You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm hacking on a system call emulator for a 68K operating system, and I was considering running each 68K process in a host process thread. However, most of Musashi relies on global state such that only one Musashi instance can really be used at a single time, though it does appear multiple non-concurrent emulation contexts are supported via context switching.
This context should really be an argument within the API and internals, and only defer to global state when NULL is passed instead of a real context as with m68k_get_reg(). This does mean a binary- and source-compatibility break, though a combination of naming tweaks (such as using M68K_ instead of m68k_ for the context-taking APIs) and either #defines or stubs that define the old API in terms of the new API with a NULL context would obviate that.
The text was updated successfully, but these errors were encountered:
I'm hacking on a system call emulator for a 68K operating system, and I was considering running each 68K process in a host process thread. However, most of Musashi relies on global state such that only one Musashi instance can really be used at a single time, though it does appear multiple non-concurrent emulation contexts are supported via context switching.
This context should really be an argument within the API and internals, and only defer to global state when
NULL
is passed instead of a real context as withm68k_get_reg()
. This does mean a binary- and source-compatibility break, though a combination of naming tweaks (such as usingM68K_
instead ofm68k_
for the context-taking APIs) and either#define
s or stubs that define the old API in terms of the new API with aNULL
context would obviate that.The text was updated successfully, but these errors were encountered: