Tags: stephie0128/ChakraCore
Tags
[MERGE chakra-core#5444 @atulkatti] JULY 2018 Security Update Merge pull request chakra-core#5444 from atulkatti:servicing/1807_1.10 JULY 2018 Security Update that addresses the following issues in ChakraCore.dll: [CVE-2018-8275], [CVE-2018-8276], [CVE-2018-8279], [CVE-2018-8280], [CVE-2018-8283], [CVE-2018-8286], [CVE-2018-8287], [CVE-2018-8288], [CVE-2018-8290], [CVE-2018-8291], [CVE-2018-8294], [CVE-2018-8298]
[MERGE chakra-core#5385 @MSLaguana] Preparing initial release of Chak… …raCore 1.10 Merge pull request chakra-core#5385 from MSLaguana:rel1.10
[MERGE chakra-core#5298 @leirocks] ChakraCore 2018-06 security updates Merge pull request chakra-core#5298 from leirocks:servicing/1806_1.8
[MERGE chakra-core#5116 @MSLaguana] ChakraCore 2018-05 security updates Merge pull request chakra-core#5116 from MSLaguana:servicing/1805
[MERGE chakra-core#4963 @rajatd] ChakraCore 2018-04 security updates Merge pull request chakra-core#4963 from rajatd:servicing/1804
[MERGE chakra-core#4812 @akroshg] ChakraCore 2018-03 Security updates Merge pull request chakra-core#4812 from akroshg:test1803_1 Pushing 18-03 changes.
[MERGE chakra-core#4535 @meg-gupta] Fix setting hasBailout when there… … are inlined functions in try/catch Merge pull request chakra-core#4535 from meg-gupta:fixCatchEatingUpEx Fixes OS#15078638 When we bailout executing trycode from within OP_TryCatch, we complete the execution of the current function enclosing the try/catch in the interpreter. If there was an exception within the try region, it is caught and handled accordingly in ProcessTryHandlerBailOut which reconstructs try/catch/finally frames when we bailout midway executing code within OP_TryCatch. If there was an exception outside the try region, the catch of the OP_TryCatch ends up catching it, because it happens to be on the callstack. For this we use the hasBailOut bit which is per function, so we know that this exception has to be passed above. When we have inlined functions inside the try, and for bailouts inside the inlined code, we do not set the hasBailedOut bit, so that the enclosing functions catch in OP_TryCatch catches it. But when we bailout from inlined code inside try, we execute inlined code, as well as the enclosing function's code in the interpreter. We will be execution code past the try/catch of the current function in the interpreter. Now if any code outside the eh region throws, we will catch that in OP_TryCatch which happens to be on the callstack. And we will end up handling it instead of passing above because we have not set the hasBailedOutBit from the bailout point. This change fixes this issue. We pass a pointer to the hasBailedOutBit and set it once we have finished executing the inlined frames and are ready to process the interpreter frame of the current function.
[MERGE chakra-core#4512 @MSLaguana] Updating intl bytecode headers Merge pull request chakra-core#4512 from MSLaguana:updateBytecode Bytecode headers had not been updated since a (minor) source code change.
[MERGE chakra-core#4411 @MikeHolman] 17-12 Security Update Merge pull request chakra-core#4411 from MikeHolman:1712 17-12 Security Update that addresses the following issues in ChakraCore: CVE-2017-11889 CVE-2017-11893 CVE-2017-11894 CVE-2017-11905 CVE-2017-11908 CVE-2017-11909 CVE-2017-11910 CVE-2017-11911 CVE-2017-11912 CVE-2017-11914 CVE-2017-11916 CVE-2017-11918 CVE-2017-11919 CVE-2017-11930
PreviousNext