Created attachment 1827258 [details] crash.png Description of problem: gmpopenh264 crashes for all videos on all sites e.g. https://tekeye.uk/html/html5-video-test-page Version-Release number of selected component (if applicable): firefox-92.0.1-1.fc35.x86_64.rpm How reproducible: Steps to Reproduce: 1. open any h264 video e.g. https://tekeye.uk/html/html5-video-test-page 2. gmpopenh264 crashes
openh264.x86_64 2.1.1-2.fc35 @fedora-cisco-openh264
This is waiting on https://pagure.io/releng/issue/10294
I think this should be a 0-day blocker (as openh264 is not on the install image, but we'd need the fix available for the release). Moving it to openh264 and proposing as a blocker so that we can track it for the release.
This is currently waiting on Cisco to upload the binaries: 16:07 < mboddu> kalev: The person responsible for it on PTO last few weeks and she should be back next week, so waiting for her to come back from PTO 16:07 < kalev> mboddu: ahh, thanks 16:08 < mboddu> kalev: FYI, I will be on PTO on Monday and maybe on Tuesday too, so... it might take longer 16:09 * kalev nods.
*** Bug 2008455 has been marked as a duplicate of this bug. ***
I cannot confirm this behaviour as all videos on the mentioned page are working normally. I have the same versions of the packages: * openh264-2.1.1-2.fc35.x86_64 * firefox-92.0.1-1.fc35.x86_64
Do you have mozilla-openh264 installed as well?
I only see this on my AMD (with amdgpu) box. I wonder if it's a hardware acceleration thing... openh264-2.1.1-2.fc35.x86_64 firefox-93.0-1.fc35.x86_64
I tested here and on the site you pointed I opened all videos succesfully https://tekeye.uk/html/html5-video-test-page Operating System: Fedora Linux 35 KDE Plasma Version: 5.22.5 KDE Frameworks Version: 5.87.0 Qt Version: 5.15.2 Kernel Version: 5.14.10-300.fc35.x86_64 (64-bit) Graphics Platform: Wayland Processors: 8 × Intel® Core™ i7-3632QM CPU @ 2.20GHz Memory: 15.4 GiB of RAM Graphics Processor: Mesa Intel® HD Graphics 4000 firefox-93.0-1.fc35.x86_64 openh264-2.1.1-2.fc35.x86_64
complement my comment: rpm -qa|grep openh264 openh264-2.1.1-2.fc35.x86_64 mozilla-openh264-2.1.1-2.fc35.x86_64 gstreamer1-plugin-openh264-1.18.2-1.fc35.x86_64
(In reply to Lukas Ruzicka from comment #6) > I cannot confirm this behaviour as all videos on the mentioned page are > working normally. I have the same versions of the packages: > > * openh264-2.1.1-2.fc35.x86_64 > * firefox-92.0.1-1.fc35.x86_64 I see a bunch of crashes with the same package versions. The first test video on the test page does not play because the Firefox plugin container is crashing. FWIW I also have amdgpu, although I'm not sure why that would be relevant since there should be no GPU access involved here. Please make sure you don't have rpmfusion enabled. I imagine it could use codecs from there in preference to openh264. Core was generated by `/usr/lib64/firefox/plugin-container /usr/lib64/mozilla/plugins/gmp-gmpopenh264/'. Program terminated with signal SIGSEGV, Segmentation fault. #0 mozilla::gmp::GMPChild::ProcessingError (this=<optimized out>, aCode=<optimized out>, aReason=<optimized out>) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/dom/media/gmp/GMPChild.cpp:605 Downloading source file /usr/src/debug/firefox-92.0-3.fc35.x86_64/dom/media/gmp/GMPChild.cpp... 605 MOZ_CRASH("aborting because of MsgProcessingError"); [Current thread is 1 (Thread 0x7f80f58960c0 (LWP 30884))] (gdb) bt full #0 mozilla::gmp::GMPChild::ProcessingError(mozilla::ipc::HasResultCodes::Result, char const*) (this=<optimized out>, aCode=<optimized out>, aReason=<optimized out>) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/dom/media/gmp/GMPChild.cpp:605 #1 0x00007f80f8c18ed1 in mozilla::ipc::IPCResult::Fail(mozilla::NotNull<mozilla::ipc::IProtocol*>, char const*, char const*) (actor=actor@entry=..., where=where@entry=0x7f80fdeaa2a7 "AnswerStartPlugin", why=why@entry=0x7f80fdeaa2e4 "Failed to load GMP.") at /usr/src/debug/firefox-92.0-3.fc35.x86_64/objdir/dist/include/nsTString.h:161 errorMsg = {<nsTAutoStringN<char, 16>> = {<nsTString<char>> = {<nsTSubstring<char>> = {<mozilla::detail::nsTStringRepr<char>> = {mData = 0x7f80f562ae88 "AnswerStartPlugin Failed to load GMP.\n", mLength = 38, mDataFlags = 5, mClassFlags = 3}, static kMaxCapacity = 2147483637}, <No data fields>}, mInlineCapacity = 15, mStorage = "\000\177\000\000\272ΤnqU\000\000\067\000\000"}, <No data fields>} #2 0x00007f80f9986b51 in mozilla::gmp::GMPChild::AnswerStartPlugin(nsTString<char16_t> const&) (this=this@entry=0x7f80f565a820, aAdapter=...) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/objdir/dist/include/mozilla/NotNull.h:225 __FUNCTION__ = "AnswerStartPlugin" libPath = {<nsTSubstring<char>> = {<mozilla::detail::nsTStringRepr<char>> = {mData = 0x7f80f568bf08 "/usr/lib64/mozilla/plugins/gmp-gmpopenh264/system-installed/libgmpopenh264.so", mLength = 77, mDataFlags = 5, mClassFlags = mozilla::detail::StringClassFlags::NULL_TERMINATED}, static kMaxCapacity = 2147483637}, <No data fields>} __func__ = "AnswerStartPlugin" platformAPI = 0x7f80f562aa40 adapter = 0x0 #3 0x00007f80f8cb07af in mozilla::gmp::PGMPChild::OnCallReceived(IPC::Message const&, IPC::Message*&) (this=0x7f80f565a820, msg__=..., reply__=@0x7fff55bbe830: 0x0) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/objdir/ipc/ipdl/PGMPChild.cpp:644 raiiObject628 = {mProfilingStack = 0x7f80f5625400} iter__ = {iter_ = {mSegment = 0, mData = 0x7f80f562e5ac "", mDataEnd = 0x7f80f562e5ac "", mAbsoluteOffset = 44}, start_ = {mValue = 10852276143935}} adapter = {<nsTSubstring<char16_t>> = {<mozilla::detail::nsTStringRepr<char16_t>> = {mData = 0x7f80fcd92b84 <gNullChar> u"", mLength = 0, mDataFlags = mozilla::detail::StringDataFlags::TERMINATED, mClassFlags = mozilla::detail::StringClassFlags::NULL_TERMINATED}, <No data fields>}, <No data fields>} id__ = 2147483647 route__ = <optimized out> #4 0x00007f80f8c16a88 in mozilla::ipc::MessageChannel::DispatchInterruptMessage(mozilla::ipc::ActorLifecycleProxy*, IPC::Message&&, unsigned long) (this=this@entry=0x7f80f565a8e0, aProxy=aProxy@entry=0x7f80f5625520, aMsg=..., stackDepth=stackDepth@entry=0) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/objdir/dist/include/mozilla/UniquePtrExtensions.h:182 reply = {mTuple = {<mozilla::detail::CompactPairHelper<IPC::Message*, mozilla::DefaultDelete<IPC::Message>, (mozilla::detail::StorageType)1, (mozilla::detail::StorageType)0>> = {<mozilla::DefaultDelete<IPC::Message>> = {<No data fields>}, mFirstA = 0x0}, <No data fields>}} rv = <optimized out> lock = {mMonitor = <optimized out>} #5 0x00007f80f8c0ecd9 in mozilla::ipc::MessageChannel::DispatchMessage(IPC::Message&&) (this=0x7f80f565a8e0, aMsg=...) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/glue/MessageChannel.cpp:1976 unlock = {mMonitor = 0x7f80f5657190} transaction = {mChan = 0x7f80f565a8e0, mActive = false, mOutgoing = false, mNestedLevel = 1, mSeqno = 2, mTransaction = 0, mNext = 0x0, mReply = {mTuple = {<mozilla::detail::CompactPairHelper<IPC::Message*, mozilla::DefaultDelete<IPC::Message>, (mozilla::detail::StorageType)1, (mozilla::detail::StorageType)0>> = {<mozilla::DefaultDelete<IPC::Message>> = {<No data fields>}, mFirstA = 0x0}, <No data fields>}}} id = 0 listenerProxy = {mRawPtr = 0x7f80f5625520} nojsapi = {<mozilla::detail::MaybeStorage<mozilla::dom::AutoNoJSAPI, false>> = {<mozilla::detail::MaybeStorageBase<mozilla::dom::AutoNoJSAPI, false>> = {mStorage = {val = {<mozilla::dom::ScriptSettingsStackEntry> = {mGlobalObject = {<nsCOMPtr_base> = {mRawPtr = 0x7f80f5700018}, <No data fields>}, mType = 4117757976, mOlder = 0x7f80f5600000}, <JSAutoNullableRealm> = {cx_ = 0x55716ea48880 <free(void*)+352>, oldRealm_ = 0x2e940}, mCx = 0x7f80f57001e8, mExecutionYield = {mExecutionGrantingManager = {mRawPtr = 0x0}}}}}, mIsSome = 0 '\000'}, <mozilla::detail::Maybe_CopyMove_Enabler<mozilla::dom::AutoNoJSAPI, false, true, true>> = {<No data fields>}, <No data fields>} reply = {mTuple = {<mozilla::detail::CompactPairHelper<IPC::Message*, mozilla::DefaultDelete<IPC::Message>, (mozilla::detail::StorageType)1, (mozilla::detail::StorageType)0>> = {<mozilla::DefaultDelete<IPC::Message>> = {<No data fields>}, mFirstA = 0x0}, <No data fields>}} #6 0x00007f80faaee820 in mozilla::ipc::MessageChannel::RunMessage(mozilla::ipc::MessageChannel::MessageTask&) (aTask=..., this=0x7f80f565a8e0) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/glue/MessageChannel.cpp:1826 msg = @0x7f80f568e5d0: {<mojo::core::ports::UserMessage> = {_vptr.UserMessage = 0x7f80ff3b81a0 <vtable for IPC::Message+16>, type_info_ = 0x7f80fdc9cd29 <IPC::Message::kUserMessageTypeInfo>}, <Pickle> = {static kSegmentAlignment = 8, buffers_ = {<InfallibleAllocPolicy> = {<No data fields>}, mOwning = true, mSegments = {<InfallibleAllocPolicy> = {<No data fields>}, static kElemIsPod = false, static kMaxInlineBytes = 999, static kInlineCapacity = 1, mBegin = 0x7f80f568e600, mLength = 1, mTail = {<mozilla::Vector<mozilla::BufferList<InfallibleAllocPolicy>::Segment, 1, InfallibleAllocPolicy>::CapacityAndReserved> = {mCapacity = 1}, mBytes = "\200\345b\365\200\177\000\000,\000\000\000\000\000\000\000X\000\000\000\000\000\000"}}, mSize = 44, mStandardCapacity = 4096}, header_ = 0x7f80f562e580, header_size_ = 36}, static kUserMessageTypeInfo = {<No data fields>}, file_descriptor_set_ = {mRawPtr = 0x0}, attached_ports_ = {<nsTArray_Impl<mozilla::ipc::ScopedPort, nsTArrayInfallibleAllocator>> = {<nsTArray_base<nsTArrayInfallibleAllocator, nsTArray_RelocateUsingMemutils>> = {mHdr = 0x7f80fcdb3690 <sEmptyTArrayHeader>}, <nsTArray_TypedBase<mozilla::ipc::ScopedPort, nsTArray_Impl<mozilla::ipc::ScopedPort, nsTArrayInfallibleAllocator> >> = {<nsTArray_SafeElementAtHelper<mozilla::ipc::ScopedPort, nsTArray_Impl<mozilla::ipc::ScopedPort, nsTArrayInfallibleAllocator> >> = {<detail::nsTArray_CopyDisabler> = {<No data fields>}, <No data fields>}, <No data fields>}, <No data fields>}, <No data fields>}, create_time_ = {mValue = 0}} lock = {mMonitor = 0x7f80f5657190} #7 mozilla::ipc::MessageChannel::MessageTask::Run() (this=0x7f80f568e580) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/glue/MessageChannel.cpp:1857 