|
 | mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp (diff) |
 | mlir/test/Dialect/Affine/canonicalize.mlir (diff) |
 | mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td (diff) |
 | mlir/test/Transforms/canonicalize.mlir (diff) |
 | mlir/test/Dialect/Tensor/canonicalize.mlir (diff) |
|
 | polly/include/polly/Support/ISLTools.h (diff) |
Commit
312257688eb0c09a8e6915ced2acdf0bcbbad353
by mgorny[lldb] [Process] Introduce protocol extension support API
Introduce a NativeProcessProtocol API for indicating support for protocol extensions and enabling them. LLGS calls GetSupportedExtensions() method on the process factory to determine which extensions are supported by the plugin. If the future is both supported by the plugin and reported as supported by the client, LLGS enables it and reports to the client as supported by the server.
The extension is enabled on the process instance by calling SetEnabledExtensions() method. This is done after qSupported exchange (if the debugger is attached to any process), as well as after launching or attaching to a new inferior.
The patch adds 'fork' extension corresponding to 'fork-events+' qSupported feature and 'vfork' extension for 'vfork-events+'. Both features rely on 'multiprocess+' being supported as well.
Differential Revision: https://reviews.llvm.org/D100153
|
 | lldb/packages/Python/lldbsuite/test/dotest.py (diff) |
 | lldb/test/API/tools/lldb-server/TestLldbGdbServer.py (diff) |
 | lldb/packages/Python/lldbsuite/test/tools/lldb-server/gdbremote_testcase.py (diff) |
 | lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.cpp (diff) |
 | lldb/include/lldb/Host/common/NativeProcessProtocol.h (diff) |
 | lldb/packages/Python/lldbsuite/test/test_categories.py (diff) |
 | lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.h (diff) |
Commit
6c37984ebaf4ee01df6a9b3f78e45f70dcd6fb33
by mgorny[lldb] [gdb-remote server] Introduce new stop reasons for fork and vfork
Introduce three new stop reasons for fork, vfork and vforkdone events. This includes server support for serializing fork/vfork events into gdb-remote protocol. The stop infos for the two base events take a pair of PID and TID for the newly forked process.
Differential Revision: https://reviews.llvm.org/D100196
|
 | lldb/include/lldb/API/SBThreadPlan.h (diff) |
 | lldb/include/lldb/API/SBThread.h (diff) |
 | lldb/include/lldb/lldb-enumerations.h (diff) |
 | lldb/docs/python_api_enums.rst (diff) |
 | lldb/source/Target/StackFrameList.cpp (diff) |
 | lldb/include/lldb/Host/Debug.h (diff) |
 | lldb/tools/lldb-vscode/LLDBUtils.cpp (diff) |
 | lldb/packages/Python/lldbsuite/test/lldbutil.py (diff) |
 | lldb/tools/lldb-vscode/JSONUtils.cpp (diff) |
 | lldb/source/Target/Thread.cpp (diff) |
 | lldb/bindings/interface/SBThread.i (diff) |
 | lldb/examples/python/performance.py (diff) |
 | lldb/bindings/interface/SBThreadPlan.i (diff) |
 | lldb/source/API/SBThread.cpp (diff) |
 | lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.cpp (diff) |
 | lldb/source/Target/Process.cpp (diff) |
Commit
bbae0c1f7b4f60e9b8ac2be24e35bec79d9b7b01
by mgorny[lldb] [llgs] Support owning and detaching extra processes
Add a NativeDelegate API to pass new processes (forks) to LLGS, and support detaching them via the 'D' packet. A 'D' packet without a specific PID detaches all processes, otherwise it detaches either the specified subprocess or the main process, depending on the passed PID.
Differential Revision: https://reviews.llvm.org/D100191
|
 | lldb/source/Utility/StringExtractorGDBRemote.cpp (diff) |
 | lldb/test/API/tools/lldb-server/TestGdbRemoteFork.py |
 | lldb/test/API/tools/lldb-server/main.cpp (diff) |
 | lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.h (diff) |
 | lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.cpp (diff) |
 | lldb/include/lldb/Host/common/NativeProcessProtocol.h (diff) |
 | lldb/unittests/TestingSupport/Host/NativeProcessTestUtils.h (diff) |
Commit
fd0af0cf08284de79fe1a5bcfdc2dad83794dcfe
by mgorny[lldb] [Process/Linux] Report fork/vfork stop reason
Enable reporting fork/vfork events to the server when supported. At this moment, this is used only to test the server code, as real client does not report fork-events and vfork-events as supported.
Differential Revision: https://reviews.llvm.org/D100208
|
 | lldb/source/Plugins/Process/Linux/NativeProcessLinux.h (diff) |
 | lldb/source/Plugins/Process/Linux/NativeThreadLinux.cpp (diff) |
 | lldb/packages/Python/lldbsuite/test/dotest.py (diff) |
 | lldb/source/Plugins/Process/Linux/NativeProcessLinux.cpp (diff) |
 | lldb/source/Plugins/Process/Linux/NativeThreadLinux.h (diff) |
Commit
286677870b306582321dcddb1531d8860bc446db
by llvm-project[Polly][ManualOpt] Match interpretation of unroll metadata to LoopUnrolls's.
We previously had a different interpretation of unroll transformation attributes than how LoopUnroll interpreted it. In particular, llvm.loop.unroll.enable was needed explicitly to enable it and disabling metadata was ignored. Additionally, it required that either full unrolling or an unroll factor to be specified or fail otherwise. An unroll factor is still required, but the transformation is ignored with the hope that LoopUnroll is going to apply the unrolling, since Polly currently does not implement an heuristic.
Fixes llvm.org/PR50109
|
 | polly/test/ScheduleOptimizer/ManualOptimization/unroll_full.ll (diff) |
 | polly/test/ScheduleOptimizer/ManualOptimization/unroll_heuristic.ll |
 | polly/lib/Support/ScopHelper.cpp (diff) |
 | polly/include/polly/Support/ScopHelper.h (diff) |
 | polly/test/ScheduleOptimizer/ManualOptimization/unroll_disable.ll |
 | polly/lib/Transform/ManualOptimizer.cpp (diff) |
 | polly/test/ScheduleOptimizer/ManualOptimization/unroll_partial.ll (diff) |
|
 | llvm/test/Transforms/InstCombine/ctpop.ll (diff) |
|
 | llvm/test/Transforms/GVN/PRE/rle.ll (diff) |
 | llvm/lib/Transforms/Scalar/GVN.cpp (diff) |
 | llvm/include/llvm/Analysis/MemoryDependenceAnalysis.h (diff) |
 | llvm/lib/Analysis/MemoryDependenceAnalysis.cpp (diff) |
Commit
4e2d5fcf71d2995c98184293f8925731a6b0f30d
by thakis[lld/mac] add test coverage for -sectcreate and -order_file with --reproduce
Would've caught the (since fixed) regression in D97610.
No behavior change.
Differential Revision: https://reviews.llvm.org/D101218
|
 | lld/test/MachO/reproduce.s (diff) |
Commit
4ca0fbfabd7a8ba0f7a3a6c98c0ba194fb1c1506
by thakis[lld/mac] simplify export-options.s test a bit
- the macro seems needlessly clever -- shorter and imho clearer without it - give all filenames an extension so they look like filenames - rename .private_extern symbol from _private to _private_extern to prepare for follow-up that adds a truly private symbol
No behavior change.
Differential Revision: https://reviews.llvm.org/D101222
|
 | lld/test/MachO/export-options.s (diff) |