s its own item as shown
in this testpmd flow command example:
flow create 0 ingress pattern eth / e_tag in_ecid_base is 42 / end actions
drop / end
Note, all multibyte values are in network order like other protocol header
definitions.
[1] http://dpdk.org/ml/archives/dev/2016-December/053181.html
Message ID: 20161223081310.gh10...@6wind.com
--
Adrien Mazarguil
6WIND
; /**< Flow ID. */
};
Like for E-Tag, applications are responsibile for breaking down and filling
individual fields properly.
[1] http://dpdk.org/ml/archives/dev/2016-December/053181.html
Message ID: 20161223081310.gh10...@6wind.com
--
Adrien Mazarguil
6WIND
> +WERROR_FLAGS += -diag-disable 188
>
> ifeq ($(RTE_DEVEL_BUILD),y)
> WERROR_FLAGS += -Werror-all
> --
> 2.9.3
I also think this warning may be useful but is not worth the trouble in many
cases, thus:
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
On Wed, Jan 04, 2017 at 10:53:50AM +0100, Simon Horman wrote:
> On Thu, Dec 22, 2016 at 01:48:04PM +0100, Adrien Mazarguil wrote:
> > On Wed, Dec 21, 2016 at 05:19:16PM +0100, Simon Horman wrote:
> > > On Fri, Dec 16, 2016 at 05:24:57PM +0100, Adrien Mazarguil wrote:
[...]
>
> VLAN, VXLAN are still not supported.
> >
> > Signed-off-by: Nelio Laranjeiro
> > Acked-by: Adrien Mazarguil
>
> <...>
>
> > +static int
> > +priv_flow_validate(struct priv *priv,
> > + const struct rte_flow_attr *a
ents.
Well, I do not agree with Konstantin (no one saw this coming eh?) and do not
think you need to update your series again.
PMD-specific symbols have nothing to do in the global namespace in my
opinion, they are not versioned and may evolve without notice. Neither
applications nor the bonding PMD can rely on them. That's the trade-off.
Therefore until APIs are made global, the safe compromise is to define
neutral, reserved symbols that any PMD can use to implement their own
temporary APIs for testing purposes. These can be renamed later without
changing their value as long as a single PMD uses them.
--
Adrien Mazarguil
6WIND
On Thu, Jan 05, 2017 at 03:12:01AM +, Zhao1, Wei wrote:
> Hi, adrien
>
> > -Original Message-
> > From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
> > Sent: Tuesday, January 3, 2017 10:08 PM
> > To: Zhao1, