l for debugging DPDK-related issues.
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK-ADVANCED.md | 16
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/INSTALL.DPDK-ADVANCED.md b/INSTALL.DPDK-ADVANCED.md
index 555f026..f6f0b5b 100644
--- a/INSTALL.DPDK-ADVANCED.m
.
Resolve occurrences of each, and any other minor style infractions.
Signed-off-by: Mark Kavanagh
---
lib/netdev-dpdk.c | 81 ++-
1 file changed, 44 insertions(+), 37 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index 819a84
Pointers to struct rte_mbuf are typically denoted within functions as
'pkt'; similarly, arrays of, and pointer-to-pointer to, struct rte_mbuf
are denoted by 'pkts'.
Update discrepancies to the above convention for consistency.
Signed-off-by: Mark Kavanagh
---
li
Clean up coding style issues and inconsistencies that have built up in
netdev-dpdk.c
Mark Kavanagh (2):
netdev-dpdk: consistent naming for mbuf variables
netdev-dpdk: Fix coding style
lib/netdev-dpdk.c | 115 +-
1 file changed, 61
r.
Signed-off-by: Mark Kavanagh
Signed-off-by: Ilya Maximets
[diproiet...@vmware.com rebased]
Signed-off-by: Daniele Di Proietto
---
v7:
- add 'Signed-off-by' for Ilya Maximets (i.maxim...@samsung.com)
v6:
- include device name in netdev_dpdk_set_mtu error log
- resolve mi
From: Daniele Di Proietto
Every provider silently drops the const attribute when converting the
parameter to the appropriate subclass. Might as well drop the const
attribute from the parameter, since this is a "set" function.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 2 +-
From: Daniele Di Proietto
Also, netdev-dummy needs to call netdev_change_seq_changed() in
set_mtu().
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 5 -
tests/ofproto-dpif.at | 30 ++
2 files changed, 34 insertions(+), 1 deletion(-)
diff --git
From: Daniele Di Proietto
This will allow run() and wait() methods to be shared between different
classes and still perform class-specific work.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-bsd.c | 6 +++---
lib/netdev-dummy.c| 4 ++--
lib/netdev-linux.c| 6 +++---
lib/net
From: Daniele Di Proietto
"internal" netdevs are treated specially in OVS (e.g. for MTU), but
the dummy datapath remaps both "system" and "internal" devices to the
same "dummy" netdev class, so there's no way to discern those in tests.
This commit adds a new "dummy-internal" netdev type, which w
From: Daniele Di Proietto
The 'mtu_request' column can be used to set the MTU of a specific
interface.
This column is useful because it will allow changing the MTU of DPDK
devices (implemented in a future commit), which are not accessible
outside the ovs-vswitchd process, but it can be used for
From: Daniele Di Proietto
Interfaces with type "internal" end up having a netdev with type "tap"
in the dpif-netdev datapath, so a strcmp will fail to match internal
interfaces.
We can translate the types with ofproto_port_open_type() before calling
strcmp to fix this.
This fixes a minor issue
From: Daniele Di Proietto
Every provider silently drops the const attribute when converting the
parameter to the appropriate subclass. Might as well drop the const
attribute from the parameter, since this is a "set" function.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 2 +-
r.
Signed-off-by: Mark Kavanagh
[diproiet...@vmware.com rebased]
Signed-off-by: Daniele Di Proietto
---
v6:
- include device name in netdev_dpdk_set_mtu error log
- resolve minor coding standards infractions
v5:
- rename dpdk_mp_configure to netdev_dpdk_mempool_configure
- c
From: Daniele Di Proietto
Interfaces with type "internal" end up having a netdev with type "tap"
in the dpif-netdev datapath, so a strcmp will fail to match internal
interfaces.
We can translate the types with ofproto_port_open_type() before calling
strcmp to fix this.
This fixes a minor issue
From: Daniele Di Proietto
Also, netdev-dummy needs to call netdev_change_seq_changed() in
set_mtu().
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 5 -
tests/ofproto-dpif.at | 30 ++
2 files changed, 34 insertions(+), 1 deletion(-)
diff --git
From: Daniele Di Proietto
"internal" netdevs are treated specially in OVS (e.g. for MTU), but
the dummy datapath remaps both "system" and "internal" devices to the
same "dummy" netdev class, so there's no way to discern those in tests.
This commit adds a new "dummy-internal" netdev type, which w
From: Daniele Di Proietto
The 'mtu_request' column can be used to set the MTU of a specific
interface.
