xsk: add generic XSk &xdp_buff -> skb conversion

Same as with converting &xdp_buff to skb on Rx, the code which allocates
a new skb and copies the XSk frame there is identical across the
drivers, so make it generic. This includes copying all the frags if they
are present in the original buff.
System percpu page_pools greatly improve XDP_PASS performance on XSk:
instead of page_alloc() + page_free(), the net core recycles the same
pages, so the only overhead left is memcpy()s. When the Page Pool is
not compiled in, the whole function is a return-NULL (but it always
gets selected when eBPF is enabled).
Note that the passed buff gets freed if the conversion is done w/o any
error, assuming you don't need this buffer after you convert it to an
skb.

Reviewed-by: Maciej Fijalkowski <[email protected]>
Signed-off-by: Alexander Lobakin <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

Author: alobakin

include/net/xdp.h

@@ -337,6 +337,7 @@ void xdp_warn(const char *msg, const char *func, const int line);
 #define XDP_WARN(msg) xdp_warn(msg, __func__, __LINE__)
 
 struct sk_buff *xdp_build_skb_from_buff(const struct xdp_buff *xdp);
+struct sk_buff *xdp_build_skb_from_zc(struct xdp_buff *xdp);
 struct xdp_frame *xdp_convert_zc_to_xdp_frame(struct xdp_buff *xdp);
 struct sk_buff *__xdp_build_skb_from_frame(struct xdp_frame *xdpf,
 					   struct sk_buff *skb,

net/core/xdp.c

@@ -684,6 +684,118 @@ struct sk_buff *xdp_build_skb_from_buff(const struct xdp_buff *xdp)
 }
 EXPORT_SYMBOL_GPL(xdp_build_skb_from_buff);
 
+/**
+ * xdp_copy_frags_from_zc - copy frags from XSk buff to skb
+ * @skb: skb to copy frags to
+ * @xdp: XSk &xdp_buff from which the frags will be copied
+ * @pp: &page_pool backing page allocation, if available
+ *
+ * Copy all frags from XSk &xdp_buff to the skb to pass it up the stack.
+ * Allocate a new buffer for each frag, copy it and attach to the skb.
+ *
+ * Return: true on success, false on netmem allocation fail.
+ */
+static noinline bool xdp_copy_frags_from_zc(struct sk_buff *skb,
+					    const struct xdp_buff *xdp,
+					    struct page_pool *pp)
+{
+	struct skb_shared_info *sinfo = skb_shinfo(skb);
+	const struct skb_shared_info *xinfo;
+	u32 nr_frags, tsize = 0;
+	bool pfmemalloc = false;
+
+	xinfo = xdp_get_shared_info_from_buff(xdp);
+	nr_frags = xinfo->nr_frags;
+
+	for (u32 i = 0; i < nr_frags; i++) {
+		u32 len = skb_frag_size(&xinfo->frags[i]);
+		u32 offset, truesize = len;
+		netmem_ref netmem;
+
+		netmem = page_pool_dev_alloc_netmem(pp, &offset, &truesize);
+		if (unlikely(!netmem)) {
+			sinfo->nr_frags = i;
+			return false;
+		}
+
+		memcpy(__netmem_address(netmem),
+		       __netmem_address(xinfo->frags[i].netmem),
+		       LARGEST_ALIGN(len));
+		__skb_fill_netmem_desc_noacc(sinfo, i, netmem, offset, len);
+
+		tsize += truesize;
+		pfmemalloc |= netmem_is_pfmemalloc(netmem);
+	}
+
+	xdp_update_skb_shared_info(skb, nr_frags, xinfo->xdp_frags_size,
+				   tsize, pfmemalloc);
+
+	return true;
+}
+
+/**
+ * xdp_build_skb_from_zc - create an skb from XSk &xdp_buff
+ * @xdp: source XSk buff
+ *
+ * Similar to xdp_build_skb_from_buff(), but for XSk frames. Allocate an skb
+ * head, new buffer for the head, copy the data and initialize the skb fields.
+ * If there are frags, allocate new buffers for them and copy.
+ * Buffers are allocated from the system percpu pools to try recycling them.
+ * If new skb was built successfully, @xdp is returned to XSk pool's freelist.
+ * On error, it remains untouched and the caller must take care of this.
+ *
+ * Return: new &sk_buff on success, %NULL on error.
+ */
+struct sk_buff *xdp_build_skb_from_zc(struct xdp_buff *xdp)
+{
+	struct page_pool *pp = this_cpu_read(system_page_pool);
+	const struct xdp_rxq_info *rxq = xdp->rxq;
+	u32 len = xdp->data_end - xdp->data_meta;
+	u32 truesize = xdp->frame_sz;
+	struct sk_buff *skb;
+	int metalen;
+	void *data;
+
+	if (!IS_ENABLED(CONFIG_PAGE_POOL))
+		return NULL;
+
+	data = page_pool_dev_alloc_va(pp, &truesize);
+	if (unlikely(!data))
+		return NULL;
+
+	skb = napi_build_skb(data, truesize);
+	if (unlikely(!skb)) {
+		page_pool_free_va(pp, data, true);
+		return NULL;
+	}
+
+	skb_mark_for_recycle(skb);
+	skb_reserve(skb, xdp->data_meta - xdp->data_hard_start);
+
+	memcpy(__skb_put(skb, len), xdp->data_meta, LARGEST_ALIGN(len));
+
+	metalen = xdp->data - xdp->data_meta;
+	if (metalen > 0) {
+		skb_metadata_set(skb, metalen);
+		__skb_pull(skb, metalen);
+	}
+
+	skb_record_rx_queue(skb, rxq->queue_index);
+
+	if (unlikely(xdp_buff_has_frags(xdp)) &&
+	    unlikely(!xdp_copy_frags_from_zc(skb, xdp, pp))) {
+		napi_consume_skb(skb, true);
+		return NULL;
+	}
+
+	xsk_buff_free(xdp);
+
+	skb->protocol = eth_type_trans(skb, rxq->dev);
+
+	return skb;
+}
+EXPORT_SYMBOL_GPL(xdp_build_skb_from_zc);
+
 struct sk_buff *__xdp_build_skb_from_frame(struct xdp_frame *xdpf,
 					   struct sk_buff *skb,
 					   struct net_device *dev)