198518 2019-06-04 00:57:14 -0700 REGRESSION(r245586): static assertion failed: Match result and EncodedMatchResult should be the same size 2019-06-26 01:28:34 -0700 1 1 1 Unclassified WebKit WebKitGTK Other Unspecified Unspecified RESOLVED FIXED P2 Normal --- 1 berto mcatanzaro bugs-noreply cgarcia commit-queue erack ews-watchlist keith_miller mark.lam mcatanzaro msaboff saam tzagallo ysuzuki oldest_to_newest 1541477 0 berto 2019-06-04 00:57:14 -0700 The build fails in most architectures (exceptions: i386, x86_64, aarch64). ------------------------------------------------------- These errors are from mips64: In file included from ../Source/JavaScriptCore/runtime/RegExp.h:25, from ../Source/JavaScriptCore/runtime/RegExpCachedResult.h:30, from ../Source/JavaScriptCore/runtime/RegExpGlobalData.h:28, from ../Source/JavaScriptCore/runtime/JSGlobalObject.h:42, from ../Source/JavaScriptCore/runtime/ExecutableBase.h:33, from ../Source/JavaScriptCore/runtime/ScriptExecutable.h:28, from ../Source/JavaScriptCore/runtime/GlobalExecutable.h:29, from ../Source/JavaScriptCore/runtime/EvalExecutable.h:29, from ../Source/JavaScriptCore/runtime/DirectEvalExecutable.h:28, from ../Source/JavaScriptCore/bytecode/DirectEvalCodeCache.h:31, from ../Source/JavaScriptCore/bytecode/CodeBlock.h:42, from ../Source/JavaScriptCore/bytecode/StructureStubInfo.h:28, from ../Source/JavaScriptCore/bytecode/BytecodeDumper.h:33, from DerivedSources/JavaScriptCore/BytecodeStructs.h:32, from ../Source/JavaScriptCore/llint/LLIntOffsetsExtractor.cpp:31: ../Source/JavaScriptCore/runtime/MatchResult.h: At global scope: ../Source/JavaScriptCore/runtime/MatchResult.h:82:35: error: static assertion failed: Match result and EncodedMatchResult should be the same size static_assert(sizeof(MatchResult) == sizeof(EncodedMatchResult), "Match result and EncodedMatchResult should be the same size"); DerivedSources/ForwardingHeaders/wtf/StdLibExtras.h: In instantiation of ‘ToType WTF::bitwise_cast(FromType) [with ToType = JSC::MatchResult; FromType = long unsigned int]’: ../Source/JavaScriptCore/runtime/MatchResult.h:55:48: required from here DerivedSources/ForwardingHeaders/wtf/StdLibExtras.h:136:36: error: static assertion failed: bitwise_cast size of FromType and ToType must be equal! static_assert(sizeof(FromType) == sizeof(ToType), "bitwise_cast size of FromType and ToType must be equal!"); ------------------------------------------------------- This one is from armhf: In file included from DerivedSources/ForwardingHeaders/wtf/StdLibExtras.h:32, from DerivedSources/ForwardingHeaders/wtf/FastMalloc.h:25, from ../Source/WebCore/config.h:50, from ../Source/WebCore/rendering/style/StyleGeneratedImage.cpp:24, from DerivedSources/WebCore/unified-sources/UnifiedSource-043dd90b-23.cpp:1: ../Source/WebCore/rendering/style/StyleRareInheritedData.cpp:78:47: error: static assertion failed: StyleRareInheritedData_should_bit_pack COMPILE_ASSERT(sizeof(StyleRareInheritedData) <= sizeof(GreaterThanOrSameSizeAsStyleRareInheritedData), StyleRareInheritedData_should_bit_pack); 1541526 1 mcatanzaro 2019-06-04 05:24:30 -0700 (In reply to Alberto Garcia from comment #0) > This one is from armhf: > > In file included from DerivedSources/ForwardingHeaders/wtf/StdLibExtras.h:32, > from DerivedSources/ForwardingHeaders/wtf/FastMalloc.h:25, > from ../Source/WebCore/config.h:50, > from > ../Source/WebCore/rendering/style/StyleGeneratedImage.cpp:24, > from > DerivedSources/WebCore/unified-sources/UnifiedSource-043dd90b-23.cpp:1: > ../Source/WebCore/rendering/style/StyleRareInheritedData.cpp:78:47: error: > static assertion failed: StyleRareInheritedData_should_bit_pack > COMPILE_ASSERT(sizeof(StyleRareInheritedData) <= > sizeof(GreaterThanOrSameSizeAsStyleRareInheritedData), > StyleRareInheritedData_should_bit_pack); This is bug #198274. For GNOME, we are hitting this on aarch64 as well. Let's use this bug for the first issue. 1545231 2 erack 2019-06-17 04:56:52 -0700 Dumb question, is this fixed in webkitgtk-2.25.2? 1545233 3 erack 2019-06-17 05:00:51 -0700 CI on master indicates it isn't. 1545576 4 berto 2019-06-18 01:12:32 -0700 (In reply to Eike Rathke from comment #2) > Dumb question, is this fixed in webkitgtk-2.25.2? It isn't, in Debian it fails in (at least) ppc64, ppc64el and s390x. https://buildd.debian.org/status/package.php?p=webkit2gtk&suite=experimental 1547468 5 372782 erack 2019-06-24 12:10:47 -0700 Created attachment 372782 proposed patch It is unclear to me why in Source/JavaScriptCore/runtime/MatchResult.h the EncodedMatchResult should depend on CPU, but clearly with using EncodedMatchResult = uint64_t; and static_assert(sizeof(MatchResult) == sizeof(EncodedMatchResult)) and the two size_t member variables of MatchResult that can compile only if size_t is 32-bit and the member variables aligned. So the straight forward change would be to do exactly that, but I don't know if there is a reason to have CPU type involved here (because of JIT asm). Attached is a tad ugly patch which on ppc64le at least compiles. Does that suit the purpose? If yes, I can create a ready-to-commit patch. 1547477 6 372782 mcatanzaro 2019-06-24 12:24:22 -0700 Comment on attachment 372782 proposed patch You can set r? to request review and cq? to request commit-queue. But the r+ and cq+ flags are reserved for WebKit reviewers and committers. 