libstdc++: async: tolerate slightly shorter sleep
Commit Message
Even without frequent signals interrupting nanosleep, sleep_for on
rtems on qemu wakes up too early by a predictable margin of less than
0,3%, which some async tests complain about the too-short wait times.
Allow the tests to tolerate a little sleep deprivation.
Regstrapped on x86_64-linux-gnu, also tested with a cross to
aarch64-rtems6. Ok to install?
for libstdc++-v3/ChangeLog
* testsuite/30_threads/async/async.cc: Tolerate early wakeup.
---
libstdc++-v3/testsuite/30_threads/async/async.cc | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
Comments
On Jun 22, 2022, Alexandre Oliva <oliva@adacore.com> wrote:
> Regstrapped on x86_64-linux-gnu, also tested with a cross to
> aarch64-rtems6. Ok to install?
The early wakeups are fixed for rtems6.1, so the same question raised at
https://gcc.gnu.org/pipermail/gcc-patches/2022-June/597102.html apply to
this one:
libstdc++: xfail nanosleep tests on rtems
From: Alexandre Oliva <oliva@adacore.com>
Since it has been determined that nanosleep may return slightly too
early on RTEMS, due to clock resolution differences, expect
30_thread/async tests that have detected too-early wakeups to fail on
RTEMS targets.
for libstdc++-v3/ChangeLog
* testsuite/30_threads/async/async.cc: xfail on RTEMS.
TN: V608-048
---
libstdc++-v3/testsuite/30_threads/async/async.cc | 1 +
1 file changed, 1 insertion(+)
diff --git a/libstdc++-v3/testsuite/30_threads/async/async.cc b/libstdc++-v3/testsuite/30_threads/async/async.cc
index 38943ff1a9a5e..e0b731186c459 100644
--- a/libstdc++-v3/testsuite/30_threads/async/async.cc
+++ b/libstdc++-v3/testsuite/30_threads/async/async.cc
@@ -2,6 +2,7 @@
// { dg-additional-options "-pthread" { target pthread } }
// { dg-require-effective-target c++11 }
// { dg-require-gthreads "" }
+// { dg-xfail-if "nanosleep may wake up too early" { *-*-rtems* } }
// Copyright (C) 2010-2022 Free Software Foundation, Inc.
//
On Thu, 23 Jun 2022 at 12:38, Alexandre Oliva via Libstdc++
<libstdc++@gcc.gnu.org> wrote:
>
> On Jun 22, 2022, Alexandre Oliva <oliva@adacore.com> wrote:
>
> > Regstrapped on x86_64-linux-gnu, also tested with a cross to
> > aarch64-rtems6. Ok to install?
>
> The early wakeups are fixed for rtems6.1, so the same question raised at
> https://gcc.gnu.org/pipermail/gcc-patches/2022-June/597102.html apply to
> this one:
Looks like I never reviewed this one, sorry.
The patch to xfail this test for rtems is OK.
>
> libstdc++: xfail nanosleep tests on rtems
>
> From: Alexandre Oliva <oliva@adacore.com>
>
> Since it has been determined that nanosleep may return slightly too
> early on RTEMS, due to clock resolution differences, expect
> 30_thread/async tests that have detected too-early wakeups to fail on
> RTEMS targets.
>
>
> for libstdc++-v3/ChangeLog
>
> * testsuite/30_threads/async/async.cc: xfail on RTEMS.
>
> TN: V608-048
> ---
> libstdc++-v3/testsuite/30_threads/async/async.cc | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/libstdc++-v3/testsuite/30_threads/async/async.cc b/libstdc++-v3/testsuite/30_threads/async/async.cc
> index 38943ff1a9a5e..e0b731186c459 100644
> --- a/libstdc++-v3/testsuite/30_threads/async/async.cc
> +++ b/libstdc++-v3/testsuite/30_threads/async/async.cc
> @@ -2,6 +2,7 @@
> // { dg-additional-options "-pthread" { target pthread } }
> // { dg-require-effective-target c++11 }
> // { dg-require-gthreads "" }
> +// { dg-xfail-if "nanosleep may wake up too early" { *-*-rtems* } }
>
> // Copyright (C) 2010-2022 Free Software Foundation, Inc.
