159931 2016-07-19 11:27:50 -0700 Fix a linking failure caused by NetworkCache::Data::~Data() 2026-03-02 02:05:54 -0800 1 1 1 Unclassified WebKit New Bugs WebKit Nightly Build PC Linux RESOLVED CONFIGURATION CHANGED P2 Normal --- 1 changseok changseok achristensen aperez cdumez cgarcia clopez commit-queue darin jfernandez oldest_to_newest 1212438 0 changseok 2016-07-19 11:27:50 -0700 A linking failure happens when building with clang 3.6. >[2068/2122] Linking CXX shared library lib/libwebkit2gtk-4.0.so.37.14.2 >FAILED: : && /home/changseok/Projects/api-sanitizer/bin/llvm/bin/clang++ -fPIC -Wno-error -std=c++1y -fcolor-diagnostics -Qunused-arguments -g -Wl,--no-undefined -L/home/changseok/Projects/WebKit/WebKitBuild/DependenciesGTK/Root/lib -fuse-ld=gold -Wl,--disable-new-dtags -fuse-ld=gold -Wl,--disable-new-dtags -shared -Wl,-soname,libwebkit2gtk-4.0.so.37 -o lib/libwebkit2gtk-4.0.so.37.14.2 @CMakeFiles/WebKit2.rsp && : >../../Source/WebKit2/NetworkProcess/cache/NetworkCacheSubresourcesEntry.cpp:76: error: undefined reference to 'WebKit::NetworkCache::Data::~Data()' >clang-3.6: error: linker command failed with exit code 1 (use -v to see invocation) 1212439 1 284022 changseok 2016-07-19 11:32:51 -0700 Created attachment 284022 Patch 1212441 2 284023 changseok 2016-07-19 11:36:26 -0700 Created attachment 284023 Patch 1212442 3 284023 cdumez 2016-07-19 11:42:25 -0700 Comment on attachment 284023 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=284023&action=review > Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109 > + virtual ~Data() { } Why would you make this class virtual?? 1212450 4 284023 cdumez 2016-07-19 11:53:34 -0700 Comment on attachment 284023 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=284023&action=review >> Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109 >> + virtual ~Data() { } > > Why would you make this class virtual?? does "~Data() = default;" fix your build error? 1212499 5 changseok 2016-07-19 13:37:27 -0700 (In reply to comment #4) > Comment on attachment 284023 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=284023&action=review > > >> Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109 > >> + virtual ~Data() { } > > > > Why would you make this class virtual?? That is meaningless here. Removed. > > does "~Data() = default;" fix your build error? No, it doesn't work. It spawns other compile errors like... ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheBlobStorage.cpp:99:29: error: attempt to use a deleted function auto existingData = mapFile(blobPath.data()); ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:446:19: error: attempt to use a deleted function auto record = cacheEntry->encodeAsStorageRecord(); 1212505 6 284043 changseok 2016-07-19 13:40:22 -0700 Created attachment 284043 Patch 1212851 7 284043 commit-queue 2016-07-20 13:04:03 -0700 Comment on attachment 284043 Patch Clearing flags on attachment: 284043 Committed r203467: <http://trac.webkit.org/changeset/203467> 1212852 8 commit-queue 2016-07-20 13:04:07 -0700 All reviewed patches have been landed. Closing bug. 1212860 9 mcatanzaro 2016-07-20 13:20:22 -0700 *** Bug 159264 has been marked as a duplicate of this bug. *** 1212885 10 darin 2016-07-20 14:53:50 -0700 Why is this change needed? This kind of change should have no effect. Is this a compiler bug? 1212887 11 darin 2016-07-20 14:56:56 -0700 (In reply to comment #5) > > does "~Data() = default;" fix your build error? > No, it doesn't work. It spawns other compile errors like... > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheBlobStorage.cpp:99:29: > error: attempt to use a deleted function > auto existingData = mapFile(blobPath.data()); > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:446:19: error: > attempt to use a deleted function > auto record = cacheEntry->encodeAsStorageRecord(); Those sound like what you would get if you tried "~Data() = delete;", not "~Data() = default;". This change is highly suspect. Explicitly defining an empty inline constructor should have no effect! 