1547486 7 372782 erack 2019-06-24 12:36:51 -0700 Comment on attachment 372782 proposed patch Meh, misclicked, was supposed to be 'r?' 1547497 8 372782 mcatanzaro 2019-06-24 12:54:04 -0700 Comment on attachment 372782 proposed patch View in context: https://bugs.webkit.org/attachment.cgi?id=372782&action=review > Source/JavaScriptCore/runtime/MatchResult.h 2019-06-24 18:36:36.947345256 +0000:43 > #if CPU(ARM64) || CPU(X86_64) > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 1 > using EncodedMatchResult = MatchResult; > -#else > +#elif (defined(SIZE_WIDTH) && (SIZE_WIDTH == 32)) || (SIZE_MAX == 0xffffffff) > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 0 > using EncodedMatchResult = uint64_t; > +#else > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 1 > +using EncodedMatchResult = MatchResult; > #endif Hm, well I see this should fix the problem. Thanks. But it's really messy and honestly not great. Looking at the commit that introduced this build failure, r245586 "Cleanup Yarr regexp code around paren contexts," I see the code is rather complicated and I'm not confident in how to fix this nicely without reverting that change, but that commit was clearly incorrect. It assumes size_t is 32-bit except on x86_64 and aarch64, which is of course not right. The code only works if size_t happens to be the same as uint32_t, so I guess it should use uint32_t if that's what's desired. However, the constructor uses size_t and changing it to use uint32_t would require auditing a large number of callsites, so that should be left to JSC developers. I've just visited bug #198063 to ask Keith to comment here. Let's see what he thinks. 1547498 9 372782 mcatanzaro 2019-06-24 12:55:22 -0700 Comment on attachment 372782 proposed patch Thanks again. BTW, the patch should have a ChangeLog entry when setting r? which you can create with: Tools/Scripts/prepare-ChangeLog -b 198518 (If it's just a WIP patch without a changelog, don't set r? yet.) 1547501 10 372782 mark.lam 2019-06-24 13:03:20 -0700 Comment on attachment 372782 proposed patch View in context: https://bugs.webkit.org/attachment.cgi?id=372782&action=review Also missing ChangeLog. > Source/JavaScriptCore/runtime/MatchResult.h 2019-06-24 18:36:36.947345256 +0000:43 > #if CPU(ARM64) || CPU(X86_64) > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 1 > using EncodedMatchResult = MatchResult; > -#else > +#elif (defined(SIZE_WIDTH) && (SIZE_WIDTH == 32)) || (SIZE_MAX == 0xffffffff) > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 0 > using EncodedMatchResult = uint64_t; > +#else > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 1 > +using EncodedMatchResult = MatchResult; > #endif This fix is invalid. The issue is that MatchResult contains 2 size_t values. This means, on 32-bit platforms, sizeof(MatchResult) is 64-bit, but on 64-bit platforms, sizeof(MatchResult) is 128-bit. The proper fix is simply: #if CPU(ADDRESS32) using EncodedMatchResult = uint64_t; static_assert(sizeof(size_t) == sizeof(uint32_t), ""); #else