Commit Graph
526 Commits
Author SHA1 Message Date
Ryan Houdek 5566b4455b IR: Change VFCMPScalarInsert to use IR::OpSize 2024-10-27 18:35:03 -07:00
Ryan Houdek fed2c13521 IR: Change VFToIScalarInsert to use IR::OpSize 2024-10-27 18:32:11 -07:00
Ryan Houdek 5626f4e50a IR: Change VSToFVectorInsert to use IR::OpSize 2024-10-27 18:27:39 -07:00
Ryan Houdek 081907e168 IR: Change VFAddScalarInsert to use IR::OpSize 2024-10-27 18:14:37 -07:00
Ryan Houdek 1a115a8ce6 IR: Change FCmp to use IR::OpSize 2024-10-27 18:06:34 -07:00
Ryan Houdek 1bde30a196 IR: Change Float_ToGPR_S to use IR::OpSize 2024-10-27 18:02:45 -07:00
Ryan Houdek 764aacaa8f IR: Change VExtractToGPR to use IR::OpSize 2024-10-27 17:58:53 -07:00
Ryan Houdek b31ce13f68 IR: Change Pop to use IR::OpSize 2024-10-27 17:41:30 -07:00
Ryan Houdek 260d3b0b4e IR: Change Push to use IR::OpSize 2024-10-27 17:39:21 -07:00
Ryan Houdek c8c7ffbf05 IR: Change VBroadcastFromMem to use IR::OpSize 2024-10-27 17:35:56 -07:00
Ryan Houdek 4b03185b77 IR: Change VStoreVectorElement to use IR::OpSize 2024-10-27 17:32:25 -07:00
Ryan Houdek 014917301a IR: Change StoreMem to use IR::OpSize 2024-10-27 17:14:18 -07:00
Ryan Houdek 07f8a4eadd IR: Change LoadMem to use IR::OpSize 2024-10-27 16:33:14 -07:00
Ryan Houdek ece89ddeab IR: Change LoadContextIndexed to use IR::OpSize 2024-10-27 15:50:03 -07:00
Ryan Houdek 2f9b0de742 IR: Change StoreContext to use IR::OpSize 2024-10-27 15:46:32 -07:00
Ryan Houdek 40fd4bbb66 IR: Change LoadContext to use IR::OpSize 2024-10-27 15:42:07 -07:00
Ryan Houdek e8baf4a28c OpcodeDispatcher: Ensure IR ops use OpSize
NFC
2024-10-27 14:11:25 -07:00
Ryan Houdek f143462ebe OpcodeDispatcher: Minor optimization to small pushf
The push operation already truncates the result, there's no need to bfe
it. Noticed this while cleaning up in #4134. Removes one instruction for
16-bit and 32-bit pushf instructions.
2024-10-25 15:41:22 -07:00
Paulo Matos 5f6c0d2245 X87 code simplification
Merges some of the code from reduced precision into the main path
since they are practically the same.
2024-10-24 18:15:49 +02:00
Alyssa Rosenzweig 58a3d174ec OpcodeDispatcher: explain why we provide defined bsf behaviour
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-10-22 13:36:24 -04:00
Alyssa Rosenzweig 9c605e7333 OpcodeDispatcher: optimize bsf/bsr
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-10-22 13:31:49 -04:00
Alyssa Rosenzweig 1fd7e88ffd OpcodeDispatcher: optimize cmp in cmpxchg
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-10-22 13:29:47 -04:00
Alyssa Rosenzweig c5e7da0631 OpcodeDispatcher: optimize more cmpxchg mask
none of it matters.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-10-22 13:29:47 -04:00
Alyssa Rosenzweig 68f58e415f OpcodeDispatcher: optimize cmpxchg masking
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-10-22 13:29:47 -04:00
Paulo Matos 10ec6b63b6 Fix FXTRACT for 0.0 and -0.0
Fixes fxtract by returning the correct values for 0.0 and -0.0. We moved the split of fxtract into _sig and _exp, to the opcode dispatcher, to ease some comparisons.

