steam-frame/FEX-Emu--FEX
1
0 Fork 0
mirror of https://github.com/FEX-Emu/FEX.git synced 2026-10-06 14:00:16 +02:00
Code Issues Packages Projects Releases 66 Wiki Activity
15,067 Commits 11 Branches 70 Tags
023d0cdaa6cfaeaaa364be2bad15487faf71717a
Commit Graph
15067 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Ryan Houdek 023d0cdaa6 Merge pull request #5927 from Plagman/plagman/mono_more_caching 
DiskCache: also patch branches for more anon hit rate
 2026-09-07 14:09:35 -07:00
Ryan Houdek 5a4f84cad9 Merge pull request #5870 from cjacek/fmt-update 
External: Update fmt to upstream master to fix building with LLVM 23
 2026-09-07 12:24:52 -07:00
Jacek Caban eb27391954 External: Update fmt to upstream master to fix building with LLVM 23 
Includes https://github.com/fmtlib/fmt/pull/4868.
 2026-09-07 16:23:16 +02:00
Pierre-Loup A. Griffais ffaa5d6316 InstcountCI: Update 2026-09-07 00:56:39 -07:00
Pierre-Loup A. Griffais 8bf22b6382 DiskCache: add plumbing for guest-patching for branches 
To catch many conditional jumps found in Mono JITted code.
 2026-09-07 00:47:47 -07:00
Ryan Houdek 42c663269b Merge pull request #5910 from Plagman/plagman/mono_initial_caching 
DiskCache: detect inline data, hash around it and patch it on Lookup
 2026-09-06 23:45:17 -07:00
Pierre-Loup A. Griffais 983c5defc1 InstcountCI: Update 2026-09-06 23:25:53 -07:00
Pierre-Loup A. Griffais c92805475d DiskCache: detect inline data, hash around it and patch it on Lookup 
Only detect certain kinds of mov reg,immediate so far, which was the bulk of
Mono JIT activity.
 2026-09-06 23:25:29 -07:00
Ryan Houdek 50e6eee95a Merge pull request #5931 from Plagman/plagman/key_flags_mr 
DiskCache: key fixes
 2026-09-06 22:32:00 -07:00
Pierre-Loup A. Griffais 31cda95039 DiskCache: also fold FileId into the key for file-backed blocks 2026-09-06 21:47:18 -07:00
Pierre-Loup A. Griffais 970ce4bc13 DiskCache: fold more codegen-affecting environment into key 
Fixes a hang during load with Three Point Hospital
 2026-09-06 21:40:54 -07:00
Ryan Houdek a77d00acae Merge pull request #5929 from Plagman/plagman/smc_mr 
DiskCache: SMC regression fixes
 2026-09-06 03:24:15 -07:00
Pierre-Loup A. Griffais 5d11b8b0d5 DiskCache: SMC regression fixes 
One from a bad conflict resolution :/
 2026-09-06 02:00:05 -07:00
Ryan Houdek bf70dee3ae Merge pull request #5928 from Plagman/plagman/jittail_mr 
Frontend: fix nondeterminism by rolling back DecodedMaxAddress on error
 2026-09-05 22:39:05 -07:00
Pierre-Loup A. Griffais 4a065ad1a2 InstcountCI: Update 2026-09-05 21:58:07 -07:00
Pierre-Loup A. Griffais 96d955e7b1 Frontend: fix nondeterminism by rolling back DecodedMaxAddress on error 
Otherwise GuestSize ends up different across compiles of the same valid code
extents, including the copy of it baked in JITCodeTail of the host code itself.
 2026-09-05 21:53:30 -07:00
LC abec314720 Merge pull request #5926 from Sonicadvance1/234 
Frontend: Stop asserting on too long of an instruction
 2026-09-05 03:29:29 -04:00
LC ec63be4028 Merge pull request #5925 from Sonicadvance1/233 
Wine: Change behaviour around SMC and NtReadFile
 2026-09-05 03:28:17 -04:00
Ryan Houdek 179b4423a7 Wine: Change behaviour around SMC and NtReadFile 
JIT invalidation around syscalls and kernel calls is mighty fickle.
This is why the previous code path was born. In order to properly handle
this case we need strong coordination between Wine and FEX around
invalidating code and memory protections. This doesn't quite exist today
so we're kind of stuck with a kludge solution. Instead of forcing the
Persona 5 code invalidation on to every process, only do it on P5R.

This fixes a hang in msiexec with PhysX trying to do a blocking read and
jitting code or creating threads, while also maintaining the P5R
approach. The full comment is in the file about the reasoning.
 2026-09-04 20:08:18 -07:00
Ryan Houdek 9956a528b9 Frontend: Stop asserting on too long of an instruction 
Return the same as the NonExecutableRange for now. I have a unittest
that is mostly correct with this change, just need to fixup some of the
supplementary data.

For now push this change to get a bad assert removed that can happen
during code discovery. I'll fix up the remaining details afterwards.
 2026-09-04 20:05:18 -07:00
Ryan Houdek e3c6b37f70 Merge pull request #5924 from Sonicadvance1/232 
Common/Config: Fixes Cache directory when `STEAM_COMPAT_SHADER_PATH` on Win32
 2026-09-04 16:08:44 -07:00
LC b78bde7ec8 Merge pull request #5923 from Sonicadvance1/231 
OpcodeDispatcher: Explicitly handle no-nop vector moves
 2026-09-04 17:54:36 -04:00
Ryan Houdek 6ff7e82e3b Merge pull request #5918 from Plagman/plagman/validation_mr 
DiskCache: optional validation-only mode
 2026-09-04 14:34:38 -07:00
Ryan Houdek 37675e0351 Common/Config: Fixes Cache directory when STEAM_COMPAT_SHADER_PATH on Win32 
Moves cache data from %APPDATA% to wherever `STEAM_COMPAT_SHADER_PATH`
points. Turns out this was easier than expected.

