netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH net-next v5 0/4] virtio_net: rx enable premapped mode by default
@ 2024-05-11  3:14 Xuan Zhuo
  2024-05-11  3:14 ` [PATCH net-next v5 1/4] virtio_ring: enable premapped mode whatever use_dma_api Xuan Zhuo
                   ` (6 more replies)
  0 siblings, 7 replies; 20+ messages in thread
From: Xuan Zhuo @ 2024-05-11  3:14 UTC (permalink / raw)
  To: netdev
  Cc: Michael S. Tsirkin, Jason Wang, Xuan Zhuo, David S. Miller,
	Eric Dumazet, Jakub Kicinski, Paolo Abeni, virtualization

Actually, for the virtio drivers, we can enable premapped mode whatever
the value of use_dma_api. Because we provide the virtio dma apis.
So the driver can enable premapped mode unconditionally.

This patch set makes the big mode of virtio-net to support premapped mode.
And enable premapped mode for rx by default.

Based on the following points, we do not use page pool to manage these
    pages:

    1. virtio-net uses the DMA APIs wrapped by virtio core. Therefore,
       we can only prevent the page pool from performing DMA operations, and
       let the driver perform DMA operations on the allocated pages.
    2. But when the page pool releases the page, we have no chance to
       execute dma unmap.
    3. A solution to #2 is to execute dma unmap every time before putting
       the page back to the page pool. (This is actually a waste, we don't
       execute unmap so frequently.)
    4. But there is another problem, we still need to use page.dma_addr to
       save the dma address. Using page.dma_addr while using page pool is
       unsafe behavior.
    5. And we need space the chain the pages submitted once to virtio core.

    More:
        https://lore.kernel.org/all/CACGkMEu=Aok9z2imB_c5qVuujSh=vjj1kx12fy9N7hqyi+M5Ow@mail.gmail.com/

Why we do not use the page space to store the dma?
    http://lore.kernel.org/all/CACGkMEuyeJ9mMgYnnB42=hw6umNuo=agn7VBqBqYPd7GN=+39Q@mail.gmail.com

Please review.

v5: 1. Fix the comments from @Larysa Zaremba
        http://lore.kernel.org/all/20240508063718.69806-1-xuanzhuo@linux.alibaba.com

v4:
    1. For the conflict, switch to the net-next branch

v3:
    1. big mode still use the mode that virtio core does the dma map/unmap

v2:
    1. make gcc happy in page_chain_get_dma()
        http://lore.kernel.org/all/202404221325.SX5ChRGP-lkp@intel.com

v1:
    1. discussed for using page pool
    2. use dma sync to replace the unmap for the first page

Thanks.






Xuan Zhuo (4):
  virtio_ring: enable premapped mode whatever use_dma_api
  virtio_net: big mode skip the unmap check
  virtio_net: rx remove premapped failover code
  virtio_net: remove the misleading comment

 drivers/net/virtio_net.c     | 90 +++++++++++++++---------------------
 drivers/virtio/virtio_ring.c |  7 +--
 2 files changed, 38 insertions(+), 59 deletions(-)

--
2.32.0.3.g01195cf9f


^ permalink raw reply	[flat|nested] 20+ messages in thread
* [PATCH RFC 1/3] Revert "virtio_net: rx remove premapped failover code"
@ 2024-08-14  6:59 Michael S. Tsirkin
  2024-08-15 15:27 ` [PATCH RFC 0/3] Revert "virtio_net: rx enable premapped mode by default" Michael S. Tsirkin
  0 siblings, 1 reply; 20+ messages in thread
From: Michael S. Tsirkin @ 2024-08-14  6:59 UTC (permalink / raw)
  To: linux-kernel, netdev
  Cc: Xuan Zhuo, Jason Wang, David S. Miller, Eric Dumazet,
	Jakub Kicinski, Paolo Abeni, virtualization, Darren Kenny,
	Boris Ostrovsky, Si-Wei Liu, Eugenio Pérez

This reverts commit defd28aa5acb0fd7c15adc6bc40a8ac277d04dea.

leads to crashes with no ACCESS_PLATFORM when
sysctl net.core.high_order_alloc_disable=1

