66230 2011-08-15 08:46:56 -0700 Also release media resources when media is completely loaded. 2012-06-24 00:59:51 -0700 1 1 1 Unclassified WebKit Media 528+ (Nightly build) Unspecified Unspecified RESOLVED DUPLICATE 89720 P2 Normal --- 66687 0 zhenghao webkit-unassigned abarth ap eric.carlson hclam jer.noble oldest_to_newest 451077 0 zhenghao 2011-08-15 08:46:56 -0700 DRT will not release resources of media player in the old page when opening a new page. It will only release resources when GC is needed (see stacktrace GC). But actually DRT will try to release media resources explicitly if it is not completely loaded (see stacktrace EXPLICIT). It may not be a problem on other platforms, but it will cause unstable behavior on platforms with limited resources e.g. mobile phones, especially when running consecutive tests in DRT. If we release media which is not completely loaded, it's better to also release completely loaded media. GC: #0 webkit_glue::WebMediaPlayerImpl::Destroy (this=0x7fffea161930) at Source/WebKit/chromium/webkit/glue/webmediaplayer_impl.cc:1004 #1 0x000000000195e01c in webkit_glue::WebMediaPlayerImpl::~WebMediaPlayerImpl (this=0x7fffea161930, __in_chrg=<value optimized out>) at Source/WebKit/chromium/webkit/glue/webmediaplayer_impl.cc:424 #2 0x0000000000506477 in WTF::deleteOwnedPtr<WebKit::WebMediaPlayer> (ptr=0x7fffea161930) at Source/JavaScriptCore/wtf/OwnPtrCommon.h:65 #3 0x0000000000505eb5 in WTF::OwnPtr<WebKit::WebMediaPlayer>::~OwnPtr (this=0x7fffea0a8660, __in_chrg=<value optimized out>) at Source/JavaScriptCore/wtf/OwnPtr.h:54 #4 0x0000000000503aea in WebKit::WebMediaPlayerClientImpl::~WebMediaPlayerClientImpl (this=0x7fffea0a8640, __in_chrg=<value optimized out>) at Source/WebKit/chromium/src/WebMediaPlayerClientImpl.cpp:108 #5 0x0000000000ff31e2 in WTF::deleteOwnedPtr<WebCore::MediaPlayerPrivateInterface> (ptr=0x7fffea0a8640) at Source/JavaScriptCore/wtf/OwnPtrCommon.h:65 #6 0x0000000000ff2f63 in WTF::OwnPtr<WebCore::MediaPlayerPrivateInterface>::~OwnPtr (this=0x7fffea104af0, __in_chrg=<value optimized out>) at Source/JavaScriptCore/wtf/OwnPtr.h:54 #7 0x0000000000ff0f1e in WebCore::MediaPlayer::~MediaPlayer (this=0x7fffea104aa0, __in_chrg=<value optimized out>) at Source/WebCore/platform/graphics/MediaPlayer.cpp:318 #8 0x0000000000ebd891 in WTF::deleteOwnedPtr<WebCore::MediaPlayer> (ptr=0x7fffea104aa0) at Source/JavaScriptCore/wtf/OwnPtrCommon.h:65 #9 0x0000000000ebd4ef in WTF::OwnPtr<WebCore::MediaPlayer>::~OwnPtr (this=0x7ffff7e8cb40, __in_chrg=<value optimized out>) at Source/JavaScriptCore/wtf/OwnPtr.h:54 #10 0x0000000000eb41eb in WebCore::HTMLMediaElement::~HTMLMediaElement (this=0x7ffff7e8c8c0, __in_chrg=<value optimized out>) at Source/WebCore/html/HTMLMediaElement.cpp:201 #11 0x000000000211445b in WebCore::HTMLAudioElement::~HTMLAudioElement (this=0x7ffff7e8c8c0, __in_chrg=<value optimized out>) at Source/WebCore/html/HTMLAudioElement.h:38 #12 0x0000000000d943fc in WebCore::TreeShared<WebCore::ContainerNode>::removedLastRef (this=0x7ffff7e8c8c8) at Source/WebCore/platform/TreeShared.h:118 #13 0x0000000000476865 in WebCore::TreeShared<WebCore::ContainerNode>::deref (this=0x7ffff7e8c8c8) at Source/WebCore/platform/TreeShared.h:79 #14 0x0000000001569a72 in WebCore::DOMDataStore::weakNodeCallback (value=..., domObject=0x7ffff7e8c8c0) at Source/WebCore/bindings/v8/DOMDataStore.cpp:153 #15 0x0000000000a47232 in v8::internal::GlobalHandles::Node::PostGarbageCollectionProcessing (this=0x7ffff7e5cc00, isolate=0x7ffff7e44000, global_handles=0x7ffff7e74000) at Source/WebKit/chromium/v8/src/global-handles.cc:233 #16 0x0000000000a45b9d in v8::internal::GlobalHandles::PostGarbageCollectionProcessing (this=0x7ffff7e74000, collector=v8::internal::MARK_COMPACTOR) at Source/WebKit/chromium/v8/src/global-handles.cc:555 #17 0x0000000000a57ad0 in v8::internal::Heap::PerformGarbageCollection (this=0x7ffff7e44098, collector=v8::internal::MARK_COMPACTOR, tracer=0x7fffffffaa40) at Source/WebKit/chromium/v8/src/heap.cc:771 #18 0x0000000000a57089 in v8::internal::Heap::CollectGarbage (this=0x7ffff7e44098, space=v8::internal::NEW_SPACE, collector=v8::internal::MARK_COMPACTOR) at Source/WebKit/chromium/v8/src/heap.cc:508 #19 0x0000000000a11789 in v8::internal::Heap::CollectGarbage (this=0x7ffff7e44098, space=v8::internal::NEW_SPACE) at Source/WebKit/chromium/v8/src/heap-inl.h:427 #20 0x0000000000a4acea in v8::internal::SetProperty (object=..., key=..., value=..., attributes=READ_ONLY, strict_mode=v8::internal::kNonStrictMode) at Source/WebKit/chromium/v8/src/handles.cc:269 #21 0x0000000000b883cc in