steam-frame/FEX-Emu--FEX
1
0 Fork 0
mirror of https://github.com/FEX-Emu/FEX.git synced 2026-10-14 07:00:43 +02:00
Code Issues Packages Projects Releases 67 Wiki Activity
5,648 Commits 11 Branches 71 Tags
bf4c5797db509610839ca381cfa93e3f1300a031
Commit Graph
2119 Commits
Author SHA1 Message Date
Ryan Houdek bf4c5797db ARMEmitter: Removes some warnings that cropped up 2023-01-18 17:48:44 -08:00
Mai 4aa984aed9 Merge pull request #2322 from Sonicadvance1/opdispatcher_helpers 
OpDispatcher: Fixes a few missing GPR/XMM helper usages
 2023-01-19 00:05:11 +00:00
Mai 95e544c840 Merge pull request #2342 from Sonicadvance1/more_asimd_ops_pt2 
ArmEmitter: Adds two more classes of ASIMD instructions
 2023-01-18 20:33:40 +00:00
Mai 81e0ac7e0b Merge pull request #2331 from Sonicadvance1/more_asimd_ops 
ArmEmitter: Adds three more classes of ASIMD instructions
 2023-01-18 20:32:46 +00:00
Mai f8d92aa121 Merge pull request #2329 from Sonicadvance1/fix_cache_invalidation 
Arm64: Fixes incorrect operation for CacheLineClear
 2023-01-18 20:29:44 +00:00
Mai ee58c5de1d Merge pull request #2315 from Sonicadvance1/add_negative_unittests 
unittests: Adds negative integer x87 tests
 2023-01-18 20:27:59 +00:00
Mai 565ed450aa Merge pull request #2310 from Sonicadvance1/aarch64_move_to_switch 
Arm64: Use switch statement for op handlers instead of jump table
 2023-01-18 20:26:16 +00:00
Mai 90bcb8c70b Merge pull request #2309 from Sonicadvance1/remove_header 
Emitter: Remove unused header
 2023-01-18 20:25:07 +00:00
Ryan Houdek 9c93c6ffcd Merge pull request #2317 from Sonicadvance1/fix_spill_register 
Arm64: Fix SpillRegister C&P error
 2023-01-17 12:56:14 -08:00
Ryan Houdek 0ef8574a56 ARMEmitter: Adds two instruction classes 2023-01-15 20:16:10 -08:00
Ryan Houdek f1e1eaa8e5 ArmEmitter: Adds four missing ASIMD Shift by Imm ops 2023-01-15 20:16:10 -08:00
Ryan Houdek f614fc6fac Merge pull request #2338 from Sonicadvance1/optimize_cpuid 
CPUID: Optimize initialization
 2023-01-14 14:15:35 -08:00
Ryan Houdek d9a1bb9c35 CPUID: Optimize initialization 
Map lookup was quite expensive, switched over to three small vectors
that are constexpr instead.

Some file querying and parsing was fairly slow as well. Optimized to
make that CPU time to go away.

This improves initialization time of CPUIDEmu by 33%
 2023-01-14 13:37:26 -08:00
Ryan Houdek d69afaf925 MContext: Insert a stack cookie with assertions enabled 
Pulled from #2176.
Ensures that when we are handling signals we are actually restoring a
stack state that is what we expect..

While this could randomly intersect with other stack data, it is highly
unlikely and will still capture incorrect stack frames otherwise.

Keeps it out of release build to ensure we aren't sticking random data
in the stack when it wouldn't have even been checked.
 2023-01-14 11:59:10 -08:00
Ryan Houdek d787a38744 ArmEmitter: Adds three more classes of ASIMD instructions 
Adds three classes:
- Advanced SIMD three same (FP16)
- Advanced SIMD two-register miscellaneous (FP16)
- Advanced SIMD three-register extension

A handful of the three-register extension unit tests are disabled
because the vixl disassembler doesn't support them.

Only six more classes of ASIMD operations remaining once this is merged.
 2023-01-12 13:38:48 -08:00
Ryan Houdek b2f7f526f8 Arm64: Fixes incorrect operation for CacheLineClear 
CIVAU does Clean+Invalidate to `Point Of Unification`
CIVAC does Clean+Invalidate to `Point of Coherency`

`Point of Unification` means to L2/L3, so unification of core
visibility.

`Point of Coherency` means SLC/RAM, All cores, DNA engines, etc must be
coherent.
 2023-01-11 19:53:33 -08:00
Ryan Houdek bd55ed51b0 OpDispatcher: Moves a few missing XMM loadstores to helper usage 
These were missed initially, these need to all be using the helper for
future optimizations.
 2023-01-09 08:23:26 -08:00
Ryan Houdek 6d912be31e OpDispatcher: Moves a few missing GPR loadstores to helper usage 
These were missed initially, these need to all be using the helper for
future optimizations.
 2023-01-09 08:23:26 -08:00
CallumDev 806587d6ae Fix FPREM flags calculation in F64 2023-01-09 22:21:23 +10:30
Ryan Houdek 4daf2f0793 Jit64: Fixes incorrect sign extension 
We were accidentally zero extending.
 2023-01-08 13:29:38 -08:00
Ryan Houdek 5aacdd744c Arm64: Fixes incorrect sign extension 
We were accidentally zero extending.
 2023-01-08 13:25:25 -08:00
Ryan Houdek c7262120a6 Arm64: Fix SpillRegister C&P error 
Was using the wrong sized registers in spill which was breaking Steam.
Oops.
 2023-01-08 12:47:02 -08:00
Ryan Houdek c2325e1772 Merge pull request #2314 from CallumDev/f64-integer-fix 
F64: Fix integer immediates for add,mul,div,sub
 2023-01-08 10:42:21 -08:00
CallumDev 9373fa0c06 F64: Fix integer immediates for add,mul,div,sub 2023-01-09 01:29:17 +10:30
Ryan Houdek ca9400ba52 Arm64: Fixes large offset spill slots 
Found an application today (hashtree tests) that causes us to spill a
large amount of values on to the stack.

