[PATCH net-next v17 07/14] memory-provider: dmabuf devmem memory provider

[Mina Almasry]

Implement a memory provider that allocates dmabuf devmem in the form of
net_iov.

The provider receives a reference to the struct netdev_dmabuf_binding
via the pool->mp_priv pointer. The driver needs to set this pointer for
the provider in the net_iov.

The provider obtains a reference on the netdev_dmabuf_binding which
guarantees the binding and the underlying mapping remains alive until
the provider is destroyed.

Usage of PP_FLAG_DMA_MAP is required for this memory provide such that
the page_pool can provide the driver with the dma-addrs of the devmem.

Support for PP_FLAG_DMA_SYNC_DEV is omitted for simplicity & p.order !=
0.

Signed-off-by: Willem de Bruijn <willemb@xxxxxxxxxx>
Signed-off-by: Kaiyuan Zhang <kaiyuanz@xxxxxxxxxx>
Signed-off-by: Mina Almasry <almasrymina@xxxxxxxxxx>
Reviewed-by: Pavel Begunkov <asml.silence@xxxxxxxxx>

---

v17:
- Use ASSERT_RTNL (Jakub)

v16:
- Add DEBUG_NET_WARN_ON_ONCE(!rtnl_is_locked()), to catch cases if
  page_pool_init without rtnl_locking when the queue is provided. In
  this case, the queue configuration may be changed while we're initing
  the page_pool, which could be a race.

v13:
- Return on warning (Pavel).
- Fixed pool->recycle_stats not being freed on error (Pavel).
- Applied reviewed-by from Pavel.

v11:
- Rebase to not use the ops. (Christoph)

v8:
- Use skb_frag_size instead of frag->bv_len to fix patch-by-patch build
  error

v6:
- refactor new memory provider functions into net/core/devmem.c (Pavel)

v2:
- Disable devmem for p.order != 0

v1:
- static_branch check in page_is_page_pool_iov() (Willem & Paolo).
- PP_DEVMEM -> PP_IOV (David).
- Require PP_FLAG_DMA_MAP (Jakub).

---
 include/net/mp_dmabuf_devmem.h |  44 ++++++++++++++
 include/net/netmem.h           |  15 +++++
 include/net/page_pool/types.h  |   2 +
 net/core/devmem.c              |  78 +++++++++++++++++++++++++
 net/core/page_pool.c           | 102 ++++++++++++++++++++++++---------
 net/core/page_pool_priv.h      |   3 +
 6 files changed, 218 insertions(+), 26 deletions(-)
 create mode 100644 include/net/mp_dmabuf_devmem.h

diff --git a/include/net/mp_dmabuf_devmem.h b/include/net/mp_dmabuf_devmem.h
new file mode 100644
index 0000000000000..300a2356eed00
--- /dev/null
+++ b/include/net/mp_dmabuf_devmem.h
@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Dmabuf device memory provider.
+ *
+ * Authors:	Mina Almasry <almasrymina@xxxxxxxxxx>
+ *
+ */
+#ifndef _NET_MP_DMABUF_DEVMEM_H
+#define _NET_MP_DMABUF_DEVMEM_H
+
+#include <net/netmem.h>
+
+#if defined(CONFIG_DMA_SHARED_BUFFER) && defined(CONFIG_GENERIC_ALLOCATOR)
+int mp_dmabuf_devmem_init(struct page_pool *pool);
+
+netmem_ref mp_dmabuf_devmem_alloc_netmems(struct page_pool *pool, gfp_t gfp);
+
+void mp_dmabuf_devmem_destroy(struct page_pool *pool);
+
+bool mp_dmabuf_devmem_release_page(struct page_pool *pool, netmem_ref netmem);
+#else
+static inline int mp_dmabuf_devmem_init(struct page_pool *pool)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline netmem_ref mp_dmabuf_devmem_alloc_netmems(struct page_pool *pool,
+							gfp_t gfp)
+{
+	return 0;
+}
+
+static inline void mp_dmabuf_devmem_destroy(struct page_pool *pool)
+{
+}
+
+static inline bool mp_dmabuf_devmem_release_page(struct page_pool *pool,
+						 netmem_ref netmem)
+{
+	return false;
+}
+#endif
+
+#endif /* _NET_MP_DMABUF_DEVMEM_H */
diff --git a/include/net/netmem.h b/include/net/netmem.h
index 99531780e53af..34dd146f9e940 100644
--- a/include/net/netmem.h
+++ b/include/net/netmem.h
@@ -100,6 +100,21 @@ static inline struct page *netmem_to_page(netmem_ref netmem)
 	return (__force struct page *)netmem;
 }
 