Reported-by: Si-Wei Liu <si-wei.liu@oracle.com>
Message-ID: <8b20cc28-45a9-4643-8e87-ba164a540c0a@oracle.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 drivers/net/virtio_net.c | 89 +++++++++++++++++++++++-----------------
 1 file changed, 52 insertions(+), 37 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index fd3d7e926022..4f7e686b8bf9 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -348,6 +348,9 @@ struct receive_queue {
 
 	/* Record the last dma info to free after new pages is allocated. */
 	struct virtnet_rq_dma *last_dma;
+
+	/* Do dma by self */
+	bool do_dma;
 };
 
 /* This structure can contain rss message with maximum settings for indirection table and keysize
@@ -848,7 +851,7 @@ static void *virtnet_rq_get_buf(struct receive_queue *rq, u32 *len, void **ctx)
 	void *buf;
 
 	buf = virtqueue_get_buf_ctx(rq->vq, len, ctx);
-	if (buf)
+	if (buf && rq->do_dma)
 		virtnet_rq_unmap(rq, buf, *len);
 
 	return buf;
@@ -861,6 +864,11 @@ static void virtnet_rq_init_one_sg(struct receive_queue *rq, void *buf, u32 len)
 	u32 offset;
 	void *head;
 
+	if (!rq->do_dma) {
+		sg_init_one(rq->sg, buf, len);
+		return;
+	}
+
 	head = page_address(rq->alloc_frag.page);
 
 	offset = buf - head;
@@ -886,42 +894,44 @@ static void *virtnet_rq_alloc(struct receive_queue *rq, u32 size, gfp_t gfp)
 
 	head = page_address(alloc_frag->page);
 
-	dma = head;
+	if (rq->do_dma) {
+		dma = head;
 
-	/* new pages */
-	if (!alloc_frag->offset) {
-		if (rq->last_dma) {
-			/* Now, the new page is allocated, the last dma
-			 * will not be used. So the dma can be unmapped
-			 * if the ref is 0.
+		/* new pages */
+		if (!alloc_frag->offset) {
+			if (rq->last_dma) {
+				/* Now, the new page is allocated, the last dma
+				 * will not be used. So the dma can be unmapped
+				 * if the ref is 0.
+				 */
+				virtnet_rq_unmap(rq, rq->last_dma, 0);
+				rq->last_dma = NULL;
+			}
+
+			dma->len = alloc_frag->size - sizeof(*dma);
+
+			addr = virtqueue_dma_map_single_attrs(rq->vq, dma + 1,
+							      dma->len, DMA_FROM_DEVICE, 0);
+			if (virtqueue_dma_mapping_error(rq->vq, addr))
+				return NULL;
+
+			dma->addr = addr;
+			dma->need_sync = virtqueue_dma_need_sync(rq->vq, addr);
+
+			/* Add a reference to dma to prevent the entire dma from
+			 * being released during error handling. This reference
+			 * will be freed after the pages are no longer used.
 			 */
-			virtnet_rq_unmap(rq, rq->last_dma, 0);
-			rq->last_dma = NULL;
+			get_page(alloc_frag->page);
+			dma->ref = 1;
+			alloc_frag->offset = sizeof(*dma);
+
+			rq->last_dma = dma;
 		}
 
-		dma->len = alloc_frag->size - sizeof(*dma);
-
-		addr = virtqueue_dma_map_single_attrs(rq->vq, dma + 1,
-						      dma->len, DMA_FROM_DEVICE, 0);
-		if (virtqueue_dma_mapping_error(rq->vq, addr))
-			return NULL;
-
-		dma->addr = addr;
-		dma->need_sync = virtqueue_dma_need_sync(rq->vq, addr);
-
-		/* Add a reference to dma to prevent the entire dma from
-		 * being released during error handling. This reference
-		 * will be freed after the pages are no longer used.
-		 */
-		get_page(alloc_frag->page);
-		dma->ref = 1;
-		alloc_frag->offset = sizeof(*dma);
-
-		rq->last_dma = dma;
+		++dma->ref;
 	}
 
-	++dma->ref;
-
 	buf = head + alloc_frag->offset;
 
 	get_page(alloc_frag->page);
@@ -938,9 +948,12 @@ static void virtnet_rq_set_premapped(struct virtnet_info *vi)
 	if (!vi->mergeable_rx_bufs && vi->big_packets)
 		return;
 
