[ovs-dev,v3] dpdk: add support for v2.1.0
diff mbox

Message ID 1441370157-20852-1-git-send-email-timox.puha@intel.com
State Accepted
Headers show

Commit Message

Timo Puha Sept. 4, 2015, 12:35 p.m. UTC
Update relevant artifacts to add support for DPDK v2.1.0
 - INSTALL.DPDK.md
 - acinclude.m4: Change DPDK library name
 - netdev-dpdk: Limit minimum mbuf size to to adapt to DPDK bug fix that
   changes the treatment of the requested mbuf size
 - build.sh: Change DPDK version number

Note that this breaks compatibility with DPDK v2.0.0 although only
for the library name change.

Note that throughput for vhost ports with mergeable buffers is reduced
about 10% due to a necessary bug fix in DPDK vhost code.

Signed-off-by: Mark Kavanagh <mark.b.kavanagh@intel.com>
Signed-off-by: Michal Weglicki <michalx.weglicki@intel.com>
Signed-off-by: Timo Puha <timox.puha@intel.com>
---
 .travis/build.sh  |  2 +-
 INSTALL.DPDK.md   | 17 ++++++-----------
 acinclude.m4      |  2 +-
 lib/netdev-dpdk.c | 12 +++++++++---
 4 files changed, 17 insertions(+), 16 deletions(-)

Comments

Timo Puha Sept. 8, 2015, 6:56 a.m. UTC | #1
Hi,

> From: Puha, TimoX
> Subject: [PATCH v3] dpdk: add support for v2.1.0
> 
> Update relevant artifacts to add support for DPDK v2.1.0
>  - INSTALL.DPDK.md
>  - acinclude.m4: Change DPDK library name
>  - netdev-dpdk: Limit minimum mbuf size to to adapt to DPDK bug fix that
>    changes the treatment of the requested mbuf size
>  - build.sh: Change DPDK version number
> 
> Note that this breaks compatibility with DPDK v2.0.0 although only
> for the library name change.
> 
> Note that throughput for vhost ports with mergeable buffers is reduced
> about 10% due to a necessary bug fix in DPDK vhost code.
> 
> Signed-off-by: Mark Kavanagh <mark.b.kavanagh@intel.com>
> Signed-off-by: Michal Weglicki <michalx.weglicki@intel.com>
> Signed-off-by: Timo Puha <timox.puha@intel.com>

Any comments on this patch version?  This v3 only corrects a stray whitespace compared to v2, but v2 accommodates all the comments for v1 as far as I understood them right. 

-Timo

Patch
diff mbox

diff --git a/.travis/build.sh b/.travis/build.sh
index e90f4d0..3cadbf0 100755
--- a/.travis/build.sh
+++ b/.travis/build.sh
@@ -71,7 +71,7 @@  fi
 
 if [ "$DPDK" ]; then
     if [ -z "$DPDK_VER" ]; then
-        DPDK_VER="2.0.0"
+        DPDK_VER="2.1.0"
     fi
     install_dpdk $DPDK_VER
     if [ "$CC" = "clang" ]; then
diff --git a/INSTALL.DPDK.md b/INSTALL.DPDK.md
index 35dd9a0..8834c1d 100644
--- a/INSTALL.DPDK.md
+++ b/INSTALL.DPDK.md
@@ -16,7 +16,7 @@  OVS needs a system with 1GB hugepages support.
 Building and Installing:
 ------------------------
 
-Required: DPDK 2.0
+Required: DPDK 2.1
 Optional (if building with vhost-cuse): `fuse`, `fuse-devel` (`libfuse-dev`
 on Debian/Ubuntu)
 
@@ -24,7 +24,7 @@  on Debian/Ubuntu)
   1. Set `$DPDK_DIR`
 
      ```
-     export DPDK_DIR=/usr/src/dpdk-2.0
+     export DPDK_DIR=/usr/src/dpdk-2.1
      cd $DPDK_DIR
      ```
 
