On 29 June 2012 13:42, Andreas Färber <afaer...@suse.de> wrote: > Am 28.06.2012 18:46, schrieb Peter Maydell: >> On 20 June 2012 18:28, Rabin Vincent <ra...@rab.in> wrote: >>> +int cpu_write_elf64_note(write_core_dump_function f, CPUArchState *env, >>> + int cpuid, void *opaque) >> >> Should these APIs really be taking a CPUArchState* rather rather than >> an ARMCPU* ? (Andreas?) > > I'm missing some context here. This is a new API? Where is it being used?
There's already an x86 implementation so the API is not new (see memory_mapping.h). I was just wondering if it was something we ought to be converting. > If it applies to multiple architectures and has separate > architecture-specific implementations then CPUState argument please. If > it's an ARM-internal API then, yes, ARMCPU please. If it's using common > CPU fields in common code then CPUArchState is still the most fitting today. I think that it's in the former category but there's no documentation :-( -- PMM
