to be computed *at all*.

It seems that at least with this configuration the `MemRegion::getAsOffset` was called every single time.

Please have a look at the patch file, and the results - I might messed something up :s

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D86295/new/

https://reviews.llvm.org/D86295