From patchwork Tue Apr 25 07:36:06 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Xuan Zhuo X-Patchwork-Id: 13222924 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id ACA58C77B7F for ; Tue, 25 Apr 2023 07:37:20 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233312AbjDYHhS (ORCPT ); Tue, 25 Apr 2023 03:37:18 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45000 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233587AbjDYHgg (ORCPT ); Tue, 25 Apr 2023 03:36:36 -0400 Received: from out30-110.freemail.mail.aliyun.com (out30-110.freemail.mail.aliyun.com [115.124.30.110]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 667EFC152; Tue, 25 Apr 2023 00:36:23 -0700 (PDT) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R171e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018045176;MF=xuanzhuo@linux.alibaba.com;NM=1;PH=DS;RN=14;SR=0;TI=SMTPD_---0VgzCnj2_1682408178; Received: from localhost(mailfrom:xuanzhuo@linux.alibaba.com fp:SMTPD_---0VgzCnj2_1682408178) by smtp.aliyun-inc.com; Tue, 25 Apr 2023 15:36:19 +0800 From: Xuan Zhuo To: virtualization@lists.linux-foundation.org Cc: "Michael S. Tsirkin" , Jason Wang , Xuan Zhuo , Alexei Starovoitov , Daniel Borkmann , "David S. Miller" , Jesper Dangaard Brouer , John Fastabend , netdev@vger.kernel.org, bpf@vger.kernel.org, Christoph Hellwig , Jakub Kicinski , Alexander Lobakin Subject: [PATCH vhost v7 04/11] virtio_ring: split: support premapped Date: Tue, 25 Apr 2023 15:36:06 +0800 Message-Id: <20230425073613.8839-5-xuanzhuo@linux.alibaba.com> X-Mailer: git-send-email 2.32.0.3.g01195cf9f In-Reply-To: <20230425073613.8839-1-xuanzhuo@linux.alibaba.com> References: <20230425073613.8839-1-xuanzhuo@linux.alibaba.com> MIME-Version: 1.0 X-Git-Hash: c518d34a1cf7 Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org virtio core only supports virtual addresses, dma is completed in virtio core. In some scenarios (such as the AF_XDP), the memory is allocated and DMA mapping is completed in advance, so it is necessary for us to support passing the DMA address to virtio core. Drives can use sg->dma_address to pass the mapped dma address to virtio core. If one sg->dma_address is used then all sgs must use sg->dma_address, otherwise all must be null when passing it to the APIs of virtio. Signed-off-by: Xuan Zhuo --- drivers/virtio/virtio_ring.c | 31 +++++++++++++++++++++++-------- 1 file changed, 23 insertions(+), 8 deletions(-) diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 62db16a15ae7..055c1c069fff 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -67,9 +67,13 @@ #define LAST_ADD_TIME_INVALID(vq) #endif +#define VRING_STATE_F_MAP_INTERNAL BIT(0) + struct vring_desc_state_split { void *data; /* Data for callback. */ struct vring_desc *indir_desc; /* Indirect descriptor, if any. */ + u32 flags; /* State flags. */ + u32 padding; }; struct vring_desc_state_packed { @@ -448,7 +452,7 @@ static void vring_unmap_one_split_indirect(const struct vring_virtqueue *vq, } static unsigned int vring_unmap_one_split(const struct vring_virtqueue *vq, - unsigned int i) + unsigned int i, bool dma_map_internal) { struct vring_desc_extra *extra = vq->split.desc_extra; u16 flags; @@ -465,6 +469,9 @@ static unsigned int vring_unmap_one_split(const struct vring_virtqueue *vq, (flags & VRING_DESC_F_WRITE) ? DMA_FROM_DEVICE : DMA_TO_DEVICE); } else { + if (!dma_map_internal) + goto out; + dma_unmap_page(vring_dma_dev(vq), extra[i].addr, extra[i].len, @@ -615,7 +622,7 @@ static inline int virtqueue_add_split(struct virtqueue *_vq, struct scatterlist *sg; struct vring_desc *desc; unsigned int i, n, avail, descs_used, prev; - bool indirect; + bool indirect, dma_map_internal; int head; START_USE(vq); @@ -668,7 +675,8 @@ static inline int virtqueue_add_split(struct virtqueue *_vq, return -ENOSPC; } - if (virtqueue_map_sgs(vq, sgs, total_sg, out_sgs, in_sgs)) + dma_map_internal = !sgs[0]->dma_address; + if (dma_map_internal && virtqueue_map_sgs(vq, sgs, total_sg, out_sgs, in_sgs)) goto err_map; for (n = 0; n < out_sgs; n++) { @@ -735,6 +743,8 @@ static inline int virtqueue_add_split(struct virtqueue *_vq, else vq->split.desc_state[head].indir_desc = ctx; + vq->split.desc_state[head].flags = dma_map_internal ? VRING_STATE_F_MAP_INTERNAL : 0; + /* Put entry in available array (but don't update avail->idx until they * do sync). */ avail = vq->split.avail_idx_shadow & (vq->split.vring.num - 1); @@ -759,7 +769,8 @@ static inline int virtqueue_add_split(struct virtqueue *_vq, return 0; unmap_release: - virtqueue_unmap_sgs(vq, sgs, total_sg, out_sgs, in_sgs); + if (dma_map_internal) + virtqueue_unmap_sgs(vq, sgs, total_sg, out_sgs, in_sgs); err_map: if (indirect) @@ -805,20 +816,22 @@ static void detach_buf_split(struct vring_virtqueue *vq, unsigned int head, { unsigned int i, j; __virtio16 nextflag = cpu_to_virtio16(vq->vq.vdev, VRING_DESC_F_NEXT); + bool dma_map_internal; /* Clear data ptr. */ vq->split.desc_state[head].data = NULL; + dma_map_internal = !!(vq->split.desc_state[head].flags & VRING_STATE_F_MAP_INTERNAL); /* Put back on free list: unmap first-level descriptors and find end */ i = head; while (vq->split.vring.desc[i].flags & nextflag) { - vring_unmap_one_split(vq, i); + vring_unmap_one_split(vq, i, dma_map_internal); i = vq->split.desc_extra[i].next; vq->vq.num_free++; } - vring_unmap_one_split(vq, i); + vring_unmap_one_split(vq, i, dma_map_internal); vq->split.desc_extra[i].next = vq->free_head; vq->free_head = head; @@ -840,8 +853,10 @@ static void detach_buf_split(struct vring_virtqueue *vq, unsigned int head, VRING_DESC_F_INDIRECT)); BUG_ON(len == 0 || len % sizeof(struct vring_desc)); - for (j = 0; j < len / sizeof(struct vring_desc); j++) - vring_unmap_one_split_indirect(vq, &indir_desc[j]); + if (dma_map_internal) { + for (j = 0; j < len / sizeof(struct vring_desc); j++) + vring_unmap_one_split_indirect(vq, &indir_desc[j]); + } kfree(indir_desc); vq->split.desc_state[head].indir_desc = NULL;