 Jianchun Xu
|
972009a89e
ChakraCore fix for servicing release 18-02B: CVE-2018-0858
|
há 8 anos atrás |
 Hitesh Kanwathirtha
|
4378b6a971
Devirtualize FunctionInfo
|
há 8 anos atrás |
 Rajat Dua
|
f8acf7f377
FromVar hardening
|
há 8 anos atrás |
|
Rajat Dua
|
f08408b524
[CVE-2017-11808] Type confusion in FrameDisplay items may lead to oob read/write
|
há 8 anos atrás |
 Sandeep Agarwal
|
e2b35ab97b
[CVE-2017-11809] Edge - Chakra: Accesses to uninitialized pointers in StackScriptFunction::BoxState::Box.
|
há 8 anos atrás |
 Louis Lafreniere
|
6e39312394
Initialize locals
|
há 8 anos atrás |
 Lei Shi
|
ad40625c48
Merge branch 'master' into users/leish/wb3merge
|
há 9 anos atrás |
 Paul Leathers
|
0824e05a0c
2 very simple redeferral fixes: 1. Now that the ScopeSlots structure uses a FunctionInfo*, make sure that the FunctionInfo points to a full FunctionBody before attempting to box a StackScriptFunction. 2. In addition to changing the type's entry point when we redefer, make sure we change a ScriptFunctionType's entry point info's type. This is necessary in the case of a cross-site call to a redeferred function.
|
há 9 anos atrás |
|
Lei Shi
|
578f5b959f
fix for the special constructor for functionBody (which enforced write barrier)
|
há 9 anos atrás |
|
Lei Shi
|
1c7dc3a7c2
some write barrier fixes
|
há 9 anos atrás |
|
Jianchun Xu
|
4c9b2e23ea
Merge branch 'master' into swb
|
há 9 anos atrás |
|
Paul Leathers
|
40d7a92d42
Replace FunctionBody reference in ScopeSlots with FunctionInfo: The ScopeSlots structure used to hold closure-captured variables has a reference to the scope in the form of a FunctionBody*. But that FunctionBody may become a stale reference after redeferral. Replace with an immutable FunctionInfo*.
|
há 9 anos atrás |
|
Jianchun Xu
|
3d72466243
Merge remote-tracking branch 'master' into swb
|
há 9 anos atrás |
|
Paul Leathers
|
a384d2e3a3
Redefer function bodies that are not currently being executed and are
|
há 9 anos atrás |
|
Rajat Dua
|
afefdfb779
Fix a conditional check when determining whether we have an inlinee to box in a native frame
|
há 9 anos atrás |
|
Rajat Dua
|
c143db5f0f
Bug fixes - marking src of CheckObjType jit-optimized, right bailout offset on InlineeEnd, and around boxing of inlinees in a jitted loop body
|
há 9 anos atrás |
 Mark Marron
|
648181a5e4
Merge from Time-Travel Debugging with history squash.
|
há 9 anos atrás |
|
Paul Leathers
|
cb554395f5
In jitted loop bodies, we may have stack-allocated closures but not create stack-allocated functions inside block scopes. Let BoxScopeSlots call BoxStackFunction for functions not on stack to make sure that any stack closures get boxed.
|
há 10 anos atrás |
|
Paul Leathers
|
c252469853
Remove an assert that doesn't hold if a boxable stack function is found in a frame display.
|
há 10 anos atrás |
|
Jianchun Xu
|
cec0e9a84f
replace wchar_t/L"..." with char16/_u("...")
|
há 10 anos atrás |
|
Paul Leathers
|
10aae6f0aa
Protect ForEachStackNestedFunc iteration with a trivial check on whether the current FunctionBody is blocked from having stack nested functions.
|
há 10 anos atrás |
|
Paul Leathers
|
d04387101b
We need to box frame display and scope slots in a native frame before calling BoxNativeFrame so that the boxed versions can be found and written back to their known locations on the frame.
|
há 10 anos atrás |
|
Jianchun Xu
|
8567cfe6c1
update all references to use correct CamelCase file names
|
há 10 anos atrás |
|
Jianchun Xu
|
47eab483b9
change top-level dir Lib back to lib
|
há 10 anos atrás |
|
Jianchun Xu
|
4e05cc2969
rename /lib/ files to CamelCase
|
há 10 anos atrás |
|
Lei Shi
|
5619992ce8
Move rarely used pointer from FunctionBody to a seperate structure.
|
há 10 anos atrás |
 ChakraBot
|
5d8406741f
Initial commit
|
há 10 anos atrás |