Basically moving disk cache from:
- compatdata/<AppID>/pfx/drive_c/users/steamuser/AppData/Local/fex-emu/DiskCache
to:
- shadercache/<AppID>/fex-emu/DiskCache/

Fixes #5913
 2026-09-04 14:20:57 -07:00
Ryan Houdek 38c9f14dbd OpcodeDispatcher: Explicitly handle no-nop vector moves 
Just be very explicit about this rather than questioning why MMX moves
can't be a nop in the regular vector move code.
Doesn't change codegen so binarycacheversion doesn't need to change.
 2026-09-04 12:19:21 -07:00
Ryan Houdek 7187bc0e59 Merge pull request #5922 from simon902/MOVVectorUnalignedOp_Selfmove 
MOVVectorUnalignedOp: Fix same-register MMX writes being treated as a Nop
 2026-09-04 12:07:19 -07:00
Simon Scherer c5700cb0cc InstcountCI: Update 2026-09-04 17:03:46 +02:00
Simon Scherer 8ed4c52d58 OpcodeDispatcher: Don't skip the MMX state transition for same-register writes in MOVVectorUnalignedOp 2026-09-04 17:00:45 +02:00
Simon Scherer bd65c3e568 unittests/ASM: Test PFRCPIT1 self-move 2026-09-04 16:58:10 +02:00
LC 7052f91495 Merge pull request #5919 from simon902/fst_tag_word 
Mark tag word valid for FST ST(0) self-store
 2026-09-04 06:34:08 -04:00
Simon Scherer 040d3115d2 InstcountCI: Update 2026-09-04 11:53:13 +02:00
Simon Scherer 7236c2bf0a Fix x87 tag word not updating on FST ST(0) self-store 2026-09-04 11:51:12 +02:00
Simon Scherer ed3fa2f1c6 unittests/ASM: Test FST ST(0) tag word update 2026-09-04 11:50:21 +02:00
LC a454845ffd Merge pull request #5909 from simon902/fxch_tag_word 
Mark both operands valid in the tag word for FXCH
 2026-09-04 02:50:17 -04:00
Pierre-Loup A. Griffais 40bfff985f DiskCache: optional validation-only mode 
Diffs guest code in lookup miss hash mismatches and host code after lookup hit.
 2026-09-03 23:31:05 -07:00
Simon Scherer 96d65cd6bc InstcountCI: Update 2026-09-04 08:30:03 +02:00
Simon Scherer e4e45bcc59 x87StackOptimizationPass: Set both operands as valid in the tag word for FXCH 2026-09-04 08:26:17 +02:00
Simon Scherer 45fbab212a unittests/ASM: Test FXCH for tag word 2026-09-04 08:26:17 +02:00
Ryan Houdek 8243b93e4d Merge pull request #5917 from lioncash/wfe 
System_Tests: Enable WFET/WFIT tests
 2026-09-03 23:18:10 -07:00
Ryan Houdek 52cc5689db Merge pull request #5916 from lioncash/bf 
ARMEmitter: Fill in bfloat API holes
 2026-09-03 23:16:20 -07:00
LC cfef3d2cbf Merge pull request #5892 from simon902/FDECSTP_FDECSTP_C1 
Clear C1 for fdecstp and fincstp
 2026-09-04 01:57:31 -04:00
Simon Scherer fd64ea54fe InstcountCI: Update 2026-09-04 07:32:09 +02:00
Simon Scherer d1b2333195 OpcodeDispatcher: Clear C1 to 0 for X87ModifySTP (fdecstp and fincstp) 2026-09-04 07:22:05 +02:00
Simon Scherer 97fbd11177 unittests/ASM: Test fdecstp and fincstp for C1 clear 2026-09-04 07:22:05 +02:00
LC e7836f26dd Merge pull request #5914 from Sonicadvance1/230 
Core: Adds more validation around pool buffer ownership
 2026-09-03 23:13:43 -04:00
Ryan Houdek cf6da6ac26 Merge pull request #5915 from lioncash/vixl 
Externals: Update vixl
 2026-09-03 19:33:19 -07:00
Ryan Houdek a0cb91daa4 Core: Adds more validation around pool buffer ownership 
All buffers should be disowned leaving their respective compilation
sites, and reowning a buffer should never have the flag already be
owned.

Throw an assert in both cases because that would be a programming error
and result in some squirrely buffer handling
 2026-09-03 14:39:01 -07:00
Ryan Houdek e010e8adf0 Merge pull request #5897 from Plagman/plagman/anon_key_mr 
DiskCache: initial anon caching
 2026-09-03 14:38:26 -07:00
LC 7fe7c39a00 Merge pull request #5911 from javelina-pkwy/fix/windows-syscalls-arm64ec 
Windows: Fix syscall regression
 2026-09-03 14:48:02 -04:00
Justin Becker b03dc847f5 Fix syscall regression 2026-09-03 11:09:19 -07:00