page_pool: devmem support

Convert netmem to be a union of struct page and struct netmem. Overload
the LSB of struct netmem* to indicate that it's a net_iov, otherwise
it's a page.

Currently these entries in struct page are rented by the page_pool and
used exclusively by the net stack:

struct {
	unsigned long pp_magic;
	struct page_pool *pp;
	unsigned long _pp_mapping_pad;
	unsigned long dma_addr;
	atomic_long_t pp_ref_count;
};

Mirror these (and only these) entries into struct net_iov and implement
netmem helpers that can access these common fields regardless of
whether the underlying type is page or net_iov.

Implement checks for net_iov in netmem helpers which delegate to mm
APIs, to ensure net_iov are never passed to the mm stack.

Signed-off-by: Mina Almasry <almasrymina@google.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Link: https://patch.msgid.link/20240910171458.219195-6-almasrymina@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

include/net/netmem.h

@@ -8,12 +8,52 @@
 #ifndef _NET_NETMEM_H
 #define _NET_NETMEM_H
 
+#include <linux/mm.h>
+#include <net/net_debug.h>
+
 /* net_iov */
 
+DECLARE_STATIC_KEY_FALSE(page_pool_mem_providers);
+
+/*  We overload the LSB of the struct page pointer to indicate whether it's
+ *  a page or net_iov.
+ */
+#define NET_IOV 0x01UL
+
 struct net_iov {
+	unsigned long __unused_padding;
+	unsigned long pp_magic;
+	struct page_pool *pp;
 	struct dmabuf_genpool_chunk_owner *owner;
+	unsigned long dma_addr;
+	atomic_long_t pp_ref_count;
 };
 
+/* These fields in struct page are used by the page_pool and net stack:
+ *
+ *        struct {
+ *                unsigned long pp_magic;
+ *                struct page_pool *pp;
+ *                unsigned long _pp_mapping_pad;
+ *                unsigned long dma_addr;
+ *                atomic_long_t pp_ref_count;
+ *        };
+ *
+ * We mirror the page_pool fields here so the page_pool can access these fields
+ * without worrying whether the underlying fields belong to a page or net_iov.
+ *
+ * The non-net stack fields of struct page are private to the mm stack and must
+ * never be mirrored to net_iov.
+ */
+#define NET_IOV_ASSERT_OFFSET(pg, iov)             \
+	static_assert(offsetof(struct page, pg) == \
+		      offsetof(struct net_iov, iov))
+NET_IOV_ASSERT_OFFSET(pp_magic, pp_magic);
+NET_IOV_ASSERT_OFFSET(pp, pp);
+NET_IOV_ASSERT_OFFSET(dma_addr, dma_addr);
+NET_IOV_ASSERT_OFFSET(pp_ref_count, pp_ref_count);
+#undef NET_IOV_ASSERT_OFFSET
+
 /* netmem */
 
 /**
@@ -27,20 +67,37 @@ struct net_iov {
  */
 typedef unsigned long __bitwise netmem_ref;
 
+static inline bool netmem_is_net_iov(const netmem_ref netmem)
+{
+	return (__force unsigned long)netmem & NET_IOV;
+}
+
 /* This conversion fails (returns NULL) if the netmem_ref is not struct page
  * backed.
- *
- * Currently struct page is the only possible netmem, and this helper never
- * fails.
  */
 static inline struct page *netmem_to_page(netmem_ref netmem)
 {
+	if (WARN_ON_ONCE(netmem_is_net_iov(netmem)))
+		return NULL;
+
 	return (__force struct page *)netmem;
 }
 
-/* Converting from page to netmem is always safe, because a page can always be
+static inline struct net_iov *netmem_to_net_iov(netmem_ref netmem)
- * a 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;
@@ -48,17 +105,70 @@ static inline netmem_ref page_to_netmem(struct page *page)
 
 static inline int netmem_ref_count(netmem_ref netmem)
 {
+	/* The non-pp refcount of net_iov is always 1. On net_iov, we only
+	 * support pp refcounting which uses the pp_ref_count field.
+	 */
+	if (netmem_is_net_iov(netmem))
+		return 1;
+
 	return page_ref_count(netmem_to_page(netmem));
 }
 
-static inline unsigned long netmem_to_pfn(netmem_ref netmem)
+static inline unsigned long netmem_pfn_trace(netmem_ref netmem)
 {
+	if (netmem_is_net_iov(netmem))
+		return 0;
+
 	return page_to_pfn(netmem_to_page(netmem));
 }
 