lock = {mMonitor = 0x7f80f5657190} #8 0x00007f80faae41e0 in MessageLoop::RunTask(already_AddRefed<nsIRunnable>) (aTask=..., this=0x7fff55bbeb80) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/chromium/src/base/message_loop.cc:454 log = {mWillRunAgain = false} pending_task = {task = {<nsCOMPtr_base> = {mRawPtr = 0x0}, <No data fields>}, delayed_run_time = {ticks_ = 0}, sequence_num = 0, nestable = true} #9 MessageLoop::DeferOrRunPendingTask(MessageLoop::PendingTask&&) (pending_task=..., this=<optimized out>) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/chromium/src/base/message_loop.cc:463 pending_task = {task = {<nsCOMPtr_base> = {mRawPtr = 0x0}, <No data fields>}, delayed_run_time = {ticks_ = 0}, sequence_num = 0, nestable = true} #10 MessageLoop::DoWork() (this=0x7fff55bbeb80) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/chromium/src/base/message_loop.cc:538 pending_task = {task = {<nsCOMPtr_base> = {mRawPtr = 0x0}, <No data fields>}, delayed_run_time = {ticks_ = 0}, sequence_num = 0, nestable = true} #11 0x00007f80faea22ef in base::MessagePumpDefault::Run(base::MessagePump::Delegate*) (this=0x7f80f5602910, delegate=0x7fff55bbeb80) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/chromium/src/base/message_pump_default.cc:35 did_work = <optimized out> raiiObject22 = {mProfilingStack = 0x7f80f5625400} loop = <optimized out> hangMonitor = {mThread = {mRawPtr = 0x0}, static kNoTimeout = 0} #12 0x00007f80faea052c in MessageLoop::RunInternal() (this=0x7fff55bbeb80) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/objdir/dist/include/mozilla/RefPtr.h:280 save_state = {<MessageLoop::RunState> = {run_depth = 1, quit_received = false}, loop_ = 0x7fff55bbeb80, previous_state_ = 0x0} #13 MessageLoop::RunHandler() (this=0x7fff55bbeb80) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/chromium/src/base/message_loop.cc:324 save_state = {<MessageLoop::RunState> = {run_depth = 1, quit_received = false}, loop_ = 0x7fff55bbeb80, previous_state_ = 0x0} #14 MessageLoop::Run() (this=this@entry=0x7fff55bbeb80) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/chromium/src/base/message_loop.cc:306 save_state = {<MessageLoop::RunState> = {run_depth = 1, quit_received = false}, loop_ = 0x7fff55bbeb80, previous_state_ = 0x0} #15 0x00007f80fbac9e80 in XRE_InitChildProcess(int, char**, XREChildData const*) (aArgc=2, aArgv=0x7fff55bbef38, aChildData=<optimized out>) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/toolkit/xre/nsEmbedFunctions.cpp:749 process = {mTuple = {<mozilla::detail::CompactPairHelper<mozilla::ipc::ProcessChild*, mozilla::DefaultDelete<mozilla::ipc::ProcessChild>, (mozilla::detail::StorageType)1, (mozilla::detail::StorageType)0>> = {<mozilla::DefaultDelete<mozilla::ipc::ProcessChild>> = {<No data fields>}, mFirstA = 0x7f80f565a800}, <No data fields>}} ioInterposerGuard = {<mozilla::detail::MaybeStorage<mozilla::IOInterposerInit, false>> = {<mozilla::detail::MaybeStorageBase<mozilla::IOInterposerInit, false>> = {mStorage = {val = {<No data fields>}}}, mIsSome = <optimized out>}, <mozilla::detail::Maybe_CopyMove_Enabler<mozilla::IOInterposerInit, false, true, true>> = {<No data fields>}, <No data fields>} uiMessageLoop = {<base::MessagePump::Delegate> = {_vptr.Delegate = 0x7f80ff3b7f10 <vtable for MessageLoop+16>}, type_ = MessageLoop::TYPE_DEFAULT, id_ = 1, work_queue_ = std::queue