ILSpy

Commit Graph

Author	SHA1	Message	Date
Siegfried Pammer	7502e521c1	Improve documentation of ExpressionTransforms.TransformCatchVariable	6 years ago
Siegfried Pammer	44c044aa33	Fix #1635 : DynamicInvokeConstructorInstruction Did report the wrong StackType for value types and unknown types.	6 years ago
Siegfried Pammer	85d1d21b35	Fix possible NRE in DynamicCallSiteTransform.ExtractArgumentInfo.	6 years ago
Siegfried Pammer	587f61be19	Fix #1280 : Transform RuntimeHelpersInitializeArray calls without receiving assignments.	6 years ago
Siegfried Pammer	87353aac5a	Fix #1624 : Stack type mismatch in expression trees.	6 years ago
Siegfried Pammer	4246a178f4	Remove unused parameter.	6 years ago
Siegfried Pammer	fdf4228a17	Fix #1623 : Nest local functions correctly, if captured variables are used.	6 years ago
Siegfried Pammer	3ad8c88d39	Fix #1622 : Transform local functions calls with named arguments correctly	6 years ago
Siegfried Pammer	72c3e828f3	Check IsPure before Match.	6 years ago
Daniel Grunwald	2b9a40371b	Remove the dead LoopingTransform. Make ILInstruction.IsDirty debug-only, as it is only used for assertions now.	6 years ago
Siegfried Pammer	c477ccf68e	Add SemanticHelper.IsPure check as per code-review.	6 years ago
Siegfried Pammer	0d92ee805c	#1349 : Insert conv instruction in expression tree, if inst.ResultType != typeHint.GetStackType()	6 years ago
Siegfried Pammer	0ca5ace524	#1349 : Fix assertions triggered by incorrect local function decompilation.	6 years ago
Siegfried Pammer	667ac1a980	Fix #1538 : Insert expression.tree.cast instructions to coerce references to outer expression tree parameter definitions to their actual type.	6 years ago
Siegfried Pammer	a3dbc27f76	Fix #1573 : if the expected type of the this argument of an expression tree call is unknown, insert an explicit conv *->unknown instruction	6 years ago
Siegfried Pammer	679623e4c3	Fix #1609 : Decompilation of cached delegate-construction with unknown delegate type.	6 years ago
Daniel Grunwald	fc73851bb4	#1610 : Slightly more aggressive copy propagation. This helps clean up the mess left behind when stack slots are not eliminated by the normal transforms. We previously didn't do this because aggressive copy propagation could confuse the normal transforms; but this is no longer an issue with the new pass ordering.	6 years ago
Siegfried Pammer	ec18094c65	Implement transformation of delegate construction with ldvirtftn.	6 years ago
Siegfried Pammer	65f9fe1076	Remove unused code.	6 years ago
Siegfried Pammer	fd24b5ea26	Simplify check for addresses	6 years ago
Siegfried Pammer	c0f954aaa6	Simplify nullable value types throw expression pattern.	6 years ago
Siegfried Pammer	50c4fa1c13	Add two more patterns with throw expressions.	6 years ago
Siegfried Pammer	1e3e8fdac1	Allow ref-locals to be initialized from nested field addresses.	6 years ago
Daniel Grunwald	df13f0ce01	Fix #1605 : Inline ref locals more aggressively; this is necessary for VB compound assignments.	6 years ago
Daniel Grunwald	0e0179edff	Improve NullPropagation (?.) when fields of value-type are involved.	6 years ago
Daniel Grunwald	7f8689c464	Allow inlining of compiler-generated value-type-temporaries when field accesses are involved.	6 years ago
Siegfried Pammer	7d4b4c6433	Implement NullCoalescingTransform with value types.	6 years ago
Daniel Grunwald	752f0de978	Further adjustments to transforms now that copy propagation no longer runs.	6 years ago
Siegfried Pammer	2ca8c31d8e	Fix ArgumentException: ignore duplicate keys after variable splitting in TransformDisplayClassUsage	6 years ago
Siegfried Pammer	adc443e4e2	Fix dynamic compound assignments	6 years ago
Daniel Grunwald	97d490a3df	Slightly generalize the post-increment transform so that it no longer depends on copy propagation.	6 years ago
Siegfried Pammer	25f625af99	Move copy propagation after expression and statement transforms.	6 years ago
Siegfried Pammer	7d62a7e032	Reset HasInitialValue flag after TransformDisplayClassUsage.	6 years ago