+static inline struct net_iov *__netmem_clear_lsb(netmem_ref netmem)
+{
+	return (struct net_iov *)((__force unsigned long)netmem & ~NET_IOV);
+}
+
+static inline struct page_pool *netmem_get_pp(netmem_ref netmem)
+{
+	return __netmem_clear_lsb(netmem)->pp;
+}
+
+static inline atomic_long_t *netmem_get_pp_ref_count_ref(netmem_ref netmem)
+{
+	return &__netmem_clear_lsb(netmem)->pp_ref_count;
+}
+
+static inline bool netmem_is_pref_nid(netmem_ref netmem, int pref_nid)
+{
+	/* NUMA node preference only makes sense if we're allocating
+	 * system memory. Memory providers (which give us net_iovs)
+	 * choose for us.
+	 */
+	if (netmem_is_net_iov(netmem))
+		return true;
+
+	return page_to_nid(netmem_to_page(netmem)) == pref_nid;
+}
+
 static inline netmem_ref netmem_compound_head(netmem_ref netmem)
 {
+	/* niov are never compounded */
+	if (netmem_is_net_iov(netmem))
+		return netmem;
+
 	return page_to_netmem(compound_head(netmem_to_page(netmem)));
 }
 
+static inline void *netmem_address(netmem_ref netmem)
+{
+	if (netmem_is_net_iov(netmem))
+		return NULL;
+
+	return page_address(netmem_to_page(netmem));
+}
+
+static inline unsigned long netmem_get_dma_addr(netmem_ref netmem)
+{
+	return __netmem_clear_lsb(netmem)->dma_addr;
+}
+
 #endif /* _NET_NETMEM_H */

include/net/page_pool/helpers.h

@@ -216,7 +216,7 @@ page_pool_get_dma_dir(const struct page_pool *pool)
 
 static inline void page_pool_fragment_netmem(netmem_ref netmem, long nr)
 {
-	atomic_long_set(&netmem_to_page(netmem)->pp_ref_count, nr);
+	atomic_long_set(netmem_get_pp_ref_count_ref(netmem), nr);
 }
 
 /**
@@ -244,7 +244,7 @@ static inline void page_pool_fragment_page(struct page *page, long nr)
 
 static inline long page_pool_unref_netmem(netmem_ref netmem, long nr)
 {
-	struct page *page = netmem_to_page(netmem);
+	atomic_long_t *pp_ref_count = netmem_get_pp_ref_count_ref(netmem);
 	long ret;
 
 	/* If nr == pp_ref_count then we have cleared all remaining
@@ -261,19 +261,19 @@ static inline long page_pool_unref_netmem(netmem_ref netmem, long nr)
 	 * initially, and only overwrite it when the page is partitioned into
 	 * more than one piece.
 	 */
-	if (atomic_long_read(&page->pp_ref_count) == nr) {
+	if (atomic_long_read(pp_ref_count) == nr) {
 		/* As we have ensured nr is always one for constant case using
 		 * the BUILD_BUG_ON(), only need to handle the non-constant case
 		 * here for pp_ref_count draining, which is a rare case.
 		 */
 		BUILD_BUG_ON(__builtin_constant_p(nr) && nr != 1);
 		if (!__builtin_constant_p(nr))
-			atomic_long_set(&page->pp_ref_count, 1);
+			atomic_long_set(pp_ref_count, 1);
 
 		return 0;
 	}
 
-	ret = atomic_long_sub_return(nr, &page->pp_ref_count);
+	ret = atomic_long_sub_return(nr, pp_ref_count);
 	WARN_ON(ret < 0);
 
 	/* We are the last user here too, reset pp_ref_count back to 1 to
@@ -282,7 +282,7 @@ static inline long page_pool_unref_netmem(netmem_ref netmem, long nr)
 	 * page_pool_unref_page() currently.
 	 */
 	if (unlikely(!ret))
-		atomic_long_set(&page->pp_ref_count, 1);
+		atomic_long_set(pp_ref_count, 1);
 
 	return ret;
 }
@@ -401,9 +401,7 @@ static inline void page_pool_free_va(struct page_pool *pool, void *va,
 
 static inline dma_addr_t page_pool_get_dma_addr_netmem(netmem_ref netmem)
 {
-	struct page *page = netmem_to_page(netmem);
+	dma_addr_t ret = netmem_get_dma_addr(netmem);
-
-	dma_addr_t ret = page->dma_addr;
 
 	if (PAGE_POOL_32BIT_ARCH_WITH_64BIT_DMA)
 		ret <<= PAGE_SHIFT;
@@ -423,24 +421,6 @@ static inline dma_addr_t page_pool_get_dma_addr(const struct page *page)
 	return page_pool_get_dma_addr_netmem(page_to_netmem((struct page *)page));
 }
 