wrapping: std::deque with 0 elements, delayed_work_queue_ = std::priority_queue wrapping: std::vector of length 0, capacity 0, deferred_non_nestable_work_queue_ = std::queue wrapping: std::deque with 0 elements, pump_ = {mRawPtr = 0x7f80f5602910}, destruction_observers_ = {observers_ = std::vector of length 1, capacity 1 = {0x7f80f562df20}, notify_depth_ = 0, type_ = base::ObserverList<MessageLoop::DestructionObserver, false>::NOTIFY_ALL}, nestable_tasks_allowed_ = false, exception_restoration_ = false, thread_name_ = "", incoming_queue_ = std::queue wrapping: std::deque with 0 elements, incoming_queue_lock_ = {<mozilla::OffTheBooksMutex> = {<mozilla::detail::MutexImpl> = {platformData_ = {0x0, 0x0, 0x3, 0x0, 0x0}}, <mozilla::BlockingResourceBase> = {<No data fields>}, <No data fields>}, <No data fields>}, state_ = 0x7fff55bbeaa0, run_depth_base_ = 1, shutting_down_ = false, transient_hang_timeout_ = 0, permanent_hang_timeout_ = 0, next_sequence_num_ = 0, mEventTarget = {mRawPtr = 0x7f80f562df10}} raiiObject401 = {mProfilingStack = 0x0} raiiObject403 = {<No data fields>} raiiObject404 = {mProfilingStack = 0x7f80f5625400} exceptionHandlerIsSet = true parentPIDString = <optimized out> end = 0x7fff55bbf433 "" parentPID = <optimized out> exitManager = {lock_ = {<mozilla::OffTheBooksMutex> = {<mozilla::detail::MutexImpl> = {platformData_ = {0x0, 0x0, 0x3, 0x0, 0x0}}, <mozilla::BlockingResourceBase> = {<No data fields>}, <No data fields>}, <No data fields>}, stack_ = std::stack wrapping: std::deque with 0 elements, next_manager_ = 0x0} rv = <optimized out> uiLoopType = <optimized out> #16 0x000055716ea54a4b in content_process_main(mozilla::Bootstrap*, int, char**) (bootstrap=0x7f80f562c530, argc=4, argv=0x7fff55bbef38) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/app/../contentproc/plugin-container.cpp:57 childData = {<No data fields>} rv = <optimized out> #17 0x000055716ea46283 in main(int, char**) (argc=5, argv=0x7fff55bbef38) at /usr/src/debug/firefox-92.0-3.fc35.x86_64/ipc/app/MozillaRuntimeMain.cpp:72 ret = <optimized out>
(In reply to Michael Catanzaro from comment #11) > (In reply to Lukas Ruzicka from comment #6) > > I cannot confirm this behaviour as all videos on the mentioned page are > > working normally. I have the same versions of the packages: > > > > * openh264-2.1.1-2.fc35.x86_64 > > * firefox-92.0.1-1.fc35.x86_64 > > I see a bunch of crashes with the same package versions. The first test > video on the test page does not play because the Firefox plugin container is > crashing. FWIW I also have amdgpu, although I'm not sure why that would be > relevant since there should be no GPU access involved here. > > Please make sure you don't have rpmfusion enabled. I imagine it could use > codecs from there in preference to openh264. > > Core was generated by `/usr/lib64/firefox/plugin-container > /usr/lib64/mozilla/plugins/gmp-gmpopenh264/'. > Program terminated with signal SIGSEGV, Segmentation fault. > #0 mozilla::gmp::GMPChild::ProcessingError (this=<optimized out>, > aCode=<optimized out>, aReason=<optimized out>) at > /usr/src/debug/firefox-92.0-3.fc35.x86_64/dom/media/gmp/GMPChild.cpp:605 > Downloading source file > /usr/src/debug/firefox-92.0-3.fc35.x86_64/dom/media/gmp/GMPChild.cpp... Yes, I tested on a new VM F35 kde upgraded today here, only with oficial fedora packages, and have the same errors as you.
