From patchwork Tue Mar 21 15:42:24 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Stefano Garzarella X-Patchwork-Id: 13182898 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 99875C6FD1D for ; Tue, 21 Mar 2023 15:44:24 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231812AbjCUPoW (ORCPT ); Tue, 21 Mar 2023 11:44:22 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52888 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231791AbjCUPoN (ORCPT ); Tue, 21 Mar 2023 11:44:13 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 61D5D38471 for ; Tue, 21 Mar 2023 08:43:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1679413384; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=bIv3PP4yBpsiDfFQUDhlNlVTwLloVQU9kKPFwpUK4lo=; b=IQLHMY2fmubLNZJeovua8l85WNKwWjb2REixGSjQDnjZsBPdlNBcKDo/lgMNHmMSlmYPEK 0YyhIhpkhxq8N3YKkqgnkrGdCrFWHvzHzorGzTOmo+FXWD32ln0dRtYYozKjwsSixOXNqh 19bV9qQgBMAaZIf9B9dqy5DqHpdpnuA= Received: from mail-wm1-f69.google.com (mail-wm1-f69.google.com [209.85.128.69]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-523-xRx3WqMoMCi0CK75S2AddA-1; Tue, 21 Mar 2023 11:43:03 -0400 X-MC-Unique: xRx3WqMoMCi0CK75S2AddA-1 Received: by mail-wm1-f69.google.com with SMTP id iv18-20020a05600c549200b003ee21220fccso1633351wmb.1 for ; Tue, 21 Mar 2023 08:43:03 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1679413380; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=bIv3PP4yBpsiDfFQUDhlNlVTwLloVQU9kKPFwpUK4lo=; b=Ip09X8MnZ4+hmQa0R6bvrg8e0DCicOLdQSCW3PiEM91ED8ZT1yrtNilcyV+NzyHazi uJ2nTrO0YNa/VIKEV1KFlUInUzAPXlLOHjFTei6Sl3Y1b27f3d5LJSwiqHj8WYX7s+LJ rCxptod24P1E6CfF0Ra0amsHDSfdZpcPsodJd3j/leJ99A1Y4KlNEz17DDMQ5mYCk2lU Na9p1hTro/8Nfb9DqxOkzlrrZR2KEsOa6GDwLqSKbWOL3tU/oejnLjNvqgiMNiTKqZYp Pcye+vIq8Dssm0YEYXH7FpsvL8sMZ+NpSrnM37fFo8et/gVs7TBLen3xLDglYFuQ9Efo lh2A== X-Gm-Message-State: AO0yUKX4wLSRwfT8di3wSsXnQlhQmxdFpW53mIVTP5R5iA+BtsCuBcWi vvl0NIqRhiv6F5XmAl7MN+Jr5HjiozCgq9g38CBgR1oRipuTLXIhKXFvfbW1su5uwtaXXiddxXq /oVrXscrjEDSY9WBc X-Received: by 2002:a5d:4d0f:0:b0:2cd:e089:398d with SMTP id z15-20020a5d4d0f000000b002cde089398dmr2806132wrt.5.1679413380266; Tue, 21 Mar 2023 08:43:00 -0700 (PDT) X-Google-Smtp-Source: AK7set8cYR72hdvy3mHJEzDhnqtBs/xqBE/H6lo8y8NxTbvPjr87rHIO4A/V1Ti6cu3VeSVcB8D4+A== X-Received: by 2002:a5d:4d0f:0:b0:2cd:e089:398d with SMTP id z15-20020a5d4d0f000000b002cde089398dmr2806126wrt.5.1679413379955; Tue, 21 Mar 2023 08:42:59 -0700 (PDT) Received: from step1.redhat.com (host-82-57-51-170.retail.telecomitalia.it. [82.57.51.170]) by smtp.gmail.com with ESMTPSA id p5-20020adfce05000000b002d64fcb362dsm6340125wrn.111.2023.03.21.08.42.58 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 21 Mar 2023 08:42:59 -0700 (PDT) From: Stefano Garzarella To: virtualization@lists.linux-foundation.org Cc: stefanha@redhat.com, "Michael S. Tsirkin" , Andrey Zhadchenko , eperezma@redhat.com, netdev@vger.kernel.org, Jason Wang , kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Stefano Garzarella Subject: [PATCH v3 4/8] vringh: support VA with iotlb Date: Tue, 21 Mar 2023 16:42:24 +0100 Message-Id: <20230321154228.182769-5-sgarzare@redhat.