@@ -33,11 +33,6 @@  on Debian/Ubuntu)
 
      `CONFIG_RTE_BUILD_COMBINE_LIBS=y`
 
-     Update `config/common_linuxapp` so that DPDK is built with vhost
-     libraries.
-
-     `CONFIG_RTE_LIBRTE_VHOST=y`
-
      Then run `make install` to build and install the library.
      For default install without IVSHMEM:
 
@@ -112,7 +107,7 @@  Using the DPDK with ovs-vswitchd:
      3. Bind network device to vfio-pci:
         `$DPDK_DIR/tools/dpdk_nic_bind.py --bind=vfio-pci eth1`
 
-3. Mount the hugetable filsystem
+3. Mount the hugetable filesystem
 
    `mount -t hugetlbfs -o pagesize=1G none /dev/hugepages`
 
@@ -315,7 +310,7 @@  the vswitchd.
 DPDK vhost:
 -----------
 
-DPDK 2.0 supports two types of vhost:
+DPDK 2.1 supports two types of vhost:
 
 1. vhost-user
 2. vhost-cuse
@@ -336,7 +331,7 @@  with OVS.
 DPDK vhost-user Prerequisites:
 -------------------------
 
-1. DPDK 2.0 with vhost support enabled as documented in the "Building and
+1. DPDK 2.1 with vhost support enabled as documented in the "Building and
    Installing section"
 
 2. QEMU version v2.1.0+
@@ -418,7 +413,7 @@  with OVS.
 DPDK vhost-cuse Prerequisites:
 -------------------------
 
-1. DPDK 2.0 with vhost support enabled as documented in the "Building and
+1. DPDK 2.1 with vhost support enabled as documented in the "Building and
    Installing section"
    As an additional step, you must enable vhost-cuse in DPDK by setting the
    following additional flag in `config/common_linuxapp`:
diff --git a/acinclude.m4 b/acinclude.m4
index 45cfaf6..90bb708 100644
--- a/acinclude.m4
+++ b/acinclude.m4
@@ -172,7 +172,7 @@  AC_DEFUN([OVS_CHECK_DPDK], [
 
     DPDK_INCLUDE=$RTE_SDK/include
     DPDK_LIB_DIR=$RTE_SDK/lib
-    DPDK_LIB="-lintel_dpdk"
+    DPDK_LIB="-ldpdk"
     DPDK_EXTRA_LIB=""
 
     AC_COMPILE_IFELSE(
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index 3444bb1..4f46acd 100644
--- a/lib/netdev-dpdk.c
+++ b/lib/netdev-dpdk.c
@@ -65,11 +65,17 @@  static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 20);
 /*
  * need to reserve tons of extra space in the mbufs so we can align the
  * DMA addresses to 4KB.
+ * The minimum mbuf size is limited to avoid scatter behaviour and drop in
+ * performance for standard Ethernet MTU.
  */
-
 #define MTU_TO_MAX_LEN(mtu)  ((mtu) + ETHER_HDR_LEN + ETHER_CRC_LEN)
-#define MBUF_SIZE(mtu)       (MTU_TO_MAX_LEN(mtu) + (512) + \
-                             sizeof(struct rte_mbuf) + RTE_PKTMBUF_HEADROOM)
+#define MBUF_SIZE_MTU(mtu)   (MTU_TO_MAX_LEN(mtu)        \
+                              + sizeof(struct dp_packet) \
+                              + RTE_PKTMBUF_HEADROOM)
+#define MBUF_SIZE_DRIVER     (2048                       \
+                              + sizeof (struct rte_mbuf) \
+                              + RTE_PKTMBUF_HEADROOM)
+#define MBUF_SIZE(mtu)       MAX(MBUF_SIZE_MTU(mtu), MBUF_SIZE_DRIVER)
 
 /* Max and min number of packets in the mempool.  OVS tries to allocate a
  * mempool with MAX_NB_MBUF: if this fails (because the system doesn't have