Commit Graph
2018 Commits
Author SHA1 Message Date
Alyssa Rosenzweig 4a2dee873d OpcodeDispatcher: use ArithRef for small rotates
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig 589906e6e6 OpcodeDispatcher: generalize AF on constants optimization
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig f0fcf6d9e6 OpcodeDispatcher: optimize SVE vmovmskpd
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig 1591ced5a7 OpcodeDispatcher: use ArithRef for ADC/SBB flags
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig 0d235d63d0 OpcodeDispatcher: use ArithRef for bit tests
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig 820d5c1447 OpcodeDispatcher: use ArithRef for rotates
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig 8093e8f5f1 OpcodeDispatcher: optimize LoadDir
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig 802857e411 OpcodeDispatcher: add ArithRef helpers
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig bf3a1839a2 OpcodeDispatcher: transform DF ourselves
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Alyssa Rosenzweig ad7844d7da OpcodeDispatcher: do not generate useless VMov
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-06-02 14:15:51 -04:00
Tony Wasserka a792dd0703 CPUBackend: Clean up CodeBuffer size limits 2025-06-01 22:45:50 +02:00
Tony Wasserka a9a6a645bf Arm64Emitter: Disable PC-relative constant encoding
This no longer works since the JIT output is now relocated before execution.
2025-06-01 22:45:50 +02:00
Tony Wasserka 7c93becd5f JIT: Increase estimate for CodeBuffer space use
The previous bound was exceeded during Steam startup before.
2025-06-01 22:45:50 +02:00
Tony Wasserka 4bbaef58e9 JIT: Re-enable parallel compilation by compiling to a temporary buffer 2025-06-01 22:45:50 +02:00
Tony Wasserka 95791a985a Core: Minimize the time CodeBufferWriteMutex is held 2025-06-01 22:45:50 +02:00
Tony Wasserka 0dfefe9730 Core: Re-check LookupCache before running compiler backend
This further reduces lock contention by skipping the backend phase in case
another thread raced the active one for the same block.
2025-06-01 22:45:49 +02:00
Tony Wasserka 8481c797df FEXCore/JIT: Extend LookupCache lock to all of ExitFunctionLink 2025-06-01 22:44:49 +02:00
Tony Wasserka a503b5e20b Rename CodeBufferManager reference 2025-06-01 22:44:49 +02:00
Tony Wasserka 4078840ef1 Core: Reduce JIT time by sharing CodeBuffers between threads
This is changes the interface of CodeBuffer to that of a partially persistent
data structure based on reference counting:
- Exactly one CodeBuffer is now designated as "active", which means data can
  be *appended* to it
- Lossy modifications to the active CodeBuffer will not invalidate any data
  in use by other threads, which enables save sharing across threads
- Instead, such lossy modifications trigger a new "version" of the data in
  the modifying thread. Old versions of the CodeBuffer persist as read-only
  data for use by the other threads.
- The other threads can update their version of the CodeBuffer. This will
  decrease the reference count and eventually trigger deallocation of the
  old version
2025-06-01 22:44:49 +02:00
Tony Wasserka ab51958b26 Move AllocateNewCodeBuffer from CPUBackend to a new CodeBufferManager interface 2025-06-01 22:42:55 +02:00
Tony Wasserka 3376587b6a CPUBackend: Manage CodeBuffers using shared_ptr
This is required for sharing CodeBuffers between threads anyway, but it also
allows use of the constructor/destructor to manage memory automatically.
2025-06-01 22:42:55 +02:00
Tony Wasserka 6681d7dcf9 LookupCache: Split L3 cache into a dedicated interface
This data isn't really a cache, since the JIT is directly responsible of
writing its contents. Instead it be considered the source to populate the
L1/L2 caches from.

Furthermore, splitting off this data allows it to be shared across threads
in the future without affecting L1/L2 caches.
2025-06-01 22:42:55 +02:00
Tony Wasserka 34224481c6 LookupCache: Prefer empty() over a size check 2025-06-01 22:42:55 +02:00
Ryan Houdek cdea8d7f74 Merge pull request #4593 from alyssarosenzweig/ici/zeroing-sub-regs
InstructionCountCI: add more cases for mov 0/~0
2025-05-29 12:26:15 -07:00
Ryan Houdek ef6dc3d802 Merge pull request #4592 from alyssarosenzweig/opt/x87-tag
OpcodeDispatcher: optimize X87FTWTag
2025-05-29 12:26:04 -07:00
Alyssa Rosenzweig 0fbe69ebcf OpcodeDispatcher: optimize xor-with-self flags
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-28 13:53:38 -04:00
Alyssa Rosenzweig ece817c691 OpcodeDispatcher: optimize logical flags
seems to be strictly better.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-28 13:34:50 -04:00
Alyssa Rosenzweig 3fbc8204b7 OpcodeDispatcher: clean up logical flags
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-28 13:28:31 -04:00
Alyssa Rosenzweig b8dd5d95b0 OpcodeDispatcher: optimize X87FTWTag
using bit twiddling tricks :-)

