From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 0E66F181B8F for ; Wed, 8 Jan 2025 11:35:27 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=170.10.129.124 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1736336130; cv=none; b=ayMSpiz5BhHuSE4MowZeSbhIcNhF5cQ2vyGps9YKlO5EkZ6npBQ8u6nunX1PRTi8snhIJLPbT/1ukQorY2nbmZXtmsDrKNIvqlaNd4Q4zJ/AO4z0DJjYB20QzgoeIgi2tEcBYIrhUxGMsf2m6mcIO1fGMphxtl9nKGjDnefxneg= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1736336130; c=relaxed/simple; bh=U0y7FwcwyD44HJYQrcVy1dEqp8zW7hw0F2XTaKo2RxE=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version: Content-Type:Content-Disposition:In-Reply-To; b=MerJTHpivFtgzFqaW9L2UemOZPAP806n35P40UvBkbwNxMXmK1FvZB1O7X35UnoSEEGL5xdtSQYIhGALuE3IYaY5yMevyg0ZrWHzdZEhKYFoEurf3aLygO6VtIMlSJWD5bzQvaDaayzrS+p31wXdNwbcsPo1EkYfc/4FD3+5FOY= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com; spf=pass smtp.mailfrom=redhat.com; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b=fsBKNE4g; arc=none smtp.client-ip=170.10.129.124 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="fsBKNE4g" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1736336127; 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: in-reply-to:in-reply-to:references:references; bh=1+XOF/Yu0g2fsX7t9xPysH0JaEinWV6AydJhgfHcaq0=; b=fsBKNE4gKP70JED+0QyOSCRH75aOsqtxpVKs6Wk6FXeleTG8NSe93Fvt5vu4mxXD/rFynI UXPny0aCpRafnKxpu8q7tcw+FTzYnGYtSa7tjUVm3PRcjqKYrwyJJu6ieIVzd0/uD8mbEK zsx/Di0dKfE7G75HTzwXztMcoNG7qGY= Received: from mail-wr1-f70.google.com (mail-wr1-f70.google.com [209.85.221.70]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-425-OVScFbodOFiTujwYoyhrQg-1; Wed, 08 Jan 2025 06:35:23 -0500 X-MC-Unique: OVScFbodOFiTujwYoyhrQg-1 X-Mimecast-MFC-AGG-ID: OVScFbodOFiTujwYoyhrQg Received: by mail-wr1-f70.google.com with SMTP id ffacd0b85a97d-3860bc1d4f1so8964013f8f.2 for ; Wed, 08 Jan 2025 03:35:23 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1736336122; x=1736940922; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=1+XOF/Yu0g2fsX7t9xPysH0JaEinWV6AydJhgfHcaq0=; b=G9SRVLnWlg4FDql8EvT2uKcjZDIHlM8lEy7BK7W1BfYlvuytEY1EbXEADGQgajaw54 aFWtt+oIzpWtNlckC9fAlFsogTcvrxvRxtDN6sDiSFZCRbY20kgE3Osg5wPs9Y7dxhI+ jbStHHmWJE+qmGipKr5sqWCiAfod2NKeWZOxAsUFwce4CcwOOazQh732YQxg3mgqO0kP 8noLTc+glfTrB8IAbP+36ksSQKvaiS1w0UAQucFc9zET56lGepU2LJNmdiNlQQbigC5f nMBI0JC2mYr3XrM29lwbOkxz+KG4E/tKERdKZTNE7QbOCbSQxJ2rGAMY1dwWDI0gDuI2 D/ww== X-Forwarded-Encrypted: i=1; AJvYcCUGEaKmYzyBW63PJXCm/0NrkHORNS9Cwj/tYKVC6nbreaVQ7QNlDtV4LNSRqds5BjBZHJKR0jLSiDDbxTk=@vger.kernel.org X-Gm-Message-State: AOJu0Yyml24rmTm/OX9E//atv8WAgNsjWv9LaZY3tO9wJGNKtg6oTP5r Yl/Wm3fjxiBSG4L/ErHgXT9qjJ2U/uDlfPalXS8DSVbA6uE4+pnwDCN7fS5Uu9LynTAofJR3mtH Egd7TNWZCeVp14Md0SWdq/9oTTqBoIeXTZkgkAaP6ZgHxvjC6KKMswTgKPjP67w== X-Gm-Gg: ASbGncvCz2BgTUY+3/idlZTeiKGfyjSINhw4jgKai5ECpFFRf2/sA/xmaYOaXWV7bbN