+static inline struct net_iov *netmem_to_net_iov(netmem_ref netmem)
+{
+	if (netmem_is_net_iov(netmem))
+		return (struct net_iov *)((__force unsigned long)netmem &
+					  ~NET_IOV);
+
+	DEBUG_NET_WARN_ON_ONCE(true);
+	return NULL;
+}
+
+static inline netmem_ref net_iov_to_netmem(struct net_iov *niov)
+{
+	return (__force netmem_ref)((unsigned long)niov | NET_IOV);
+}
+
 static inline netmem_ref page_to_netmem(struct page *page)
 {
 	return (__force netmem_ref)page;
diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h
index 4afd6dd56351a..e01ccb42e582f 100644
--- a/include/net/page_pool/types.h
+++ b/include/net/page_pool/types.h
@@ -52,6 +52,7 @@ struct pp_alloc_cache {
  * @nid:	NUMA node id to allocate from pages from
  * @dev:	device, for DMA pre-mapping purposes
  * @napi:	NAPI which is the sole consumer of pages, otherwise NULL
+ * @queue:	struct netdev_rx_queue this page_pool is being created for.
  * @dma_dir:	DMA mapping direction
  * @max_len:	max DMA sync memory size for PP_FLAG_DMA_SYNC_DEV
  * @offset:	DMA sync address offset for PP_FLAG_DMA_SYNC_DEV
@@ -66,6 +67,7 @@ struct page_pool_params {
 		int		nid;
 		struct device	*dev;
 		struct napi_struct *napi;
+		struct netdev_rx_queue *queue;
 		enum dma_data_direction dma_dir;
 		unsigned int	max_len;
 		unsigned int	offset;
diff --git a/net/core/devmem.c b/net/core/devmem.c
index befff59a2ee64..f8e78907c6c68 100644
--- a/net/core/devmem.c
+++ b/net/core/devmem.c
@@ -17,8 +17,11 @@
 #include <linux/genalloc.h>
 #include <linux/dma-buf.h>
 #include <net/devmem.h>
+#include <net/mp_dmabuf_devmem.h>
 #include <net/netdev_queues.h>
 
+#include "page_pool_priv.h"
+
 /* Device memory support */
 
 #if defined(CONFIG_DMA_SHARED_BUFFER) && defined(CONFIG_GENERIC_ALLOCATOR)
@@ -297,4 +300,79 @@ void dev_dmabuf_uninstall(struct net_device *dev)
 					xa_erase(&binding->bound_rxqs, xa_idx);
 	}
 }
+
+/*** "Dmabuf devmem memory provider" ***/
+
+int mp_dmabuf_devmem_init(struct page_pool *pool)
+{
+	struct net_devmem_dmabuf_binding *binding = pool->mp_priv;
+
+	if (!binding)
+		return -EINVAL;
+
+	if (!pool->dma_map)
+		return -EOPNOTSUPP;
+
+	if (pool->dma_sync)
+		return -EOPNOTSUPP;
+
+	if (pool->p.order != 0)
+		return -E2BIG;
+
+	net_devmem_dmabuf_binding_get(binding);
+	return 0;
+}
+
+netmem_ref mp_dmabuf_devmem_alloc_netmems(struct page_pool *pool, gfp_t gfp)
+{
+	struct net_devmem_dmabuf_binding *binding = pool->mp_priv;
+	netmem_ref netmem;
+	struct net_iov *niov;
+	dma_addr_t dma_addr;
+
+	niov = net_devmem_alloc_dmabuf(binding);
+	if (!niov)
+		return 0;
+
+	dma_addr = net_devmem_get_dma_addr(niov);
+
+	netmem = net_iov_to_netmem(niov);
+
+	page_pool_set_pp_info(pool, netmem);
+
+	if (page_pool_set_dma_addr_netmem(netmem, dma_addr))
+		goto err_free;
+
+	pool->pages_state_hold_cnt++;
+	trace_page_pool_state_hold(pool, netmem, pool->pages_state_hold_cnt);
+	return netmem;
+
+err_free:
+	net_devmem_free_dmabuf(niov);
+	return 0;
+}
+
+void mp_dmabuf_devmem_destroy(struct page_pool *pool)
+{
+	struct net_devmem_dmabuf_binding *binding = pool->mp_priv;
+
+	net_devmem_dmabuf_binding_put(binding);
+}
+
+bool mp_dmabuf_devmem_release_page(struct page_pool *pool, netmem_ref netmem)
+{
+	if (WARN_ON_ONCE(!netmem_is_net_iov(netmem)))
+		return false;
+
+	if (WARN_ON_ONCE(atomic_long_read(netmem_get_pp_ref_count_ref(netmem)) !=
+		     1))
+		return false;
+
+	page_pool_clear_pp_info(netmem);
+
+	net_devmem_free_dmabuf(netmem_to_net_iov(netmem));
+
+	/* We don't want the page pool put_page()ing our net_iovs. */
+	return false;
+}
 #endif
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index c5c303746d494..b70f779cb7856 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -13,6 +13,7 @@
 
