From: Takeshi Yoshimura <tyos@jp.ibm.com>
To: dev@dpdk.org
Cc: stable@dpdk.org, Thomas Monjalon <thomas@monjalon.net>,
anatoly.burakov@intel.com, chaozhu@linux.vnet.ibm.com,
Takeshi Yoshimura <t.yoshimura8869@gmail.com>,
Takeshi Yoshimura <tyos@jp.ibm.com>
Subject: [dpdk-dev] [PATCH v2] eal/vfio: fix sPAPR IOMMU mapping
Date: Tue, 7 Aug 2018 11:35:06 +0900 [thread overview]
Message-ID: <20180807023506.1714-1-tyos@jp.ibm.com> (raw)
In-Reply-To: <20180807023211.1627-1-tyos@jp.ibm.com>
Commit 73a639085938 ("vfio: allow to map other memory regions")
introduced a bug in sPAPR IOMMU mapping. The commit removed necessary
ioctl with VFIO_IOMMU_SPAPR_REGISTER_MEMORY. Also, vfio_spapr_map_walk
should call vfio_spapr_dma_do_map instead of vfio_spapr_dma_mem_map.
Fixes: 73a639085938 ("vfio: allow to map other memory regions")
Cc: stable@dpdk.org
Signed-off-by: Takeshi Yoshimura <tyos@jp.ibm.com>
---
v2: Added Cc in message
lib/librte_eal/linuxapp/eal/eal_vfio.c | 23 +++++++++++++++--------
1 file changed, 15 insertions(+), 8 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_vfio.c b/lib/librte_eal/linuxapp/eal/eal_vfio.c
index c68dc38e0..68e862946 100644
--- a/lib/librte_eal/linuxapp/eal/eal_vfio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_vfio.c
@@ -1145,8 +1145,22 @@ vfio_spapr_dma_do_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
struct vfio_iommu_type1_dma_map dma_map;
struct vfio_iommu_type1_dma_unmap dma_unmap;
int ret;
+ struct vfio_iommu_spapr_register_memory reg = {
+ .argsz = sizeof(reg),
+ .flags = 0
+ };
+ reg.vaddr = (uintptr_t) vaddr;
+ reg.size = len;
if (do_map != 0) {
+ ret = ioctl(vfio_container_fd,
+ VFIO_IOMMU_SPAPR_REGISTER_MEMORY, ®);
+ if (ret) {
+ RTE_LOG(ERR, EAL, " cannot register vaddr for IOMMU, "
+ "error %i (%s)\n", errno, strerror(errno));
+ return -1;
+ }
+
memset(&dma_map, 0, sizeof(dma_map));
dma_map.argsz = sizeof(struct vfio_iommu_type1_dma_map);
dma_map.vaddr = vaddr;
@@ -1163,13 +1177,6 @@ vfio_spapr_dma_do_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
}
} else {
- struct vfio_iommu_spapr_register_memory reg = {
- .argsz = sizeof(reg),
- .flags = 0
- };
- reg.vaddr = (uintptr_t) vaddr;
- reg.size = len;
-
ret = ioctl(vfio_container_fd,
VFIO_IOMMU_SPAPR_UNREGISTER_MEMORY, ®);
if (ret) {
@@ -1201,7 +1208,7 @@ vfio_spapr_map_walk(const struct rte_memseg_list *msl __rte_unused,
{
int *vfio_container_fd = arg;
- return vfio_spapr_dma_mem_map(*vfio_container_fd, ms->addr_64, ms->iova,
+ return vfio_spapr_dma_do_map(*vfio_container_fd, ms->addr_64, ms->iova,
ms->len, 1);
}
--
2.15.1
next prev parent reply other threads:[~2018-08-07 2:35 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-06 8:43 [dpdk-dev] [PATCH] " Takeshi Yoshimura
2018-08-06 9:50 ` Thomas Monjalon
2018-08-07 8:39 ` Burakov, Anatoly
2018-08-07 2:28 ` Takeshi T Yoshimura
2018-08-07 2:32 ` Takeshi Yoshimura
2018-08-07 2:35 ` Takeshi Yoshimura [this message]
2018-10-28 21:21 ` [dpdk-dev] [PATCH v2] " Thomas Monjalon
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=20180807023506.1714-1-tyos@jp.ibm.com \
--to=tyos@jp.ibm.com \
--cc=anatoly.burakov@intel.com \
--cc=chaozhu@linux.vnet.ibm.com \
--cc=dev@dpdk.org \
--cc=stable@dpdk.org \
--cc=t.yoshimura8869@gmail.com \
--cc=thomas@monjalon.net \
/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).