New FS-UAE branches with updated emulation code from WinUAE

[FrodeSolheim]

Please note; the winuae4210 branch has been merged into master. Before doing that, master was branched into fs-uae-3.0, so further 2.9dev / 3.0 releases will be made from this branch. 3.1dev releases will be made from the master branch.

[BastyCDGS]

Great work! Thank you!

BTW, here is the GDB crash dump when trying to enable clipboard sharing with OS4.1FE:

Assembly:

0x0000000000d90660 fs_semaphore_post+0 mov    (%rdi),%rdi
0x0000000000d90663 fs_semaphore_post+3 jmpq   0x409490 <SDL_SemPost@plt>

registers:

   rax 0x0000000000000001      rbx 0x0000000000000000      rcx 0x0000000000000004      rdx 0x00007fff44eb0000  
   rsi 0x0000000000000001      rdi 0x00eb004000000000      rbp 0x0000000005a6c200      rsp 0x00007fffe90a8d48  
    r8 0x00000000ffff3054       r9 0x0000000000000004      r10 0x0000000000000000      r11 0x0000000000000000  
   r12 0x0000000000000000      r13 0x0000000000000000      r14 0x00000000076fec50      r15 0x00007fffffffda90  
   rip 0x0000000000d90660   eflags [ IF RF ]                cs 0x00000033               ss 0x0000002b          
    ds 0x00000000               es 0x00000000               fs 0x00000000               gs 0x00000000

fs_semaphore_post (semaphore=0xeb004000000000) at libfsemu/src/thread.c:306
306         return SDL_SemPost(semaphore->semaphore);
>>> bt
#0  fs_semaphore_post (semaphore=0xeb004000000000) at libfsemu/src/thread.c:306
#1  0x0000000000748511 in put_byte (b=<optimized out>, addr=<optimized out>) at ./src/include/uae/memory.h:650
#2  op_117c_0_ff (opcode=<optimized out>) at gen/cpuemu_0.cpp:7369
#3  0x00000000005bd1bb in m68k_run_2 () at src/newcpu.cpp:6521
#4  0x00000000005ba6a2 in m68k_go (may_quit=may_quit@entry=1) at src/newcpu.cpp:6778
#5  0x000000000055f085 in do_start_program () at src/main.cpp:1075
#6  0x000000000056120c in do_start_program () at src/main.cpp:1062
#7  start_program () at src/main.cpp:1091
#8  real_main2 (argv=0x7fffe90a8e10, argc=1) at src/main.cpp:1238
#9  real_main (argc=argc@entry=1, argv=argv@entry=0x7fffe90a8e10) at src/main.cpp:1254
#10 0x00000000005c7305 in amiga_main () at src/od-fs/libamiga.cpp:317
#11 0x0000000000412ca9 in main_function () at src/fs-uae/main.c:748
#12 0x0000000000d7c26d in emulation_thread (data=0x412ca0 <main_function>) at libfsemu/src/emu/emu.c:412
#13 0x00007ffff792e195 in ?? () from /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0
#14 0x00007ffff5be56db in start_thread (arg=0x7fffe90a9700) at pthread_create.c:463
#15 0x00007ffff590e88f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

[ReadOnlyCat]

Originally Posted by BastyCDGS

I have found a quote in the deep web which might help. Another possibility would be to look into the wine code how it translates the beam racing code from DX to OpenGL.

I realize that this would be pile even more work on Frode's shoulders but on macOS there may be a chance that the Metal API exposes these kind of things?
I am not familiar with it so I have no clue if that is the case but since the API was pushed with game developers in mind, it seems likely that sync-status retrieval was included in some form.

[FrodeSolheim]

This thread will soon be replaced with a proper thread for the new development series. I have of course taken noticed of the reported crash reports and will still be reviewing those.

Originally Posted by ReadOnlyCat

I realize that this would be pile even more work on Frode's shoulders but on macOS there may be a chance that the Metal API exposes these kind of things?
I am not familiar with it so I have no clue if that is the case but since the API was pushed with game developers in mind, it seems likely that sync-status retrieval was included in some form.

It could actually be, but I can almost guarantee that I'm not writing nor directly supporting Metal. (I'm quite annoyed that Apple did their own thing instead of supporting Vulcan, or provide proper and updated OpenGL drivers).