-	for (i = 0; i < vi->max_queue_pairs; i++)
-		/* error should never happen */
-		BUG_ON(virtqueue_set_dma_premapped(vi->rq[i].vq));
+	for (i = 0; i < vi->max_queue_pairs; i++) {
+		if (virtqueue_set_dma_premapped(vi->rq[i].vq))
+			continue;
+
+		vi->rq[i].do_dma = true;
+	}
 }
 
 static void virtnet_rq_unmap_free_buf(struct virtqueue *vq, void *buf)
@@ -2036,7 +2049,8 @@ static int add_recvbuf_small(struct virtnet_info *vi, struct receive_queue *rq,
 
 	err = virtqueue_add_inbuf_ctx(rq->vq, rq->sg, 1, buf, ctx, gfp);
 	if (err < 0) {
-		virtnet_rq_unmap(rq, buf, 0);
+		if (rq->do_dma)
+			virtnet_rq_unmap(rq, buf, 0);
 		put_page(virt_to_head_page(buf));
 	}
 
@@ -2150,7 +2164,8 @@ static int add_recvbuf_mergeable(struct virtnet_info *vi,
 	ctx = mergeable_len_to_ctx(len + room, headroom);
 	err = virtqueue_add_inbuf_ctx(rq->vq, rq->sg, 1, buf, ctx, gfp);
 	if (err < 0) {
-		virtnet_rq_unmap(rq, buf, 0);
+		if (rq->do_dma)
+			virtnet_rq_unmap(rq, buf, 0);
 		put_page(virt_to_head_page(buf));
 	}
 
@@ -5231,7 +5246,7 @@ static void free_receive_page_frags(struct virtnet_info *vi)
 	int i;
 	for (i = 0; i < vi->max_queue_pairs; i++)
 		if (vi->rq[i].alloc_frag.page) {
-			if (vi->rq[i].last_dma)
+			if (vi->rq[i].do_dma && vi->rq[i].last_dma)
 				virtnet_rq_unmap(&vi->rq[i], vi->rq[i].last_dma, 0);
 			put_page(vi->rq[i].alloc_frag.page);
 		}
-- 
MST


^ permalink raw reply related	[flat|nested] 20+ messages in thread
end of thread, other threads:[~2024-08-20  6:21 UTC | newest]

Thread overview: 20+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-05-11  3:14 [PATCH net-next v5 0/4] virtio_net: rx enable premapped mode by default Xuan Zhuo
2024-05-11  3:14 ` [PATCH net-next v5 1/4] virtio_ring: enable premapped mode whatever use_dma_api Xuan Zhuo
2024-08-13 19:28   ` Si-Wei Liu
2024-08-13 19:46     ` Michael S. Tsirkin
2024-08-14  3:39       ` Si-Wei Liu
2024-08-14  7:00         ` Michael S. Tsirkin
2024-08-17 13:20     ` Xuan Zhuo
2024-08-20  1:06       ` Si-Wei Liu
2024-08-20  6:19         ` Xuan Zhuo
2024-05-11  3:14 ` [PATCH net-next v5 2/4] virtio_net: big mode skip the unmap check Xuan Zhuo
2024-05-11  3:14 ` [PATCH net-next v5 3/4] virtio_net: rx remove premapped failover code Xuan Zhuo
2024-05-11  3:14 ` [PATCH net-next v5 4/4] virtio_net: remove the misleading comment Xuan Zhuo
2024-05-14  0:20 ` [PATCH net-next v5 0/4] virtio_net: rx enable premapped mode by default patchwork-bot+netdevbpf
2024-08-14  6:59 ` [PATCH RFC 0/3] Revert "virtio_net: rx enable premapped mode by default" Michael S. Tsirkin
2024-08-15  7:14 ` Linux regression tracking (Thorsten Leemhuis)
2024-08-15 10:22   ` Darren Kenny
2024-08-16  5:03     ` Linux regression tracking (Thorsten Leemhuis)
2024-08-15 15:23   ` Michael S. Tsirkin
2024-08-15 15:28     ` Michael S. Tsirkin
  -- strict thread matches above, loose matches on Subject: below --
2024-08-14  6:59 [PATCH RFC 1/3] Revert "virtio_net: rx remove premapped failover code" Michael S. Tsirkin
2024-08-15 15:27 ` [PATCH RFC 0/3] Revert "virtio_net: rx enable premapped mode by default" Michael S. Tsirkin

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).