Em seg, 24 de jun de 2019 às 07:21, Eelco Chaudron
<echaudro@xxxxxxxxxx> escreveu:
> /* Stuff the ring with as much frames as possible */
> stock_frames = xsk_ring_prod__free(&xsk->umem->fq);
> stock_frames = MIN(stock_frames, xsk_umem_free_frames(xsk));
>
> if (stock_frames > 0) {
>
> ret = xsk_ring_prod__reserve(&xsk->umem->fq, stock_frames,
> &idx_fq);
> while (ret != stock_frames)
> ret = xsk_ring_prod__reserve(&xsk->umem->fq, rcvd,
> &idx_fq);
once you've checked before the number of free entries, this while is
still needed?
> Where xsk_ring_prod__free() is as follows (sent patch upstream for
> adding it as an API):
>
> static inline __u32 xsk_ring_prod__free(struct xsk_ring_prod *r)
> {
> r->cached_cons = *r->consumer + r->size;
> return r->cached_cons - r->cached_prod;
> }
this is basically xsk_prod_nb_free, but always triggering the cache
refresh, right?
>
> And the following is an API around my UMEM frame handling:
>
> xsk_umem_free_frames() -> How many buffers do I have available on my
> stack
>
> xsk_free_umem_frame() -> return a buffer to my buffer stack
>
> xsk_alloc_umem_frame() -> Get a buffer from my buffer stack
>
Is this API using another local ring or stack buffer or is it using
some clever trick I'm not figuring out?
I'll try this approach, thanks!
|