[Bug tree-optimization/107895] New: mt19937 bad performance on LP64

[Bug tree-optimization/107895] New: mt19937 bad performance on LP64

[jengelh at inai dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107895

            Bug ID: 107895
           Summary: mt19937 bad performance on LP64
           Product: gcc
           Version: 12.2.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jengelh at inai dot de
  Target Milestone: ---

Input
=====

#include <random>
#include <unistd.h>
static std::mt19937 rng;
int main() {
        for (size_t j = 0; j < 0x8000; ++j) {
                uint32_t numbers[65536];
                for (size_t i = 0; i < std::size(numbers); ++i)
                        numbers[i] = rng();
                // ensure number generation is not all optimized away
                write(STDOUT_FILENO, numbers, sizeof(numbers));
        }
}


Observed
========

Target: x86_64-suse-linux
gcc version 12.2.1 20221020 [revision 0aaef83351473e8f4eb774f8f999bbe87a4866d7]
(SUSE Linux)

$ g++ x.cpp -O2 && time ./a.out >/dev/zero

                  -m32    -m64
===============  =====  ======
std::mt19937      3.9s   11.5s
std::mt19937_64  14.0s   11.6s
===============  =====  ======
error ±0.1s

With -ftree-loop-if-convert [Bug #80520], but still not at -m32 levels:

+-ftree-          -m32    -m64
===============  =====  ======
std::mt19937      3.9s    5.2s
std::mt19937_64  14.0s    5.4s
===============  =====  ======
error ±0.1s


Expected
========

Expected to see <= 4.7s on -m64 at all times. (3.9 + ~20% margin for wider
transfers CPU<->caches/RAM)

The -m64 versions should have somewhat equal runtime or faster runtime (because
more registers, more opportunities); concerns like https://gmplib.org/32vs64
apply to old CPUs, but I do not think it's indicative of how contemporary
x86_64 systems perform.


Additional information
======================

CPUs:
 "11th Gen Intel(R) Core(TM) i5-1135G7 @ 2.40GHz"
    [fam 6 model 140 stepping 1 microcode 0xa4] and
 "AMD Ryzen 7 3700X 8-Core Processor"
    [fam 23 model 113 stepping 0 microcode 0x8701013]
    (about 3.0 and 10.2 seconds runtime, respectively)

[Bug tree-optimization/107895] mt19937 bad performance on LP64

[jengelh at inai dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107895

--- Comment #1 from Jan Engelhardt <jengelh at inai dot de> ---
clang-15.0.5+gnustdlibc timing distribution.

            -m32  -m64
mt19937      6.0   4.7
mt19937_64   9.2   4.7

[Bug tree-optimization/107895] mt19937 bad performance on LP64

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107895

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |missed-optimization
             Target|                            |x86_64-*-*

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Sounds like an if-conversion issue, thus RTL?  Btw, we inline a wrapper
doing

  if (..->_M_p > 623)
    mersenne_twister_engine ();
  else
   {
     some inlined stuff, incrementing _M_p
   }

but the inlined stuff is already fully if-converted and thus not the
timing critical part?

[Bug tree-optimization/107895] mt19937 bad performance on LP64

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107895

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
                 CC|                            |marxin at gcc dot gnu.org
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2022-11-28

--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
Btw. the benchmark was sped up by r13-739-g793f847ba7dbe763 from 8.4 -> 5.3s on
AMD Ryzen 9 5950X 16-Core Processor with -O2 options.