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20230321154228.182769-1-sgarzare@redhat.com> References: <20230321154228.182769-1-sgarzare@redhat.com> MIME-Version: 1.0 Content-type: text/plain Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org vDPA supports the possibility to use user VA in the iotlb messages. So, let's add support for user VA in vringh to use it in the vDPA simulators. Signed-off-by: Stefano Garzarella Acked-by: Eugenio Pérez --- Notes: v3: - refactored avoiding code duplication [Eugenio] v2: - replace kmap_atomic() with kmap_local_page() [see previous patch] - fix cast warnings when build with W=1 C=1 include/linux/vringh.h | 5 +- drivers/vdpa/mlx5/net/mlx5_vnet.c | 2 +- drivers/vdpa/vdpa_sim/vdpa_sim.c | 4 +- drivers/vhost/vringh.c | 153 +++++++++++++++++++++++------- 4 files changed, 127 insertions(+), 37 deletions(-) diff --git a/include/linux/vringh.h b/include/linux/vringh.h index 1991a02c6431..d39b9f2dcba0 100644 --- a/include/linux/vringh.h +++ b/include/linux/vringh.h @@ -32,6 +32,9 @@ struct vringh { /* Can we get away with weak barriers? */ bool weak_barriers; + /* Use user's VA */ + bool use_va; + /* Last available index we saw (ie. where we're up to). */ u16 last_avail_idx; @@ -279,7 +282,7 @@ void vringh_set_iotlb(struct vringh *vrh, struct vhost_iotlb *iotlb, spinlock_t *iotlb_lock); int vringh_init_iotlb(struct vringh *vrh, u64 features, - unsigned int num, bool weak_barriers, + unsigned int num, bool weak_barriers, bool use_va, struct vring_desc *desc, struct vring_avail *avail, struct vring_used *used); diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c index 520646ae7fa0..dfd0e000217b 100644 --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c @@ -2537,7 +2537,7 @@ static int setup_cvq_vring(struct mlx5_vdpa_dev *mvdev) if (mvdev->actual_features & BIT_ULL(VIRTIO_NET_F_CTRL_VQ)) err = vringh_init_iotlb(&cvq->vring, mvdev->actual_features, - MLX5_CVQ_MAX_ENT, false, + MLX5_CVQ_MAX_ENT, false, false, (struct vring_desc *)(uintptr_t)cvq->desc_addr, (struct vring_avail *)(uintptr_t)cvq->driver_addr, (struct vring_used *)(uintptr_t)cvq->device_addr); diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index eea23c630f7c..47cdf2a1f5b8 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -60,7 +60,7 @@ static void vdpasim_queue_ready(struct vdpasim *vdpasim, unsigned int idx) struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx]; uint16_t last_avail_idx = vq->vring.last_avail_idx; - vringh_init_iotlb(&vq->vring, vdpasim->features, vq->num, true, + vringh_init_iotlb(&vq->vring, vdpasim->features, vq->num, true, false, (struct vring_desc *)(uintptr_t)vq->desc_addr, (struct vring_avail *) (uintptr_t)vq->driver_addr, @@ -92,7 +92,7 @@ static void vdpasim_vq_reset(struct vdpasim *vdpasim, vq->cb = NULL; vq->private = NULL; vringh_init_iotlb(&vq->vring, vdpasim->dev_attr.supported_features, - VDPASIM_QUEUE_MAX, false, NULL, NULL, NULL); + VDPASIM_QUEUE_MAX, false, false, NULL, NULL, NULL); vq->vring.notify = NULL; } diff --git a/drivers/vhost/vringh.c b/drivers/vhost/vringh.c index 0ba3ef809e48..72c88519329a 100644 --- a/drivers/vhost/vringh.c +++ b/drivers/vhost/vringh.c @@ -1094,10 +1094,18 @@ EXPORT_SYMBOL(vringh_need_notify_kern); #if IS_REACHABLE(CONFIG_VHOST_IOTLB) +struct iotlb_vec { + union { + struct iovec *iovec; + struct bio_vec *bvec; + } iov; + size_t count; + bool is_iovec; +}; + static int iotlb_translate(const struct