This column is useful because it will allow changing the MTU of DPDK
devices (implemented in a future commit), which are not accessible
outside the ovs-vswitchd process, but it can be used for
From: Daniele Di Proietto
This will allow run() and wait() methods to be shared between different
classes and still perform class-specific work.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-bsd.c | 6 +++---
lib/netdev-dummy.c| 4 ++--
lib/netdev-linux.c| 6 +++---
lib/net
From: Daniele Di Proietto
Every provider silently drops the const attribute when converting the
parameter to the appropriate subclass. Might as well drop the const
attribute from the parameter, since this is a "set" function.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 2 +-
From: Daniele Di Proietto
"internal" netdevs are treated specially in OVS (e.g. for MTU), but
the dummy datapath remaps both "system" and "internal" devices to the
same "dummy" netdev class, so there's no way to discern those in tests.
This commit adds a new "dummy-internal" netdev type, which w
r.
Signed-off-by: Mark Kavanagh
[diproiet...@vmware.com rebased]
Signed-off-by: Daniele Di Proietto
---
v5:
- rename dpdk_mp_configure to netdev_dpdk_mempool_configure
- consolidate socket_id and mtu changes within
netdev_dpdk_mempool_configure
- add lower bounds check for use
From: Daniele Di Proietto
The 'mtu_request' column can be used to set the MTU of a specific
interface.
This column is useful because it will allow changing the MTU of DPDK
devices (implemented in a future commit), which are not accessible
outside the ovs-vswitchd process, but it can be used for
From: Daniele Di Proietto
This will allow run() and wait() methods to be shared between different
classes and still perform class-specific work.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-bsd.c | 6 +++---
lib/netdev-dummy.c| 4 ++--
lib/netdev-linux.c| 6 +++---
lib/net
From: Daniele Di Proietto
Also, netdev-dummy needs to call netdev_change_seq_changed() in
set_mtu().
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 5 -
tests/ofproto-dpif.at | 30 ++
2 files changed, 34 insertions(+), 1 deletion(-)
diff --git
From: Daniele Di Proietto
Interfaces with type "internal" end up having a netdev with type "tap"
in the dpif-netdev datapath, so a strcmp will fail to match internal
interfaces.
We can translate the types with ofproto_port_open_type() before calling
strcmp to fix this.
This fixes a minor issue
From: Daniele Di Proietto
"internal" netdevs are treated specially in OVS (e.g. for MTU), but
the dummy datapath remaps both "system" and "internal" devices to the
same "dummy" netdev class, so there's no way to discern those in tests.
This commit adds a new "dummy-internal" netdev type, which w
r.
Signed-off-by: Mark Kavanagh
[diproiet...@vmware.com rebased]
Signed-off-by: Daniele Di Proietto
---
v4:
- restore error reporting in *_reconfigure functions (for
non-mtu-configuration based errors)
- remove 'goto' in the event of dpdk_mp_configure failure
- remo
From: Daniele Di Proietto
Also, netdev-dummy needs to call netdev_change_seq_changed() in
set_mtu().
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 5 -
tests/ofproto-dpif.at | 30 ++
2 files changed, 34 insertions(+), 1 deletion(-)
diff --git
From: Daniele Di Proietto
This will allow run() and wait() methods to be shared between different
classes and still perform class-specific work.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-bsd.c | 6 +++---
lib/netdev-dummy.c| 4 ++--
lib/netdev-linux.c| 6 +++---
lib/net
From: Daniele Di Proietto
Every provider silently drops the const attribute when converting the
parameter to the appropriate subclass. Might as well drop the const
attribute from the parameter, since this is a "set" function.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 2 +-
From: Daniele Di Proietto
The 'mtu_request' column can be used to set the MTU of a specific
interface.
This column is useful because it will allow changing the MTU of DPDK
devices (implemented in a future commit), which are not accessible
outside the ovs-vswitchd process, but it can be used for
From: Daniele Di Proietto
Interfaces with type "internal" end up having a netdev with type "tap"
in the dpif-netdev datapath, so a strcmp will fail to match internal
interfaces.
We can translate the types with ofproto_port_open_type() before calling
strcmp to fix this.
This fixes a minor issue
From: Daniele Di Proietto
"internal" netdevs are treated specially in OVS (e.g. for MTU), but
the dummy datapath remaps both "system" and "internal" devices to the
same "dummy" netdev class, so there's no way to discern those in tests.
