Commit Graph

7468 Commits

Author SHA1 Message Date
Joel Linn da9c90835b [ImGui] Use new key API 2022-12-28 14:16:32 -06:00
Joel Linn b641e39c4d [ImGui] Use new ImageButton signature 2022-12-28 14:16:32 -06:00
Joel Linn d6b5cbd634 [ImGui] Fix deprecated SetCursorPos use 2022-12-28 14:16:32 -06:00
Joel Linn 29985f69e1 [ImGui] Fix fsr sharpness slider scaling
https://github.com/ocornut/imgui/issues/3361#issuecomment-1287045159
2022-12-28 14:16:32 -06:00
Joel Linn c8a39bad29 [ImGui] Update to v1.89 2022-12-28 14:16:32 -06:00
Joel Linn f452d6a007 [UI] Fix UB (moved mem) in file picker
- References to vector data become UB after vector size changes.
- Add one extra level of indirection to pin the wide string memory
  location regardless of vector memory
2022-12-28 14:16:32 -06:00
Joel Linn 7877331d8a [ImGui] Use ImDrawCmd::IdxOffset field
c80e8b964c
https://github.com/ocornut/imgui/issues/4845#issuecomment-1003329113
2022-12-28 14:16:32 -06:00
Joel Linn 8f88bb237e [ImGui] Fix empty IDs 2022-12-28 14:16:32 -06:00
Joel Linn b107af68bc [ImGui] Fix removed flags 2022-12-28 14:16:32 -06:00
Joel Linn f04cfb3b65 [ImGui] Update to v1.88 2022-12-28 14:16:32 -06:00
Gliniak 9f0d3d4c5b [Kernel] Simple support for loading achievement data 2022-12-25 16:11:59 +01:00
Gliniak a25ea03a28 [Threading] Revert sleepdelay0_for_maybeyield back to 0 2022-12-21 20:12:40 +01:00
Gliniak 783845c56e [XAM] Cleanup in xam_input.
- Added some missing xinput flags
- Simplified XamInputGetCapabilities
- Replaced hardcoded checks to flag
2022-12-20 19:33:55 +01:00
Radosław Gliński 78ff40af90
Merge pull request #107 from InvoxiPlayGames/get_capabilities_flags_fix
fix XamInputGetCapabilities breaking subtypes
2022-12-20 09:05:13 +01:00
InvoxiPlayGames d13c3b7306 fix XamInputGetCapabilities breaking subtypes 2022-12-19 04:09:27 +00:00
Radosław Gliński 8c43160fc6
Merge pull request #106 from Bo98/canary_burnout5_pr1
CPU threading improvements, Kernel game region improvements
2022-12-18 11:16:53 +01:00
Radosław Gliński f55defc1ab
Merge pull request #105 from Bo98/canary_burnout5_pr2
xb & build improvements, store recent.toml in storage root
2022-12-18 11:07:16 +01:00
Bo Anderson 24e7e5429a xb: allow py.exe launcher 2022-12-17 13:37:10 -05:00
Bo Anderson df368691cd xenia-build: find clang-format in VS installation 2022-12-17 13:36:59 -05:00
Bo Anderson 6c5423ceed app/emulator_window: store recent.toml in storage root
This makes it consistent with the config file.
2022-12-17 13:36:43 -05:00
Bo Anderson 422ee1fbdc kernal/xam/xam_info: base game region on user_country
It is debatable whether this is correct in the general case.

There's nothing really wrong with 0xFFFF logically.
Burnout Paradise however bases its in-game language on this and does not recognise 0xFFFF.
The game uses Japanese in the default case.

I've avoided the "rest of Asia" code since Burnout Paradise seems to use a different value (0x01F8) for that than what I expected (0x01FC).
2022-12-17 13:34:49 -05:00
Bo Anderson 2a77ed7246 cpu/backend/x64/x64_tracers: fix thread matching 2022-12-17 13:34:48 -05:00
Bo Anderson 9489161d0b kernel: separate host objects from regular handle range
Burnout Paradise statically expects certain thread handle values based on how many objects it knows it is allocating ahead of time.
From this, it calculates an ID by subtracting the thread handle from a base handle of what it expects the first such thread to be assigned.
The value is statically declared in the executable and is not determined automatically.

