* [dpdk-dev] [PATCH] vfio: use contiguous mapping for IOVA as VA mode
@ 2019-07-23 10:01 Anatoly Burakov
2019-07-23 10:30 ` Sirvys, Andrius
0 siblings, 1 reply; 3+ messages in thread
From: Anatoly Burakov @ 2019-07-23 10:01 UTC (permalink / raw)
To: dev; +Cc: andrius.sirvys, stable
When using IOVA as VA mode, there is no need to map segments
page by page. This normally isn't a problem, but it becomes one
when attempting to use DPDK in no-huge mode, where VFIO subsystem
simply runs out of space to store mappings.
Fix this for x86 by triggering different callbacks based on whether
IOVA as VA mode is enabled.
Fixes: 73a639085938 ("vfio: allow to map other memory regions")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
---
lib/librte_eal/linux/eal/eal_vfio.c | 20 ++++++++++++++++++++
1 file changed, 20 insertions(+)
diff --git a/lib/librte_eal/linux/eal/eal_vfio.c b/lib/librte_eal/linux/eal/eal_vfio.c
index ed04231b1..501c74f23 100644
--- a/lib/librte_eal/linux/eal/eal_vfio.c
+++ b/lib/librte_eal/linux/eal/eal_vfio.c
@@ -1231,6 +1231,19 @@ rte_vfio_get_group_num(const char *sysfs_base,
return 1;
}
+static int
+type1_map_contig(const struct rte_memseg_list *msl, const struct rte_memseg *ms,
+ size_t len, void *arg)
+{
+ int *vfio_container_fd = arg;
+
+ if (msl->external)
+ return 0;
+
+ return vfio_type1_dma_mem_map(*vfio_container_fd, ms->addr_64, ms->iova,
+ len, 1);
+}
+
static int
type1_map(const struct rte_memseg_list *msl, const struct rte_memseg *ms,
void *arg)
@@ -1300,6 +1313,13 @@ vfio_type1_dma_mem_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
static int
vfio_type1_dma_map(int vfio_container_fd)
{
+ if (rte_eal_iova_mode() == RTE_IOVA_VA) {
+ /* with IOVA as VA mode, we can get away with mapping contiguous
+ * chunks rather than going page-by-page.
+ */
+ return rte_memseg_contig_walk(type1_map_contig,
+ &vfio_container_fd);
+ }
return rte_memseg_walk(type1_map, &vfio_container_fd);
}
--
2.17.1
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [dpdk-dev] [PATCH] vfio: use contiguous mapping for IOVA as VA mode
2019-07-23 10:01 [dpdk-dev] [PATCH] vfio: use contiguous mapping for IOVA as VA mode Anatoly Burakov
@ 2019-07-23 10:30 ` Sirvys, Andrius
2019-07-23 18:48 ` Thomas Monjalon
0 siblings, 1 reply; 3+ messages in thread
From: Sirvys, Andrius @ 2019-07-23 10:30 UTC (permalink / raw)
To: Burakov, Anatoly, dev; +Cc: stable
Subject: [PATCH] vfio: use contiguous mapping for IOVA as VA mode
When using IOVA as VA mode, there is no need to map segments page by page. This normally isn't a problem, but it becomes one when attempting to use DPDK in no-huge mode, where VFIO subsystem simply runs out of space to store mappings.
Fix this for x86 by triggering different callbacks based on whether IOVA as VA mode is enabled.
Fixes: 73a639085938 ("vfio: allow to map other memory regions")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
---
lib/librte_eal/linux/eal/eal_vfio.c | 20 ++++++++++++++++++++
1 file changed, 20 insertions(+)
diff --git a/lib/librte_eal/linux/eal/eal_vfio.c b/lib/librte_eal/linux/eal/eal_vfio.c
index ed04231b1..501c74f23 100644
--- a/lib/librte_eal/linux/eal/eal_vfio.c
+++ b/lib/librte_eal/linux/eal/eal_vfio.c
@@ -1231,6 +1231,19 @@ rte_vfio_get_group_num(const char *sysfs_base,
return 1;
}
+static int
+type1_map_contig(const struct rte_memseg_list *msl, const struct rte_memseg *ms,
+ size_t len, void *arg)
+{
+ int *vfio_container_fd = arg;
+
+ if (msl->external)
+ return 0;
+
+ return vfio_type1_dma_mem_map(*vfio_container_fd, ms->addr_64, ms->iova,
+ len, 1);
+}
+
static int
type1_map(const struct rte_memseg_list *msl, const struct rte_memseg *ms,
void *arg)
@@ -1300,6 +1313,13 @@ vfio_type1_dma_mem_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova, static int vfio_type1_dma_map(int vfio_container_fd) {
+ if (rte_eal_iova_mode() == RTE_IOVA_VA) {
+ /* with IOVA as VA mode, we can get away with mapping contiguous
+ * chunks rather than going page-by-page.
+ */
+ return rte_memseg_contig_walk(type1_map_contig,
+ &vfio_container_fd);
+ }
return rte_memseg_walk(type1_map, &vfio_container_fd); }
Tested-by: Andrius Sirvys <andrius.sirvys@intel.com>
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [dpdk-dev] [PATCH] vfio: use contiguous mapping for IOVA as VA mode
2019-07-23 10:30 ` Sirvys, Andrius
@ 2019-07-23 18:48 ` Thomas Monjalon
0 siblings, 0 replies; 3+ messages in thread
From: Thomas Monjalon @ 2019-07-23 18:48 UTC (permalink / raw)
To: Burakov, Anatoly; +Cc: dev, Sirvys, Andrius, stable
23/07/2019 12:30, Sirvys, Andrius:
> When using IOVA as VA mode, there is no need to map segments page by page. This normally isn't a problem, but it becomes one when attempting to use DPDK in no-huge mode, where VFIO subsystem simply runs out of space to store mappings.
>
> Fix this for x86 by triggering different callbacks based on whether IOVA as VA mode is enabled.
>
> Fixes: 73a639085938 ("vfio: allow to map other memory regions")
> Cc: stable@dpdk.org
>
> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
>
> Tested-by: Andrius Sirvys <andrius.sirvys@intel.com>
Applied, thanks
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2019-07-23 18:48 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-07-23 10:01 [dpdk-dev] [PATCH] vfio: use contiguous mapping for IOVA as VA mode Anatoly Burakov
2019-07-23 10:30 ` Sirvys, Andrius
2019-07-23 18:48 ` Thomas Monjalon
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).