On 11/5/24 16:28, Alexander Lobakin wrote:
...
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index a813d30d2135..c21c5b9edc68 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -284,10 +284,11 @@ static int page_pool_init(struct page_pool *pool,
rxq = __netif_get_rx_queue(pool->slow.netdev,
pool->slow.queue_idx);
pool->mp_priv = rxq->mp_params.mp_priv;
+ pool->mp_ops = rxq->mp_params.mp_ops;
}
- if (pool->mp_priv) {
- err = mp_dmabuf_devmem_init(pool);
+ if (pool->mp_ops) {
+ err = pool->mp_ops->init(pool);
Can't we just switch-case instead of indirect calls?
IO_URING is bool, it can't be a module, which means its functions will
be available here when it's enabled. These ops are easy to predict (no
ops, dmabuf, io_uring), so this really looks like an enum with 3 entries
+ switch-case ("regular" path is out if this switch-case under likely etc).
Because it better frames the provider api and doesn't require
io_uring calls sticking off the net code, i.e. decouples subsystems
better, that's while these calls are not in the hot path (in case of
io_uring it's ammortised). But you're right that it can be turned into
a switch, I just don't think it's better, and that's how it was done
in the original patch.
if (err) {
pr_warn("%s() mem-provider init failed %d\n", __func__,
err);
@@ -584,8 +585,8 @@ netmem_ref page_pool_alloc_netmem(struct page_pool *pool, gfp_t gfp)
return netmem;
Thanks,
Olek
--
Pavel Begunkov