The host objects in the handle range made these thread handles higher than what the game expects.
Removing these, and allowing 0xF8000000 to be assigned, allows the thread handles to fit perfectly in the range the game expects.

It is not clear what handle range the host objects should be taking. For now though, they're 0-based rather than 0xF8000000-based.
2022-12-17 13:34:47 -05:00
Radosław Gliński da2710f18c
Merge pull request #101 from AdrianCassar/canary_experimental
Fixed crashes while loading titles
2022-12-16 20:21:37 +01:00
MoistyMarley 4f165825ea Fixed three crashes and refactored
While a title is running any attempt to open another title will result in a crash this commit prevents these crashes via File->Open and File->Open Recent->Title

Opening a recent title with an invalid path caused a crash.
2022-12-16 18:39:42 +00:00
chrisps 24bb4644ac
Merge pull request #104 from chrisps/command_processor_optimizations
Fix fetch issue introduced yesterday
2022-12-15 09:09:40 -08:00
chss95cs@gmail.com 3952997116 Fix issue introduced yesterday where the final fetch constant would never be marked as written
Reorganized SystemPageFlags for sharedmemory, each field now goes into its own array, the three arrays are page aligned in a single virtual allocation
Refactored sharedmemory a bit, use tzcnt if available when finding ranges (faster on pre-zen4 amd cpus)
2022-12-15 08:35:36 -08:00
chrisps b0268ab876
Merge pull request #103 from chrisps/command_processor_optimizations
Command processor optimizations
2022-12-14 12:16:22 -08:00
chrisps 8c6cefcce0
Merge branch 'xenia-canary:canary_experimental' into command_processor_optimizations 2022-12-14 11:35:13 -08:00
chss95cs@gmail.com f931c34ecb Cleaned up for commit, moved WriteRegistersFromMemCommonSense code into WriteRegistersFromMem
optimized copy_and_swap_32_unaligned further
2022-12-14 11:34:33 -08:00
chss95cs@gmail.com 754293ffc3 Fix mistake with fetch constant dirty mask 2022-12-14 10:13:13 -08:00
chss95cs@gmail.com 7a0fd0f32a Remove MaybeYields when vsync is off 2022-12-14 09:56:33 -08:00
chss95cs@gmail.com 080b6f4cbd Partially vectorized GetScissor (loading and unpacking the bitfields from the registers is still scalar) 2022-12-14 09:33:14 -08:00
chss95cs@gmail.com ab6d9dade0 add avx2 codepath for copy_and_swap_32_unaligned
use the new writerange approach in WriteRegisterRangeFromMem_WithKnownBound
2022-12-14 07:53:21 -08:00
Gliniak b2dd489151 [APU] Set first frame offset for next buffer + Note about edgecase 2022-12-13 22:57:36 +01:00
Gliniak e00feb7b0f [APU] Fixed incorrect frame count + removed hopefully useless check from now 2022-12-13 21:47:35 +01:00
chss95cs@gmail.com 82dcf3f951 faster/more compact MatchValueAndRef
Made commandprocessor GetcurrentRingReadcount inline, it was made noinline to match PGO decisions but i think PGO can make extra reg allocation decisions that make this inlining choice yield gains, whereas if we do it manually we lose a tiny bit of performance

Working on a more compact vectorized version of GetScissor to save icache on cmd processor thread
Add WriteRegisterForceinline, will probably end up amending to remove it

add ERMS path for vastcpy

Adding WriteRegistersFromMemCommonSense (name will be changed later), name is because i realized my approach with optimizing writeregisters has been backwards,  instead of handling all checks more quickly within the loop that writes the registers, we need a different loop for each range with unique handling. we also manage to hoist a lot of the logic out of the loops

Use 100ns delay for MaybeYield, noticed we often return almost immediately from the syscall so we end up wasting some cpu, instead we give up the cpu for the min waitable time (on my system, this is 0.5 ms)