v8::internal::Runtime::SetObjectProperty (isolate=0x7ffff7e44000, object=..., key=..., value=..., attr=READ_ONLY, strict_mode=v8::internal::kNonStrictMode) at Source/WebKit/chromium/v8/src/runtime.cc:4135 #22 0x0000000000b88b62 in v8::internal::Runtime_SetProperty (args=..., isolate=0x7ffff7e44000) at Source/WebKit/chromium/v8/src/runtime.cc:4270 EXPLICIT: #10 0x002fb5a0 in ~MediaPlayer (this=0x12adb18, __in_chrg=<value optimized out>) at third_party/WebKit/Source/WebCore/platform/graphics/MediaPlayer.cpp:318 #11 0x0000cb2a in WTF::deleteOwnedPtr<CppBoundClass::Callback> (ptr=0x12adb18) at third_party/WebKit/Source/JavaScriptCore/wtf/OwnPtrCommon.h:65 #12 0x002a6f20 in WTF::OwnPtr<WebCore::MediaPlayer>::clear (this=0x123b840) at third_party/WebKit/Source/JavaScriptCore/wtf/OwnPtr.h:99 #13 WebCore::HTMLMediaElement::userCancelledLoad (this=0x123b840) at third_party/WebKit/Source/WebCore/html/HTMLMediaElement.cpp:2343 #14 0x002a9212 in WebCore::HTMLMediaElement::stop (this=0x123b840) at third_party/WebKit/Source/WebCore/html/HTMLMediaElement.cpp:2388 #15 0x0028e07c in WebCore::ScriptExecutionContext::stopActiveDOMObjects (this=0x1247c58) at third_party/WebKit/Source/WebCore/dom/ScriptExecutionContext.cpp:271 #16 0x00276930 in WebCore::Document::detach (this=0x1247bd0) at third_party/WebKit/Source/WebCore/dom/Document.cpp:1787 #17 0x003edc1e in WebCore::Frame::setView (this=0x12334c8, view=...) at third_party/WebKit/Source/WebCore/page/Frame.cpp:275 #18 0x00024104 in WebKit::WebFrameImpl::createFrameView (this=0x1233378) at third_party/WebKit/Source/WebKit/chromium/src/WebFrameImpl.cpp:2308 #19 0x0003a71c in WebKit::FrameLoaderClientImpl::makeDocumentView (this=<value optimized out>) at third_party/WebKit/Source/WebKit/chromium/src/FrameLoaderClientImpl.cpp:254 #20 WebKit::FrameLoaderClientImpl::transitionToCommittedForNewPage (this=<value optimized out>) at third_party/WebKit/Source/WebKit/chromium/src/FrameLoaderClientImpl.cpp:1395 #21 0x003c8ecc in WebCore::FrameLoader::transitionToCommitted (this=0x1233508, cachedPage=...) at third_party/WebKit/Source/WebCore/loader/FrameLoader.cpp:1892 #22 0x003c8fca in WebCore::FrameLoader::commitProvisionalLoad (this=0x1233508) at third_party/WebKit/Source/WebCore/loader/FrameLoader.cpp:1739 #23 0x003c056a in WebCore::DocumentLoader::commitIfReady (this=<value optimized out>) at third_party/WebKit/Source/WebCore/loader/DocumentLoader.cpp:279 #24 0x003c058c in WebCore::DocumentLoader::commitLoad (this=0x12c22d8, data=0x12c9b20 "<body>\n<p>Test that Audio(\"url\") constructor loads the specified resource.</p>\n<script src=media-file.js></script>\n<script src=video-test.js></script>\n<script>\n var audioFile = findMediaFile(\"audio"..., length=506) at third_party/WebKit/Source/WebCore/loader/DocumentLoader.cpp:300 #25 0x003c060e in WebCore::DocumentLoader::receivedData (this=0x12c22d8, data=0x12c9b20 "<body>\n<p>Test that Audio(\"url\") constructor loads the specified resource.</p>\n<script src=media-file.js></script>\n<script src=video-test.js></script>\n<script>\n var audioFile = findMediaFile(\"audio"..., length=506) at third_party/WebKit/Source/WebCore/loader/DocumentLoader.cpp:334 #26 0x003cdc6e in WebCore::MainResourceLoader::addData (this=0x12c3100, data=0x12c9b20 "<body>\n<p>Test that Audio(\"url\") constructor loads the specified resource.</p>\n<script src=media-file.js></script>\n<script src=video-test.js></script>\n<script>\n var audioFile = findMediaFile(\"audio"..., length=506, allAtOnce=<value optimized out>) at third_party/WebKit/Source/WebCore/loader/MainResourceLoader.cpp:168 #27 0x003d2270 in WebCore::ResourceLoader::didReceiveData (this=0x12c3100, data=0x12c9b20 "<body>\n<p>Test that Audio(\"url\") constructor loads the specified resource.</p>\n<script src=media-file.js></script>\n<script src=video-test.js></script>\n<script>\n var audioFile = findMediaFile(\"audio"..., length=506, encodedDataLength=<value optimized out>, allAtOnce=false) at third_party/WebKit/Source/WebCore/loader/ResourceLoader.cpp:304 #28 0x003cdc4a in WebCore::MainResourceLoader::didReceiveData (this=0x12c3100, data=0x12c9b20 "<body>\n<p>Test that Audio(\"url\") constructor loads the specified resource.</p>\n<script src=media-file.js></script>\n<script src=video-test.js></script>\n<script>\n var audioFile = findMediaFile(\"audio"..., length=506, encodedDataLength=-1, allAtOnce=false) at third_party/WebKit/Source/WebCore/loader/MainResourceLoader.cpp:464 #29 0x003d2098 in WebCore::ResourceLoader::didReceiveData (this=0x12c3100, data=0x12c9b20 "<body>\n<p>Test that Audio(\"url\") constructor loads the specified resource.