[Bug sanitizer/112981] New: [13/14 Regression] Big compile time and run time regression in libasan with g:f732bf6a603721f61102a08ad2d023c7c2670870

[Bug sanitizer/112981] New: [13/14 Regression] Big compile time and run time regression in libasan with g:f732bf6a603721f61102a08ad2d023c7c2670870

[tnfchris at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112981

            Bug ID: 112981
           Summary: [13/14 Regression] Big compile time and run time
                    regression in libasan with
                    g:f732bf6a603721f61102a08ad2d023c7c2670870
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Keywords: compile-time-hog
          Severity: normal
          Priority: P3
         Component: sanitizer
          Assignee: unassigned at gcc dot gnu.org
          Reporter: tnfchris at gcc dot gnu.org
                CC: dodji at gcc dot gnu.org, dvyukov at gcc dot gnu.org,
                    jakub at gcc dot gnu.org, kcc at gcc dot gnu.org, marxin at gcc dot gnu.org
  Target Milestone: ---
            Target: aarch64-*

Starting with the sync with upstream in
g:f732bf6a603721f61102a08ad2d023c7c2670870

f732bf6a603721f61102a08ad2d023c7c2670870 is the first bad commit
commit f732bf6a603721f61102a08ad2d023c7c2670870
Author: Martin Liska <mliska@suse.cz>
Date:   Tue May 3 12:56:26 2022 +0200

    libsanitizer: merge from upstream
(0a1bcab9f3bf75c4c5d3e53bafb3eeb80320af46).

there's a massive compile time and performance overhead (6x slower) when using
-fsanitize=address.

The following testcase (due to the dependencies provided as cmake project)
illustrates this:

-- main.cpp:

#include <catch2/generators/catch_generators.hpp>
#include <catch2/catch_test_macros.hpp>

TEST_CASE("Test") {
  for (int i= 0; i<1000000; i++) {
    REQUIRE(1 + 2 == 3);
  }
}

-- CMakeList.txt

cmake_minimum_required(VERSION 3.5)
project(slow_asan)

add_compile_options(-fsanitize=address)
add_link_options(-fsanitize=address)

Include(FetchContent)
FetchContent_Declare(
    Catch2
    SYSTEM
    GIT_REPOSITORY https://github.com/catchorg/Catch2.git
    GIT_TAG v3.3.0
)

FetchContent_MakeAvailable(Catch2)

add_executable(tests main.cpp)
target_link_libraries(tests PRIVATE Catch2::Catch2WithMain)

--- command

cmake -DCMAKE_C_COMPILER=${install_dir}/bin/gcc
-DCMAKE_CXX_COMPILER=${install_dir}/bin/g++ -DCMAKE_BUILD_TYPE=Release . &&
make VERBOSE=1 -j && timeout 10.0s ./tests

will take significantly longer than in GCC 12 (compiles at -O3).

I have not yet tracked down if the compile hog is a GCC issue or not.

[Bug sanitizer/112981] [13/14 Regression] Big compile time and run time regression in libasan with g:f732bf6a603721f61102a08ad2d023c7c2670870

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112981

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING
   Last reconfirmed|                            |2023-12-12
     Ever confirmed|0                           |1

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Can you provide the preprocessed source before and after the commit? There
should be no compile time change due to that commit except link time.

[Bug sanitizer/112981] [13/14 Regression] Big compile time and run time regression in libasan with g:f732bf6a603721f61102a08ad2d023c7c2670870

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112981

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           See Also|                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=110835

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
My bet this is a dup of bug 110835

[Bug sanitizer/112981] [13/14 Regression] Big compile time and run time regression in libasan with g:f732bf6a603721f61102a08ad2d023c7c2670870

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112981

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |13.3

[Bug sanitizer/112981] [13/14 Regression] Big compile time and run time regression in libasan with g:f732bf6a603721f61102a08ad2d023c7c2670870

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112981

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |RESOLVED
         Resolution|---                         |DUPLICATE

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Yes this is a dup.

*** This bug has been marked as a duplicate of bug 110835 ***

[Bug sanitizer/112981] [13/14 Regression] Big compile time and run time regression in libasan with g:f732bf6a603721f61102a08ad2d023c7c2670870

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112981

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #3)
> Yes this is a dup.

I looked into what catch2 was doing and yes it is doing the whole throw and
catch (and using rethrow_exception too in some cases).