Commit Graph
15 Commits
Author SHA1 Message Date
Scott Mansell cb57797550 PassManager: Lookup pass by name 2021-09-13 11:46:11 +12:00
Ryan Houdek f161e3bfb0 Massive amount of IWYU cleanup
This isn't quite a 100% clean sweep of IWYU.
There are some false positives where clang fails.
Additionally there are still a few missed in the frontend side of things
that I didn't get to
2021-08-28 00:32:15 -07:00
Lioncash 120ba3d171 PassManager: std::move exit handler function
Avoids allocations if the internal std::function buffer ever happens to
be large enough to warrant it.
2021-06-11 22:05:17 -04:00
Lioncash 6084bdf982 Passes: Return by unique_ptr where applicable
Same behavior, but makes the ownership intentions explicit in the
interface.
2021-06-11 22:05:15 -04:00
Ryan Houdek 520dfd7edf Adds Long Divide removal pass
If the long divides don't have anything in the upper bits then they can be optimized away.
Teeworlds, SuperTuxKart, and RRootage were between 50%-55% getting removed
2021-04-24 23:10:59 -07:00
Stefanos Kornilios Mitsis Poiitidis 942b8549c6 Docs: Add tags to the source code 2021-03-30 12:21:18 +03:00
Stefanos Kornilios Mitsis Poiitidis 9c5a1fd5ea SRA: Static Register Allocation, arm64 only 2021-01-05 17:14:57 +02:00
Ryan Houdek 37983006f0 Moves FEXCore's SignalDelegator and SyscallHandler to the frontend
Sadly these things can't be split without breaking functionality so it
turns in to a bit of a mess.

SyscallHandler is very much something that is a Linux only construct and
shouldn't be in FEXCore itself. Lets the frontend register a
Syscallhandler with FEXCore. FEXCore itself is then aware of the current
syscall ABI and handles the ABI in an optimal fashion.
So it is not a 100% clean break otherwise we would lose performance.

The SignalDelegator then needs to move to the frontend since the
SyscallHandler requires it for signal based syscalls.
The CPU backend signal handling still needs to happen in FEXCore because
it is a very tight coupling with the CPU backend.
Once we need to support more Signal handling we can give the backends
cleaner support to select which specific OS handler to handle.
2020-12-13 22:38:04 -08:00
Ryan Houdek 17402aa02d Remove duplicated compaction pass
RA Pass and passmanager compaction pass don't need to be independent.
This saves some memory.

In the future we should allow passes to fetch passes by name, that will
be part of the PassManager improvement in the future
2020-12-03 19:58:47 -08:00
Stefanos Kornilios Mitsis Poiitidis 93e372c7a4 IR/JITs: Inline Constants Implementation
Based on @phire's idea, Inline constants are constants that can be used as imms in instructions.

This
- Adds OP_INLINECONSTANT to represent them in the IR
- Modifies the ConstProp pass to transform OP_CONSTANT to OP_INLINECONSTANT based on per-platform heuristics
- Modifies the backends to use imms
- Supported opcodes: Add, Sub, Lsl, Lsr, Asr, Ror, Rol
2020-10-22 18:49:17 +03:00
Ryan Houdek 74a310eeb2 Enables assertions option from cmake 2020-08-29 23:35:26 -07:00
Ryan Houdek fd857e4966 Makes threaded compilation not rely on a global mutex
As we run more threaded applications it is becoming apparent that this
needs to be fixed now.
This duplicates the FrontendDecoder and Passmanager per thread so they
no longer block each other while compiling.
Increases a bit of memory usage but completely worth it.
2020-05-27 10:18:02 -07:00
Ryan Houdek 718b3e6b4c Refactors PassManager to ingest IREmitter class instead of OpDispatcher
The Passmanager passes need very little or zero x86 knowledge to do
their work. They don't need the x86 OpDispatcher handling code at all.
Convert this entirely over to using the IREmitter class so it can
actually optimize IR code from the IRLoader frontend
2020-05-09 12:14:59 -07:00
Ryan Houdek 37e2a2bb6f Adds validation passes after regular passes
This is required for the forthcoming IRValidation pass change that
validates RA is working
2020-04-17 18:09:46 -07:00
Stefanos Kornilios Mitsis Poiitidis e9ea4cbb76 FEXCore: Move to Externals/FEXCore 2020-03-06 09:07:37 +02:00