-static inline bool page_pool_set_dma_addr_netmem(netmem_ref netmem,
-						 dma_addr_t addr)
-{
-	struct page *page = netmem_to_page(netmem);
-
-	if (PAGE_POOL_32BIT_ARCH_WITH_64BIT_DMA) {
-		page->dma_addr = addr >> PAGE_SHIFT;
-
-		/* We assume page alignment to shave off bottom bits,
-		 * if this "compression" doesn't work we need to drop.
-		 */
-		return addr != (dma_addr_t)page->dma_addr << PAGE_SHIFT;
-	}
-
-	page->dma_addr = addr;
-	return false;
-}
-
 /**
  * page_pool_dma_sync_for_cpu - sync Rx page for CPU after it's written by HW
  * @pool: &page_pool the @page belongs to
@@ -463,11 +443,6 @@ static inline void page_pool_dma_sync_for_cpu(const struct page_pool *pool,
 				      page_pool_get_dma_dir(pool));
 }
 
-static inline bool page_pool_set_dma_addr(struct page *page, dma_addr_t addr)
-{
-	return page_pool_set_dma_addr_netmem(page_to_netmem(page), addr);
-}
-
 static inline bool page_pool_put(struct page_pool *pool)
 {
 	return refcount_dec_and_test(&pool->user_cnt);

include/trace/events/page_pool.h

@@ -57,12 +57,12 @@ TRACE_EVENT(page_pool_state_release,
 		__entry->pool		= pool;
 		__entry->netmem		= (__force unsigned long)netmem;
 		__entry->release	= release;
-		__entry->pfn		= netmem_to_pfn(netmem);
+		__entry->pfn		= netmem_pfn_trace(netmem);
 	),
 
-	TP_printk("page_pool=%p netmem=%p pfn=0x%lx release=%u",
+	TP_printk("page_pool=%p netmem=%p is_net_iov=%lu pfn=0x%lx release=%u",
 		  __entry->pool, (void *)__entry->netmem,
-		  __entry->pfn, __entry->release)
+		  __entry->netmem & NET_IOV, __entry->pfn, __entry->release)
 );
 
 TRACE_EVENT(page_pool_state_hold,
@@ -83,12 +83,12 @@ TRACE_EVENT(page_pool_state_hold,
 		__entry->pool	= pool;
 		__entry->netmem	= (__force unsigned long)netmem;
 		__entry->hold	= hold;
-		__entry->pfn	= netmem_to_pfn(netmem);
+		__entry->pfn	= netmem_pfn_trace(netmem);
 	),
 
-	TP_printk("page_pool=%p netmem=%p pfn=0x%lx hold=%u",
+	TP_printk("page_pool=%p netmem=%p is_net_iov=%lu, pfn=0x%lx hold=%u",
 		  __entry->pool, (void *)__entry->netmem,
-		  __entry->pfn, __entry->hold)
+		  __entry->netmem & NET_IOV, __entry->pfn, __entry->hold)
 );
 
 TRACE_EVENT(page_pool_update_nid,

net/core/devmem.c

@@ -18,6 +18,7 @@
 #include <trace/events/page_pool.h>
 
 #include "devmem.h"
+#include "page_pool_priv.h"
 
 /* Device memory support */
 
@@ -82,6 +83,10 @@ net_devmem_alloc_dmabuf(struct net_devmem_dmabuf_binding *binding)
 	index = offset / PAGE_SIZE;
 	niov = &owner->niovs[index];
 
+	niov->pp_magic = 0;
+	niov->pp = NULL;
+	atomic_long_set(&niov->pp_ref_count, 0);
+
 	return niov;
 }
 
