new flavors:
no_mmx - do not build with mmx support
no_comp - do not build with compressed hard drive support
Also new in this version is the vga font is compiled into the binary, no
more 'xset fp+ ..' or editing XF86Config to run bochs!
Changes in 2.0 (December 21, 2002):
Since the change log is hundreds of lines long, here is a very brief summary.
- 2x emulation speedup!!!
- added plugin devices and guis. Now you can compile with many more
options, and choose between them at runtime.
- added emulation of AMD x86-64, MMX, SSE, SSE2 instructions
- add wxWindows port (a graphical configuration interface and display lib)
and SVGAlib port (full screen display for Linux without X11)
- improvements in many I/O devices: for example up to 8 hard disks/cdroms,
TUN/TAP network interface, 360k floppies,
- improved MacOSX/Carbon interface and updated MacOS9 port
- GDB remote stub, allows symbolic debugging with Bochs simulation.
- support for up to 32gig hard disk images
networking working, I'm committing this from inside bochs!
There are too many changes to mention, not the least of which is
two hard drive support, smp (up to 16 cpus), networking, and much,
much more ..
o too many changes to mention specifically, but amongst them:
- better error reporting
- openbsd boots, freebsd boots
- new 'term' flavor, I added a textmode interface, the only drawback
is that control-c is intercepted, and you must kill -9 it from
another tty if it doesn't exit normally
- new flavors:
- debugger -> debug (for shortness)
- i386
- pci
- smp
- net (compiles, does not yet provide networking, if anyone
wants to help, I'd be ELATED to have this fixed!)
- no_x11 (made it actually work)
- term (make sure to use 'log: /dev/null' or something else
on the cmdline, otherwise, debugging and the screen will
start getting quite confusing!)
- better error reporting (and no error reporting where it is not
necessary; a typicall boot of cdrom28.fs is a screenful of output now)
- in general, if the hardware reaches an unknown state, fail
gracefully instead of calling exit(1) .. are we trying to work or
are we trying to .. fail? I like working myself.
