For now the optimization is supported only when no SMP is compiled in because it doesn't handle cross-modifying code.
The current stack page will cache also current pageWriteStamp for that page and could skip pageWriteStamp access if possible.
Any code fetch access missing trace cache will invalidate current stack page.
Code fetch accesses from another SMP threads should do the same to support SMP.
Next step:
- support SMP
- support pageWriteStamp access skipping for all other memory writes from all segments
The sanity check would help to detect real Bochs crash issue under Win x64 with MSDEV
configure script under Mingw env decided that SIZEOF_INT_P == 4 which is terribly wrong for 64-bit host.
implemented important CPU statistics which were used for Bochs CPU model performance analysis.
old statistics code from paging.cc and cpu.cc is replaced with new infrastructure.
In order to enale statitics collection in Bochs CPU:
- Enable statistics @ compilation time in cpu/cpustats.h
- Dump statistics periodically by adding -dumpstats N into Bochs command line
Do not clear IA32_FEATURE_CTRL MSR on soft reset (will clear the VMX lock bit)
On real HW XSAVE/XRSTOR which is not 4-byte aligned cause #AC(0) intead of #GP(0) when alignment check is enabled
Debugger: if AVX-512 if not supported by current configuration do not print high256 of vector registers and zmm15..zmm31 in AVX command
Implement VBROADCASTF64x4, VBROADCASTF32x4, VBROADCASTFI64x4, VBROADCASTI32x4 AVX-512 instructions
Fetchdecode optimizations and bugfixes
The Windows version looks almost stable, but the GTK version fails in some cases.
That's why the classic wx debugger is still available if BX_DEBUGGER_GUI is set to 0.
- added function close_debug_dialog() to handle the simulation stop case in wx
- disable all the wx debugger related code if BX_DEBUGGER_GUI is set to 1
- added enhanced debugger specific init code similar to the code in sdl.cc
- include debugger related resources on Windows
- TODO: make the GTK / wxGTK case stable and remove the wx debugger
Minor speedup (of 1-2%) was observed due to new implementation
Remove obsolete dbg_take_irq function and dbg_force_interrupt function from CPU code, the functions were not working properly anyway
