linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stefano Garzarella <sgarzare@redhat.com>
To: virtualization@lists.linux-foundation.org
Cc: Xie Yongji <xieyongji@bytedance.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Stefano Garzarella <sgarzare@redhat.com>,
	Laurent Vivier <lvivier@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	linux-kernel@vger.kernel.org, Max Gurtovoy <mgurtovoy@nvidia.com>,
	Jason Wang <jasowang@redhat.com>,
	kvm@vger.kernel.org
Subject: [PATCH RFC v2 01/10] vdpa_sim: use iova module to allocate IOVA addresses
Date: Thu, 28 Jan 2021 15:41:18 +0100	[thread overview]
Message-ID: <20210128144127.113245-2-sgarzare@redhat.com> (raw)
In-Reply-To: <20210128144127.113245-1-sgarzare@redhat.com>

The identical mapping used until now created issues when mapping
different virtual pages with the same physical address.
To solve this issue, we can use the iova module, to handle the IOVA
allocation.
For simplicity we use an IOVA allocator with byte granularity.

We add two new functions, vdpasim_map_range() and vdpasim_unmap_range(),
to handle the IOVA allocation and the registration into the IOMMU/IOTLB.
These functions are used by dma_map_ops callbacks.

Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
v2:
- used ULONG_MAX instead of ~0UL [Jason]
- fixed typos in comment and patch description [Jason]
---
 drivers/vdpa/vdpa_sim/vdpa_sim.h |   2 +
 drivers/vdpa/vdpa_sim/vdpa_sim.c | 108 +++++++++++++++++++------------
 drivers/vdpa/Kconfig             |   1 +
 3 files changed, 69 insertions(+), 42 deletions(-)

diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.h b/drivers/vdpa/vdpa_sim/vdpa_sim.h
index 6d75444f9948..cd58e888bcf3 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim.h
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim.h
@@ -6,6 +6,7 @@
 #ifndef _VDPA_SIM_H
 #define _VDPA_SIM_H
 
+#include <linux/iova.h>
 #include <linux/vringh.h>
 #include <linux/vdpa.h>
 #include <linux/virtio_byteorder.h>
@@ -57,6 +58,7 @@ struct vdpasim {
 	/* virtio config according to device type */
 	void *config;
 	struct vhost_iotlb *iommu;
+	struct iova_domain iova;
 	void *buffer;
 	u32 status;
 	u32 generation;
diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c
index d5942842432d..2183a833fcf4 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim.c
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c
@@ -17,6 +17,7 @@
 #include <linux/vringh.h>
 #include <linux/vdpa.h>
 #include <linux/vhost_iotlb.h>
+#include <linux/iova.h>
 
 #include "vdpa_sim.h"
 
@@ -128,30 +129,57 @@ static int dir_to_perm(enum dma_data_direction dir)
 	return perm;
 }
 
+static dma_addr_t vdpasim_map_range(struct vdpasim *vdpasim, phys_addr_t paddr,
+				    size_t size, unsigned int perm)
+{
+	struct iova *iova;
+	dma_addr_t dma_addr;
+	int ret;
+
+	/* We set the limit_pfn to the maximum (ULONG_MAX - 1) */
+	iova = alloc_iova(&vdpasim->iova, size, ULONG_MAX - 1, true);
+	if (!iova)
+		return DMA_MAPPING_ERROR;
+
+	dma_addr = iova_dma_addr(&vdpasim->iova, iova);
+
+	spin_lock(&vdpasim->iommu_lock);
+	ret = vhost_iotlb_add_range(vdpasim->iommu, (u64)dma_addr,
+				    (u64)dma_addr + size - 1, (u64)paddr, perm);
+	spin_unlock(&vdpasim->iommu_lock);
+
+	if (ret) {
+		__free_iova(&vdpasim->iova, iova);
+		return DMA_MAPPING_ERROR;
+	}
+
+	return dma_addr;
+}
+
+static void vdpasim_unmap_range(struct vdpasim *vdpasim, dma_addr_t dma_addr,
+				size_t size)
+{
+	spin_lock(&vdpasim->iommu_lock);
+	vhost_iotlb_del_range(vdpasim->iommu, (u64)dma_addr,
+			      (u64)dma_addr + size - 1);
+	spin_unlock(&vdpasim->iommu_lock);
+
+	free_iova(&vdpasim->iova, iova_pfn(&vdpasim->iova, dma_addr));
+}
+
 static dma_addr_t vdpasim_map_page(struct device *dev, struct page *page,
 				   unsigned long offset, size_t size,
 				   enum dma_data_direction dir,
 				   unsigned long attrs)
 {
 	struct vdpasim *vdpasim = dev_to_sim(dev);
-	struct vhost_iotlb *iommu = vdpasim->iommu;
-	u64 pa = (page_to_pfn(page) << PAGE_SHIFT) + offset;
-	int ret, perm = dir_to_perm(dir);
+	phys_addr_t paddr = page_to_phys(page) + offset;
+	int perm = dir_to_perm(dir);
 
 	if (perm < 0)
 		return DMA_MAPPING_ERROR;
 
