Mercurial > repos > blastem
diff backend.h @ 2666:38c281ef57b0
Memory access optimizaiton in new 68K core that gives a modest speed bump on average and will allow low-cost watchpoints
| author | Michael Pavone <pavone@retrodev.com> |
|---|---|
| date | Fri, 07 Mar 2025 23:40:58 -0800 |
| parents | bf4f1a8d1d48 |
| children | 7e86ec94c899 |
line wrap: on
line diff
--- a/backend.h Fri Mar 07 21:45:53 2025 -0800 +++ b/backend.h Fri Mar 07 23:40:58 2025 -0800 @@ -86,6 +86,11 @@ typedef uint8_t * (*native_addr_func)(void * context, uint32_t address); +typedef uint16_t (*interp_read_16)(uint32_t address, void *context, void *data); +typedef uint8_t (*interp_read_8)(uint32_t address, void *context, void *data); +typedef void (*interp_write_16)(uint32_t address, void *context, uint16_t value, void *data); +typedef void (*interp_write_8)(uint32_t address, void *context, uint8_t value, void *data); + deferred_addr * defer_address(deferred_addr * old_head, uint32_t address, uint8_t *dest); void remove_deferred_until(deferred_addr **head_ptr, deferred_addr * remove_to); void process_deferred(deferred_addr ** head_ptr, void * context, native_addr_func get_native); @@ -110,6 +115,10 @@ memmap_chunk const *find_map_chunk(uint32_t address, cpu_options *opts, uint16_t flags, uint32_t *size_sum); uint32_t chunk_size(cpu_options *opts, memmap_chunk const *chunk); uint32_t ram_size(cpu_options *opts); +interp_read_16 get_interp_read_16(void *context, cpu_options *opts, uint32_t start, uint32_t end, void **data_out); +interp_read_8 get_interp_read_8(void *context, cpu_options *opts, uint32_t start, uint32_t end, void **data_out); +interp_write_16 get_interp_write_16(void *context, cpu_options *opts, uint32_t start, uint32_t end, void **data_out); +interp_write_8 get_interp_write_8(void *context, cpu_options *opts, uint32_t start, uint32_t end, void **data_out); #endif //BACKEND_H_