@@ -269,6 +274,8 @@ net_devmem_bind_dmabuf(struct net_device *dev, unsigned int dmabuf_fd,
 		for (i = 0; i < owner->num_niovs; i++) {
 			niov = &owner->niovs[i];
 			niov->owner = owner;
+			page_pool_set_dma_addr_netmem(net_iov_to_netmem(niov),
+						      net_devmem_get_dma_addr(niov));
 		}
 
 		virtual += len;

net/core/netmem_priv.h

@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef __NETMEM_PRIV_H
+#define __NETMEM_PRIV_H
+
+static inline unsigned long netmem_get_pp_magic(netmem_ref netmem)
+{
+	return __netmem_clear_lsb(netmem)->pp_magic;
+}
+
+static inline void netmem_or_pp_magic(netmem_ref netmem, unsigned long pp_magic)
+{
+	__netmem_clear_lsb(netmem)->pp_magic |= pp_magic;
+}
+
+static inline void netmem_clear_pp_magic(netmem_ref netmem)
+{
+	__netmem_clear_lsb(netmem)->pp_magic = 0;
+}
+
+static inline void netmem_set_pp(netmem_ref netmem, struct page_pool *pool)
+{
+	__netmem_clear_lsb(netmem)->pp = pool;
+}
+
+static inline void netmem_set_dma_addr(netmem_ref netmem,
+				       unsigned long dma_addr)
+{
+	__netmem_clear_lsb(netmem)->dma_addr = dma_addr;
+}
+#endif

net/core/page_pool.c

@@ -24,8 +24,11 @@
 
 #include <trace/events/page_pool.h>
 
+#include "netmem_priv.h"
 #include "page_pool_priv.h"
 
+DEFINE_STATIC_KEY_FALSE(page_pool_mem_providers);
+
 #define DEFER_TIME (msecs_to_jiffies(1000))
 #define DEFER_WARN_INTERVAL (60 * HZ)
 
@@ -358,7 +361,7 @@ static noinline netmem_ref page_pool_refill_alloc_cache(struct page_pool *pool)
 		if (unlikely(!netmem))
 			break;
 
-		if (likely(page_to_nid(netmem_to_page(netmem)) == pref_nid)) {
+		if (likely(netmem_is_pref_nid(netmem, pref_nid))) {
 			pool->alloc.cache[pool->alloc.count++] = netmem;
 		} else {
 			/* NUMA mismatch;
@@ -454,10 +457,8 @@ static bool page_pool_dma_map(struct page_pool *pool, netmem_ref netmem)
 
 static void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem)
 {
-	struct page *page = netmem_to_page(netmem);
+	netmem_set_pp(netmem, pool);
-
+	netmem_or_pp_magic(netmem, PP_SIGNATURE);
-	page->pp = pool;
-	page->pp_magic |= 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
@@ -472,10 +473,8 @@ static void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem)
 
 static void page_pool_clear_pp_info(netmem_ref netmem)
 {
-	struct page *page = netmem_to_page(netmem);
+	netmem_clear_pp_magic(netmem);
-
+	netmem_set_pp(netmem, NULL);
-	page->pp_magic = 0;
-	page->pp = NULL;
 }
 
 static struct page *__page_pool_alloc_page_order(struct page_pool *pool,
@@ -692,8 +691,9 @@ static bool page_pool_recycle_in_cache(netmem_ref netmem,
 
 static bool __page_pool_page_can_be_recycled(netmem_ref netmem)
 {
-	return page_ref_count(netmem_to_page(netmem)) == 1 &&
+	return netmem_is_net_iov(netmem) ||
-	       !page_is_pfmemalloc(netmem_to_page(netmem));
+	       (page_ref_count(netmem_to_page(netmem)) == 1 &&
+		!page_is_pfmemalloc(netmem_to_page(netmem)));
 }
 
 /* If the page refcnt == 1, this will try to recycle the page.
@@ -728,6 +728,7 @@ __page_pool_put_page(struct page_pool *pool, netmem_ref netmem,
 		/* Page found as candidate for recycling */
 		return netmem;
 	}
+
 	/* Fallback/non-XDP mode: API user have elevated refcnt.
 	 *
 	 * Many drivers split up the page into fragments, and some
@@ -949,7 +950,7 @@ static void page_pool_empty_ring(struct page_pool *pool)
 	/* Empty recycle ring */
 	while ((netmem = (__force netmem_ref)ptr_ring_consume_bh(&pool->ring))) {
 		/* Verify the refcnt invariant of cached pages */
-		if (!(page_ref_count(netmem_to_page(netmem)) == 1))
+		if (!(netmem_ref_count(netmem) == 1))
 			pr_crit("%s() page_pool refcnt %d violation\n",
 				__func__, netmem_ref_count(netmem));
 

net/core/page_pool_priv.h

@@ -3,10 +3,36 @@
 #ifndef __PAGE_POOL_PRIV_H
 #define __PAGE_POOL_PRIV_H
 
+#include <net/page_pool/helpers.h>
+
+#include "netmem_priv.h"
+
 s32 page_pool_inflight(const struct page_pool *pool, bool strict);
 
 int page_pool_list(struct page_pool *pool);
 void page_pool_detached(struct page_pool *pool);
 void page_pool_unlist(struct page_pool *pool);
 