lTaN1iTcXKuLsSH7fb6X5pE8wXiFrTPfAptXlDLb55GMuPxU76xrzopT4+fNeS93XGjmw7aGFLR MS05uWC0uNQAZLOdBwEUTkYB+3sGrXg0LhCmrtSANkt8KlxKAiBG/L8kWq4bPCTLsiSenv5rI6v JbPtzd2fK2c1gt01SpcGZU8F1B6XIrr3D93xhbZVuU3jyrYQrE= X-Received: by 2002:a5d:6da4:0:b0:386:3356:f3ac with SMTP id ffacd0b85a97d-38a8730cf2cmr2072561f8f.26.1736336122483; Wed, 08 Jan 2025 03:35:22 -0800 (PST) X-Google-Smtp-Source: AGHT+IGSxEwNeaWY0iq88jJcFtUsYY7LdG5hvMnzg7jF1DNAqRGfd3L2Ij4vGfuJoUWTe6dQ/ngDfw== X-Received: by 2002:a5d:6da4:0:b0:386:3356:f3ac with SMTP id ffacd0b85a97d-38a8730cf2cmr2072534f8f.26.1736336122037; Wed, 08 Jan 2025 03:35:22 -0800 (PST) Received: from redhat.com ([2a02:14f:175:d62d:93ef:d7e2:e7da:ed72]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-436e2dc12dasm17964855e9.10.2025.01.08.03.35.20 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 08 Jan 2025 03:35:21 -0800 (PST) Date: Wed, 8 Jan 2025 06:35:17 -0500 From: "Michael S. Tsirkin" To: Jason Wang Cc: xuanzhuo@linux.alibaba.com, eperezma@redhat.com, virtualization@lists.linux.dev, linux-kernel@vger.kernel.org Subject: Re: [PATCH] virtio_ring: skip cpu sync when mapping fails Message-ID: <20250108063310-mutt-send-email-mst@kernel.org> References: <20241111025538.2837-1-jasowang@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20241111025538.2837-1-jasowang@redhat.com> On Mon, Nov 11, 2024 at 10:55:38AM +0800, Jason Wang wrote: > There's no need to sync DMA for CPU on mapping errors. So this patch > skips the CPU sync in the error handling path of DMA mapping. > > Signed-off-by: Jason Wang So as I said, I do not get why we are optimizing error paths. The commit log at least needs to be improved to document the motivation. > --- > drivers/virtio/virtio_ring.c | 98 +++++++++++++++++++++--------------- > 1 file changed, 57 insertions(+), 41 deletions(-) > > diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c > index be7309b1e860..b422b5fb22db 100644 > --- a/drivers/virtio/virtio_ring.c > +++ b/drivers/virtio/virtio_ring.c > @@ -441,8 +441,10 @@ static void virtqueue_init(struct vring_virtqueue *vq, u32 num) > */ > > static void vring_unmap_one_split_indirect(const struct vring_virtqueue *vq, > - const struct vring_desc *desc) > + const struct vring_desc *desc, > + bool skip_sync) > { > + unsigned long attrs = skip_sync ? DMA_ATTR_SKIP_CPU_SYNC : 0; > u16 flags; If you really feel we must do it, just pass attrs directly so we do not get an extra branch. Also makes for a more readable code. > if (!vq->do_unmap) > @@ -450,16 +452,18 @@ static void vring_unmap_one_split_indirect(const struct vring_virtqueue *vq, > > flags = virtio16_to_cpu(vq->vq.vdev, desc->flags); > > - dma_unmap_page(vring_dma_dev(vq), > - virtio64_to_cpu(vq->vq.vdev, desc->addr), > - virtio32_to_cpu(vq->vq.vdev, desc->len), > - (flags & VRING_DESC_F_WRITE) ? > - DMA_FROM_DEVICE : DMA_TO_DEVICE); > + dma_unmap_page_attrs(vring_dma_dev(vq), > + virtio64_to_cpu(vq->vq.vdev, desc->addr), > + virtio32_to_cpu(vq->vq.vdev, desc->len), > + (flags & VRING_DESC_F_WRITE) ? > + DMA_FROM_DEVICE : DMA_TO_DEVICE, > + attrs); > } > > static unsigned int vring_unmap_one_split(const struct vring_virtqueue *vq, > - unsigned int i) > + unsigned int i, bool skip_sync) > { > + unsigned long attrs = skip_sync ? DMA_ATTR_SKIP_CPU_SYNC : 0; > struct vring_desc_extra *extra = vq->split.desc_extra; > u16 flags; > > @@ -469,20 +473,22 @@ static unsigned int vring_unmap_one_split(const struct vring_virtqueue *vq, > if (!vq->use_dma_api) > goto out; > > - dma_unmap_single(vring_dma_dev(vq), > - extra[i].addr, > - extra[i].len, > - (flags & VRING_DESC_F_WRITE) ? > - DMA_FROM_DEVICE : DMA_TO_DEVICE); > + dma_unmap_single_attrs(vring_dma_dev(vq), > + extra[i].addr, > + extra[i].len, > + (flags & VRING_DESC_F_WRITE) ? > + DMA_FROM_DEVICE : DMA_TO_DEVICE, > + attrs); > } else { > if (!vq->do_unmap) > goto out; > > - dma_unmap_page(vring_dma_dev(vq), > - extra[i].addr, > - extra[i].len, > - (flags & VRING_DESC_F_WRITE) ? > - DMA_FROM_DEVICE : DMA_TO_DEVICE); > + dma_unmap_page_attrs(vring_dma_dev(vq), > + extra[i].addr, > + extra[i].len, > + (flags & VRING_DESC_F_WRITE) ? > + DMA_FROM_DEVICE : DMA_TO_DEVICE, > + attrs); > } > > out: > @@ -717,10 +723,10 @@ static inline int virtqueue_add_split(struct virtqueue *_vq, > if (i == err_idx) > break; > if (indirect) { > - vring_unmap_one_split_indirect(vq, &desc[i]); > + vring_unmap_one_split_indirect(vq, &desc[i], true); > i = virtio16_to_cpu(_vq->vdev, desc[i].next); > } else > - i = vring_unmap_one_split(vq, i); > + i = vring_unmap_one_split(vq, i, true); > } > > free_indirect: > @@ -775,12 +781,12 @@ static void detach_buf_split(struct vring_virtqueue *vq, unsigned int head, > i = head; > > while (vq->split.vring.desc[i].flags & nextflag) { > - vring_unmap_one_split(vq, i); > + vring_unmap_one_split(vq, i, false); > i = vq->split.desc_extra[i].next; > vq->vq.num_free++; > } > > - vring_unmap_one_split(vq, i); > + vring_unmap_one_split(vq, i, false); > vq->split.desc_extra[i].next = vq->free_head; > vq->free_head = head; > > @@ -804,7 +810,8 @@ static void detach_buf_split(struct vring_virtqueue *vq, unsigned int head, > > if (vq->do_unmap) { > for (j = 0; j < len / sizeof(struct vring_desc); j++) > - vring_unmap_one_split_indirect(vq, &indir_desc[j]); > + vring_unmap_one_split_indirect(vq, > + &indir_desc[j], false); > } > > kfree(indir_desc); > @@ -1221,8 +1228,10 @@ static u16 packed_last_used(u16 last_used_idx) > } > > static void vring_unmap_extra_packed(const struct vring_virtqueue *vq, > - const struct vring_desc_extra *extra) > + const struct vring_desc_extra *extra, > + bool skip_sync) > { > + unsigned long attrs = skip_sync ? DMA_ATTR_SKIP_CPU_SYNC : 0; > u16 flags; > > flags = extra->flags; > @@ -1231,24 +1240,28 @@ static void