</p>\n<script src=media-file.js></script>\n<script src=video-test.js></script>\n<script>\n var audioFile = findMediaFile(\"audio"..., length=506, encodedDataLength=-1) at third_party/WebKit/Source/WebCore/loader/ResourceLoader.cpp:462 #30 0x0057306e in WebCore::ResourceHandleInternal::didReceiveData (this=0x12c3870, data=0x12c9b20 "<body>\n<p>Test that Audio(\"url\") constructor loads the specified resource.</p>\n<script src=media-file.js></script>\n<script src=video-test.js></script>\n<script>\n var audioFile = findMediaFile(\"audio"..., dataLength=506, encodedDataLength=-1) at third_party/WebKit/Source/WebKit/chromium/src/ResourceHandle.cpp:172 #31 0x005372a0 in webkit_glue::WebURLLoaderImpl::Context::OnReceivedData (this=0x128a148, data=0x12c9b20 "<body>\n<p>Test that Audio(\"url\") constructor loads the specified resource.</p>\n<script src=media-file.js></script>\n<script src=video-test.js></script>\n<script>\n var audioFile = findMediaFile(\"audio"..., data_length=506, encoded_data_length=<value optimized out>) at webkit/glue/weburlloader_impl.cc:619 #32 0x0056c376 in NotifyReceivedData (this=<value optimized out>, bytes_read=506) at webkit/tools/test_shell/simple_resource_loader_bridge.cc:280 #33 0x0056a62c in DispatchToMethod<<unnamed>::RequestProxy, void (<unnamed>::RequestProxy::*)(int), int> (this=0x1fa) at ./base/tuple.h:551 #34 Run (this=0x1fa) at ./base/task.h:338 #35 0x0012651c in Run (this=0x12c8c40) at base/message_loop.cc:109 #36 0x001265c6 in DoInvoke (base=<value optimized out>) at ./base/bind_internal.h:595 451082 1 103918 zhenghao 2011-08-15 08:57:17 -0700 Created attachment 103918 Proposed patch. 451320 2 ap 2011-08-15 15:18:34 -0700 Does this mean that the resource will be unavailable in browser when going back to a page that contains it? 451338 3 eric.carlson 2011-08-15 15:42:13 -0700 (In reply to comment #2) > Does this mean that the resource will be unavailable in browser when going back to a page that contains it? Yes it does. With this patch leaving a page will cause all HTMLMediaElement state to be reset and the media engine to be deleted, so when a user goes back the media engine must be recreated, media data will reloaded, all events will be retired, etc. The existing logic is from http://www.w3.org/TR/html5/video.html#dfnReturnLink-2, see the paragraph after "If the media data fetching process is aborted by the user". 451341 4 103918 eric.carlson 2011-08-15 15:45:03 -0700 Comment on attachment 103918 Proposed patch. Changing the behavior everywhere, in a way that breaks compatibility with the spec, is the wrong way to fix a problem that only happens in DRT on some platforms. 451668 5 hclam 2011-08-16 09:31:58 -0700 (In reply to comment #4) > (From update of attachment 103918 [details]) > Changing the behavior everywhere, in a way that breaks compatibility with the spec, is the wrong way to fix a problem that only happens in DRT on some platforms. I suppose that a media resource never is loaded unless it's from a local file system. This effectively means that in all cases except for file:// URLs MediaPlayer is deleted. If this is the final result then why do we treat file:// differently? This seems to fall into a grey area where the spec doesn't specify the behavior regarding media resources / cache when user leave the page. I think we should just treat the logic for file:// the same as other URLs. Logic in this piece of code seems to be coming the old spec where there was a loaded event. 451714 6 eric.carlson 2011-08-16 10:20:29 -0700 (In reply to comment #5) > (In reply to comment #4) > > (From update of attachment 103918 [details] [details]) > > Changing the behavior everywhere, in a way that breaks compatibility with the spec, is the wrong way to fix a problem that only happens in DRT on some platforms. > > I suppose that a media resource never is loaded unless it's from a local file system. This effectively means that in all cases except for file:// URLs MediaPlayer is deleted. > No, what makes you say that? HTMLMediaElement::m_completelyLoaded is set when the media engine reports that the networkState is MediaPlayer::Loaded. This can potentially happen for any file, it is up to the media engine. > If this is the final result then why do we treat file:// differently? This seems to fall into a grey area where the spec doesn't specify the behavior regarding media resources / cache when user leave the page. > I don't think the spec is at all vague. It says to run the steps we have in HTMLMediaElement::userCancelledLoad if the "fetching process