+static inline bool
+page_pool_set_dma_addr_netmem(netmem_ref netmem, dma_addr_t addr)
+{
+	if (PAGE_POOL_32BIT_ARCH_WITH_64BIT_DMA) {
+		netmem_set_dma_addr(netmem, addr >> PAGE_SHIFT);
+
+		/* We assume page alignment to shave off bottom bits,
+		 * if this "compression" doesn't work we need to drop.
+		 */
+		return addr != (dma_addr_t)netmem_get_dma_addr(netmem)
+				       << PAGE_SHIFT;
+	}
+
+	netmem_set_dma_addr(netmem, addr);
+	return false;
+}
+
+static inline bool page_pool_set_dma_addr(struct page *page, dma_addr_t addr)
+{
+	return page_pool_set_dma_addr_netmem(page_to_netmem(page), addr);
+}
+
 #endif

net/core/skbuff.c

@@ -88,6 +88,7 @@
 #include <linux/textsearch.h>
 
 #include "dev.h"
+#include "netmem_priv.h"
 #include "sock_destructor.h"
 
 #ifdef CONFIG_SKB_EXTENSIONS
@@ -920,9 +921,9 @@ static void skb_clone_fraglist(struct sk_buff *skb)
 		skb_get(list);
 }
 
-static bool is_pp_page(struct page *page)
+static bool is_pp_netmem(netmem_ref netmem)
 {
-	return (page->pp_magic & ~0x3UL) == PP_SIGNATURE;
+	return (netmem_get_pp_magic(netmem) & ~0x3UL) == PP_SIGNATURE;
 }
 
 int skb_pp_cow_data(struct page_pool *pool, struct sk_buff **pskb,
@@ -1020,9 +1021,7 @@ EXPORT_SYMBOL(skb_cow_data_for_xdp);
 #if IS_ENABLED(CONFIG_PAGE_POOL)
 bool napi_pp_put_page(netmem_ref netmem)
 {
-	struct page *page = netmem_to_page(netmem);
+	netmem = netmem_compound_head(netmem);
-
-	page = compound_head(page);
 
 	/* page->pp_magic is OR'ed with PP_SIGNATURE after the allocation
 	 * in order to preserve any existing bits, such as bit 0 for the
@@ -1031,10 +1030,10 @@ bool napi_pp_put_page(netmem_ref netmem)
 	 * and page_is_pfmemalloc() is checked in __page_pool_put_page()
 	 * to avoid recycling the pfmemalloc page.
 	 */
-	if (unlikely(!is_pp_page(page)))
+	if (unlikely(!is_pp_netmem(netmem)))
 		return false;
 
-	page_pool_put_full_netmem(page->pp, page_to_netmem(page), false);
+	page_pool_put_full_netmem(netmem_get_pp(netmem), netmem, false);
 
 	return true;
 }
@@ -1061,7 +1060,7 @@ static bool skb_pp_recycle(struct sk_buff *skb, void *data)
 static int skb_pp_frag_ref(struct sk_buff *skb)
 {
 	struct skb_shared_info *shinfo;
-	struct page *head_page;
+	netmem_ref head_netmem;
 	int i;
 
 	if (!skb->pp_recycle)
@@ -1070,11 +1069,11 @@ static int skb_pp_frag_ref(struct sk_buff *skb)
 	shinfo = skb_shinfo(skb);
 
 	for (i = 0; i < shinfo->nr_frags; i++) {
-		head_page = compound_head(skb_frag_page(&shinfo->frags[i]));
+		head_netmem = netmem_compound_head(shinfo->frags[i].netmem);
-		if (likely(is_pp_page(head_page)))
+		if (likely(is_pp_netmem(head_netmem)))
-			page_pool_ref_page(head_page);
+			page_pool_ref_netmem(head_netmem);
 		else
-			page_ref_inc(head_page);
+			page_ref_inc(netmem_to_page(head_netmem));
 	}
 	return 0;
 }