vring_unmap_extra_packed(const struct vring_virtqueue *vq, > if (!vq->use_dma_api) > return; > > - dma_unmap_single(vring_dma_dev(vq), > - extra->addr, extra->len, > - (flags & VRING_DESC_F_WRITE) ? > - DMA_FROM_DEVICE : DMA_TO_DEVICE); > + dma_unmap_single_attrs(vring_dma_dev(vq), > + extra->addr, extra->len, > + (flags & VRING_DESC_F_WRITE) ? > + DMA_FROM_DEVICE : DMA_TO_DEVICE, > + attrs); > } else { > if (!vq->do_unmap) > return; > > - dma_unmap_page(vring_dma_dev(vq), > - extra->addr, extra->len, > - (flags & VRING_DESC_F_WRITE) ? > - DMA_FROM_DEVICE : DMA_TO_DEVICE); > + dma_unmap_page_attrs(vring_dma_dev(vq), > + extra->addr, extra->len, > + (flags & VRING_DESC_F_WRITE) ? > + DMA_FROM_DEVICE : DMA_TO_DEVICE, > + attrs); > } > } > > static void vring_unmap_desc_packed(const struct vring_virtqueue *vq, > - const struct vring_packed_desc *desc) > + const struct vring_packed_desc *desc, > + bool skip_sync) > { > + unsigned long attrs = skip_sync ? DMA_ATTR_SKIP_CPU_SYNC : 0; > u16 flags; > > if (!vq->do_unmap) > @@ -1256,11 +1269,12 @@ static void vring_unmap_desc_packed(const struct vring_virtqueue *vq, > > flags = le16_to_cpu(desc->flags); > > - dma_unmap_page(vring_dma_dev(vq), > - le64_to_cpu(desc->addr), > - le32_to_cpu(desc->len), > - (flags & VRING_DESC_F_WRITE) ? > - DMA_FROM_DEVICE : DMA_TO_DEVICE); > + dma_unmap_page_attrs(vring_dma_dev(vq), > + le64_to_cpu(desc->addr), > + le32_to_cpu(desc->len), > + (flags & VRING_DESC_F_WRITE) ? > + DMA_FROM_DEVICE : DMA_TO_DEVICE, > + attrs); > } > > static struct vring_packed_desc *alloc_indirect_packed(unsigned int total_sg, > @@ -1389,7 +1403,7 @@ static int virtqueue_add_indirect_packed(struct vring_virtqueue *vq, > err_idx = i; > > for (i = 0; i < err_idx; i++) > - vring_unmap_desc_packed(vq, &desc[i]); > + vring_unmap_desc_packed(vq, &desc[i], true); > > free_desc: > kfree(desc); > @@ -1539,7 +1553,8 @@ static inline int virtqueue_add_packed(struct virtqueue *_vq, > for (n = 0; n < total_sg; n++) { > if (i == err_idx) > break; > - vring_unmap_extra_packed(vq, &vq->packed.desc_extra[curr]); > + vring_unmap_extra_packed(vq, > + &vq->packed.desc_extra[curr], true); > curr = vq->packed.desc_extra[curr].next; > i++; > if (i >= vq->packed.vring.num) > @@ -1619,7 +1634,8 @@ static void detach_buf_packed(struct vring_virtqueue *vq, > curr = id; > for (i = 0; i < state->num; i++) { > vring_unmap_extra_packed(vq, > - &vq->packed.desc_extra[curr]); > + &vq->packed.desc_extra[curr], > + false); > curr = vq->packed.desc_extra[curr].next; > } > } > @@ -1636,7 +1652,7 @@ static void detach_buf_packed(struct vring_virtqueue *vq, > len = vq->packed.desc_extra[id].len; > for (i = 0; i < len / sizeof(struct vring_packed_desc); > i++) > - vring_unmap_desc_packed(vq, &desc[i]); > + vring_unmap_desc_packed(vq, &desc[i], false); > } > kfree(desc); > state->indir_desc = NULL; > -- > 2.31.1