is aborted by the user". To me this means the steps should not be run if the fetching process is not aborted, hence if the data has all been fetched we do not run it. > I think we should just treat the logic for file:// the same as other URLs. Logic in this piece of code seems to be coming the old spec where there was a loaded event. As noted this is not caused by file: urls, and it does not come from the old spec. I think this is a usability issue. When navigating back to a page it is a good thing if the page can reload without having to refetch the media, eg. the user's place in the media is not reset so they can resume watching/listening where they were, etc. If you don't like this behavior, change your media engine so the networkState never reaches MediaPlayer::Loaded. 451728 7 hclam 2011-08-16 10:36:04 -0700 (In reply to comment #6) > (In reply to comment #5) > > (In reply to comment #4) > > > (From update of attachment 103918 [details] [details] [details]) > > > Changing the behavior everywhere, in a way that breaks compatibility with the spec, is the wrong way to fix a problem that only happens in DRT on some platforms. > > > > I suppose that a media resource never is loaded unless it's from a local file system. This effectively means that in all cases except for file:// URLs MediaPlayer is deleted. > > > > No, what makes you say that? > > HTMLMediaElement::m_completelyLoaded is set when the media engine reports that the networkState is MediaPlayer::Loaded. This can potentially happen for any file, it is up to the media engine. > > > If this is the final result then why do we treat file:// differently? This seems to fall into a grey area where the spec doesn't specify the behavior regarding media resources / cache when user leave the page. > > > I don't think the spec is at all vague. It says to run the steps we have in HTMLMediaElement::userCancelledLoad if the "fetching process is aborted by the user". To me this means the steps should not be run if the fetching process is not aborted, hence if the data has all been fetched we do not run it. > > > I think we should just treat the logic for file:// the same as other URLs. Logic in this piece of code seems to be coming the old spec where there was a loaded event. > > As noted this is not caused by file: urls, and it does not come from the old spec. > > I think this is a usability issue. When navigating back to a page it is a good thing if the page can reload without having to refetch the media, eg. the user's place in the media is not reset so they can resume watching/listening where they were, etc. > > If you don't like this behavior, change your media engine so the networkState never reaches MediaPlayer::Loaded. If the media resource is loaded from a data: URL, going to a loaded state is the right thing to do since there's no network activity and no progress event. When a site loads some data: URL objects for games, and user leaves the page that would leave a lot of media engines remain in memory, which can be quite heavy. Currently there's no way for a media engine to know that it can clean up safely once it reached loaded state, what about we call cancelLoad() and let the media engine to decide what to do with its internal resources? 451745 8 eric.carlson 2011-08-16 11:03:58 -0700 (In reply to comment #7) > If the media resource is loaded from a data: URL, going to a loaded state is the right thing to do since there's no network activity and no progress event. When a site loads some data: URL objects for games, and user leaves the page that would leave a lot of media engines remain in memory, which can be quite heavy. > > Currently there's no way for a media engine to know that it can clean up safely once it reached loaded state, what about we call cancelLoad() and let the media engine to decide what to do with its internal resources? That isn't a good idea, cancelLoad() is currently called from HTMLMediaElement::prepareForLoad to tell a media engine to clean up in preparation for a new url. Apple's media engines dispose of all assets in cancelLoad, so all state is lost. I assume at least some others do the same. We could add a new media engine function that is called when its document becomes active/inactive. 452181 9 104161 zhenghao 2011-08-17 02:38:50 -0700 Created attachment 104161 Proposed patch 2. Good idea. Actually, I noticed Document::detach() said: // Send out documentWillBecomeInactive() notifications to registered elements, // in order to stop media elements documentWillBecomeInactive(); But HTMLMediaElement has not overridden related methods. Please help review my new patch. 