vringh *vrh, u64 addr, u64 len, u64 *translated, - struct bio_vec iov[], - int iov_size, u32 perm) + struct iotlb_vec *ivec, u32 perm) { struct vhost_iotlb_map *map; struct vhost_iotlb *iotlb = vrh->iotlb; @@ -1107,9 +1115,9 @@ static int iotlb_translate(const struct vringh *vrh, spin_lock(vrh->iotlb_lock); while (len > s) { - u64 size, pa, pfn; + u64 size; - if (unlikely(ret >= iov_size)) { + if (unlikely(ret >= ivec->count)) { ret = -ENOBUFS; break; } @@ -1124,10 +1132,22 @@ static int iotlb_translate(const struct vringh *vrh, } size = map->size - addr + map->start; - pa = map->addr + addr - map->start; - pfn = pa >> PAGE_SHIFT; - bvec_set_page(&iov[ret], pfn_to_page(pfn), min(len - s, size), - pa & (PAGE_SIZE - 1)); + if (ivec->is_iovec) { + struct iovec *iovec = ivec->iov.iovec; + + iovec[ret].iov_len = min(len - s, size); + iovec[ret].iov_base = (void __user *)(unsigned long) + (map->addr + addr - map->start); + } else { + u64 pa = map->addr + addr - map->start; + u64 pfn = pa >> PAGE_SHIFT; + struct bio_vec *bvec = ivec->iov.bvec; + + bvec_set_page(&bvec[ret], pfn_to_page(pfn), + min(len - s, size), + pa & (PAGE_SIZE - 1)); + } + s += size; addr += size; ++ret; @@ -1141,26 +1161,42 @@ static int iotlb_translate(const struct vringh *vrh, return ret; } +#define IOTLB_IOV_SIZE 16 + static inline int copy_from_iotlb(const struct vringh *vrh, void *dst, void *src, size_t len) { + struct iotlb_vec ivec; + union { + struct iovec iovec[IOTLB_IOV_SIZE]; + struct bio_vec bvec[IOTLB_IOV_SIZE]; + } iov; u64 total_translated = 0; + ivec.iov.iovec = iov.iovec; + ivec.count = IOTLB_IOV_SIZE; + ivec.is_iovec = vrh->use_va; + while (total_translated < len) { - struct bio_vec iov[16]; struct iov_iter iter; u64 translated; int ret; ret = iotlb_translate(vrh, (u64)(uintptr_t)src, len - total_translated, &translated, - iov, ARRAY_SIZE(iov), VHOST_MAP_RO); + &ivec, VHOST_MAP_RO); if (ret == -ENOBUFS) - ret = ARRAY_SIZE(iov); + ret = IOTLB_IOV_SIZE; else if (ret < 0) return ret; - iov_iter_bvec(&iter, ITER_SOURCE, iov, ret, translated); + if (ivec.is_iovec) { + iov_iter_init(&iter, ITER_SOURCE, ivec.iov.iovec, ret, + translated); + } else { + iov_iter_bvec(&iter, ITER_SOURCE, ivec.iov.bvec, ret, + translated); + } ret = copy_from_iter(dst, translated, &iter); if (ret < 0) @@ -1177,23 +1213,37 @@ static inline int copy_from_iotlb(const struct vringh *vrh, void *dst, static inline int copy_to_iotlb(const struct vringh *vrh, void *dst, void *src, size_t len) { + struct iotlb_vec ivec; + union { + struct iovec iovec[IOTLB_IOV_SIZE]; + struct bio_vec bvec[IOTLB_IOV_SIZE]; + } iov; u64 total_translated = 0; + ivec.iov.iovec = iov.iovec; + ivec.count = IOTLB_IOV_SIZE; + ivec.is_iovec = vrh->use_va; + while (total_translated < len) { - struct bio_vec iov[16]; struct iov_iter iter; u64 translated; int ret; ret = iotlb_translate(vrh, (u64)(uintptr_t)dst, len - total_translated, &translated, - iov, ARRAY_SIZE(iov), VHOST_MAP_WO); + &ivec, VHOST_MAP_WO); if (ret == -ENOBUFS) - ret = ARRAY_SIZE(iov); + ret = IOTLB_IOV_SIZE; else if (ret < 0) return ret; - iov_iter_bvec(&iter, ITER_DEST, iov, ret, translated); + if (ivec.is_iovec) { + iov_iter_init(&iter, ITER_DEST, ivec.iov.iovec, ret, + translated); + } else { + iov_iter_bvec(&iter, ITER_DEST, ivec.iov.bvec, ret, + translated); + } ret = copy_to_iter(src, translated, &iter); if (ret < 0) @@ -1210,20 +1260,37 @@ static inline int copy_to_iotlb(const struct vringh *vrh, void *dst, static inline int getu16_iotlb(const struct vringh *vrh, u16 *val, const __virtio16 *p) { - struct bio_vec iov; - void *kaddr, *from; + struct iotlb_vec ivec; + union { + struct iovec iovec[1]; + struct bio_vec bvec[1]; + } iov; + __virtio16 tmp; int ret; + ivec.iov.iovec = iov.iovec; + ivec.count = 1; + ivec.is_iovec = vrh->use_va; + /* Atomic read is needed for getu16 */ - ret = iotlb_translate(vrh, (u64)(uintptr_t)p, sizeof(*p), NULL, - &iov, 1, VHOST_MAP_RO); + ret = iotlb_translate(vrh, (u64)(uintptr_t)p, sizeof(*p), + NULL, &ivec, VHOST_MAP_RO); if (ret < 0) return ret; - kaddr = kmap_local_page(iov.bv_page); - from = kaddr + iov.bv_offset; - *val = vringh16_to_cpu(vrh, READ_ONCE(*(__virtio16 *)from)); - kunmap_local(kaddr); + if (ivec.is_iovec) { + ret = __get_user(tmp, (__virtio16 __user *)ivec.iov.iovec[0].iov_base); + if (ret) + return ret; + } else { + void *kaddr = kmap_local_page(ivec.iov.bvec[0].bv_page); + void *from = kaddr + ivec.iov.bvec[0].bv_offset; + + tmp = READ_ONCE(*(__virtio16 *)from); + kunmap_local(kaddr); + } + + *val = vringh16_to_cpu(vrh, tmp); return 0; } @@ -1231,20 +1298,37 @@ static inline int getu16_iotlb(const struct vringh *vrh, static inline int putu16_iotlb(const struct vringh *vrh, __virtio16 *p, u16 val) { - struct bio_vec iov; - void *kaddr, *to; + struct iotlb_vec ivec; + union { + struct iovec iovec; + struct bio_vec bvec; + } iov; + __virtio16 tmp; int ret; + ivec.iov.iovec = &iov.iovec; + ivec.count = 1; + ivec.is_iovec = vrh->use_va; + /* Atomic write is needed for putu16 */ - ret = iotlb_translate(vrh, (u64)(uintptr_t)p, sizeof(*p), NULL, - &iov, 1, VHOST_MAP_WO); + ret = iotlb_translate(vrh, (u64)(uintptr_t)p, sizeof(*p), + NULL, &ivec, VHOST_MAP_RO); if (ret < 0) return ret; - kaddr = kmap_local_page(iov.bv_page); - to = kaddr + iov.bv_offset; - WRITE_ONCE(*(__virtio16 *)to, cpu_to_vringh16(vrh, val)); - kunmap_local(kaddr); + tmp = cpu_to_vringh16(vrh, val); + + if (ivec.is_iovec) { + ret = __put_user(tmp, (__virtio16 __user *)ivec.iov.iovec[0].iov_base); + if (ret) + return ret; + } else { + void *kaddr = kmap_local_page(ivec.iov.bvec[0].bv_page); + void *to = kaddr + ivec.iov.bvec[0].bv_offset; + + WRITE_ONCE(*(__virtio16 *)to, tmp); + kunmap_local(kaddr); + } return 0; } @@ -1306,6 +1390,7 @@ static inline int putused_iotlb(const struct vringh *vrh, * @features: the feature bits for this ring. * @num: the number of elements. * @weak_barriers: true if we only need memory barriers, not I/O. + * @use_va: true if IOTLB contains user VA * @desc: the userpace descriptor pointer. * @avail: the userpace avail pointer. * @used: the userpace used pointer. @@ -1313,11 +1398,13 @@ static inline int putused_iotlb(const struct vringh *vrh, * Returns an error if num is invalid. */ int vringh_init_iotlb(struct vringh *vrh, u64 features, - unsigned int num, bool weak_barriers, + unsigned int num, bool weak_barriers, bool use_va, struct vring_desc *desc, struct vring_avail *avail, struct vring_used *used) { + vrh->use_va = use_va; + return vringh_init_kern(vrh, features, num, weak_barriers, desc, avail, used); }