This is implemented by pretending every 16 bytes in ROM is a different function, allowing the profiler to display cycle counts for each individual "function" in the UI
Also fixes a missing call to ResetFlags for the merge opcode for FX3
-Fixed getAccessCounters documentation (swapped param order)
-Fixed file watcher not being disabled properly when closing the script window, causing the script to run if changed on the disk, even if the script window is closed
-Prevent file watcher from triggering from writes done by the script window itself (which caused the script to be reloaded 2 or 3 times in a row)
-Fixed UI freezes when using emu.log() on a large table
-Fixed crash when 100k character limit (in the last 500 lines) for script log was reached (out of bounds read/write)
-Added "fps" value to "getState"
-Slightly improved getAccessCounters/getCdlData performance
-Added a new drawPixels API to improve performance when a script needs to draw a lot of individual pixels
Misc fixes:
-Fixed issues with the opcode tooltip not appearing for some GBA instructions (esp. when the condition code ends with an 's', e.g: rsbcs, movls, subcs)
-Fixed byte code not being lower case when the "use lower case" option is enabled
Also, minor bug fixes for WS:
-Fixed tilemap viewer for sometimes using the wrong background color
-Fixed the 4-bit RGB color values in the palette viewer having the R and B values swapped
Also fixes presets/clear buttons for NTT, Pachinko & Hypershot, and improves layout for all controller config popups (buttons/borders are aligned, etc.)
When importing labels via .dbg/.elf/etc. files, the 64kb length limit for labels was ignored.
This prevents several performance problems when a label is defined as having a length of e.g 1+mb
-Removed the confusing/hard-to-use "previous refresh value" "compare to" option
-The "last value" column now shows the values that "previous search value" will compare against
-Added "update values" button to apply an empty filter that does not hide any addresses, but updates the "last value" that will be used for subsequent filters
The header editor was treating MB units as KB, and its error checking was showing errors for MB sizes between 8 and 59 that were not a multiple of 4 because it was treating them as KB sizes.
* Code: Add auto-formatting rules for C++ projects (clang-format) and enforce code style via "clang-format" and "dotnet format"
* fix ui format warnings
+Fixed some incorrect/missing validations
+Disable save button when an error message is shown
+Fixed .nes load logic ignoring PRG/CHR sizes from NES 2.0 headers (was getting overriden by game DB)
The NES header editor is currently unable to save a ROM if it was loaded from an archive (such as .zip) or if the ROM was soft-patched. This is because the editor reloads the ROM from disk to then apply the new header and save a copy, but the C# compnent is not able to open archives or apply patches. If the ROM file is in an archive, it throws an Unexpected Error and fails.
This change instead makes the editor use the current PRG and CHR ROM. This means it doesn't need to reload the ROM, so it can save it regardless of how it was loaded. It also means that it saves any changes that have been made to the ROM in the session, either by self-flashing or by editing them via tools, which could be seen as a pro or a con. Also, because Mesen does not currently use misc ROM, it is unable to include that in the saved ROM, but this is something we could feasibly add later.
This change was written by Sour.
The game sets the sprites to obj mode 3 (by mistake?), which is an unused "stereoscopic" mode for sprites. Mode 3 seems to behave like mode 0 when the global "stereoscopic" flag is not set (previously "green swap"), so the sprites are meant to display normally in this scenario (instead of being hidden like they were before)
Caused by what looks like slight differences between how AOT treats an empty array (vs JIT builds) when it marshals the array (AOT sends a null pointer, JIT sends a pointer to an empty array)
This was causing issues in both builds (AOT crashed immediately because of the null pointer, JIT read from out-of-bounds memory)
+ Fixed VRAM read to wait until next available CPU access slot
+ Fixed NMI triggering on the wrong scanline on SMS
+ Fixed NMI issues on ColecoVision
+ Fixed random ram option not working properly on SG-1000
+ Added "Mem access" tab to tilemap viewer to see fetch patterns
-Also allows cropping the edit window when the size doesn't fit (e.g using 4x4 near the bottom of the screen can result in a 4x2 edit window, etc.)
-Fixed a display issue when the number of edited columns & rows didn't match
Prevent queueing another update while the previous update is still running, which should prevent most issues.
Also slightly improves performance for the SNES tilemap viewer (which was causing the freeze when displaying an 8bpp 1024x1024 tilemap)
+ Automatically move to current address when clicking on the source view tab (so that the active row is shown after manually loading a symbol file, etc.)
Specifically, this happened if the breakpoint was already visible on screen and was associated to an address that wasn't mapped when the source view was last updated
SourMesen