Created attachment 1831020 [details] allot of coredumps as trying to open test website
Created attachment 1831021 [details] coredump info trying to open videos on https://tekeye.uk/html/html5-video-test-page the first never hapen to play, the others play correctly. openh264-2.1.1-2.fc35.x86_64 mozilla-openh264-2.1.1-2.fc35.x86_64 firefox-93.0-2.fc35.x86_64
Created attachment 1831022 [details] video from webpage tests accessing https://tekeye.uk/html/html5-video-test-page
I also have [julius@ideapad5pro ~]$ rpm -qa|grep openh264 openh264-2.1.1-2.fc35.x86_64 mozilla-openh264-2.1.1-2.fc35.x86_64 gstreamer1-plugin-openh264-1.18.2-1.fc35.x86_64 and an AMD GPU (5800H)
AMD GPU here, too.
In my machine, all tested videos from the above mentioned page work normally with this setup: * Thinkpad T580 * VGA compatible controller: Intel Corporation UHD Graphics 620 (rev 07) * openh264-2.1.1-2.fc35.x86_64 * mozilla-openh264-2.1.1-2.fc35.x86_64 * gstreamer1-plugin-openh264-1.18.2-1.fc35.x86_64 * firefox-93.0-2.fc35.x86_64 I also did not see any problems on my desktop with Nvidia and Nouveau drivers, but I am not at the desktop at the moment, so I will retest for it when I am there again.
Oops, but I have rpmfusion enabled, so I will need to recreate the environment first to make sure. However, it can be only related to some graphics card as well.
Discussed during the 2021-10-11 blocker review meeting: [0] The decision to classify this bug as both an "Accepted0Day" was made by ticket vote of +8 (was not cleared before meeting due to blockerbugs webUI limitation). [0] https://meetbot.fedoraproject.org/fedora-blocker-review/2021-10-11/f35-blocker-review.2021-10-11-16.00.txt
The openh264-2.1.1-3.fc36 and openh264-2.1.1-3.fc35 have been uploaded to Cisco CDN and the work on releng side is done. So this ticket can be closed now.
I downloaded and installed everything from https://codecs.fedoraproject.org/openh264/35/x86_64/os/Packages/ and it is working now
There seems to be a performance regression with openh264-2.1.1-3.fc35, some videos have severely choppy playback. Can this be re-opened/passed back upstream for investigation?
(In reply to James from comment #23) > There seems to be a performance regression with openh264-2.1.1-3.fc35, some > videos have severely choppy playback. Can this be re-opened/passed back > upstream for investigation? Needs a new bug report, please. Unfortunately upstream is not active and we're going to have to solve any regressions ourselves. :/ I'm going to reopen anyway, though, because something is wrong. I cannot update to the new -3 version using 'sudo dnf upgrade --refresh openh264', and I don't know why. It sees -2 as the latest version, even though there are no mirrors in the metalink -- it's only codecs.fedoraproject.org -- so there shouldn't be any good reason for dnf to see stale packages. Very confused here. BTW, anyone know what is up with codecs.fedoraproject.org? Is this server actually run by Cisco, just under our domain name?
(In reply to Michael Catanzaro from comment #24) > I'm going to reopen anyway, though, because something is wrong. I cannot > update to the new -3 version using 'sudo dnf upgrade --refresh openh264', > and I don't know why. It sees -2 as the latest version, even though there > are no mirrors in the metalink -- it's only codecs.fedoraproject.org -- so > there shouldn't be any good reason for dnf to see stale packages. Very > confused here. Found bug #2008803.
(In reply to Michael Catanzaro from comment #25) > Found bug #2008803. I meant: found bug #2014514. (In reply to James from comment #23) > There seems to be a performance regression with openh264-2.1.1-3.fc35, some > videos have severely choppy playback. Can this be re-opened/passed back > upstream for investigation? Seems to be working well enough for me (as well as OpenH264 even has: it has always suffered from jitters, so playback has never been particularly good). Definitely needs a separate bug report. I'm going to close this one again because the fixed package is really available now (bug #2014514 is fixed) and I have verified that it fixes this crash for me.