 #include <net/page_pool/helpers.h>
 #include <net/xdp.h>
+#include <net/netdev_rx_queue.h>
 
 #include <linux/dma-direction.h>
 #include <linux/dma-mapping.h>
@@ -21,6 +22,9 @@
 #include <linux/poison.h>
 #include <linux/ethtool.h>
 #include <linux/netdevice.h>
+#include <linux/genalloc.h>
+#include <net/devmem.h>
+#include <net/mp_dmabuf_devmem.h>
 
 #include <trace/events/page_pool.h>
 
@@ -28,6 +32,7 @@
 #include "netmem_priv.h"
 
 DEFINE_STATIC_KEY_FALSE(page_pool_mem_providers);
+EXPORT_SYMBOL(page_pool_mem_providers);
 
 #define DEFER_TIME (msecs_to_jiffies(1000))
 #define DEFER_WARN_INTERVAL (60 * HZ)
@@ -190,6 +195,7 @@ static int page_pool_init(struct page_pool *pool,
 			  int cpuid)
 {
 	unsigned int ring_qsize = 1024; /* Default */
+	int err;
 
 	page_pool_struct_check();
 
@@ -271,7 +277,35 @@ static int page_pool_init(struct page_pool *pool,
 	if (pool->dma_map)
 		get_device(pool->p.dev);
 
+	if (pool->p.queue) {
+		/* We rely on rtnl_lock()ing to make sure netdev_rx_queue
+		 * configuration doesn't change while we're initializing the
+		 * page_pool.
+		 */
+		ASSERT_RTNL();
+		pool->mp_priv = pool->p.queue->mp_params.mp_priv;
+	}
+
+	if (pool->mp_priv) {
+		err = mp_dmabuf_devmem_init(pool);
+		if (err) {
+			pr_warn("%s() mem-provider init failed %d\n", __func__,
+				err);
+			goto free_ptr_ring;
+		}
+
+		static_branch_inc(&page_pool_mem_providers);
+	}
+
 	return 0;
+
+free_ptr_ring:
+	ptr_ring_cleanup(&pool->ring, NULL);
+#ifdef CONFIG_PAGE_POOL_STATS
+	if (!pool->system)
+		free_percpu(pool->recycle_stats);
+#endif
+	return err;
 }
 
 static void page_pool_uninit(struct page_pool *pool)
@@ -455,28 +489,6 @@ static bool page_pool_dma_map(struct page_pool *pool, netmem_ref netmem)
 	return false;
 }
 
-static void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem)
-{
-	netmem_set_pp(netmem, pool);
-	netmem_or_pp_magic(netmem, PP_SIGNATURE);
-
-	/* Ensuring all pages have been split into one fragment initially:
-	 * page_pool_set_pp_info() is only called once for every page when it
-	 * is allocated from the page allocator and page_pool_fragment_page()
-	 * is dirtying the same cache line as the page->pp_magic above, so
-	 * the overhead is negligible.
-	 */
-	page_pool_fragment_netmem(netmem, 1);
-	if (pool->has_init_callback)
-		pool->slow.init_callback(netmem, pool->slow.init_arg);
-}
-
-static void page_pool_clear_pp_info(netmem_ref netmem)
-{
-	netmem_clear_pp_magic(netmem);
-	netmem_set_pp(netmem, NULL);
-}
-
 static struct page *__page_pool_alloc_page_order(struct page_pool *pool,
 						 gfp_t gfp)
 {
@@ -572,7 +584,10 @@ netmem_ref page_pool_alloc_netmem(struct page_pool *pool, gfp_t gfp)
 		return netmem;
 