1212999 12 changseok 2016-07-20 19:35:19 -0700 (In reply to comment #11) > (In reply to comment #5) > > > does "~Data() = default;" fix your build error? > > No, it doesn't work. It spawns other compile errors like... > > > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheBlobStorage.cpp:99:29: > > error: attempt to use a deleted function > > auto existingData = mapFile(blobPath.data()); > > > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:446:19: error: > > attempt to use a deleted function > > auto record = cacheEntry->encodeAsStorageRecord(); > > Those sound like what you would get if you tried "~Data() = delete;", not > "~Data() = default;". > > This change is highly suspect. Explicitly defining an empty inline > constructor should have no effect! Nah.. An incorrect log mislead you. Here is correct full logs. Sorry about this confusion. ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:410:19: error: attempt to use a deleted function auto record = cacheEntry->encodeAsStorageRecord(); ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: destructor of 'Record' is implicitly deleted because field 'header' has a deleted destructor Data header; ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: '~Data' has been explicitly marked deleted here ~Data() = delete; ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:446:19: error: attempt to use a deleted function auto record = cacheEntry->encodeAsStorageRecord(); ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: destructor of 'Record' is implicitly deleted because field 'header' has a deleted destructor Data header; ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: '~Data' has been explicitly marked deleted here ~Data() = delete; ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:461:25: error: attempt to use a deleted function auto updateRecord = updateEntry->encodeAsStorageRecord(); ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: destructor of 'Record' is implicitly deleted because field 'header' has a deleted destructor Data header; ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: '~Data' has been explicitly marked deleted here ~Data() = delete; ^ In file included from ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:26: In file included from ../../Source/WebKit2/config.h:44: In file included from ../../Source/WTF/wtf/FastMalloc.h:26: In file included from ../../Source/WTF/wtf/StdLibExtras.h:31: In file included from /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/memory:81: /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/unique_ptr.h:76:2: error: attempt to use a deleted function delete __ptr; ^ /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/unique_ptr.h:236:4: note: in instantiation of member function 'std::default_delete<WebKit::NetworkCache::Entry>::operator()' requested here get_deleter()(__ptr); ^ /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/unique_ptr.h:201:50: note: in instantiation of member function 'std::unique_ptr<WebKit::NetworkCache::Entry, std::default_delete<WebKit::NetworkCache::Entry> >::~unique_ptr' requested here constexpr unique_ptr(nullptr_t) noexcept : unique_ptr() { } ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheEntry.h:92:21: note: destructor of 'Entry' is implicitly deleted because field 'm_sourceStorageRecord' has a deleted destructor Storage::Record m_sourceStorageRecord { }; ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: destructor of 'Record' is implicitly deleted