using EncodedMatchResult = MatchResult; #endif 1547505 11 keith_miller 2019-06-24 13:13:37 -0700 (In reply to Mark Lam from comment #10) > Comment on attachment 372782 [details] > proposed patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=372782&action=review > > Also missing ChangeLog. > > > Source/JavaScriptCore/runtime/MatchResult.h 2019-06-24 18:36:36.947345256 +0000:43 > > #if CPU(ARM64) || CPU(X86_64) > > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 1 > > using EncodedMatchResult = MatchResult; > > -#else > > +#elif (defined(SIZE_WIDTH) && (SIZE_WIDTH == 32)) || (SIZE_MAX == 0xffffffff) > > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 0 > > using EncodedMatchResult = uint64_t; > > +#else > > +#define ENCODEDMATCHRESULT_IS_MATCHRESULT 1 > > +using EncodedMatchResult = MatchResult; > > #endif > > This fix is invalid. > > The issue is that MatchResult contains 2 size_t values. This means, on > 32-bit platforms, sizeof(MatchResult) is 64-bit, but on 64-bit platforms, > sizeof(MatchResult) is 128-bit. > > The proper fix is simply: > > #if CPU(ADDRESS32) > using EncodedMatchResult = uint64_t; > static_assert(sizeof(size_t) == sizeof(uint32_t), ""); > #else > using EncodedMatchResult = MatchResult; > #endif I think a better assertion would be static_assert(sizeof(EncodedMatchResult) == 2 * sizeof(size_t)), which should be true on all platforms. I like the rest of your change though. The point of the EncodedMatchResult type is to ensure that the C calling convention will return the result in the two return registers rather than on the stack. 1547509 12 mcatanzaro 2019-06-24 13:19:41 -0700 Thanks Mark, Keith! I was about to post my own solution using SIZE_MAX, but yours is much better. Eike, Berto, could you please test this patch and make sure it works on the architectures you care about? 1547510 13 372790 mcatanzaro 2019-06-24 13:19:56 -0700 Created attachment 372790 Patch 1547511 14 372791 mcatanzaro 2019-06-24 13:24:41 -0700 Created attachment 372791 Patch 1547524 15 erack 2019-06-24 13:39:59 -0700 (In reply to Mark Lam from comment #10) > Also missing ChangeLog. Yes, it was a WIP patch. > The issue is that MatchResult contains 2 size_t values. This means, on > 32-bit platforms, sizeof(MatchResult) is 64-bit, but on 64-bit platforms, > sizeof(MatchResult) is 128-bit. > > The proper fix is simply: > > #if CPU(ADDRESS32) > using EncodedMatchResult = uint64_t; > static_assert(sizeof(size_t) == sizeof(uint32_t), ""); > #else > using EncodedMatchResult = MatchResult; > #endif I thought of such as well, but wasn't sure because of these odd Windows 64/32 platforms where at least a long still is only 32-bit. I'll try Michael's patch. (In reply to Keith Miller from comment #11) > The point of the EncodedMatchResult type is to ensure that the C calling > convention will return the result in the two return registers rather than on > the stack. Ah, thanks for enlightenment. 