452229 10 104161 eric.carlson 2011-08-17 07:05:14 -0700 Comment on attachment 104161 Proposed patch 2. View in context: https://bugs.webkit.org/attachment.cgi?id=104161&action=review > Source/WebCore/html/HTMLMediaElement.cpp:2654 > +void HTMLMediaElement::documentWillBecomeInactive() > +{ > +#if !ENABLE(PLUGIN_PROXY_FOR_VIDEO) > + LOG(Media, "HTMLMediaElement::documentWillBecomeInactive"); > + if (m_player) > + m_player->willBecomeInactive(); > +#endif > +} > + > +void HTMLMediaElement::documentDidBecomeActive() > +{ > +#if !ENABLE(PLUGIN_PROXY_FOR_VIDEO) > + LOG(Media, "HTMLMediaElement::documentDidBecomeActive"); > + if (m_player) > + m_player->didBecomeActive(); > +#endif > +} > + These new methods are unnecessary. documentWillBecomeInactive is called by Document::detach, *after* it calls stopActiveDOMObjects(). HTMLMediaElement is an ActiveDOMObject. > Source/WebKit/chromium/src/WebMediaPlayerClientImpl.cpp:243 > +void WebMediaPlayerClientImpl::willBecomeInactive() > +{ > + if (m_webMediaPlayer.get()) > + m_webMediaPlayer->willBecomeInactive(); > +} > + > +void WebMediaPlayerClientImpl::didBecomeActive() > +{ > + if (m_webMediaPlayer.get()) > + m_webMediaPlayer->didBecomeActive(); > +} > + I assume that the Chromium media player sets network and ready states correctly when it unloads cached media data? Where are the tests to show that the behavior is correct.? 452345 11 hclam 2011-08-17 10:41:03 -0700 (In reply to comment #6) > (In reply to comment #5) > > (In reply to comment #4) > > > (From update of attachment 103918 [details] [details] [details]) > > > Changing the behavior everywhere, in a way that breaks compatibility with the spec, is the wrong way to fix a problem that only happens in DRT on some platforms. > > > > I suppose that a media resource never is loaded unless it's from a local file system. This effectively means that in all cases except for file:// URLs MediaPlayer is deleted. > > > > No, what makes you say that? > > HTMLMediaElement::m_completelyLoaded is set when the media engine reports that the networkState is MediaPlayer::Loaded. This can potentially happen for any file, it is up to the media engine. > > > If this is the final result then why do we treat file:// differently? This seems to fall into a grey area where the spec doesn't specify the behavior regarding media resources / cache when user leave the page. > > > I don't think the spec is at all vague. It says to run the steps we have in HTMLMediaElement::userCancelledLoad if the "fetching process is aborted by the user". To me this means the steps should not be run if the fetching process is not aborted, hence if the data has all been fetched we do not run it. > > > I think we should just treat the logic for file:// the same as other URLs. Logic in this piece of code seems to be coming the old spec where there was a loaded event. > > As noted this is not caused by file: urls, and it does not come from the old spec. > > I think this is a usability issue. When navigating back to a page it is a good thing if the page can reload without having to refetch the media, eg. the user's place in the media is not reset so they can resume watching/listening where they were, etc. > > If you don't like this behavior, change your media engine so the networkState never reaches MediaPlayer::Loaded. (In reply to comment #8) > (In reply to comment #7) > > If the media resource is loaded from a data: URL, going to a loaded state is the right thing to do since there's no network activity and no progress event. When a site loads some data: URL objects for games, and user leaves the page that would leave a lot of media engines remain in memory, which can be quite heavy. > > > > Currently there's no way for a media engine to know that it can clean up safely once it reached loaded state, what about we call cancelLoad() and let the media engine to decide what to do with its internal resources? > > That isn't a good idea, cancelLoad() is currently called from HTMLMediaElement::prepareForLoad to tell a media engine to clean up in preparation for a new url. Apple's media engines dispose of all assets in cancelLoad, so all state is lost. I assume at least some others do the same. > > We could add a new media engine function that is called when its document becomes active/inactive. That seems to be a better solution in terms of flexibility for the media engine to unload resources if it chooses to. Also my other question is that instead of a m_player.clear(), should it be a cancelLoad()? Conceptually doing a destruction seems to be doing more than just cancel the resource fetch. 