Added a note about affinity mask/dynamic process affinity updates to threading_win

Add TextureFetchConstantsWritten
2022-12-13 11:25:33 -08:00
Gliniak 16580f5fae [APU] Fixed crash in error message caused by invalid arguments number 2022-12-13 11:02:59 +01:00
Gliniak 97fdf9c6dd [APU] Resolved crash related to negative amount of bits to copy
This is likely due to hitting somehow valid frame in invalid data
2022-12-13 08:46:48 +01:00
Gliniak 43d7fc5158 [APU] Shuffle checks to hopefully prevent crashing from logger 2022-12-11 21:06:47 +01:00
Gliniak c9cd6f15fc [APU] Fixed logged frame count
Until now info about frames that were provided in log was always incorrect by 1
2022-12-11 13:11:23 +01:00
Gliniak 82ccdd3db5 [APU] Misc Changes:
- Unified APU error messages
- Removed magic number from SetOffset call
- Commented out that annoying assertion from XmaContext::GetNextFrame
- Removed checks for current_input_packet_count and replaced with bool check

Not sure how to call it correctly. I know that calls with packet count == 1 is specific one
and probably handled differently. Is it streaming or how should it be called?
2022-12-11 12:22:23 +01:00
Gliniak 2c6bbf9a4a [GIT] Added recent.toml to ignored files 2022-12-11 11:11:02 +01:00
chrisps 0f94eb21c2
Merge pull request #102 from chrisps/error_modules_and_threads_plus_xmacontext_workaround
Host exception improvements, bandaid over div by 0 crash
2022-12-10 09:09:52 -08:00
chss95cs@gmail.com 7d49b97e4c Print any module name+ offset in host exception reports
print thread name in host exception reports
trying to force win32 error descriptions to english
Return if output buffer block count is 0 in XmaContext, this is an attempt to fix a divide by zero crash many users have reported
2022-12-09 12:24:06 -08:00
Gliniak 7c5da821d4 [Kernel] Fixed invalid thread pointer in KeEnableFpuExceptions 2022-12-08 21:48:13 +01:00
Radosław Gliński 747fb42bdf
Merge pull request #98 from AdrianCassar/canary_experimental
Added a hotkey to open the previously played title
2022-12-05 18:09:52 +01:00
MoistyMarley 6d2724a861 Added a hotkey to open the previously played title 2022-12-05 17:01:16 +00:00
chrisps 85723f117d
Merge pull request #99 from chrisps/stack_sync2_fence_krnl_hostexcept
Improve stack sync, kernel fixes, better host exception reporting
2022-12-04 13:50:06 -08:00
chss95cs@gmail.com a63f424c0a Directly check PEB for IsDebuggerAttached
Add constexpr getters to magicdiv class so it can be used from jitted x64/dxbc
Track the guest return address as well for guest/host sync, if multiple entries have the same guest stack find the first one with a matching guest retaddr. this fixes epic mickey 2 (which the previous guest-stack change had allowed to go ingame for a bit) and potentially also a crash in fable3.
Break if under debugger when stackpoints are overflowed

Add much more useful output for host exceptions, print out xenia_canary.exe relative offsets if exception is in module, formatmessage for ntstatus/win32err, strerror

Minor d3d12 microoptimization, instead of doing SetEventOnCompletion + WaitForSingleObject do SetEventOnCompletion w/ nullptr so that the wait happens in kernel mode, avoiding two extra context switches

add unimplemented kernel functions:
ExAllocatePoolWithTag
ObReferenceObject

ObDereferenceObject has no return value.
Log a message when ObDereferenceObject/Reference receive unregistered guest kernel objects
gave ObLookupThreadByThreadId its correct error status
hoist object_types initialization out of ObReferenceObjectByHandle
Fix out parameter values on error for a few kernel funcs
add note about msr to KeSetCurrentStackPointers
add X_STATUS_OBJECT_TYPE_MISMATCH check for xeNtSetEvent
add msr_mask field to X_KPCR
2022-12-04 12:38:19 -08:00