1547531 16 erack 2019-06-24 14:00:51 -0700 That part compiled at least on ppc64le, I'll prepare a Fedora rawhide build of 2.25.2 with the patch applied to cover the other platforms as well. 1547674 17 erack 2019-06-25 02:57:35 -0700 Ok, that did it. Thanks! 1547681 18 berto 2019-06-25 04:57:50 -0700 (In reply to Michael Catanzaro from comment #14) > Created attachment 372791 [details] > Patch I just built 2.25.2 on aarch64 with this patch. 1547693 19 372791 mcatanzaro 2019-06-25 08:04:02 -0700 Comment on attachment 372791 Patch Maybe Mark or Keith could set r+ if satisfied. Thanks again for the help! 1547700 20 372791 commit-queue 2019-06-25 09:08:33 -0700 Comment on attachment 372791 Patch Clearing flags on attachment: 372791 Committed r246792: <https://trac.webkit.org/changeset/246792> 1547701 21 commit-queue 2019-06-25 09:08:35 -0700 All reviewed patches have been landed. Closing bug. 1547985 22 erack 2019-06-26 01:28:34 -0700 CI Jenkins builds are back to normal for ppc64, ppc64le and s390x. 372782 2019-06-24 12:10:47 -0700 2019-06-24 13:19:51 -0700 proposed patch webkit-encodedmatchresult.patch text/plain 1013 erack LS0tIFNvdXJjZS9KYXZhU2NyaXB0Q29yZS9ydW50aW1lL01hdGNoUmVzdWx0LmguZW5jb2RlZG1h dGNocmVzdWx0ICAgMjAxOS0wNS0yMiAwODoxNjo0OC4wMDAwMDAwMDAgKzAwMDAKKysrIFNvdXJj ZS9KYXZhU2NyaXB0Q29yZS9ydW50aW1lL01hdGNoUmVzdWx0LmggICAgICAyMDE5LTA2LTI0IDE4 OjM2OjM2Ljk0NzM0NTI1NiArMDAwMApAQCAtMzIsMTEgKzMyLDE3IEBAIG5hbWVzcGFjZSBKU0Mg ewogCiBzdHJ1Y3QgTWF0Y2hSZXN1bHQ7CiAjaWYgQ1BVKEFSTTY0KSB8fCBDUFUoWDg2XzY0KQor I2RlZmluZSBFTkNPREVETUFUQ0hSRVNVTFRfSVNfTUFUQ0hSRVNVTFQgMQogdXNpbmcgRW5jb2Rl ZE1hdGNoUmVzdWx0ID0gTWF0Y2hSZXN1bHQ7Ci0jZWxzZQorI2VsaWYgKGRlZmluZWQoU0laRV9X SURUSCkgJiYgKFNJWkVfV0lEVEggPT0gMzIpKSB8fCAoU0laRV9NQVggPT0gMHhmZmZmZmZmZikK KyNkZWZpbmUgRU5DT0RFRE1BVENIUkVTVUxUX0lTX01BVENIUkVTVUxUIDAKIHVzaW5nIEVuY29k ZWRNYXRjaFJlc3VsdCA9IHVpbnQ2NF90OworI2Vsc2UKKyNkZWZpbmUgRU5DT0RFRE1BVENIUkVT VUxUX0lTX01BVENIUkVTVUxUIDEKK3VzaW5nIEVuY29kZWRNYXRjaFJlc3VsdCA9IE1hdGNoUmVz dWx0OwogI2VuZGlmCiAKKwogc3RydWN0IE1hdGNoUmVzdWx0IHsKICAgICBNYXRjaFJlc3VsdCgp CiAgICAgICAgIDogc3RhcnQoV1RGOjpub3RGb3VuZCkKQEAgLTUwLDcgKzU2LDcgQEAgc3RydWN0 IE1hdGNoUmVzdWx0IHsKICAgICB7CiAgICAgfQogCi0jaWYgIShDUFUoQVJNNjQpIHx8IENQVShY ODZfNjQpKQorI2lmICFFTkNPREVETUFUQ0hSRVNVTFRfSVNfTUFUQ0hSRVNVTFQKICAgICBBTFdB WVNfSU5MSU5FIE1hdGNoUmVzdWx0KEVuY29kZWRNYXRjaFJlc3VsdCBtYXRjaCkKICAgICAgICAg OiBzdGFydChiaXR3aXNlX2Nhc3Q8TWF0Y2hSZXN1bHQ+KG1hdGNoKS5zdGFydCkKICAgICAgICAg LCBlbmQoYml0d2lzZV9jYXN0PE1hdGNoUmVzdWx0PihtYXRjaCkuZW5kKQo= 372790 2019-06-24 13:19:56 -0700 2019-06-24 13:24:38 -0700 Patch bug-198518-20190624151955.patch