452364 12 eric.carlson 2011-08-17 11:00:47 -0700 (In reply to comment #11) > > Also my other question is that instead of a m_player.clear(), should it be a cancelLoad()? Conceptually doing a destruction seems to be doing more than just cancel the resource fetch. The problem that while they are conceptually very similar, clear() deletes the player so *everything* gets cleaned up and all state is reset. cancelLoad do the same, but testing that will be difficult because AFIK not all ports have leaks bots. 656175 13 abarth 2012-06-24 00:59:51 -0700 *** This bug has been marked as a duplicate of bug 89720 *** 103918 2011-08-15 08:57:17 -0700 2011-08-17 02:38:50 -0700 Proposed patch. release.patch text/plain 2421 zhenghao SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDkzMDQzKQorKysgU291cmNlL1dlYkNvcmUvQ2hhbmdl TG9nCSh3b3JraW5nIGNvcHkpCkBAIC0xLDMgKzEsMTMgQEAKKzIwMTEtMDgtMTUgIEhhbyBaaGVu ZyAgPHpoZW5naGFvQGNocm9taXVtLm9yZz4KKworICAgICAgICBBbHNvIHJlbGVhc2UgbWVkaWEg cmVzb3VyY2VzIHdoZW4gbWVkaWEgaXMgY29tcGxldGVseSBsb2FkZWQuCisgICAgICAgIGh0dHBz Oi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD02NjIzMAorCisgICAgICAgIFJldmll d2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgICogaHRtbC9IVE1MTWVkaWFFbGVtZW50 LmNwcDoKKyAgICAgICAgKFdlYkNvcmU6OkhUTUxNZWRpYUVsZW1lbnQ6OnVzZXJDYW5jZWxsZWRM b2FkKToKKwogMjAxMS0wOC0xNSAgQWRhbSBSb2JlbiAgPGFyb2JlbkBhcHBsZS5jb20+CiAKICAg ICAgICAgUmVuYW1lIGFuIGluc3RhbmNlIG9mIHBhZ2VTY2FsZUZhY3RvckNoYW5nZWQgSSBtaXNz ZWQgaW4gcjkzMDQwCkluZGV4OiBTb3VyY2UvV2ViQ29yZS9odG1sL0hUTUxNZWRpYUVsZW1lbnQu Y3BwCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT0KLS0tIFNvdXJjZS9XZWJDb3JlL2h0bWwvSFRNTE1lZGlhRWxlbWVudC5j cHAJKHJldmlzaW9uIDkzMDQyKQorKysgU291cmNlL1dlYkNvcmUvaHRtbC9IVE1MTWVkaWFFbGVt ZW50LmNwcAkod29ya2luZyBjb3B5KQpAQCAtMjMzMyw3ICsyMzMzLDcgQEAgdm9pZCBIVE1MTWVk aWFFbGVtZW50Ojp1c2VyQ2FuY2VsbGVkTG9hZAogewogICAgIExPRyhNZWRpYSwgIkhUTUxNZWRp YUVsZW1lbnQ6OnVzZXJDYW5jZWxsZWRMb2FkIik7CiAKLSAgICBpZiAobV9uZXR3b3JrU3RhdGUg PT0gTkVUV09SS19FTVBUWSB8fCBtX2NvbXBsZXRlbHlMb2FkZWQpCisgICAgaWYgKG1fbmV0d29y a1N0YXRlID09IE5FVFdPUktfRU1QVFkpCiAgICAgICAgIHJldHVybjsKIAogICAgIC8vIElmIHRo ZSBtZWRpYSBkYXRhIGZldGNoaW5nIHByb2Nlc3MgaXMgYWJvcnRlZCBieSB0aGUgdXNlcjoKQEAg LTIzNDYsMTEgKzIzNDYsMTMgQEAgdm9pZCBIVE1MTWVkaWFFbGVtZW50Ojp1c2VyQ2FuY2VsbGVk TG9hZAogICAgIG1fbG9hZFRpbWVyLnN0b3AoKTsKICAgICBtX2xvYWRTdGF0ZSA9IFdhaXRpbmdG b3JTb3VyY2U7CiAKLSAgICAvLyAyIC0gU2V0IHRoZSBlcnJvciBhdHRyaWJ1dGUgdG8gYSBuZXcg TWVkaWFFcnJvciBvYmplY3Qgd2hvc2UgY29kZSBhdHRyaWJ1dGUgaXMgc2V0IHRvIE1FRElBX0VS Ul9BQk9SVEVELgotICAgIG1fZXJyb3IgPSBNZWRpYUVycm9yOjpjcmVhdGUoTWVkaWFFcnJvcjo6 TUVESUFfRVJSX0FCT1JURUQpOworICAgIGlmICghbV9jb21wbGV0ZWx5TG9hZGVkKSB7CisgICAg ICAgIC8vIDIgLSBTZXQgdGhlIGVycm9yIGF0dHJpYnV0ZSB0byBhIG5ldyBNZWRpYUVycm9yIG9i amVjdCB3aG9zZSBjb2RlIGF0dHJpYnV0ZSBpcyBzZXQgdG8gTUVESUFfRVJSX0FCT1JURUQuCisg ICAgICAgIG1fZXJyb3IgPSBNZWRpYUVycm9yOjpjcmVhdGUoTWVkaWFFcnJvcjo6TUVESUFfRVJS X0FCT1JURUQpOwogCi0gICAgLy8gMyAtIFF1ZXVlIGEgdGFzayB0byBmaXJlIGEgc2ltcGxlIGV2 ZW50IG5hbWVkIGVycm9yIGF0IHRoZSBtZWRpYSBlbGVtZW50LgotICAgIHNjaGVkdWxlRXZlbnQo ZXZlbnROYW1lcygpLmFib3J0RXZlbnQpOworICAgICAgICAvLyAzIC0gUXVldWUgYSB0YXNrIHRv IGZpcmUgYSBzaW1wbGUgZXZlbnQgbmFtZWQgZXJyb3IgYXQgdGhlIG1lZGlhIGVsZW1lbnQuCisg ICAgICAgIHNjaGVkdWxlRXZlbnQoZXZlbnROYW1lcygpLmFib3J0RXZlbnQpOworICAgIH0KIAog ICAgIC8vIDQgLSBJZiB0aGUgbWVkaWEgZWxlbWVudCdzIHJlYWR5U3RhdGUgYXR0cmlidXRlIGhh cyBhIHZhbHVlIGVxdWFsIHRvIEhBVkVfTk9USElORywgc2V0IHRoZSAKICAgICAvLyBlbGVtZW50 J3MgbmV0d29ya1N0YXRlIGF0dHJpYnV0ZSB0byB0aGUgTkVUV09SS19FTVBUWSB2YWx1ZSBhbmQg cXVldWUgYSB0YXNrIHRvIGZpcmUgYSAKQEAgLTIzNzEsNiArMjM3Myw3IEBAIHZvaWQgSFRNTE1l ZGlhRWxlbWVudDo6dXNlckNhbmNlbGxlZExvYWQKIAogICAgIC8vIFJlc2V0IG1fcmVhZHlTdGF0 ZSBzaW5jZSBtX3BsYXllciBpcyBnb25lLgogICAgIG1fcmVhZHlTdGF0ZSA9IEhBVkVfTk9USElO RzsKKyAgICBtX2NvbXBsZXRlbHlMb2FkZWQgPSBmYWxzZTsKIH0KIAogYm9vbCBIVE1MTWVkaWFF bGVtZW50OjpjYW5TdXNwZW5kKCkgY29uc3QK 104161 2011-08-17 02:38:50 -0700 2011-08-17 07:05:14 -0700 Proposed patch 2. expose.patch text/plain 7611 zhenghao SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDkzMTk3KQorKysgU291cmNlL1dlYkNvcmUvQ2hhbmdl TG9nCSh3b3JraW5nIGNvcHkpCkBAIC0xLDMgKzEsMjUgQEAKKzIwMTEtMDgtMTcgIEhhbyBaaGVu ZyAgPHpoZW5naGFvQGNocm9taXVtLm9yZz4KKworICAgICAgICBBZGQgbWV0aG9kcyB0byBpbmZv cm0gbWVkaWEgZW5naW5lIHdoZW4gaXRzIGRvY3VtZW50IGJlY29tZXMgYWN0aXZlL2luYWN0aXZl LgorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9NjYyMzAK KworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBFeHBvc2Ug bWV0aG9kcyB3aXRoIHdoaWNoIGVhY2ggbWVkaWEgZW5naW5lIGNvdWxkIGRlY2lkZSB3aGF0IHRv