In particular, we don't want to assume that when we sleep for N ms that will
actually take about Nms of time. Instead, record how much time it took, and
make sure that our thread metric accounting accounts for all that time, but not
any of the extra time it shouldn't account for.
Differential Revision: https://phabricator.services.mozilla.com/D59656
--HG--
extra : moz-landing-system : lando
The idea is to use the slice-accounting setup from the performance counters for
idle tasks as well, and fix the various bugs we have when nested event loops
are involved.
The bit in nsThread::SizeOfEventQueues that counted memory for
mCurrentPerformanceCounter was wrong all along, I think: the docgroup that owns
the performance counter should account for it.
Differential Revision: https://phabricator.services.mozilla.com/D59596
--HG--
extra : moz-landing-system : lando
The increments are incrementing the pointers, not the values.
This was caught by a GCC warning.
Depends on D59646
Differential Revision: https://phabricator.services.mozilla.com/D59647
--HG--
extra : moz-landing-system : lando
This removes various unused `#include "nsAutoPtr.h"` in `xpcom/`. Additionally
adds a few includes to the media stack.
Differential Revision: https://phabricator.services.mozilla.com/D58282
--HG--
extra : moz-landing-system : lando
The inclusions were removed with the following very crude script and the
resulting breakage was fixed up by hand. The manual fixups did either
revert the changes done by the script, replace a generic header with a more
specific one or replace a header with a forward declaration.
find . -name "*.idl" | grep -v web-platform | grep -v third_party | while read path; do
interfaces=$(grep "^\(class\|interface\).*:.*" "$path" | cut -d' ' -f2)
if [ -n "$interfaces" ]; then
if [[ "$interfaces" == *$'\n'* ]]; then
regexp="\("
for i in $interfaces; do regexp="$regexp$i\|"; done
regexp="${regexp%%\\\|}\)"
else
regexp="$interfaces"
fi
interface=$(basename "$path")
rg -l "#include.*${interface%%.idl}.h" . | while read path2; do
hits=$(grep -v "#include.*${interface%%.idl}.h" "$path2" | grep -c "$regexp" )
if [ $hits -eq 0 ]; then
echo "Removing ${interface} from ${path2}"
grep -v "#include.*${interface%%.idl}.h" "$path2" > "$path2".tmp
mv -f "$path2".tmp "$path2"
fi
done
fi
done
Differential Revision: https://phabricator.services.mozilla.com/D55444
--HG--
extra : moz-landing-system : lando
Most of these tests have been disabled for a long time; they run well
in the current test environment.
Differential Revision: https://phabricator.services.mozilla.com/D46642
--HG--
extra : moz-landing-system : lando
EXIT_FAILURE is 'implementation defined' but can be defined to be 1.
In our case, pingsender exits with EXIT_FAILURE but nsIProcess wasn't
reporting it as failure because it thought failures were always negative.
Differential Revision: https://phabricator.services.mozilla.com/D45038
--HG--
extra : moz-landing-system : lando
This matches the requirement imposed by the Container concept that its iterator type
should be convertible to its const_iterator type.
Differential Revision: https://phabricator.services.mozilla.com/D44673
--HG--
extra : moz-landing-system : lando
This condition is intermittently failing, but it's hard to tell from the
failure which part of the condition is failing. Split the condition
apart so ideally we'll get some better error messages. While we're
doing that, use a more explicit `EXPECT_LT` for the first half of the
condition for even better error messages.
Differential Revision: https://phabricator.services.mozilla.com/D40051
--HG--
extra : moz-landing-system : lando
Gtest jobs on the osx1014 try hardware are taking longer than anticipated to finish up, and ThreadPool.ShutdownWithTimeout tests need to wait an extra 50ms to let all threads finish on time. On the long term, when the underlying issue is fixed thread timeouts will need to be readjusted back to 300ms.
Differential Revision: https://phabricator.services.mozilla.com/D38452
--HG--
extra : moz-landing-system : lando
Giving each logical test its own counter variable will help tests to not
interfere with each other. We also make one infinite looping thread
exit so we're not chewing CPU after these tests finish.
Differential Revision: https://phabricator.services.mozilla.com/D36522
--HG--
extra : moz-landing-system : lando
Increased slack boundaries to account for issues with timer bounds in macos 1014. Lower bound slack = 60ms and upper bound slack = 1200ms. The intent is to get gtests working in the short term while timer delays are fixed, then roll the slack boundary back to 30ms.
Differential Revision: https://phabricator.services.mozilla.com/D36419
--HG--
extra : moz-landing-system : lando
The first test does not really need to check that we spend less than
200ms, and we can give more time to the scheduler in the second
test, as long as we check that is less than the recursive dispatched
event
Differential Revision: https://phabricator.services.mozilla.com/D34988
--HG--
extra : moz-landing-system : lando
Boris Zbarsky
Emilio Cobos Álvarez
Eric Rahm
Nathan Froyd
emalysz
Simon Giesecke
Andreas Farre
Mihai Alexandru Michis
Gabriele Svelto
Kershaw Chang
Cosmin Sabou
Dorel Luca
Noemi Erli
Sylvestre Ledru
Geoff Brown
Henri Sivonen
Tom Ritter
Razvan Maries
Tom Ritter
Coroiu Cristina
Hiroyuki Ikezoe
Ciure Andrei
Kristen Wright
Barret Rennie
Victor Porof
Tarek Ziadé
Andrea Marchesini
Dave Townsend
Nicklas Boman