-	/* For simplicity, use identical mapping to avoid e.g iova
-	 * allocator.
-	 */
-	spin_lock(&vdpasim->iommu_lock);
-	ret = vhost_iotlb_add_range(iommu, pa, pa + size - 1,
-				    pa, dir_to_perm(dir));
-	spin_unlock(&vdpasim->iommu_lock);
-	if (ret)
-		return DMA_MAPPING_ERROR;
-
-	return (dma_addr_t)(pa);
+	return vdpasim_map_range(vdpasim, paddr, size, perm);
 }
 
 static void vdpasim_unmap_page(struct device *dev, dma_addr_t dma_addr,
@@ -159,12 +187,8 @@ static void vdpasim_unmap_page(struct device *dev, dma_addr_t dma_addr,
 			       unsigned long attrs)
 {
 	struct vdpasim *vdpasim = dev_to_sim(dev);
-	struct vhost_iotlb *iommu = vdpasim->iommu;
 
-	spin_lock(&vdpasim->iommu_lock);
-	vhost_iotlb_del_range(iommu, (u64)dma_addr,
-			      (u64)dma_addr + size - 1);
-	spin_unlock(&vdpasim->iommu_lock);
+	vdpasim_unmap_range(vdpasim, dma_addr, size);
 }
 
 static void *vdpasim_alloc_coherent(struct device *dev, size_t size,
@@ -172,27 +196,22 @@ static void *vdpasim_alloc_coherent(struct device *dev, size_t size,
 				    unsigned long attrs)
 {
 	struct vdpasim *vdpasim = dev_to_sim(dev);
-	struct vhost_iotlb *iommu = vdpasim->iommu;
-	void *addr = kmalloc(size, flag);
-	int ret;
+	phys_addr_t paddr;
+	void *addr;
 
-	spin_lock(&vdpasim->iommu_lock);
+	addr = kmalloc(size, flag);
 	if (!addr) {
 		*dma_addr = DMA_MAPPING_ERROR;
-	} else {
-		u64 pa = virt_to_phys(addr);
-
-		ret = vhost_iotlb_add_range(iommu, (u64)pa,
-					    (u64)pa + size - 1,
-					    pa, VHOST_MAP_RW);
-		if (ret) {
-			*dma_addr = DMA_MAPPING_ERROR;
-			kfree(addr);
-			addr = NULL;
-		} else
-			*dma_addr = (dma_addr_t)pa;
+		return NULL;
+	}
+
+	paddr = virt_to_phys(addr);
+
+	*dma_addr = vdpasim_map_range(vdpasim, paddr, size, VHOST_MAP_RW);
+	if (*dma_addr == DMA_MAPPING_ERROR) {
+		kfree(addr);
+		return NULL;
 	}
-	spin_unlock(&vdpasim->iommu_lock);
 
 	return addr;
 }
@@ -202,14 +221,10 @@ static void vdpasim_free_coherent(struct device *dev, size_t size,
 				  unsigned long attrs)
 {
 	struct vdpasim *vdpasim = dev_to_sim(dev);
-	struct vhost_iotlb *iommu = vdpasim->iommu;
 
-	spin_lock(&vdpasim->iommu_lock);
-	vhost_iotlb_del_range(iommu, (u64)dma_addr,
-			      (u64)dma_addr + size - 1);
-	spin_unlock(&vdpasim->iommu_lock);
+	vdpasim_unmap_range(vdpasim, dma_addr, size);
 
-	kfree(phys_to_virt((uintptr_t)dma_addr));
+	kfree(vaddr);
 }
 
 static const struct dma_map_ops vdpasim_dma_ops = {
@@ -271,6 +286,13 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr)
 	for (i = 0; i < dev_attr->nvqs; i++)
 		vringh_set_iotlb(&vdpasim->vqs[i].vring, vdpasim->iommu);
 
