Re: [Qemu-devel] [PATCH 1/4] memory: make memory API parsable by gtkdoc-scan

[Anthony Liguori]

On 12/14/2011 04:23 PM, Stefan Weil wrote:
> Am 14.12.2011 22:51, schrieb Anthony Liguori:
> > Look carefully at:
> >
> > http://qemu.weilnetz.de/gtkdoc/QEMU-Memory-API.html#MemoryRegionOps
> >
> > vs:
> >
> > http://wiki.qemu.org/docs-internal/QEMU-Memory-API.html#MemoryRegionOps
> >
> > There's a significant difference :-)
> >
> > Regards,
> >
> > Anthony Liguori
>
> I tried the following declaration:
>
> typedef struct sMemoryRegionOps {
> uint64_t (*read)(void *opaque,
> target_phys_addr_t addr,
> unsigned size);
> void (*write)(void *opaque,
> target_phys_addr_t addr,
> uint64_t data,
> unsigned size);
>
> enum device_endian endianness;
>
> MemoryRegionGuestConstraints valid;
> MemoryRegionInternalConstraints impl;
>
> const MemoryRegionPortio *old_portio;
> const MemoryRegionMmio old_mmio;
> } MemoryRegionOps;
>
> See the result here:
> http://qemu.weilnetz.de/gtkdoc4/QEMU-Memory-API.html#MemoryRegionOps

Interesting, but it wouldn't be possible to do a forward declaration?

I think my patch to gtk-doc (make _ optional) seems reasonable and I think 
that's a bit nicer than doing struct sCamelCase too.

That doesn't help with C++ compatibility but now that it is not in favor of my 
argument, I no longer care about it ;-) (j/k)

Regards,

Anthony Liguori

> Regards,
> Stefan Weil