We were encoding larger offsets than what unsigned offset load and store
can handle.

If the offset is too large for the loadstore, use a temporary to put the
offset in to first.
 2023-01-07 18:03:24 -08:00
Ryan Houdek 9a748c020d Arm64: Use switch statement for op handlers instead of jump table 
Removes some startup time where we are copying nearly a page worth of
16byte vtable pointers at startup.

Also allows the compiler to choose to inline functions if it wants to.
 2023-01-06 17:44:28 -08:00
Ryan Houdek 842e36e9b2 Emitter: Remove unused header 2023-01-06 10:34:41 -08:00
Ryan Houdek ffd9bb547d Arm64: Convert ARM Emitter over to new emitter 
Not yet complete. Missing a full SVE implementation and needs
testing/validation.
 2023-01-04 05:30:01 -08:00
Ryan Houdek 7a6ef8821f Arm64: Adds new ARM emitter 
Still needs more work.
Missing operations, cleanup, validation
Notably SVE is missing large chunks.
 2023-01-04 05:30:01 -08:00
Ryan Houdek 3904a5264f Merge pull request #2306 from lioncash/perm 
OpcodeDispatcher: Handle immediate variants of VPERMILPD/VPERMILPS
 2022-12-31 21:31:59 -08:00
lioncash b95c1719c3 OpcodeDispatcher: Handle VPERMILPS (immediate) 2023-01-01 05:15:25 +00:00
lioncash dfb3f31453 OpcodeDispatcher: Handle VPERMILPD (immediate) 2023-01-01 04:59:48 +00:00
lioncash 8031f76642 OpcodeDispatcher: Handle VMASKMOVDQU 2023-01-01 04:18:46 +00:00
lioncash ae8a5fa98d OpcodeDispatcher: Handle VPHSUBD 2023-01-01 03:10:21 +00:00
lioncash 6914598f9a OpcodeDispatcher: Handle VPHSUBW 2023-01-01 02:42:05 +00:00
lioncash 450aedc8b6 x86_64: Fix 256-bit UnZip/UnZip2 
We weren't swapping the elements so that the operation acts like the two
vectors are concatenated.
 2023-01-01 02:42:05 +00:00
lioncash 1e221210a2 OpcodeDispatcher: Move PHSUB impl to helper function 
This will be used for the AVX variants.
 2023-01-01 02:42:03 +00:00
Ryan Houdek 58ec2b2d7f Merge pull request #2303 from lioncash/swizz 
OpcodeDispatcher: Zip elements instead of for loop insertion in PHSUB
 2022-12-31 16:09:13 -08:00
lioncash 438adf2f45 x86_64/VectorOps: Handle OpSize==8 case in VUnZip/VUnZip2 
Allows the x86 side of things to execute the new codepath in PHSUB
 2022-12-31 23:30:59 +00:00
lioncash 9707e9a4df OpcodeDispatcher: Zip elements instead of for loop in PHSUB 
Makes this much nicer for 128-bit and soon-to-be 256-bit vectors.
 2022-12-31 23:30:38 +00:00
Ryan Houdek 9b8c92e275 Merge pull request #2302 from lioncash/dpp 
OpcodeDispatcher: Handle VDPPD/VDPPS
 2022-12-31 15:01:52 -08:00
Ryan Houdek 6caf764b7c Merge pull request #2301 from lioncash/insps 
OpcodeDispatcher: Handle VINSERTPS
 2022-12-31 14:59:11 -08:00
lioncash 769c548ba4 OpcodeDispatcher: Handle VDPPD 
x86 just doesn't have a 256-bit version of this op.
 2022-12-31 21:13:20 +00:00
lioncash dae1676e4a OpcodeDispatcher: Handle VDPPS 2022-12-31 21:01:48 +00:00
lioncash 74526d1f02 OpcodeDispatcher: Move DPP op impl into helper function 
This will be used to implement the AVX variants.
 2022-12-31 21:01:26 +00:00
lioncash 8c005db81c OpcodeDispatcher: Handle VINSERTPS 2022-12-31 20:23:22 +00:00
lioncash 32b70c8590 OpcodeDispatcher: Move InsertPS impl to helper function 
This will be reused for VINSERTPS
 2022-12-31 20:23:18 +00:00
lioncash bdda14eb75 OpcodeDispatcher: Handle VMOVMSKPD 2022-12-31 19:24:14 +00:00
lioncash b8c0b0c267 OpcodeDispatcher: Handle VMOVMSKPS 2022-12-31 19:24:11 +00:00
lioncash ad7dc6ca0a OpcodeDispatcher: Move ADDSUB table entries 
Makes them numerically ordered.
 2022-12-31 19:07:19 +00:00