This commit adds a new "dummy-internal" netdev type, which w
From: Daniele Di Proietto
This will allow run() and wait() methods to be shared between different
classes and still perform class-specific work.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-bsd.c | 6 +++---
lib/netdev-dummy.c| 4 ++--
lib/netdev-linux.c| 6 +++---
lib/net
From: Daniele Di Proietto
Every provider silently drops the const attribute when converting the
parameter to the appropriate subclass. Might as well drop the const
attribute from the parameter, since this is a "set" function.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 2 +-
r.
Signed-off-by: Mark Kavanagh
[diproiet...@vmware.com rebased]
Signed-off-by: Daniele Di Proietto
---
v3:
- replace netdev_dpdk.last_mtu with local variable
- add comment for dpdk_mp_configure
v2:
- rebase to HEAD of master
- fall back to previous 'good' MTU
From: Daniele Di Proietto
Also, netdev-dummy needs to call netdev_change_seq_changed() in
set_mtu().
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 5 -
tests/ofproto-dpif.at | 30 ++
2 files changed, 34 insertions(+), 1 deletion(-)
diff --git
From: Daniele Di Proietto
The 'mtu_request' column can be used to set the MTU of a specific
interface.
This column is useful because it will allow changing the MTU of DPDK
devices (implemented in a future commit), which are not accessible
outside the ovs-vswitchd process, but it can be used for
From: Daniele Di Proietto
Interfaces with type "internal" end up having a netdev with type "tap"
in the dpif-netdev datapath, so a strcmp will fail to match internal
interfaces.
We can translate the types with ofproto_port_open_type() before calling
strcmp to fix this.
This fixes a minor issue
From: Daniele Di Proietto
Also, netdev-dummy needs to call netdev_change_seq_changed() in
set_mtu().
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dummy.c| 5 -
tests/ofproto-dpif.at | 30 ++
2 files changed, 34 insertions(+), 1 deletion(-)
diff --git
r.
Signed-off-by: Mark Kavanagh
[diproiet...@vmware.com rebased]
Signed-off-by: Daniele Di Proietto
---
Previous: http://openvswitch.org/pipermail/dev/2016-July/076845.html
v2->v1:
- rebase to HEAD of master
- fall back to previous 'good' MTU if reconfigure fails
From: Daniele Di Proietto
Every provider silently drops the const attribute when converting the
parameter to the appropriate subclass. Might as well drop the const
attribute from the parameter, since this is a "set" function.
Signed-off-by: Daniele Di Proietto
---
v2->v1: add missing 'Signed-o
From: Daniele Di Proietto
This will allow run() and wait() methods to be shared between different
classes and still perform class-specific work.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-bsd.c | 6 +++---
lib/netdev-dummy.c| 4 ++--
lib/netdev-linux.c| 6 +++---
lib/net
From: Daniele Di Proietto
"internal" netdevs are treated specially in OVS (e.g. for MTU), but
the dummy datapath remaps both "system" and "internal" devices to the
same "dummy" netdev class, so there's no way to discern those in tests.
This commit adds a new "dummy-internal" netdev type, which w
From: Daniele Di Proietto
The 'mtu_request' column can be used to set the MTU of a specific
interface.
This column is useful because it will allow changing the MTU of DPDK
devices (implemented in a future commit), which are not accessible
outside the ovs-vswitchd process, but it can be used for
From: Daniele Di Proietto
Interfaces with type "internal" end up having a netdev with type "tap"
in the dpif-netdev datapath, so a strcmp will fail to match internal
interfaces.
We can translate the types with ofproto_port_open_type() before calling
strcmp to fix this.
This fixes a minor issue
memory pool is no longer in use).
Signed-off-by: Mark Kavanagh
---
v2->v1: rebase to head of master, and remove 'RFC' tag
lib/netdev-dpdk.c | 29 +++--
1 file changed, 15 insertions(+), 14 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
memory pool is no longer in use).
Until DPDK v16.07 is supported, this patch may be considered RFC.
Signed-off-by: Mark Kavanagh
---
lib/netdev-dpdk.c | 29 ++---
1 file changed, 14 insertions(+), 15 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index
Commit 64839cf43 applies batch objects to netdev-providers, but
some comments were not updated accordingly. Fix these:
- replace 'pkts' with 'batch'
- replace '*cnt' with 'batch->count'
- replace MAX_RX_BATCH with NETDEV_MAX_BURST
- remove
This patch constitutes a response to a request on ovs-discuss
(http://openvswitch.org/pipermail/discuss/2016-May/021261.html), and
is only for consideration in the testing scenario documented therein.