> //
>
>
> --
> Alexandre Oliva, happy hacker https://FSFLA.org/blogs/lxo/
> Free Software Activist GNU Toolchain Engineer
> Disinformation flourishes because many people care deeply about injustice
> but very few check the facts. Ask me about <https://stallmansupport.org>
>
On Wed, 12 Oct 2022 at 12:41, Jonathan Wakely wrote:
>
> On Thu, 23 Jun 2022 at 12:38, Alexandre Oliva via Libstdc++
> <libstdc++@gcc.gnu.org> wrote:
> >
> > On Jun 22, 2022, Alexandre Oliva <oliva@adacore.com> wrote:
> >
> > > Regstrapped on x86_64-linux-gnu, also tested with a cross to
> > > aarch64-rtems6. Ok to install?
> >
> > The early wakeups are fixed for rtems6.1, so the same question raised at
> > https://gcc.gnu.org/pipermail/gcc-patches/2022-June/597102.html apply to
> > this one:
>
> Looks like I never reviewed this one, sorry.
>
> The patch to xfail this test for rtems is OK.
It's also fine if you just want to drop this patch for the same reason
as https://gcc.gnu.org/pipermail/gcc-patches/2022-June/597105.html
(I'm just going through old patch submissions that never got acked or
nacked and this was one of them.)
On Oct 12, 2022, Jonathan Wakely <jwakely@redhat.com> wrote:
> On Wed, 12 Oct 2022 at 12:41, Jonathan Wakely wrote:
>>
>> On Thu, 23 Jun 2022 at 12:38, Alexandre Oliva via Libstdc++
>> <libstdc++@gcc.gnu.org> wrote:
>> >
>> > On Jun 22, 2022, Alexandre Oliva <oliva@adacore.com> wrote:
>> >
>> > > Regstrapped on x86_64-linux-gnu, also tested with a cross to
>> > > aarch64-rtems6. Ok to install?
>> >
>> > The early wakeups are fixed for rtems6.1, so the same question raised at
>> > https://gcc.gnu.org/pipermail/gcc-patches/2022-June/597102.html apply to
>> > this one:
>>
>> Looks like I never reviewed this one, sorry.
>>
>> The patch to xfail this test for rtems is OK.
> It's also fine if you just want to drop this patch for the same reason
> as https://gcc.gnu.org/pipermail/gcc-patches/2022-June/597105.html
Yeah, nanosleep is fixed, no need for this one, thanks, withdrawn.
@@ -104,7 +104,8 @@ void test03()
VERIFY( status == std::future_status::ready );
auto const elapsed = CLOCK::now() - start;
- VERIFY( elapsed >= std::chrono::seconds(2) );
+ auto const tolerance = std::chrono::milliseconds(6);
+ VERIFY( elapsed + tolerance >= std::chrono::seconds(2) );
VERIFY( elapsed < std::chrono::seconds(5) );
}
@@ -169,7 +170,8 @@ void test_pr91486_wait_for()
auto status = f1.wait_for(wait_time);
auto const elapsed_steady = chrono::steady_clock::now() - start_steady;
- VERIFY( elapsed_steady >= std::chrono::seconds(1) );
+ auto const tolerance = std::chrono::milliseconds(3);
+ VERIFY( elapsed_steady + tolerance >= std::chrono::seconds(1) );
}
// This is a clock with a very recent epoch which ensures that the difference
@@ -222,7 +224,8 @@ void test_pr91486_wait_until()
auto const elapsed_steady = chrono::steady_clock::now() - start_steady;
// This checks that we didn't come back too soon
- VERIFY( elapsed_steady >= std::chrono::seconds(1) );
+ auto const tolerance = std::chrono::milliseconds(3);
+ VERIFY( elapsed_steady + tolerance >= std::chrono::seconds(1) );
// This checks that wait_until didn't busy wait checking the clock more times
// than necessary.