public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug libstdc++/101825] New: Atomic wait returns without the value having changed
@ 2021-08-09  7:47 tobias.schmidt at in dot tum.de
  2021-08-09  9:59 ` [Bug libstdc++/101825] " redi at gcc dot gnu.org
  2022-07-28 16:06 ` redi at gcc dot gnu.org
  0 siblings, 2 replies; 3+ messages in thread
From: tobias.schmidt at in dot tum.de @ 2021-08-09  7:47 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101825

            Bug ID: 101825
           Summary: Atomic wait returns without the value having changed
           Product: gcc
           Version: 11.1.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libstdc++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: tobias.schmidt at in dot tum.de
  Target Milestone: ---

We noticed a possible bug in gcc for std::atomic<T>::wait, where the function
returns with the atomic still having the old value.

The following combination of notify_all and wait is used to execute
initialization code in one thread while the other threads wait:

auto expected = Empty;
// Only one thread should do the initialization
if (state.compare_exchange_strong(expected, Initialize)) {
    // Do some work

    state.store(Ready);
    state.notify_all();
}
// Other threads wait here
state.wait(Initialize);

State is a 8-bit atomic value initialized with Empty and it's value always
changes from Empty to Initialize and then to Ready. Except the given code
snippet we never write to the value.

Very rarely one of the waiting threads leaves the wait function while the state
value is still Initialize. The bug seems to depend on the time it takes to
perform the work, i.e., how long the other threads have to wait. Therefore, we
were not able to reproduce the bug in a smaller example.


However, replacing state.wait(Initialize) with 

while (variant.state.load() == Initialize) {}

solves the problem. Thus, we suspect that the wait function returns at some
point without checking the state's value first.


GCC Version:
Using built-in specs.
COLLECT_GCC=gcc-11
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/11/lto-wrapper
OFFLOAD_TARGET_NAMES=nvptx-none:amdgcn-amdhsa
OFFLOAD_TARGET_DEFAULT=1
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Ubuntu
11.1.0-1ubuntu1~21.04' --with-bugurl=file:///usr/share/doc/gcc-11/README.Bugs
--enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --prefix=/usr
--with-gcc-major-version-only --program-suffix=-11
--program-prefix=x86_64-linux-gnu- --enable-shared --enable-linker-build-id
--libexecdir=/usr/lib --without-included-gettext --enable-threads=posix
--libdir=/usr/lib --enable-nls --enable-bootstrap --enable-clocale=gnu
--enable-libstdcxx-debug --enable-libstdcxx-time=yes
--with-default-libstdcxx-abi=new --enable-gnu-unique-object
--disable-vtable-verify --enable-plugin --enable-default-pie --with-system-zlib
--enable-libphobos-checking=release --with-target-system-zlib=auto
--enable-objc-gc=auto --enable-multiarch --disable-werror --enable-cet
--with-arch-32=i686 --with-abi=m64 --with-multilib-list=m32,m64,mx32
--enable-multilib --with-tune=generic
--enable-offload-targets=nvptx-none=/build/gcc-11-RPS7jb/gcc-11-11.1.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-RPS7jb/gcc-11-11.1.0/debian/tmp-gcn/usr
--without-cuda-driver --enable-checking=release --build=x86_64-linux-gnu
--host=x86_64-linux-gnu --target=x86_64-linux-gnu
--with-build-config=bootstrap-lto-lean --enable-link-serialization=2
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 11.1.0 (Ubuntu 11.1.0-1ubuntu1~21.04)

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [Bug libstdc++/101825] Atomic wait returns without the value having changed
  2021-08-09  7:47 [Bug libstdc++/101825] New: Atomic wait returns without the value having changed tobias.schmidt at in dot tum.de
@ 2021-08-09  9:59 ` redi at gcc dot gnu.org
  2022-07-28 16:06 ` redi at gcc dot gnu.org
  1 sibling, 0 replies; 3+ messages in thread
From: redi at gcc dot gnu.org @ 2021-08-09  9:59 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101825

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2021-08-09

--- Comment #1 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Is this PR 100334? Please check 11.2

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [Bug libstdc++/101825] Atomic wait returns without the value having changed
  2021-08-09  7:47 [Bug libstdc++/101825] New: Atomic wait returns without the value having changed tobias.schmidt at in dot tum.de
  2021-08-09  9:59 ` [Bug libstdc++/101825] " redi at gcc dot gnu.org
@ 2022-07-28 16:06 ` redi at gcc dot gnu.org
  1 sibling, 0 replies; 3+ messages in thread
From: redi at gcc dot gnu.org @ 2022-07-28 16:06 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101825

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> ---
(In reply to Jonathan Wakely from comment #1)
> Is this PR 100334? Please check 11.2

Ping

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2022-07-28 16:06 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-08-09  7:47 [Bug libstdc++/101825] New: Atomic wait returns without the value having changed tobias.schmidt at in dot tum.de
2021-08-09  9:59 ` [Bug libstdc++/101825] " redi at gcc dot gnu.org
2022-07-28 16:06 ` redi at gcc dot gnu.org

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).