It should not be considered for review, or submission to the OVS source
code - the proposed mech
the user-supplied value is invalid,
the MTU for the port defaults to standard Ethernet MTU (i.e. 1500B).
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 60 -
NEWS | 1 +
lib/netdev-dpdk.c | 248 +-
3 files changed
the user-supplied value is invalid,
the MTU for the port defaults to standard Ethernet MTU (i.e. 1500B).
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 60 -
NEWS | 3 +-
lib/netdev-dpdk.c | 248 +-
3 files
This patchset enables DPDK ports to support Jumbo frames
up to, and including 13K in size. Jumbo frame support is implemented
by increasing the amount of data that each mbuf can encompass, thus
allowing each mbuf segment to accomodate a single jumbo frame.
The series additionally cleans up mbuf in
Current mbuf initialization relies on magic numbers and does not
accomodate mbufs of different sizes.
Resolve this issue by ensuring that mbufs are always aligned to a 1k
boundary (a typical DPDK NIC Rx buffer alignment).
Signed-off-by: Mark Kavanagh
---
lib/netdev-dpdk.c | 87
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 59 +
1 file changed, 59 insertions(+)
diff --git a/INSTALL.DPDK.md b/INSTALL.DPDK.md
index 722fb9e..4ca98cb 100644
--- a/INSTALL.DPDK.md
+++ b/INSTALL.DPDK.md
@@ -878,6 +878,60 @@ by
Current mbuf initialization relies on magic numbers and does not
accomodate mbufs of different sizes.
Resolve this issue by ensuring that mbufs are always aligned to a 1k
boundary (a typical DPDK NIC Rx buffer alignment).
Signed-off-by: Mark Kavanagh
---
lib/netdev-dpdk.c | 51
the user-supplied value is invalid,
the MTU for the port defaults to standard Ethernet MTU (i.e. 1500B).
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 1 -
lib/netdev-dpdk.c | 259 +-
2 files changed, 159 insertions(+), 101 deletions
Signed-off-by: Mark Kavanagh
---
NEWS | 2 ++
1 file changed, 2 insertions(+)
diff --git a/NEWS b/NEWS
index 3e33871..2942ae0 100644
--- a/NEWS
+++ b/NEWS
@@ -12,6 +12,8 @@ Post-v2.5.0
Old 'other_config:n-dpdk-rxqs' is no longer supported.
- ovs-benchmark: This utilit
This series of commits enables DPDK ports to support Jumbo frames
up to, and including 13K in size. Jumbo frame support is implemented
by increasing the amount of data that each mbuf can encompass, thus
allowing each mbuf segment to accomodate a single jumbo frame.
The series additionally cleans u
Signed-off-by: Mark Kavanagh
---
NEWS | 2 ++
1 file changed, 2 insertions(+)
diff --git a/NEWS b/NEWS
index 3e33871..2942ae0 100644
--- a/NEWS
+++ b/NEWS
@@ -12,6 +12,8 @@ Post-v2.5.0
Old 'other_config:n-dpdk-rxqs' is no longer supported.
- ovs-benchmark: This utilit
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 59 +
1 file changed, 59 insertions(+)
diff --git a/INSTALL.DPDK.md b/INSTALL.DPDK.md
index 722fb9e..4ca98cb 100644
--- a/INSTALL.DPDK.md
+++ b/INSTALL.DPDK.md
@@ -878,6 +878,60 @@ by
the user-supplied value is invalid,
the MTU for the port defaults to standard Ethernet MTU (i.e. 1500B).
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 1 -
lib/netdev-dpdk.c | 259 +-
2 files changed, 159 insertions(+), 101 deletions
the user-supplied value is invalid,
the MTU for the port defaults to standard Ethernet MTU (i.e. 1500B).
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 59 +-
NEWS | 2 +
lib/netdev-dpdk.c | 347 +-
3 files changed
the user-supplied value is invalid,
the MTU for the port defaults to standard Ethernet MTU (i.e. 1500B).
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 59 -
lib/netdev-dpdk.c | 356 +-
2 files changed, 328 insertions(+), 87
lds may be upstreamed, leading to issues such as broken build,
e.g. http://openvswitch.org/pipermail/dev/2016-January/064328.html
Signed-off-by: Mark Kavanagh
---
CONTRIBUTING.md | 3 +++
1 file changed, 3 insertions(+)
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 086b6e8..fe9400a 10
x27;s minimum Rx buffer length, the frame
length is rounded up to the closest value that is.