Also removed the IR node F80XTRACTStack which is not needed anymore.
2024-10-17 09:05:10 +02:00
Paulo Matos 0d53f2b45c Implements explicit state switch between X87 and MMX
Fixes #3850
2024-10-15 17:58:53 +02:00
Alyssa Rosenzweig e2d58809ed OpcodeDispatcher: rm dead
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-10-02 13:38:48 -04:00
Alyssa Rosenzweig 967a74cda9 OpcodeDispatcher: manually inline constants
less work for constprop.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-10-02 13:38:48 -04:00
Ryan Houdek dc2a26582f FEXCore/OpcodeDispatcher: Fixes missing static on some tables
This was causing egregious stack usage in these two functions
2024-09-16 18:52:55 -07:00
Ryan Houdek c984bdb42a OpcodeDispatcher: Remove previous template instantiantions and use Bind 2024-09-16 18:52:54 -07:00
Ryan Houdek 2645b374a7 OpcodeDispatcher: Deduplicate InstallToTable helper 2024-09-16 18:52:54 -07:00
Ryan Houdek 0e57cbf5b9 OpcodeDispatcher: Remove EVEX table install
This hasn't been necessary for quite a while as we handle
invalid/unsupported opcodes in the frontend now.
2024-09-16 17:40:20 -07:00
Ryan Houdek c7413d96ad OpcodeDispatcher: Constexpr-ify VEX & VEXGroup tables 2024-09-16 17:40:20 -07:00
Ryan Houdek f3ba8cb33c OpcodeDispatcher: Constexpr-ify DDD tables 2024-09-16 17:40:20 -07:00
Ryan Houdek e714933b10 OpcodeDispatcher: Constexpr-ify H0F3A tables 2024-09-16 17:40:20 -07:00
Ryan Houdek 9a0f83cbf6 OpcodeDispatcher: Constexpr-ify H0F38 tables 2024-09-16 17:40:20 -07:00
Ryan Houdek c4d30e8437 OpcodeDispatcher: Constexpr-ify SecondaryModRM tables 2024-09-16 17:40:20 -07:00
Ryan Houdek 7b30df8a56 OpcodeDispatcher: Constexpr-ify SecondaryGroup tables 2024-09-16 17:40:20 -07:00
Ryan Houdek 86aa459dd1 OpcodeDispatcher: Constexpr-ify PrimaryGroup tables 2024-09-16 17:40:20 -07:00
Ryan Houdek 6c9a47fc71 OpcodeDispatcher: Constexpr-ify Secondary OpSizeMod tables 2024-09-16 17:40:20 -07:00
Ryan Houdek 7a72cf631b OpcodeDispatcher: Constexpr-ify Secondary RepNEMod tables 2024-09-16 17:40:20 -07:00
Ryan Houdek 3d98eef7b8 OpcodeDispatcher: Constexpr-ify Secondary RepMod tables 2024-09-16 17:40:20 -07:00
Ryan Houdek f2011b0b79 OpcodeDispatcher: Constexpr-ify Secondary base tables 2024-09-16 17:40:19 -07:00
Ryan Houdek 37a70e2ec6 FEXCore: Convert Base tables over to constexpr
Only doing the single table for review purposes. Once reviewed I will
hammer out the remaining tables.

Similar to #3320, most of the OpcodeDispatcher tables can be consteval
and made to be a compile time constant. This just requires shuffling the
code slightly. The idea is to get almost all of the table setup out of
the `InstallOpcodeHandlers` function and instead only install the
handlers that change based on 32-bit or 64-bit, just like the x86 tables
we also did.
2024-09-13 11:39:09 -07:00
Ryan Houdek a94aaca0ec OpcodeDispatcher: Add default FEX_UNREACHABLE
This can't happen but if it changes then make sure we capture it.
2024-09-08 18:03:20 -07:00
Alyssa Rosenzweig e9ab514962 IR: push parity evaluation down
so we can optimize it globally

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-09-07 08:26:04 -04:00
Billy Laws f7b911ca43 OpcodeDispatcher: Do not forbid INT 2E syscalls on 64-bit Windows
This works fine on real Windows and is relied on by wine as SystemCall
is set to 1 in KUSER_SHARED_DATA, which causes the ntdll thunks to use
it over `syscall`
2024-09-06 15:58:17 +00:00
Ryan Houdek fd4f6b8020 FEXCore: Dynamically scale TSC
When I implemented TSC scaling originally, I chose a scale factor of 128
because it basically covered the range of devices we cared about without
going too high. I also only tested devices that had a TSC scale factor
from 19.2Mhz to 34Mhz. Turns out there is hardware that also has a 48Mhz
cycle counter, which cause them to effectively have a 6.1Ghz cycle
counter, which is kind of absurd.

Instead of a fixed scale, just calculate the amount of scaling we need
to get >= the minimum threshold of 1Ghz. This will change the shift from
7 to 5 or 6 for the faster cycle counter devices.

Of course if someone wants to know the scale factor they can still use
cpuid function 15h to know it.

Fixes #4026
2024-09-03 13:41:31 -07:00
Alyssa Rosenzweig 335cd9180e OpcodeDispatcher: optimize mul rax
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
2024-08-27 12:32:43 -04:00
Mai 2829ad56a1 Merge pull request #3996 from Sonicadvance1/more_bind
OpcodeDispatcher: Convert more template handlers to Bind handlers
2024-08-23 18:27:59 -04:00