Siegfried Pammer	e5f95143cf	TransformDisplayClassUsage: set HasInitialValue = true for newly introduced locals.	6 years ago
Daniel Grunwald	6c0216bbb9	Fix assertion in ReduceNestingTransform: after copying the exitInst, it was possible that the old copy of the exitInst became unreachable.	6 years ago
Daniel Grunwald	8117dfff4a	Fix #1242 : Discard unreachable code. Unreachable code is not part of the dominator tree, which most of our transforms are based on. In particular, dominance-based loop detection runs into the problem where unreachable code might have jumps into two independent loops. In that case, it's impossible to place the unreachable code in a way that avoids assertions / generating invalid C#. We establish the invariant that all blocks in a BlockContainer must be statically reachable from the entry point (-> every block is part of the dominator tree). This means transforms no longer have to deal with special cases for unreachable code. The "Remove dead and side effect free code" option still has an effect on dead stores, but unreachable code is now always removed (previously this also was dependent on this option).	6 years ago
Daniel Grunwald	c3037afa70	Fix #1352 : Assertion in NullableLiftingTransform.DoLift	6 years ago
Daniel Grunwald	539e3a906d	Fix #1552 : user-defined and decimal increments For user-defined increments, there were problems with Roslyn was optimizing out one of the stores. The new transform FixRemainingIncrements now takes increments/decrements that were not detected by TransformAssignment and introduces a temporary variable that can be incremented. This sometimes requires un-inlining via the new ILInstruction.Extract() operation. Extract() is not supported in all possible contexts, so it is possible but unlikely that some op_Increment calls remain. For decimals, the situation is different: legacy csc actually was optimizing "d + 1m" to "op_Increment(d)", so we can get rid of any left-over increments by undoing this optimization. This now happens in ReplaceMethodCallsWithOperators.	6 years ago
Siegfried Pammer	61231b7725	Fix #1593 : TransformCollectionAndObjectInitializers uses nested indexer call as target of object initializer.	6 years ago
Daniel Grunwald	856cedc95e	#1456 : add test case; add additional checks to ensure we only transform normal locals into using/foreach-locals	6 years ago
Daniel Grunwald	beed6b5e24	Remove MarkGeneratedVariables() step in AsyncAwaitDecompiler. This was left-over from earlier versions; but ILSpy stopped caring so much about variable vs. stack slot since Roslyn started to optimize more aggressively. The change of variable type caused problems for debug information and could even cause an assertion. Closes #1456, closes #1562.	6 years ago
Siegfried Pammer	f10ab69328	Improve local-function detection to minimize false positives.	6 years ago
Siegfried Pammer	a109b77858	Refactor LocalFunctionDecompiler to allow mutually recursive local functions to be decompiled correctly.	6 years ago
Siegfried Pammer	e215f69b2d	Extend description of LocalFunctionDecompiler	6 years ago
Siegfried Pammer	37e14f43e3	Refactor/Clean up local functions representation in type system.	6 years ago
Daniel Grunwald	33c7425fa2	Fix several bugs with compound assignments on indexers. Closes #1580.	6 years ago
Siegfried Pammer	f3f38783df	ICSharpCode.Decompiler roundtrip test now completes	6 years ago
Daniel Grunwald	ccdabf2325	Fix a bunch of XmlDoc compiler warnings.	6 years ago
Daniel Grunwald	b32cae412a	Eliminate BlockKind.PostfixOperator. We can now represent all compound assignments using the CompoundAssignInstruction, including those on local variables. Closes #1583.	6 years ago
Daniel Grunwald	f2151972a0	Redesign CompoundAssignmentInstruction to support an 'Address' mode where we don't implicitly turn an LdObj into an StObj, but instead directly operate on an address. This means we can get rid of the special case in TransformDisplayClassUsage, as compound.assign can now also be used with the address of a local variable.	6 years ago

1 2 3 4 5 ...

689 Commits (7502e521c150f1a314fea01865a2f8950f0d636d)