text/plain 1879 mcatanzaro U3VidmVyc2lvbiBSZXZpc2lvbjogMjQ2NzUxCmRpZmYgLS1naXQgYS9Tb3VyY2UvSmF2YVNjcmlw dENvcmUvQ2hhbmdlTG9nIGIvU291cmNlL0phdmFTY3JpcHRDb3JlL0NoYW5nZUxvZwppbmRleCA3 NGJkMmZmYWNkZmYwZWNlNDFkYTQ4OTU5YzhiYTZjZWZkODM3YzI0Li45OGRhZjdiOTBmNDE4MmMy NWJmZTkwZjY3MzJlNTdiZmIyZDMzY2YwIDEwMDY0NAotLS0gYS9Tb3VyY2UvSmF2YVNjcmlwdENv cmUvQ2hhbmdlTG9nCisrKyBiL1NvdXJjZS9KYXZhU2NyaXB0Q29yZS9DaGFuZ2VMb2cKQEAgLTEs MyArMSwxNyBAQAorMjAxOS0wNi0yNCAgTWljaGFlbCBDYXRhbnphcm8gIDxtY2F0YW56YXJvQGln YWxpYS5jb20+CisKKyAgICAgICAgUkVHUkVTU0lPTihyMjQ1NTg2KTogc3RhdGljIGFzc2VydGlv biBmYWlsZWQ6IE1hdGNoIHJlc3VsdCBhbmQgRW5jb2RlZE1hdGNoUmVzdWx0IHNob3VsZCBiZSB0 aGUgc2FtZSBzaXplCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNn aT9pZD0xOTg1MTgKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAg ICAgICByMjQ1NTg2IG1hZGUgc29tZSBiYWQgYXNzdW1wdGlvbnMgYWJvdXQgdGhlIHNpemUgb2Yg c2l6ZV90LCB3aGljaCB3ZSBjYW4gc29sdmUgdXNpbmcgdGhlCisgICAgICAgIENQVShBRERSRVNT MzIpIGd1YXJkIHRoYXQgSSBkaWRuJ3Qga25vdyBhYm91dC4KKworICAgICAgICBUaGlzIHNvbHV0 aW9uIHdhcyBkZXZlbG9wZWQgYnkgTWFyayBMYW0gYW5kIEtlaXRoIE1pbGxlci4gSSdtIGp1c3Qg cHJlcGFyaW5nIHRoZSBwYXRjaC4KKworICAgICAgICAqIHJ1bnRpbWUvTWF0Y2hSZXN1bHQuaDoK KwogMjAxOS0wNi0yNCAgTWFyayBMYW0gIDxtYXJrLmxhbUBhcHBsZS5jb20+CiAKICAgICAgICAg QXJyYXlTbGljZSBuZWVkcyB0byBrZWVwIHRoZSBzb3VyY2UgYXJyYXkgYWxpdmUuCmRpZmYgLS1n aXQgYS9Tb3VyY2UvSmF2YVNjcmlwdENvcmUvcnVudGltZS9NYXRjaFJlc3VsdC5oIGIvU291cmNl L0phdmFTY3JpcHRDb3JlL3J1bnRpbWUvTWF0Y2hSZXN1bHQuaAppbmRleCBjNGFlY2FhYzc2YTQx Zjk0ZTE4NDlkNTVjNTJmNzkzOTJmMWRkN2MxLi45Yjg4MTk3OTQ5NzM2ZjhlZjE0YjAwNjc2OTAy ODQ1YzI0NzM4NzU2IDEwMDY0NAotLS0gYS9Tb3VyY2UvSmF2YVNjcmlwdENvcmUvcnVudGltZS9N YXRjaFJlc3VsdC5oCisrKyBiL1NvdXJjZS9KYXZhU2NyaXB0Q29yZS9ydW50aW1lL01hdGNoUmVz dWx0LmgKQEAgLTMxLDEwICszMSwxMSBAQAogbmFtZXNwYWNlIEpTQyB7CiAKIHN0cnVjdCBNYXRj aFJlc3VsdDsKLSNpZiBDUFUoQVJNNjQpIHx8IENQVShYODZfNjQpCi11c2luZyBFbmNvZGVkTWF0 Y2hSZXN1bHQgPSBNYXRjaFJlc3VsdDsKLSNlbHNlCisjaWYgQ1BVKEFERFJFU1MzMikKIHVzaW5n IEVuY29kZWRNYXRjaFJlc3VsdCA9IHVpbnQ2NF90Oworc3RhdGljX2Fzc2VydChzaXplb2YoRW5j b2RlZE1hdGNoUmVzdWx0KSA9PSAyICogc2l6ZW9mKHNpemVfdCkpOworI2Vsc2UKK3VzaW5nIEVu Y29kZWRNYXRjaFJlc3VsdCA9IE1hdGNoUmVzdWx0OwogI2VuZGlmCiAKIHN0cnVjdCBNYXRjaFJl c3VsdCB7CkBAIC01MCw3ICs1MSw3IEBAIHN0cnVjdCBNYXRjaFJlc3VsdCB7CiAgICAgewogICAg IH0KIAotI2lmICEoQ1BVKEFSTTY0KSB8fCBDUFUoWDg2XzY0KSkKKyNpZiBDUFUoQUREUkVTUzMy KQogICAgIEFMV0FZU19JTkxJTkUgTWF0Y2hSZXN1bHQoRW5jb2RlZE1hdGNoUmVzdWx0IG1hdGNo KQogICAgICAgICA6IHN0YXJ0KGJpdHdpc2VfY2FzdDxNYXRjaFJlc3VsdD4obWF0Y2gpLnN0YXJ0 KQogICAgICAgICAsIGVuZChiaXR3aXNlX2Nhc3Q8TWF0Y2hSZXN1bHQ+KG1hdGNoKS5lbmQpCg== 372791 2019-06-24 13:24:41 -0700 2019-06-25 09:08:33 -0700 Patch bug-198518-20190624152440.patch text/plain 2147 mcatanzaro U3VidmVyc2lvbiBSZXZpc2lvbjogMjQ2NzU5CmRpZmYgLS1naXQgYS9Tb3VyY2UvSmF2YVNjcmlw