+	ret = iova_cache_get();
+	if (ret)
+		goto err_iommu;
+
+	/* For simplicity we use an IOVA allocator with byte granularity */
+	init_iova_domain(&vdpasim->iova, 1, 0);
+
 	vdpasim->vdpa.dma_dev = dev;
 
 	return vdpasim;
@@ -541,6 +563,8 @@ static void vdpasim_free(struct vdpa_device *vdpa)
 	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
 
 	cancel_work_sync(&vdpasim->work);
+	put_iova_domain(&vdpasim->iova);
+	iova_cache_put();
 	kvfree(vdpasim->buffer);
 	if (vdpasim->iommu)
 		vhost_iotlb_free(vdpasim->iommu);
diff --git a/drivers/vdpa/Kconfig b/drivers/vdpa/Kconfig
index ffd1e098bfd2..21a23500f430 100644
--- a/drivers/vdpa/Kconfig
+++ b/drivers/vdpa/Kconfig
@@ -14,6 +14,7 @@ config VDPA_SIM
 	depends on RUNTIME_TESTING_MENU && HAS_DMA
 	select DMA_OPS
 	select VHOST_RING
+	select IOMMU_IOVA
 	help
 	  Enable this module to support vDPA device simulators. These devices
 	  are used for testing, prototyping and development of vDPA.
-- 
2.29.2


  reply	other threads:[~2021-01-28 14:52 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-28 14:41 [PATCH RFC v2 00/10] vdpa: add vdpa simulator for block device Stefano Garzarella
2021-01-28 14:41 ` Stefano Garzarella [this message]
2021-01-28 14:41 ` [PATCH RFC v2 02/10] vringh: add 'iotlb_lock' to synchronize iotlb accesses Stefano Garzarella
2021-01-29  7:43   ` Jason Wang
     [not found]     ` <20210129091850.gatf3ih3knw2p4l4@steredhat>
2021-02-01  6:31       ` Jason Wang
2021-01-28 14:41 ` [PATCH RFC v2 03/10] vringh: reset kiov 'consumed' field in __vringh_iov() Stefano Garzarella
2021-02-01  5:40   ` Jason Wang
2021-02-01 10:21     ` Stefano Garzarella
2021-02-02  3:26       ` Jason Wang
2021-01-28 14:41 ` [PATCH RFC v2 04/10] vringh: implement vringh_kiov_advance() Stefano Garzarella
2021-02-01  5:43   ` Jason Wang
2021-02-01 10:23     ` Stefano Garzarella
2021-01-28 14:41 ` [PATCH RFC v2 05/10] vringh: add vringh_kiov_length() helper Stefano Garzarella
2021-02-01  5:46   ` Jason Wang
2021-01-28 14:41 ` [PATCH RFC v2 06/10] vdpa_sim: cleanup kiovs in vdpasim_free() Stefano Garzarella
2021-02-01  5:47   ` Jason Wang
2021-01-28 14:41 ` [PATCH RFC v2 07/10] vdpa: Remove the restriction that only supports virtio-net devices Stefano Garzarella
2021-01-28 14:41 ` [PATCH RFC v2 08/10] vdpa: add vdpa simulator for block device Stefano Garzarella
2021-01-31 15:31   ` Max Gurtovoy
2021-02-01  8:29     ` Stefano Garzarella
2021-02-01  5:50   ` Jason Wang
2021-02-02  9:34   ` Stefan Hajnoczi
2021-02-02 15:49     ` Stefano Garzarella
2021-02-03 16:45       ` Stefan Hajnoczi
2021-02-04  8:02         ` Stefano Garzarella
2021-01-28 14:41 ` [PATCH RFC v2 09/10] vdpa_sim_blk: implement ramdisk behaviour Stefano Garzarella
2021-02-01  6:03   ` Jason Wang
2021-02-02  9:41   ` Stefan Hajnoczi
2021-01-28 14:41 ` [PATCH RFC v2 10/10] vdpa_sim_blk: handle VIRTIO_BLK_T_GET_ID Stefano Garzarella
2021-02-01  6:04   ` Jason Wang
2021-02-02  9:44   ` Stefan Hajnoczi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210128144127.113245-2-sgarzare@redhat.com \
    --to=sgarzare@redhat.com \
    --cc=jasowang@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lvivier@redhat.com \
    --cc=mgurtovoy@nvidia.com \
    --cc=mst@redhat.com \
    --cc=stefanha@redhat.com \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=xieyongji@bytedance.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).