because field 'header' has a deleted destructor Data header; ^ ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: '~Data' has been explicitly marked deleted here ~Data() = delete; ^ 1213001 13 changseok 2016-07-20 19:41:01 -0700 (In reply to comment #10) > Why is this change needed? This kind of change should have no effect. Is > this a compiler bug? That is a possible suspicion. Because this linking issue was found with clang 3.6 which is far from the latest. Let me try recent one and leave a note about it. 1213005 14 cdumez 2016-07-20 19:46:27 -0700 (In reply to comment #12) > (In reply to comment #11) > > (In reply to comment #5) > > > > does "~Data() = default;" fix your build error? > > > No, it doesn't work. It spawns other compile errors like... > > > > > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheBlobStorage.cpp:99:29: > > > error: attempt to use a deleted function > > > auto existingData = mapFile(blobPath.data()); > > > > > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:446:19: error: > > > attempt to use a deleted function > > > auto record = cacheEntry->encodeAsStorageRecord(); > > > > Those sound like what you would get if you tried "~Data() = delete;", not > > "~Data() = default;". > > > > This change is highly suspect. Explicitly defining an empty inline > > constructor should have no effect! > > Nah.. An incorrect log mislead you. Here is correct full logs. Sorry about > this confusion. > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:410:19: error: > attempt to use a deleted function > auto record = cacheEntry->encodeAsStorageRecord(); > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: > destructor of 'Record' is implicitly deleted because field 'header' has a > deleted destructor > Data header; > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: > '~Data' has been explicitly marked deleted here > ~Data() = delete; > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:446:19: error: > attempt to use a deleted function > auto record = cacheEntry->encodeAsStorageRecord(); > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: > destructor of 'Record' is implicitly deleted because field 'header' has a > deleted destructor > Data header; > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: > '~Data' has been explicitly marked deleted here > ~Data() = delete; > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:461:25: error: > attempt to use a deleted function > auto updateRecord = updateEntry->encodeAsStorageRecord(); > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: > destructor of 'Record' is implicitly deleted because field 'header' has a > deleted destructor > Data header; > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: > '~Data' has been explicitly marked deleted here > ~Data() = delete; > ^ > In file included from > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:26: > In file included from ../../Source/WebKit2/config.h:44: > In file included from ../../Source/WTF/wtf/FastMalloc.h:26: > In file included from ../../Source/WTF/wtf/StdLibExtras.h:31: > In file included from > /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/memory: > 81: > /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/ > unique_ptr.h:76:2: error: attempt to use a deleted function > delete __ptr; > ^ > /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/ > unique_ptr.h:236:4: note: in instantiation of member function > 'std::default_delete<WebKit::NetworkCache::Entry>::operator()' requested here > get_deleter()(__ptr); > ^ > /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/ > unique_ptr.h:201:50: note: in instantiation of member function > 'std::unique_ptr<WebKit::NetworkCache::Entry, > std::default_delete<WebKit::NetworkCache::Entry> >::~unique_ptr' requested > here > constexpr unique_ptr(nullptr_t) noexcept : unique_ptr() { } > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheEntry.h:92:21: note: > destructor of 'Entry' is implicitly deleted because field > 'm_sourceStorageRecord' has a deleted destructor > Storage::Record m_sourceStorageRecord { }; > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: > destructor of 'Record' is implicitly deleted because field 'header' has a > deleted destructor > Data header; > ^ > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: > '~Data' has been explicitly marked deleted here > ~Data() = delete; > ^ If you look at this last line, the compiler says you deleted the Data destructor: ~Data() = delete; The workaround I suggested was: ~Data() = default; 1213006 15 changseok 2016-07-20 19:50:04 -0700 (In reply to comment #14) > (In reply to comment #12) > > (In reply to comment #11) > > > (In reply to comment #5) > > > > > does "~Data() = default;" fix your build error? > > > > No, it doesn't work. It spawns other compile errors like... > > > > > > > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheBlobStorage.cpp:99:29: > > > > error: attempt to use a deleted function > > > > auto existingData = mapFile(blobPath.data()); > > > > > > > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:446:19: error: > > > > attempt to use a deleted function > > > > auto record = cacheEntry->encodeAsStorageRecord(); > > > > > > Those sound like what you would get if you tried "~Data() = delete;", not > > > "~Data() = default;". > > > > > > This change is highly suspect. Explicitly defining an empty inline > > > constructor should have no effect! > > > > Nah.. An incorrect log mislead you. Here is correct full logs. Sorry about > > this confusion. > > > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:410:19: error: > > attempt to use a deleted function > > auto record = cacheEntry->encodeAsStorageRecord(); > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: > > destructor of 'Record' is implicitly deleted because field 'header' has a > > deleted destructor > > Data header; > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: > > '~Data' has been explicitly marked deleted here > > ~Data() = delete; > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:446:19: error: > > attempt to use a deleted function > > auto record = cacheEntry->encodeAsStorageRecord(); > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: > > destructor of 'Record' is implicitly deleted because field 'header' has a > > deleted destructor > > Data header; > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: > > '~Data' has been explicitly marked deleted here > > ~Data() = delete; > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:461:25: error: > > attempt to use a deleted function > > auto updateRecord = updateEntry->encodeAsStorageRecord(); > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: > > destructor of 'Record' is implicitly deleted because field 'header' has a > > deleted destructor > > Data header; > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: > > '~Data' has been explicitly marked deleted here > > ~Data() = delete; > > ^ > > In file included from > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCache.cpp:26: > > In file included from ../../Source/WebKit2/config.h:44: > > In file included from ../../Source/WTF/wtf/FastMalloc.h:26: > > In file included from ../../Source/WTF/wtf/StdLibExtras.h:31: > > In file included from > > /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/memory: > > 81: > > /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/ > > unique_ptr.h:76:2: error: attempt to use a deleted function > > delete __ptr; > > ^ > > /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/ > > unique_ptr.h:236:4: note: in instantiation of member function > > 'std::default_delete<WebKit::NetworkCache::Entry>::operator()' requested here > > get_deleter()(__ptr); > > ^ > > /usr/lib/gcc/x86_64-redhat-linux/6.1.1/../../../../include/c++/6.1.1/bits/ > > unique_ptr.h:201:50: note: in instantiation of member function > > 'std::unique_ptr<WebKit::NetworkCache::Entry, > > std::default_delete<WebKit::NetworkCache::Entry> >::~unique_ptr' requested > > here > > constexpr unique_ptr(nullptr_t) noexcept : unique_ptr() { } > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheEntry.h:92:21: note: > > destructor of 'Entry' is implicitly deleted because field > > 'm_sourceStorageRecord' has a deleted destructor > > Storage::Record m_sourceStorageRecord { }; > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheStorage.h:58:14: note: > > destructor of 'Record' is implicitly deleted because field 'header' has a > > deleted destructor > > Data header; > > ^ > > ../../Source/WebKit2/NetworkProcess/cache/NetworkCacheData.h:109:5: note: > > '~Data' has been explicitly marked deleted here > > ~Data() = delete; > > ^ > > If you look at this last line, the compiler says you deleted the Data > destructor: > ~Data() = delete; > > The workaround I suggested was: > ~Data() = default; Oh.. my terrible mistake. Let me fix it. 1213007 16 cdumez 2016-07-20 19:51:09 -0700 That said, I agree with Darin that it looks like a compiler/linker bug as there is no reason we should need an explicit destructor here :/ 1213052 17 jfernandez 2016-07-20 23:05:00 -0700 (In reply to comment #16) > That said, I agree with Darin that it looks like a compiler/linker bug as > there is no reason we should need an explicit destructor here :/ I've been investigating this issue in bug #159264 and I think the proposed patch here is not correct. As darin suggested it seems a compiler bug because just doing a change like this avoids the compilation error: diff --git a/Source/WebKit2/NetworkProcess/cache/NetworkCacheSubresourcesEntry.cpp b/Source/WebKit2/NetworkProcess/cache/NetworkCacheSubresourcesEntry.cpp index 4726979..c4bea12 100644 --- a/Source/WebKit2/NetworkProcess/cache/NetworkCacheSubresourcesEntry.cpp +++ b/Source/WebKit2/NetworkProcess/cache/NetworkCacheSubresourcesEntry.cpp @@ -72,7 +72,7 @@ Storage::Record SubresourcesEntry::encodeAsStorageRecord() const encoder.encodeChecksum(); - return { m_key, m_timeStamp, { encoder.buffer(), encoder.bufferSize() } , { } }; + return { m_key, m_timeStamp, { encoder.buffer(), encoder.bufferSize() } , Data() }; 1213123 18 changseok 2016-07-21 06:57:56 -0700 O.K This needs more touch. 1213580 19 changseok 2016-07-22 08:57:22 -0700 I confirmed that clang3.6-3.9 have the same issue. I also tried ~Data() = *default*, but it was not helpful for the linking failure. Instead Javier's approach worked. If his fix mitigates everyone's concern, I am ok with it. Javier, feel free to take over this ticket from me. Of course, I can lend my hand to land your fix. 1213642 20 darin 2016-07-22 11:16:25 -0700 (In reply to comment #19) > I confirmed that clang3.6-3.9 have the same issue. I also tried ~Data() = > *default*, but it was not helpful for the linking failure. Instead Javier's > approach worked. If his fix mitigates everyone's concern, I am ok with it. Now that it’s confirmed that this is a compiler bug that we are working around, I don’t have a preference for which of the two workarounds we use. But I do want a comment indicating that this is a workaround for a clang compiler bug and I would like someone to remove this when we no longer need to support the compiler with the bug. And ideally someone would report this to the clang project and we would track when it's fixed too. 1214515 21 jfernandez 2016-07-26 05:51:12 -0700 (In reply to comment #19) > I confirmed that clang3.6-3.9 have the same issue. I also tried ~Data() = > *default*, but it was not helpful for the linking failure. Instead Javier's > approach worked. If his fix mitigates everyone's concern, I am ok with it. > > Javier, feel free to take over this ticket from me. Of course, I can lend my > hand to land your fix. I really don't mind doing it. I still failed to reproduce the clang bug with some reduced test case, though. I'm a bit busy these days, but I think I can do land a workaround during this week, if you don't do it before. Anyway, if you end up implemented a similar approach to my workaround, I think it'd be better to avoid completely the aggregate-initialization of both Data arguments; it seems the bug is bypassed by just using an explicit instantiation on of of the Data fields, but I think it'd clearer doing it with both. 1575064 22 aperez 2019-09-28 12:04:01 -0700 Provided that nowadays Clang is at version 9.0, maybe this bug is not relevant anymore. What do you think? Is this relevant anymore? 2186068 23 aperez 2026-03-02 02:05:54 -0800 Closing due to radio silence, and FWIW we currently require Clang 18 and this is not an issue anymore. 284022 2016-07-19 11:32:51 -0700 2016-07-19 11:36:21 -0700 Patch bug-159931-20160719143154.patch text/plain 1428 changseok U3VidmVyc2lvbiBSZXZpc2lvbjogMjAzMzkzCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0Mi9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViS2l0Mi9DaGFuZ2VMb2cKaW5kZXggNzJkZjcwODFlZjBjZjMy NjRmOTUxMGVkNjIwMzhmOTdlOWIwOGI1YS4uMjE4ZTM5NjcwODk3ZDZjNzk2NWI4ZTI4YWE5MTVk ZmMwNDBkYmU3OCAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJLaXQyL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE2IEBACisyMDE2LTA3LTE5ICBDaGFu Z1Nlb2sgT2ggIDxjaGFuZ3Nlb2tAd2Via2l0Lm9yZz4KKworICAgICAgICBGaXggYSBsaW5raW5n IGZhaWx1cmUgY2F1c2VkIGJ5IE5ldHdvcmtDYWNoZTo6RGF0YTo6fkRhdGEoKQorICAgICAgICBo dHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9MTU5OTMxCisKKyAgICAgICAg TGlua2luZyBpcyBmYWlsZWQgd2l0aCBjbGFuZyAzLjYuIEl0IHNheXMgV2ViS2l0OjpOZXR3b3Jr Q2FjaGU6OkRhdGE6On5EYXRhIGlzIHVuZGVmaW5lZC4KKyAgICAgICAgSXQgaXMgZml4ZWQgYnkg YWRkaW5nIGEgZGVzdHJ1Y3RvciBmb3IgTmV0d29ya0NhY2hlOjpEYXRhLiAKKworICAgICAgICBS ZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICAqIE5ldHdvcmtQcm9jZXNzL2Nh Y2hlL05ldHdvcmtDYWNoZURhdGEuaDoKKyAgICAgICAgKFdlYktpdDo6TmV0d29ya0NhY2hlOjpE YXRhOjp+RGF0YSk6CisKIDIwMTYtMDctMTggIEFsZXggQ2hyaXN0ZW5zZW4gIDxhY2hyaXN0ZW5z ZW5Ad2Via2l0Lm9yZz4KIAogICAgICAgICB3ZWJib29rbWFya3NkIG5lZWRzIHRvIHVzZSB0aGUg c2FtZSBBcHBDYWNoZSBkaXJlY3RvcnkgYXMgTW9iaWxlU2FmYXJpCmRpZmYgLS1naXQgYS9Tb3Vy Y2UvV2ViS2l0Mi9OZXR3b3JrUHJvY2Vzcy9jYWNoZS9OZXR3b3JrQ2FjaGVEYXRhLmggYi9Tb3Vy Y2UvV2ViS2l0Mi9OZXR3b3JrUHJvY2Vzcy9jYWNoZS9OZXR3b3JrQ2FjaGVEYXRhLmgKaW5kZXgg MjI1NTg5ODU0ZGE0NzA2OGZmYzQ5ZmY2MGM1OTAxNGM4OTdmYmQ0OS4uYmY0OTFjYTlhYmNhOTcy MWE2Y2E5ODE1ZWFhODRlY2VjNTUwZTc2OSAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvTmV0 d29ya1Byb2Nlc3MvY2FjaGUvTmV0d29ya0NhY2hlRGF0YS5oCisrKyBiL1NvdXJjZS9XZWJLaXQy L05ldHdvcmtQcm9jZXNzL2NhY2hlL05ldHdvcmtDYWNoZURhdGEuaApAQCAtMTA2LDYgKzEwNiw4 IEBAIHB1YmxpYzoKICAgICBEYXRhKCkgeyB9CiAgICAgRGF0YShjb25zdCB1aW50OF90Kiwgc2l6 ZV90KTsKIAorICAgIHZpcnR1YWwgfkRhdGEoKSB7IH0KKwogICAgIHN0YXRpYyBEYXRhIGVtcHR5 KCk7CiAgICAgc3RhdGljIERhdGEgYWRvcHRNYXAodm9pZCogbWFwLCBzaXplX3QsIGludCBmZCk7 CiAK 284023 2016-07-19 11:36:26 -0700 2016-07-19 13:40:17 -0700 Patch bug-159931-20160719143529.patch text/plain 1427 changseok U3VidmVyc2lvbiBSZXZpc2lvbjogMjAzMzkzCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0Mi9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViS2l0Mi9DaGFuZ2VMb2cKaW5kZXggNzJkZjcwODFlZjBjZjMy NjRmOTUxMGVkNjIwMzhmOTdlOWIwOGI1YS4uMmIxYTEyNDRiNmU4NDdjNTFjMzk5NzI0M2UxMTli