Re: [Qemu-devel] [PATCH v1 01/17] qemu/obj_pool.h: introduce object allocation pool

["Michael S. Tsirkin" <mst@redhat.com>]

On Tue, Aug 05, 2014 at 06:21:55AM -0600, Eric Blake wrote:
> On 08/05/2014 06:05 AM, Michael S. Tsirkin wrote:
> > On Tue, Aug 05, 2014 at 05:55:49AM -0600, Eric Blake wrote:
> >> On 08/04/2014 09:33 PM, Ming Lei wrote:
> >>> This patch introduces object allocation pool for speeding up
> >>> object allocation in fast path.
> >>>
> >>> Signed-off-by: Ming Lei <ming.lei@canonical.com>
> >>> ---
> >>>  include/qemu/obj_pool.h |   64 +++++++++++++++++++++++++++++++++++++++++++++++
> >>>  1 file changed, 64 insertions(+)
> >>>  create mode 100644 include/qemu/obj_pool.h
> >>>
> 
> >>> +
> >>> +    char *objs;
> >>> +} ObjPool;
> >>> +
> >>> +static inline void obj_pool_init(ObjPool *op, void *objs_buf, void **free_objs,
> >>> +                                 unsigned int obj_size, unsigned cnt)
> >>> +{
> >>> +    int i;
> >>> +
> >>> +    op->objs = (char *)objs_buf;
> >>
> >> Why the cast? This is C, not C++.
> > 
> > It's not needed in C++ either, right?
> 
> In C++, going from void* to a typed pointer requires a cast (that's why
> in C++ you see casts on malloc results).

Ah yes, I was confusing this with going from char * to void *.
You are right, thanks for the reminder.

>  In C, void* can implicitly be
> converted to any other pointer (modulo const-/volatile-correctness).

Yes: and const and voilatile safety is exactly the reason
one *shouldn't* typically cast to/from void * explicitly.

> -- 
> Eric Blake   eblake redhat com    +1-919-301-3266
> Libvirt virtualization library http://libvirt.org
>