dENvcmUvQ2hhbmdlTG9nIGIvU291cmNlL0phdmFTY3JpcHRDb3JlL0NoYW5nZUxvZwppbmRleCA3 NGJkMmZmYWNkZmYwZWNlNDFkYTQ4OTU5YzhiYTZjZWZkODM3YzI0Li45OGRhZjdiOTBmNDE4MmMy NWJmZTkwZjY3MzJlNTdiZmIyZDMzY2YwIDEwMDY0NAotLS0gYS9Tb3VyY2UvSmF2YVNjcmlwdENv cmUvQ2hhbmdlTG9nCisrKyBiL1NvdXJjZS9KYXZhU2NyaXB0Q29yZS9DaGFuZ2VMb2cKQEAgLTEs MyArMSwxNyBAQAorMjAxOS0wNi0yNCAgTWljaGFlbCBDYXRhbnphcm8gIDxtY2F0YW56YXJvQGln YWxpYS5jb20+CisKKyAgICAgICAgUkVHUkVTU0lPTihyMjQ1NTg2KTogc3RhdGljIGFzc2VydGlv biBmYWlsZWQ6IE1hdGNoIHJlc3VsdCBhbmQgRW5jb2RlZE1hdGNoUmVzdWx0IHNob3VsZCBiZSB0 aGUgc2FtZSBzaXplCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNn aT9pZD0xOTg1MTgKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAg ICAgICByMjQ1NTg2IG1hZGUgc29tZSBiYWQgYXNzdW1wdGlvbnMgYWJvdXQgdGhlIHNpemUgb2Yg c2l6ZV90LCB3aGljaCB3ZSBjYW4gc29sdmUgdXNpbmcgdGhlCisgICAgICAgIENQVShBRERSRVNT MzIpIGd1YXJkIHRoYXQgSSBkaWRuJ3Qga25vdyBhYm91dC4KKworICAgICAgICBUaGlzIHNvbHV0 aW9uIHdhcyBkZXZlbG9wZWQgYnkgTWFyayBMYW0gYW5kIEtlaXRoIE1pbGxlci4gSSdtIGp1c3Qg cHJlcGFyaW5nIHRoZSBwYXRjaC4KKworICAgICAgICAqIHJ1bnRpbWUvTWF0Y2hSZXN1bHQuaDoK KwogMjAxOS0wNi0yNCAgTWFyayBMYW0gIDxtYXJrLmxhbUBhcHBsZS5jb20+CiAKICAgICAgICAg QXJyYXlTbGljZSBuZWVkcyB0byBrZWVwIHRoZSBzb3VyY2UgYXJyYXkgYWxpdmUuCmRpZmYgLS1n aXQgYS9Tb3VyY2UvSmF2YVNjcmlwdENvcmUvcnVudGltZS9NYXRjaFJlc3VsdC5oIGIvU291cmNl L0phdmFTY3JpcHRDb3JlL3J1bnRpbWUvTWF0Y2hSZXN1bHQuaAppbmRleCBjNGFlY2FhYzc2YTQx Zjk0ZTE4NDlkNTVjNTJmNzkzOTJmMWRkN2MxLi5hZGQyNjRlOTA5NzViMjY1ZmQ1YmUxMGYyNjMy ZmQ3NWUyODZjODFjIDEwMDY0NAotLS0gYS9Tb3VyY2UvSmF2YVNjcmlwdENvcmUvcnVudGltZS9N YXRjaFJlc3VsdC5oCisrKyBiL1NvdXJjZS9KYXZhU2NyaXB0Q29yZS9ydW50aW1lL01hdGNoUmVz dWx0LmgKQEAgLTMxLDEwICszMSwxMCBAQAogbmFtZXNwYWNlIEpTQyB7CiAKIHN0cnVjdCBNYXRj aFJlc3VsdDsKLSNpZiBDUFUoQVJNNjQpIHx8IENQVShYODZfNjQpCi11c2luZyBFbmNvZGVkTWF0 Y2hSZXN1bHQgPSBNYXRjaFJlc3VsdDsKLSNlbHNlCisjaWYgQ1BVKEFERFJFU1MzMikKIHVzaW5n IEVuY29kZWRNYXRjaFJlc3VsdCA9IHVpbnQ2NF90OworI2Vsc2UKK3VzaW5nIEVuY29kZWRNYXRj aFJlc3VsdCA9IE1hdGNoUmVzdWx0OwogI2VuZGlmCiAKIHN0cnVjdCBNYXRjaFJlc3VsdCB7CkBA IC01MCw3ICs1MCw3IEBAIHN0cnVjdCBNYXRjaFJlc3VsdCB7CiAgICAgewogICAgIH0KIAotI2lm ICEoQ1BVKEFSTTY0KSB8fCBDUFUoWDg2XzY0KSkKKyNpZiBDUFUoQUREUkVTUzMyKQogICAgIEFM V0FZU19JTkxJTkUgTWF0Y2hSZXN1bHQoRW5jb2RlZE1hdGNoUmVzdWx0IG1hdGNoKQogICAgICAg ICA6IHN0YXJ0KGJpdHdpc2VfY2FzdDxNYXRjaFJlc3VsdD4obWF0Y2gpLnN0YXJ0KQogICAgICAg ICAsIGVuZChiaXR3aXNlX2Nhc3Q8TWF0Y2hSZXN1bHQ+KG1hdGNoKS5lbmQpCkBAIC03OSw2ICs3 OSw3IEBAIHN0cnVjdCBNYXRjaFJlc3VsdCB7CiAgICAgc2l6ZV90IGVuZDsKIH07CiAKK3N0YXRp Y19hc3NlcnQoc2l6ZW9mKEVuY29kZWRNYXRjaFJlc3VsdCkgPT0gMiAqIHNpemVvZihzaXplX3Qp LCAiaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTE5ODUxOCNjMTEiKTsK IHN0YXRpY19hc3NlcnQoc2l6ZW9mKE1hdGNoUmVzdWx0KSA9PSBzaXplb2YoRW5jb2RlZE1hdGNo UmVzdWx0KSwgIk1hdGNoIHJlc3VsdCBhbmQgRW5jb2RlZE1hdGNoUmVzdWx0IHNob3VsZCBiZSB0 aGUgc2FtZSBzaXplIik7CiAKIH0gLy8gbmFtZXNwYWNlIEpTQwo=