Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Make Hostx64/arm crossgen /CreatePerfMap behave the same as Hostarm/arm crossgen
|
|
|
|
Always define printf/fprintf in JIT
|
|
Fix MultiReg methods on GenTree
|
|
Improved fix for target register false dependency of lzcnt/tzcnt/popcnt
|
|
- Ensure that the registry keys are always disposed
- Use smaller subset of registry APIs
- Reduce diffs with CoreCLR/CoreFX
- Contributes to #11009 and #17899
|
|
|
|
Fixes https://github.com/dotnet/coreclr/issues/19954
- `SetTargetInterlocked` can be soon followed by `ResetTargetInterlocked`, so the assert at the end of `SetTargetInterlocked` is invalid
- Removed the assert and instead just verified that the specified target is not the default prestub target
|
|
https://github.com/dotnet/coreclr/pull/17096 completely, which only showed modest size improvement
* Removing the _version increment from Clear() entirely to bring it in line with the behavior in Remove() and to keep size gains
[tfs-changeset: 1714543]
Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
|
|
For consistency with https://github.com/dotnet/coreclr/pull/20005
|
|
Update clrjit headers files included via CMakeList.txt into clrjit.vcxproj
|
|
|
|
|
|
|
|
The Copy/Reload case was not being handled for 64-bit targets with multireg ops.
Also, the methods `IsMultiRegNode`, `GetMultiRegCount`, `GetRegByIndex` and `GetRegTypeByIndex` should be in sync.
Fix #20031
|
|
|
|
Fixes #19902
|
|
(#20011)
|
|
Reduces diffs with CoreRT
|
|
Work toward #14884
|
|
* Add EH info support to R2RDump
This change expands runtime function dump to include the exception
handling info looked up via the EXCEPTION_INFO R2R header table.
* Address Zach's PR feedback
1) Base EHClause.Length on sizeof(uint) instead of sizeof(int)
for consistency with the asctual data types in the class.
2) Fix my overlooking that Zach spotted - in the EHLookupTable
ctor, we need to assign rva2 and eh2 to rva1 and eh1 at the end
of the loop block as we need to traverse consecutive pairs
of CORCOMPILE_EXCEPTION_LOOKUP_TABLE_ENTRY elements.
* Addressed Bruce Forstall's PR feedback
1) I modified the EH info method ctors to accept the R2RReader
instead of the raw byte[] image as the reader can be used to
provide textual representations of metadata tokens.
2) I changed the mask test to a switch that throws an exception
if multiple of the three lowest bits are set which is illegal
according to Bruce.
* Addressed additional Bruce Forstall's PR feedback
Remove interpretation of ClassTokenOrFilterOffset as it's not applicable
in the FINALLY and FAULT cases.
Thanks
Tomas
|
|
Port https://github.com/dotnet/corert/pull/6341/ to CoreCLR
|
|
|
|
Move FP DIV/MUL handling to genCodeForBinary
|
|
* Dump spmi string environment variables with -d key.
Sometimes it is not clear which environment variables are set during a replay. Add a possibility to dump all active string environment variables with "-v d" key.
It is especially useful when you debug one method and can't get output that you want (for example JitDump) because altJit was set during the collection.
* Dump int spmi environment variables with -d key.
The same change but for int variables that need an additional parsing.
* Fix environment reset in spmi.
The previous changes revealed that we have a problem with environment reset. The reset was added in #13596 to support a correct replay of mch files with mc that had different env variables during the collection.
It was based on Environment that was later deprecated by #13110.
That caused the environment to be reset for each mc file. The fix updates env only when it is necessary and decreases the replay time of an mch file with 200000 methods from 500s to 370s.
* Fix dumpHelp.
Fix formating and adds note that all keys and values are case sensetive.
* Delete retired packages logic.
These packages were retired a long time ago and there is no value to have a special logic for them.
|
|
Fix inconsistent Intel hardware intrinsic APIs
|
|
We need to arrange to actually update the call objp in the SpillRetExprHelper,
otherwise the changes don't propagate back into the tree.
|
|
In PAL based builds the stdio printf is already a macro so JIT's host.h fails to replace it with logf/flogf.
|
|
|
|
* Move ResoureReader/ResourceSet to shared.
* Remove unnecessary cast to RuntimeType from ResoureReader.
|
|
* convert HandleStore::_underlyingBucket to a pointer
* undo _underlyingBucket pointer change and delete Ref_CreateHandleTableBucket
* update comment
|
|
[WIP] Fix CSE side effect and definition extraction
|
|
Value Number Enhancements:
|
|
|
|
|
|
|
|
Arm: Correctly handle multi-reg COPY
|
|
|
|
to explicit convert string to wstring to avoid warning. (#19947)
|
|
* Add test for 16472
* Replace fgMarkAddressExposedLocals
* CR: Move fgAddFieldSeqForZeroOffset to improve diff
* CR: Improve comments and various other small issues
* CR: Delete gtCheckQuirkAddrExposedLclVar
* Revert "disable tests\src\JIT\Methodical\fp\exgen\10w5d_cs_do (#19465)"
This reverts commit b0686a29be2eff059f080a66b7aa4febe55a01bc.
|