Signed-off-by: Mark Kavanagh
---
INSTALL.DPDK.md | 67 -
lib/netdev-dpdk.c | 176 ++---
2 files changed, 207 insertions(+), 36 dele
Commit 74ff329 introduces 'struct eth_addr'; this breaks compilation
for netdev-dpdk.c.
Refactor any affected code to resolve.
Signed-off-by: Mark Kavanagh
---
lib/netdev-dpdk.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dp
ed on this information, a user may determine what affinity
modifications are required to optimise performance for their
particular use case.
Signed-off-by: Mark Kavanagh
Signed-off-by: Wojciech Andralojc
---
lib/dpif-netdev.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/lib/dpif-netd
ang
- add details of vhost user support limitations to INSTALL.DPDK.md
- refactor travis/build.sh to reflect these changes
v3: -correct minor typos in commit message
Signed-off-by: Mark Kavanagh
Signed-off-by: Panu Matilainen
---
.travis/build.sh | 11 +++
ang
- add details of vhost user support imitations to INSTALL.DPDK.md
- refactor .travis/build.sh to reflect these changes
Signed-off-by: Mark Kavanagh
Signed-off-by: Panu Matilainen
---
.travis/build.sh | 11 +++
INSTALL.DPDK.md | 13 +
acinclude.m4 | 2 +-
lib/n
as to whether anyone else has observed these issues, and if so,
if they'd be willing to assist by validating this patch?
Thanks in advance.
Mark Kavanagh (1):
travis: fix errors
.travis/build.sh | 4 ++--
.travis/prepare.sh | 10 +-
2 files changed, 7 inser
Fix two issues observed in travis scripts:
- prepare.sh: add '-E' flag to ensure user environmental variables
(for example, URL of the proxy server) are passed to 'sudo'
- build.sh: add quotes around compiler variable to fix 'unary
operator expected' error
Update relevant artifacts to add support for DPDK v2.0.0
- INSTALL.DPDK.md
- travis build script
- netdev-dpdk: fix build with unified offload types in DPDK v2.0.0
Note that this breaks compatibility with DPDK v1.8.0
Signed-off-by: Mark Kavanagh
Signed-off-by: Panu Matilainen
---
.travis
7;mssse3' flag at compile time;
however, it's not possible to pass an additional flag via the command
line without overwriting an existing compiler variable.
Introduce an additonal configuration option to resolve this.
Signed-off-by: Mark Kavanagh
---
acinclude.m4 | 11 +++
conf
a_len' and 'data_off' would result
in wrap-around, and consequently, data corruption. Changes
introduced in this patch prevent this from occurring.
v1:
- update affected functions to use the correct rte_mbuf fields
- remove init function from netdev-dpdk (no longer required as
e definition of some functions (ovs_assert, for example), altering
the outcome of some unit tests. Rather, it is preferable to use a
combination of 'make distcheck' with 'make check' to cover all bases.
Signed-off-by: Mark Kavanagh
---
CONTRIBUTING.md | 5 -
1 file chang
split across multiple ofpbufs, values larger
than UINT16_MAX for 'data_len' and 'data_off' would result
in wrap-around, and consequently, data corruption. Changes
introduced in this patch prevent this from occurring.
Signed-off-by: Mark Kavanagh
Signed-off-by: Mar
ephen Finucane
Reviewed-By: Ciara Loftus
Signed-off-by: Mark Kavanagh
---
lib/netdev-dpdk.c | 18 +-
1 file changed, 1 insertion(+), 17 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index 9ac49c8..9c42756 100644
--- a/lib/netdev-dpdk.c
+++ b/lib/netdev-d
DPDK v1.8.0-rc4 makes significant changes to struct rte_mbuf, including
removal of the 'pkt' and 'data' fields. These fields are referenced
by OVS when accessing the data section of an ofpbuf.
Update any relevant functions to accomodate the new rte_mbuf layout.
Signed-o
o accomodate the new rte_mbuf layout.
Also update documentation and Travis scripts.
Mark Kavanagh (2):
netdev-dpdk: move to DPDK 1.8.0-rc4
ofpbuf: use DPDK 1.8 rte_mbuf layout
Rory Sexton (1):
netdev-dpdk: remove init function
.travis.yml | 2 +-
.travis/build.sh | 6 +++---
IN
This patch updates the documentation to reflect that DPDK 1.8.0-rc4 is
supported. Travis scripts have also been updated to reflect this.
DPDK phy and ring ports were validated against DPDK 1.8.0-rc4
Signed-off-by: Mark Kavanagh
Signed-off-by: Rory Sexton
Reviewed-By: Stephen Finucane
81 matches
Mail list logo