MWQ5MDhhN2YwNCAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJLaXQyL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE2IEBACisyMDE2LTA3LTE5ICBDaGFu Z1Nlb2sgT2ggIDxjaGFuZ3Nlb2tAd2Via2l0Lm9yZz4KKworICAgICAgICBGaXggYSBsaW5raW5n IGZhaWx1cmUgY2F1c2VkIGJ5IE5ldHdvcmtDYWNoZTo6RGF0YTo6fkRhdGEoKQorICAgICAgICBo dHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9MTU5OTMxCisKKyAgICAgICAg TGlua2luZyBpcyBmYWlsZWQgd2l0aCBjbGFuZyAzLjYuIEl0IHNheXMgV2ViS2l0OjpOZXR3b3Jr Q2FjaGU6OkRhdGE6On5EYXRhIGlzIHVuZGVmaW5lZC4KKyAgICAgICAgSXQgaXMgZml4ZWQgYnkg YWRkaW5nIGEgZGVzdHJ1Y3RvciBmb3IgTmV0d29ya0NhY2hlOjpEYXRhLgorCisgICAgICAgIFJl dmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgICogTmV0d29ya1Byb2Nlc3MvY2Fj aGUvTmV0d29ya0NhY2hlRGF0YS5oOgorICAgICAgICAoV2ViS2l0OjpOZXR3b3JrQ2FjaGU6OkRh dGE6On5EYXRhKToKKwogMjAxNi0wNy0xOCAgQWxleCBDaHJpc3RlbnNlbiAgPGFjaHJpc3RlbnNl bkB3ZWJraXQub3JnPgogCiAgICAgICAgIHdlYmJvb2ttYXJrc2QgbmVlZHMgdG8gdXNlIHRoZSBz YW1lIEFwcENhY2hlIGRpcmVjdG9yeSBhcyBNb2JpbGVTYWZhcmkKZGlmZiAtLWdpdCBhL1NvdXJj ZS9XZWJLaXQyL05ldHdvcmtQcm9jZXNzL2NhY2hlL05ldHdvcmtDYWNoZURhdGEuaCBiL1NvdXJj ZS9XZWJLaXQyL05ldHdvcmtQcm9jZXNzL2NhY2hlL05ldHdvcmtDYWNoZURhdGEuaAppbmRleCAy MjU1ODk4NTRkYTQ3MDY4ZmZjNDlmZjYwYzU5MDE0Yzg5N2ZiZDQ5Li5iZjQ5MWNhOWFiY2E5NzIx YTZjYTk4MTVlYWE4NGVjZWM1NTBlNzY5IDEwMDY0NAotLS0gYS9Tb3VyY2UvV2ViS2l0Mi9OZXR3 b3JrUHJvY2Vzcy9jYWNoZS9OZXR3b3JrQ2FjaGVEYXRhLmgKKysrIGIvU291cmNlL1dlYktpdDIv TmV0d29ya1Byb2Nlc3MvY2FjaGUvTmV0d29ya0NhY2hlRGF0YS5oCkBAIC0xMDYsNiArMTA2LDgg QEAgcHVibGljOgogICAgIERhdGEoKSB7IH0KICAgICBEYXRhKGNvbnN0IHVpbnQ4X3QqLCBzaXpl X3QpOwogCisgICAgdmlydHVhbCB+RGF0YSgpIHsgfQorCiAgICAgc3RhdGljIERhdGEgZW1wdHko KTsKICAgICBzdGF0aWMgRGF0YSBhZG9wdE1hcCh2b2lkKiBtYXAsIHNpemVfdCwgaW50IGZkKTsK IAo= 284043 2016-07-19 13:40:22 -0700 2016-07-20 13:04:03 -0700 Patch bug-159931-20160719163925.patch text/plain 1419 changseok U3VidmVyc2lvbiBSZXZpc2lvbjogMjAzMzkzCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0Mi9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViS2l0Mi9DaGFuZ2VMb2cKaW5kZXggNzJkZjcwODFlZjBjZjMy NjRmOTUxMGVkNjIwMzhmOTdlOWIwOGI1YS4uMmIxYTEyNDRiNmU4NDdjNTFjMzk5NzI0M2UxMTli MWQ5MDhhN2YwNCAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJLaXQyL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE2IEBACisyMDE2LTA3LTE5ICBDaGFu Z1Nlb2sgT2ggIDxjaGFuZ3Nlb2tAd2Via2l0Lm9yZz4KKworICAgICAgICBGaXggYSBsaW5raW5n IGZhaWx1cmUgY2F1c2VkIGJ5IE5ldHdvcmtDYWNoZTo6RGF0YTo6fkRhdGEoKQorICAgICAgICBo dHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9MTU5OTMxCisKKyAgICAgICAg TGlua2luZyBpcyBmYWlsZWQgd2l0aCBjbGFuZyAzLjYuIEl0IHNheXMgV2ViS2l0OjpOZXR3b3Jr Q2FjaGU6OkRhdGE6On5EYXRhIGlzIHVuZGVmaW5lZC4KKyAgICAgICAgSXQgaXMgZml4ZWQgYnkg YWRkaW5nIGEgZGVzdHJ1Y3RvciBmb3IgTmV0d29ya0NhY2hlOjpEYXRhLgorCisgICAgICAgIFJl dmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgICogTmV0d29ya1Byb2Nlc3MvY2Fj aGUvTmV0d29ya0NhY2hlRGF0YS5oOgorICAgICAgICAoV2ViS2l0OjpOZXR3b3JrQ2FjaGU6OkRh dGE6On5EYXRhKToKKwogMjAxNi0wNy0xOCAgQWxleCBDaHJpc3RlbnNlbiAgPGFjaHJpc3RlbnNl bkB3ZWJraXQub3JnPgogCiAgICAgICAgIHdlYmJvb2ttYXJrc2QgbmVlZHMgdG8gdXNlIHRoZSBz YW1lIEFwcENhY2hlIGRpcmVjdG9yeSBhcyBNb2JpbGVTYWZhcmkKZGlmZiAtLWdpdCBhL1NvdXJj ZS9XZWJLaXQyL05ldHdvcmtQcm9jZXNzL2NhY2hlL05ldHdvcmtDYWNoZURhdGEuaCBiL1NvdXJj ZS9XZWJLaXQyL05ldHdvcmtQcm9jZXNzL2NhY2hlL05ldHdvcmtDYWNoZURhdGEuaAppbmRleCAy MjU1ODk4NTRkYTQ3MDY4ZmZjNDlmZjYwYzU5MDE0Yzg5N2ZiZDQ5Li42MWNiODNkNmRhZmRhZmJk M2IxZGNjNjBiMWVmNmM5MDYwNGIzNmVjIDEwMDY0NAotLS0gYS9Tb3VyY2UvV2ViS2l0Mi9OZXR3 b3JrUHJvY2Vzcy9jYWNoZS9OZXR3b3JrQ2FjaGVEYXRhLmgKKysrIGIvU291cmNlL1dlYktpdDIv TmV0d29ya1Byb2Nlc3MvY2FjaGUvTmV0d29ya0NhY2hlRGF0YS5oCkBAIC0xMDYsNiArMTA2LDgg QEAgcHVibGljOgogICAgIERhdGEoKSB7IH0KICAgICBEYXRhKGNvbnN0IHVpbnQ4X3QqLCBzaXpl X3QpOwogCisgICAgfkRhdGEoKSB7IH0KKwogICAgIHN0YXRpYyBEYXRhIGVtcHR5KCk7CiAgICAg c3RhdGljIERhdGEgYWRvcHRNYXAodm9pZCogbWFwLCBzaXplX3QsIGludCBmZCk7CiAK