 	/* Slow-path: cache empty, do real allocation */
-	netmem = __page_pool_alloc_pages_slow(pool, gfp);
+	if (static_branch_unlikely(&page_pool_mem_providers) && pool->mp_priv)
+		netmem = mp_dmabuf_devmem_alloc_netmems(pool, gfp);
+	else
+		netmem = __page_pool_alloc_pages_slow(pool, gfp);
 	return netmem;
 }
 EXPORT_SYMBOL(page_pool_alloc_netmem);
@@ -608,6 +623,28 @@ s32 page_pool_inflight(const struct page_pool *pool, bool strict)
 	return inflight;
 }
 
+void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem)
+{
+	netmem_set_pp(netmem, pool);
+	netmem_or_pp_magic(netmem, PP_SIGNATURE);
+
+	/* Ensuring all pages have been split into one fragment initially:
+	 * page_pool_set_pp_info() is only called once for every page when it
+	 * is allocated from the page allocator and page_pool_fragment_page()
+	 * is dirtying the same cache line as the page->pp_magic above, so
+	 * the overhead is negligible.
+	 */
+	page_pool_fragment_netmem(netmem, 1);
+	if (pool->has_init_callback)
+		pool->slow.init_callback(netmem, pool->slow.init_arg);
+}
+
+void page_pool_clear_pp_info(netmem_ref netmem)
+{
+	netmem_clear_pp_magic(netmem);
+	netmem_set_pp(netmem, NULL);
+}
+
 static __always_inline void __page_pool_release_page_dma(struct page_pool *pool,
 							 netmem_ref netmem)
 {
@@ -636,8 +673,13 @@ static __always_inline void __page_pool_release_page_dma(struct page_pool *pool,
 void page_pool_return_page(struct page_pool *pool, netmem_ref netmem)
 {
 	int count;
+	bool put;
 
-	__page_pool_release_page_dma(pool, netmem);
+	put = true;
+	if (static_branch_unlikely(&page_pool_mem_providers) && pool->mp_priv)
+		put = mp_dmabuf_devmem_release_page(pool, netmem);
+	else
+		__page_pool_release_page_dma(pool, netmem);
 
 	/* This may be the last page returned, releasing the pool, so
 	 * it is not safe to reference pool afterwards.
@@ -645,8 +687,10 @@ void page_pool_return_page(struct page_pool *pool, netmem_ref netmem)
 	count = atomic_inc_return_relaxed(&pool->pages_state_release_cnt);
 	trace_page_pool_state_release(pool, netmem, count);
 
-	page_pool_clear_pp_info(netmem);
-	put_page(netmem_to_page(netmem));
+	if (put) {
+		page_pool_clear_pp_info(netmem);
+		put_page(netmem_to_page(netmem));
+	}
 	/* An optimization would be to call __free_pages(page, pool->p.order)
 	 * knowing page is not part of page-cache (thus avoiding a
 	 * __page_cache_release() call).
@@ -965,6 +1009,12 @@ static void __page_pool_destroy(struct page_pool *pool)
 
 	page_pool_unlist(pool);
 	page_pool_uninit(pool);
+
+	if (pool->mp_priv) {
+		mp_dmabuf_devmem_destroy(pool);
+		static_branch_dec(&page_pool_mem_providers);
+	}
+
 	kfree(pool);
 }
 
diff --git a/net/core/page_pool_priv.h b/net/core/page_pool_priv.h
index 90665d40f1eb7..e5b019ab0b735 100644
--- a/net/core/page_pool_priv.h
+++ b/net/core/page_pool_priv.h
@@ -9,4 +9,7 @@ int page_pool_list(struct page_pool *pool);
 void page_pool_detached(struct page_pool *pool);
 void page_pool_unlist(struct page_pool *pool);
 
+void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem);
+void page_pool_clear_pp_info(netmem_ref netmem);
+
 #endif
-- 
2.46.0.rc1.232.g9752f9e123-goog