IGRvIHdoZW4gdGhlCisgICAgICAgIHBhZ2UgYmVjb21lcyBhY3RpdmUvaW5hY3RpdmUuCisKKyAg ICAgICAgKiBodG1sL0hUTUxNZWRpYUVsZW1lbnQuY3BwOgorICAgICAgICAoV2ViQ29yZTo6SFRN TE1lZGlhRWxlbWVudDo6ZG9jdW1lbnRXaWxsQmVjb21lSW5hY3RpdmUpOgorICAgICAgICAoV2Vi Q29yZTo6SFRNTE1lZGlhRWxlbWVudDo6ZG9jdW1lbnREaWRCZWNvbWVBY3RpdmUpOgorICAgICAg ICAqIGh0bWwvSFRNTE1lZGlhRWxlbWVudC5oOgorICAgICAgICAqIHBsYXRmb3JtL2dyYXBoaWNz L01lZGlhUGxheWVyLmNwcDoKKyAgICAgICAgKFdlYkNvcmU6Ok1lZGlhUGxheWVyOjp3aWxsQmVj b21lSW5hY3RpdmUpOgorICAgICAgICAoV2ViQ29yZTo6TWVkaWFQbGF5ZXI6OmRpZEJlY29tZUFj dGl2ZSk6CisgICAgICAgICogcGxhdGZvcm0vZ3JhcGhpY3MvTWVkaWFQbGF5ZXIuaDoKKyAgICAg ICAgKiBwbGF0Zm9ybS9ncmFwaGljcy9NZWRpYVBsYXllclByaXZhdGUuaDoKKyAgICAgICAgKFdl YkNvcmU6Ok1lZGlhUGxheWVyUHJpdmF0ZUludGVyZmFjZTo6d2lsbEJlY29tZUluYWN0aXZlKToK KyAgICAgICAgKFdlYkNvcmU6Ok1lZGlhUGxheWVyUHJpdmF0ZUludGVyZmFjZTo6ZGlkQmVjb21l QWN0aXZlKToKKwogMjAxMS0wOC0xNyAgU2FpbGVzaCBBZ3Jhd2FsICA8c2FpbEBjaHJvbWl1bS5v cmc+CiAKICAgICAgICAgQ2hyb21pdW0gTWFjOiBGaXggaXNzdWUgd2hlcmUgc2Nyb2xsYmFyIHdv dWxkbid0IGJlIGRyYXduIHVudGlsIHBhZ2UgZmluaXNoZWQgbG9hZGluZwpJbmRleDogU291cmNl L1dlYkNvcmUvaHRtbC9IVE1MTWVkaWFFbGVtZW50LmNwcAo9PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2Uv V2ViQ29yZS9odG1sL0hUTUxNZWRpYUVsZW1lbnQuY3BwCShyZXZpc2lvbiA5MzA0MikKKysrIFNv dXJjZS9XZWJDb3JlL2h0bWwvSFRNTE1lZGlhRWxlbWVudC5jcHAJKHdvcmtpbmcgY29weSkKQEAg LTI2MzQsNiArMjYzNCwyNCBAQCB2b2lkIEhUTUxNZWRpYUVsZW1lbnQ6OndpbGxTdG9wQmVpbmdG dWxsCiAgICAgICAgIG1lZGlhQ29udHJvbHMoKS0+ZXhpdGVkRnVsbHNjcmVlbigpOwogfQogCit2 b2lkIEhUTUxNZWRpYUVsZW1lbnQ6OmRvY3VtZW50V2lsbEJlY29tZUluYWN0aXZlKCkKK3sKKyNp ZiAhRU5BQkxFKFBMVUdJTl9QUk9YWV9GT1JfVklERU8pCisgICAgTE9HKE1lZGlhLCAiSFRNTE1l ZGlhRWxlbWVudDo6ZG9jdW1lbnRXaWxsQmVjb21lSW5hY3RpdmUiKTsKKyAgICBpZiAobV9wbGF5 ZXIpCisgICAgICAgIG1fcGxheWVyLT53aWxsQmVjb21lSW5hY3RpdmUoKTsKKyNlbmRpZgorfQor Cit2b2lkIEhUTUxNZWRpYUVsZW1lbnQ6OmRvY3VtZW50RGlkQmVjb21lQWN0aXZlKCkKK3sKKyNp ZiAhRU5BQkxFKFBMVUdJTl9QUk9YWV9GT1JfVklERU8pCisgICAgTE9HKE1lZGlhLCAiSFRNTE1l ZGlhRWxlbWVudDo6ZG9jdW1lbnREaWRCZWNvbWVBY3RpdmUiKTsKKyAgICBpZiAobV9wbGF5ZXIp CisgICAgICAgIG1fcGxheWVyLT5kaWRCZWNvbWVBY3RpdmUoKTsKKyNlbmRpZgorfQorCiBQbGF0 Zm9ybU1lZGlhIEhUTUxNZWRpYUVsZW1lbnQ6OnBsYXRmb3JtTWVkaWEoKSBjb25zdAogewogICAg IHJldHVybiBtX3BsYXllciA/IG1fcGxheWVyLT5wbGF0Zm9ybU1lZGlhKCkgOiBOb1BsYXRmb3Jt TWVkaWE7CkluZGV4OiBTb3VyY2UvV2ViQ29yZS9odG1sL0hUTUxNZWRpYUVsZW1lbnQuaAo9PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09Ci0tLSBTb3VyY2UvV2ViQ29yZS9odG1sL0hUTUxNZWRpYUVsZW1lbnQuaAkocmV2aXNp b24gOTMwNDIpCisrKyBTb3VyY2UvV2ViQ29yZS9odG1sL0hUTUxNZWRpYUVsZW1lbnQuaAkod29y a2luZyBjb3B5KQpAQCAtMjMxLDYgKzIzMSw5IEBAIHByaXZhdGU6CiAgICAgdmlydHVhbCB2b2lk IGRpZEJlY29tZUZ1bGxzY3JlZW5FbGVtZW50KCk7CiAgICAgdmlydHVhbCB2b2lkIHdpbGxTdG9w QmVpbmdGdWxsc2NyZWVuRWxlbWVudCgpOwogCisgICAgdmlydHVhbCB2b2lkIGRvY3VtZW50V2ls bEJlY29tZUluYWN0aXZlKCk7CisgICAgdmlydHVhbCB2b2lkIGRvY3VtZW50RGlkQmVjb21lQWN0 aXZlKCk7CisKICAgICAvLyBBY3RpdmVET01PYmplY3QgZnVuY3Rpb25zLgogICAgIHZpcnR1YWwg Ym9vbCBjYW5TdXNwZW5kKCkgY29uc3Q7CiAgICAgdmlydHVhbCB2b2lkIHN1c3BlbmQoUmVhc29u Rm9yU3VzcGVuc2lvbik7CkluZGV4OiBTb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9N ZWRpYVBsYXllci5jcHAKPT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3Jh cGhpY3MvTWVkaWFQbGF5ZXIuY3BwCShyZXZpc2lvbiA5MzA0MikKKysrIFNvdXJjZS9XZWJDb3Jl L3BsYXRmb3JtL2dyYXBoaWNzL01lZGlhUGxheWVyLmNwcAkod29ya2luZyBjb3B5KQpAQCAtODcz LDYgKzg3MywxNiBAQCB2b2lkIE1lZGlhUGxheWVyOjpjaGFyYWN0ZXJpc3RpY0NoYW5nZWQoCiAg ICAgICAgIG1fbWVkaWFQbGF5ZXJDbGllbnQtPm1lZGlhUGxheWVyQ2hhcmFjdGVyaXN0aWNDaGFu Z2VkKHRoaXMpOwogfQogCit2b2lkIE1lZGlhUGxheWVyOjp3aWxsQmVjb21lSW5hY3RpdmUoKQor eworICAgIG1fcHJpdmF0ZS0+d2lsbEJlY29tZUluYWN0aXZlKCk7Cit9CisKK3ZvaWQgTWVkaWFQ bGF5ZXI6OmRpZEJlY29tZUFjdGl2ZSgpCit7CisgICAgbV9wcml2YXRlLT5kaWRCZWNvbWVBY3Rp dmUoKTsKK30KKwogfQogCiAjZW5kaWYKSW5kZXg6IFNvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL2dy YXBoaWNzL01lZGlhUGxheWVyLmgKPT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gU291cmNlL1dlYkNvcmUvcGxhdGZv cm0vZ3JhcGhpY3MvTWVkaWFQbGF5ZXIuaAkocmV2aXNpb24gOTMwNDIpCisrKyBTb3VyY2UvV2Vi Q29yZS9wbGF0Zm9ybS9ncmFwaGljcy9NZWRpYVBsYXllci5oCSh3b3JraW5nIGNvcHkpCkBAIC0y NzAsNiArMjcwLDggQEAgcHVibGljOgogICAgIHZvaWQgZmlyc3RWaWRlb0ZyYW1lQXZhaWxhYmxl KCk7CiAgICAgdm9pZCBjaGFyYWN0ZXJpc3RpY0NoYW5nZWQoKTsKIAorICAgIHZvaWQgd2lsbEJl Y29tZUluYWN0aXZlKCk7CisgICAgdm9pZCBkaWRCZWNvbWVBY3RpdmUoKTsKIAogICAgIHZvaWQg cmVwYWludCgpOwogCkluZGV4OiBTb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9NZWRp YVBsYXllclByaXZhdGUuaAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9n cmFwaGljcy9NZWRpYVBsYXllclByaXZhdGUuaAkocmV2aXNpb24gOTMwNDIpCisrKyBTb3VyY2Uv V2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9NZWRpYVBsYXllclByaXZhdGUuaAkod29ya2luZyBj