steam-frame/FEX-Emu--FEX
1
0 Fork 0
mirror of https://github.com/FEX-Emu/FEX.git synced 2026-10-13 18:00:44 +02:00
Code Issues Packages Projects Releases 67 Wiki Activity
9,173 Commits 11 Branches 71 Tags
84d5b3ee59ecf2922919b8f028fef59cd626db14
Commit Graph
980 Commits
Author SHA1 Message Date
Ryan Houdek 84d5b3ee59 CPUID: Enable enhanced rep movs in more situations 
Instead of only enabling enhanced rep movs if software TSO is disabled,
Enable it if software tso is disabled OR memcpysettso is disabled. This
is because now we hit the fast path when memcpysettso is disabled alone
but global TSO is disabled.

Retested Hades and performance was fine in this configuration.
 2024-04-21 18:50:17 -07:00
Ryan Houdek 376936c808 Merge pull request #3591 from alyssarosenzweig/ra/fix 
JIT: fix ShiftFlags shuffles
 2024-04-19 13:50:45 -07:00
Alyssa Rosenzweig 932b8f38f4 JIT: fix ShiftFlags shuffles 
messed up my RA.

fixes ShiftPF.asm with jit_1 with a pathological register allocation

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-18 14:11:03 -04:00
Ryan Houdek c8704a7f71 OpcodeDispatcher: Implement support for SMSW 
Found out that Far Cry uses this instruction and it is viable to use in
CPL-3. This only returns constant data but its behaviour is a little
quirky.

This instruction has a weird behaviour that the 32-bit operation does an
insert in to the 64-bit destination, which might be an Intel versus AMD
behaviour. I don't have an Intel machine available to test if that
theory is true although. This assumption would match similar behaviour
where segment registers are inserted instead of zext.

