Re: [Qemu-ppc] [Qemu-devel] [PULL 07/32] PPC: Add support for Apple gdb in gdbstub

[Alexander Graf]

On 02.07.14 05:34, Programmingkid wrote:
> > The Apple gdbstub protocol is different from the normal gdbstub protocol
> > used on PowerPC. Add support for the different variant, so that we 
> > can use
> > Apple's gdb to debug guest code.
> >
> > Keep in mind that the switch is a compile time option. We can't detect
> > during runtime whether a gdb connecting to us is an upstream gdb or an
> > Apple gdb.
>
> Wouldn't it be better to assume we are using Apple's GDB? Building 
> the powerpc-apple-darwin target of gdb is not supported.

Why would we assume that any user wants to run the broken Apple gdb?

> > //#define PPC_DUMP_CPU
> > //#define PPC_DEBUG_SPR
> > //#define PPC_DUMP_SPR_ACCESSES
> > +/* #define USE_APPLE_GDB */
>
> Users are not going to be able to find these options very easily. 
> Using a condition to test for Mac OS X would solve this problem:
>
> #if defined(__APPLE__)
>
>     #define PPC_DUMP_CPU
>     #define PPC_DEBUG_SPR
>     #define PPC_DUMP_SPR_ACCESSES
>     #define USE_APPLE_GDB
>
> #endif

Eh. These defines are there for very specific debug purposes. I don't 
see why it would make sense to enable any of them by default just 
because we happen to run on an Apple host system.

> When I uncommented the "#define PPC_DUMP_CPU" code, there were a lot 
> of build errors on Mac OS X. They all had a message like this:
>
> /target-ppc/translate_init.c:8729: error: ‘opc_handler_t’ has no 
> member named ‘oname’

Ah, the joy of #ifdef'ed code :). I guess nobody used those debug 
features in a while. Feel free to send fixes ;).

But really, there is *no* correlation at all between the fact that a 
user wants to use an Apple gdb to debug a target and that QEMU happens 
to run on an Apple host system.


Alex