b3B5KQpAQCAtNDQsNyArNDQsMTAgQEAgcHVibGljOgogCiAgICAgdmlydHVhbCB2b2lkIGxvYWQo Y29uc3QgU3RyaW5nJiB1cmwpID0gMDsKICAgICB2aXJ0dWFsIHZvaWQgY2FuY2VsTG9hZCgpID0g MDsKLSAgICAKKworICAgIHZpcnR1YWwgdm9pZCB3aWxsQmVjb21lSW5hY3RpdmUoKSB7IH07Cisg ICAgdmlydHVhbCB2b2lkIGRpZEJlY29tZUFjdGl2ZSgpIHsgfTsKKwogICAgIHZpcnR1YWwgdm9p ZCBwcmVwYXJlVG9QbGF5KCkgeyB9CiAgICAgdmlydHVhbCBQbGF0Zm9ybU1lZGlhIHBsYXRmb3Jt TWVkaWEoKSBjb25zdCB7IHJldHVybiBOb1BsYXRmb3JtTWVkaWE7IH0KICNpZiBVU0UoQUNDRUxF UkFURURfQ09NUE9TSVRJTkcpCkBAIC01Miw3ICs1NSw3IEBAIHB1YmxpYzoKICNlbmRpZgogCiAg ICAgdmlydHVhbCB2b2lkIHBsYXkoKSA9IDA7Ci0gICAgdmlydHVhbCB2b2lkIHBhdXNlKCkgPSAw OyAgICAKKyAgICB2aXJ0dWFsIHZvaWQgcGF1c2UoKSA9IDA7CiAKICAgICB2aXJ0dWFsIGJvb2wg c3VwcG9ydHNGdWxsc2NyZWVuKCkgY29uc3QgeyByZXR1cm4gZmFsc2U7IH0KICAgICB2aXJ0dWFs IGJvb2wgc3VwcG9ydHNTYXZlKCkgY29uc3QgeyByZXR1cm4gZmFsc2U7IH0KSW5kZXg6IFNvdXJj ZS9XZWJLaXQvY2hyb21pdW0vQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIFNvdXJjZS9XZWJLaXQv Y2hyb21pdW0vQ2hhbmdlTG9nCShyZXZpc2lvbiA5MzE5NykKKysrIFNvdXJjZS9XZWJLaXQvY2hy b21pdW0vQ2hhbmdlTG9nCSh3b3JraW5nIGNvcHkpCkBAIC0xLDMgKzEsMTggQEAKKzIwMTEtMDgt MTcgIEhhbyBaaGVuZyAgPHpoZW5naGFvQGNocm9taXVtLm9yZz4KKworICAgICAgICBBZGQgbWV0 aG9kcyB0byBpbmZvcm0gbWVkaWEgZW5naW5lIHdoZW4gaXRzIGRvY3VtZW50IGJlY29tZXMgYWN0 aXZlL2luYWN0aXZlLgorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5j Z2k/aWQ9NjYyMzAKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAg ICAgICAqIHB1YmxpYy9XZWJNZWRpYVBsYXllci5oOgorICAgICAgICAoV2ViS2l0OjpXZWJNZWRp YVBsYXllcjo6d2lsbEJlY29tZUluYWN0aXZlKToKKyAgICAgICAgKFdlYktpdDo6V2ViTWVkaWFQ bGF5ZXI6OmRpZEJlY29tZUFjdGl2ZSk6CisgICAgICAgICogc3JjL1dlYk1lZGlhUGxheWVyQ2xp ZW50SW1wbC5jcHA6CisgICAgICAgIChXZWJLaXQ6OldlYk1lZGlhUGxheWVyQ2xpZW50SW1wbDo6 d2lsbEJlY29tZUluYWN0aXZlKToKKyAgICAgICAgKFdlYktpdDo6V2ViTWVkaWFQbGF5ZXJDbGll bnRJbXBsOjpkaWRCZWNvbWVBY3RpdmUpOgorICAgICAgICAqIHNyYy9XZWJNZWRpYVBsYXllckNs aWVudEltcGwuaDoKKwogMjAxMS0wOC0xNiAgUGVyLUVyaWsgQnJvZGluICA8cGVyLWVyaWsuYnJv ZGluQGVyaWNzc29uLmNvbT4KIAogICAgICAgICBNYWtlIGl0IHBvc3NpYmxlIHRvIGV4cGxpY2l0 bHkgcHJldmVudCBhIHByZWZsaWdodCB2aWEgVGhyZWFkYWJsZUxvYWRlck9wdGlvbnMKSW5kZXg6 IFNvdXJjZS9XZWJLaXQvY2hyb21pdW0vcHVibGljL1dlYk1lZGlhUGxheWVyLmgKPT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PQotLS0gU291cmNlL1dlYktpdC9jaHJvbWl1bS9wdWJsaWMvV2ViTWVkaWFQbGF5ZXIuaAkocmV2 aXNpb24gOTMwNDIpCisrKyBTb3VyY2UvV2ViS2l0L2Nocm9taXVtL3B1YmxpYy9XZWJNZWRpYVBs YXllci5oCSh3b3JraW5nIGNvcHkpCkBAIC05MCw2ICs5MCw5IEBAIHB1YmxpYzoKICAgICB2aXJ0 dWFsIHZvaWQgbG9hZChjb25zdCBXZWJVUkwmKSA9IDA7CiAgICAgdmlydHVhbCB2b2lkIGNhbmNl bExvYWQoKSA9IDA7CiAKKyAgICB2aXJ0dWFsIHZvaWQgd2lsbEJlY29tZUluYWN0aXZlKCkgeyB9 OworICAgIHZpcnR1YWwgdm9pZCBkaWRCZWNvbWVBY3RpdmUoKSB7IH07CisKICAgICAvLyBQbGF5 YmFjayBjb250cm9scy4KICAgICB2aXJ0dWFsIHZvaWQgcGxheSgpID0gMDsKICAgICB2aXJ0dWFs IHZvaWQgcGF1c2UoKSA9IDA7CkluZGV4OiBTb3VyY2UvV2ViS2l0L2Nocm9taXVtL3NyYy9XZWJN ZWRpYVBsYXllckNsaWVudEltcGwuY3BwCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIFNvdXJjZS9XZWJLaXQvY2hy b21pdW0vc3JjL1dlYk1lZGlhUGxheWVyQ2xpZW50SW1wbC5jcHAJKHJldmlzaW9uIDkzMDQyKQor KysgU291cmNlL1dlYktpdC9jaHJvbWl1bS9zcmMvV2ViTWVkaWFQbGF5ZXJDbGllbnRJbXBsLmNw cAkod29ya2luZyBjb3B5KQpAQCAtMjI5LDYgKzIyOSwxOCBAQCB2b2lkIFdlYk1lZGlhUGxheWVy Q2xpZW50SW1wbDo6Y2FuY2VsTG9hCiAgICAgICAgIG1fd2ViTWVkaWFQbGF5ZXItPmNhbmNlbExv YWQoKTsKIH0KIAordm9pZCBXZWJNZWRpYVBsYXllckNsaWVudEltcGw6OndpbGxCZWNvbWVJbmFj dGl2ZSgpCit7CisgICAgaWYgKG1fd2ViTWVkaWFQbGF5ZXIuZ2V0KCkpCisgICAgICAgIG1fd2Vi TWVkaWFQbGF5ZXItPndpbGxCZWNvbWVJbmFjdGl2ZSgpOworfQorCit2b2lkIFdlYk1lZGlhUGxh eWVyQ2xpZW50SW1wbDo6ZGlkQmVjb21lQWN0aXZlKCkKK3sKKyAgICBpZiAobV93ZWJNZWRpYVBs YXllci5nZXQoKSkKKyAgICAgICAgbV93ZWJNZWRpYVBsYXllci0+ZGlkQmVjb21lQWN0aXZlKCk7 Cit9CisKICNpZiBVU0UoQUNDRUxFUkFURURfQ09NUE9TSVRJTkcpCiBQbGF0Zm9ybUxheWVyKiBX ZWJNZWRpYVBsYXllckNsaWVudEltcGw6OnBsYXRmb3JtTGF5ZXIoKSBjb25zdAogewpJbmRleDog U291cmNlL1dlYktpdC9jaHJvbWl1bS9zcmMvV2ViTWVkaWFQbGF5ZXJDbGllbnRJbXBsLmgKPT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PQotLS0gU291cmNlL1dlYktpdC9jaHJvbWl1bS9zcmMvV2ViTWVkaWFQbGF5ZXJDbGll bnRJbXBsLmgJKHJldmlzaW9uIDkzMDQyKQorKysgU291cmNlL1dlYktpdC9jaHJvbWl1bS9zcmMv V2ViTWVkaWFQbGF5ZXJDbGllbnRJbXBsLmgJKHdvcmtpbmcgY29weSkKQEAgLTgyLDYgKzgyLDgg QEAgcHVibGljOgogICAgIC8vIE1lZGlhUGxheWVyUHJpdmF0ZUludGVyZmFjZSBtZXRob2RzOgog ICAgIHZpcnR1YWwgdm9pZCBsb2FkKGNvbnN0IFdURjo6U3RyaW5nJiB1cmwpOwogICAgIHZpcnR1 YWwgdm9pZCBjYW5jZWxMb2FkKCk7CisgICAgdmlydHVhbCB2b2lkIHdpbGxCZWNvbWVJbmFjdGl2 ZSgpOworICAgIHZpcnR1YWwgdm9pZCBkaWRCZWNvbWVBY3RpdmUoKTsKICNpZiBVU0UoQUNDRUxF UkFURURfQ09NUE9TSVRJTkcpCiAgICAgdmlydHVhbCBXZWJDb3JlOjpQbGF0Zm9ybUxheWVyKiBw bGF0Zm9ybUxheWVyKCkgY29uc3Q7CiAjZW5kaWYK