Gets the game farther but then it crashes in a `___ascii_strnicmp`
function where the arguments end up being `___ascii_strnicmp(nullptr, "Color", 5);`.
 2024-04-18 07:41:39 -07:00
Paulo Matos 2b4ec88dae Whole-tree reformat 
This follows discussions from #3413.
Followup commits add clang-format file, script and blame ignore lists.
 2024-04-12 16:26:02 +02:00
Paulo Matos 6524716404 Move const to the left in preparation for reformatting 
clang-format-16 had some issues with const placement, so we are manually changing these.
 2024-04-12 16:06:57 +02:00
Ryan Houdek 271700e9f6 Merge pull request #3568 from lioncash/const 
X87: Simplify constant loading for FLD family
 2024-04-11 00:32:26 -07:00
Ryan Houdek 1ba678f631 Merge pull request #3562 from Sonicadvance1/fix_rsp_store_tso 
OpcodeDispatcher: Fixes disabling TSO access on RSP SIB stores
 2024-04-11 00:32:14 -07:00
Ryan Houdek a0f2cae1cb Merge pull request #3567 from lioncash/veczero 
IR: Remove VectorZero
 2024-04-09 20:04:26 -07:00
Ryan Houdek 66cbb66732 Merge pull request #3566 from lioncash/address 
OpcodeDispatcher: Add helper for making segment offset addresses
 2024-04-09 17:31:43 -07:00
Lioncache 4cb2432b5c OpcodeDispatcher: Make use of new x87 constants 
Now we can load these directly instead of needing to manually materialize them.
 2024-04-09 10:17:15 -04:00
Lioncache 65b5281d7c IR: Add constants for FLD variants 2024-04-09 10:13:33 -04:00
Ryan Houdek 1a8b61b9fc Merge pull request #3560 from bylaws/ec-pt6 
FEXCore: Support x64 -> arm64ec calls
 2024-04-09 07:08:38 -07:00
Ryan Houdek f0dad86332 Merge pull request #3559 from bylaws/ec-pt5 
LookupCache: Track ARM64EC page state in the code cache
 2024-04-09 07:08:29 -07:00
Mai eedb120fd0 Merge pull request #3563 from Sonicadvance1/fill_spill_pairs 
JIT: Adds support for spilling/Filling GPRPair
 2024-04-08 23:05:11 -04:00
Lioncache 98841fe07a IR: Remove VectorZero 
We have LoadNamedVectorConstant that now performs this behavior while
also being more flexible.
 2024-04-08 22:42:58 -04:00
Lioncache b0aeb501f4 OpcodeDispatcher: Add helper for making segment offset addresses 
There's quite a few places where the segment offset appending is open-coded
throughout the opcode dispatcher, but we can pull these out into a few
helpers to make the sites a little more compact and declarative.
 2024-04-08 17:50:58 -04:00
Lioncache b26bf2eaf6 DebugData: Remove header 
This isn't included or used anywhere, so it can be removed.
 2024-04-08 16:05:40 -04:00
Ryan Houdek 574fdcef32 JIT: Adds support for spilling/Filling GPRPair 
Tony noticed this last week. I encountered it this week.
Add support for spilling and filling GPR pairs.
 2024-04-08 11:55:29 -07:00
Ryan Houdek 0e93fd0f3e OpcodeDispatcher: Fixes disabling TSO access on RSP SIB stores 
GPR Direct/Indirect already had this and SIB version also already
supported on the load side. Fixes this missed behaviour.
 2024-04-08 11:32:01 -07:00
Alyssa Rosenzweig d9493e5d9b OpcodeDispatcher: fix xblock liveness in xsave/xrstr 
didn't fix this hard enough before. caught by validation.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-08 13:48:29 -04:00
Alyssa Rosenzweig eb83c9e7f2 Core: use safe CondJump for self-modifying code 
this ensures we put the StoreNZCV in the right block, which will fix validation
fails later in the series.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-08 13:48:29 -04:00
Billy Laws 526e3e654f LookupCache: Track ARM64EC page state in the code cache 
Rather than checking the actual EC bitmap in the dispatcher (~6 instrs), this
indirection through the code cache allows just 1 instr for the hot path
of calling repeated EC code/x64 code.
 2024-04-08 16:08:17 +01:00
Billy Laws 243bb45a68 FEXCore: Support x64 -> arm64ec calls 
The frontend will provide the return logic via ExitFunctionEC, which
will be jumped to whenever there is an indirect branch/return to an addr
such that RtlIsEcCode(addr) returns true.
 2024-04-06 13:20:48 +00:00
Alyssa Rosenzweig 95589f6172 OpcodeDispatcher: rm deferred variable shift flag calcs 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig 098859caf7 OpcodeDispatcher: use _ShiftFlags for ASHR 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig c632543451 OpcodeDispatcher: use _ShiftFlags for SHRD 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig 801cf72f95 OpcodeDispatcher: use _ShiftFlags for SHLD 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig 650cd2c46e OpcodeDispatcher: use _ShiftFlags for SHR 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig 5b48ce2228 OpcodeDispatcher: use _ShiftFlags for SHL 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig 2173c26fd8 OpcodeDispatcher: add HandleShift helper 
all the variable shift impls need to do this dance, make it common.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig 982391ba9d IR: add ShiftFlags op 
Generates flags for a variable shift as a dedicated IR op. This lets us optimize
around it (without generating control flow, relying on deferred flag infra,
etc). And it neatly solves our RA problem for shifts.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig 3661da1bc6 OpcodeDispatcher: calculate deferred flags before RMW on NZCV 
otherwise we might have the wrong input NZCV.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 20:34:05 -04:00
Alyssa Rosenzweig 859df5e0b2 OpcodeDispatcher: optimize shl flag 
This is something the new shift flag code will do. Backporting the opt since
that's stalled and this reduces the diff.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-05 19:38:44 -04:00
Alyssa Rosenzweig a05cc06ab4 OpcodeDispatcher: unify imm/1-bit ASHR 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig 031e756a78 OpcodeDispatcher: unify imm/1-bit SHR 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig 2a9f1ce8cb OpcodeDispatcher: unify imm/1-bit SHL 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig 8c53a9f051 OpcodeDispatcher: use LoadConstantShift for rotates 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig cf26ec7898 OpcodeDispatcher: use LoadConstantShift for SHRD 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig 582c3dae6e OpcodeDispatcher: use LoadConstantShift for SHLD 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig 2abac03ab0 OpcodeDispatcher: add LoadConstantShift helper 
shows up a bunch

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig 8cc684fa12 OpcodeDispatcher: drop misinformed comment 
tbnz only tests a single bit, not a mask.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig d92de1d947 OpcodeDispatcher: drop result masking for shifts 
flag calcs are fine with upper garbage.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-04 07:42:15 -04:00
Alyssa Rosenzweig 37f2b417e4 Merge pull request #3546 from alyssarosenzweig/flag/cleanup 
Minor cleanups around flags
 2024-04-02 11:29:19 -04:00
Alyssa Rosenzweig bd0b5eceb8 Merge pull request #3545 from alyssarosenzweig/opt/pf-scalar 
Use scalar integer code to calculate PF
 2024-04-02 11:28:53 -04:00
Alyssa Rosenzweig b632f7215c Merge pull request #3544 from alyssarosenzweig/ra/zero-multiple 
OpcodeDispatcher: drop ZeroMultipleFlags
 2024-04-02 11:27:45 -04:00
Ryan Houdek e8abc88702 Merge pull request #3542 from alyssarosenzweig/ra/rep 
Eliminate xblock liveness with rep cmp/lod/scas
 2024-04-02 04:24:24 -07:00
Ryan Houdek 29c6281e11 Merge pull request #3539 from alyssarosenzweig/ra/rol-ror2 
rewrite ROL/ROR
 2024-04-02 00:17:08 -07:00
Alyssa Rosenzweig 067a5444dc OpcodeDispatcher: use HandleNZ00Write 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-01 16:42:38 -04:00
Alyssa Rosenzweig c7f159972d OpcodeDispatcher: rm pointless NZCV loads 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
 2024-04-01 16:42:38 -04:00