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-28 13:13:16 -04:00
Ryan Houdek 7cd52febc2 FEXCore/CPUID: Remove warning
This is currently only used on win32 builds because Linux doesn't
understand TPIDRRO.
2025-05-28 09:28:41 -07:00
Alyssa Rosenzweig 6927c7577a RegisterAllocationPass: delete trivial instructions
lots of instructions only exist for RA, so RA can garbage collect them before
post-RA passes (including the JIT) deals with them. this simplifies our life
now, and makes post-RA passes a LOT simpler for little cost.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-26 15:52:23 -04:00
Alyssa Rosenzweig 61ae53cc03 IR: add paired PushTwo/PopTwo helpers
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-26 15:52:23 -04:00
Alyssa Rosenzweig 0e67f30103 OpcodeDispatcher: make Push do the right thing and use it
this both optimizes and bug-fixes pusha while deleting a snotton of code.

Closes: https://github.com/FEX-Emu/FEX/issues/4589
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-26 15:52:23 -04:00
Ryan Houdek dc9f8aa855 Merge pull request #4580 from alyssarosenzweig/ir/inline-ra
IR: Inline registers into the IR
2025-05-26 09:44:51 -07:00
StanfordZhang 23cda2c961 Update Arm64Emitter.cpp
fix callee saved floating-point arguments issue
2025-05-23 22:14:37 +08:00
Alyssa Rosenzweig 65ee1fafa8 IR: drop RegisterAllocationData
This sideband is now unused, registers are encoded directly in the IR. So we can
garbage collect all this code for quite some savings.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-19 13:42:59 -04:00
Alyssa Rosenzweig 49f8332c5b JIT: use registers directly from the IR
This is the flag day change from the series, using all the new shiny
infrastructre we added.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-19 13:42:59 -04:00
Alyssa Rosenzweig afce108ed7 JIT: make almost all the DEF_OPs common
this deduplicates a bunch of #defines, letting us change the signature easier.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-19 13:42:59 -04:00
Alyssa Rosenzweig 30f3b545af RegisterAllocationPass: use Header spill slots instead
Removes even more RAData dependence.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-19 13:42:59 -04:00
Alyssa Rosenzweig bf1597920c IR: use post-RA flag
rather than implicitly depending on the RA data.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-19 13:42:59 -04:00
Alyssa Rosenzweig a01e29ac99 IR: extend the IR header with RA info
Beyond the actual registers allocated, there are two pieces of sideband data we
store in the RAData object:

* # of spill slots (explicitly)
* whether RA has run (implicitly by the existence of RAData)

We want to get rid of RAData, so we'll move these to the header.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-19 13:42:59 -04:00
Alyssa Rosenzweig c7797d56c9 IR: don't use GetOpSize in ExitFunction
nothing else does this, and it complicates upcoming refactor to move away from
IR builder helpers doing IR dereferencing.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-15 14:51:11 -04:00
Alyssa Rosenzweig dfd1aedae5 JIT: use .ID() even less
oops, missed a "/g" with th sed.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-14 16:23:22 -04:00
Alyssa Rosenzweig ecc6fea54e JIT: stop using .ID() pattern
sed -ie 's/.ID()//' *

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-14 15:46:32 -04:00
Alyssa Rosenzweig 99446da7c1 JIT: add Reg helpers taking OrderedNodeWrappers
more ergonomic and will give us freedom to migrate things easier soon.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2025-05-14 15:43:20 -04:00
Tony Wasserka 39a5c2021e ThreadPoolAllocator: Rename FixedSizePoolAllocation to PoolBufferWithTimedRetirement
This more accurately reflects that the core feature of the helper is the
timer-based unclaiming of buffers instead of the allocation size.
2025-05-14 13:29:44 +02:00
LC 1dff7073de Merge pull request #4554 from Sonicadvance1/remove_unused_argument
NFC: FEXCore: Removes unused argument on CreateThread
2025-05-05 14:30:24 -04:00
Ryan Houdek 92a82c3134 FEXCore: Removes unused argument on CreateThread
ParentTID is purely a Linux construct and has been moved entirely to the
frontend at this point. Remove this argument which is now unused.
2025-05-05 11:17:02 -07:00
Tony Wasserka cdaa65f6fd Arm64Emitter: Fix overalignment in Align16B
Previously, 16 additional bytes were emitted if the buffer was already
aligned.
2025-05-05 16:02:42 +02:00
Ryan Houdek 7ed17f68c5 FEXCore: Remove unused InvalidateGuestCodeRange with callback 2025-05-02 01:15:50 -07:00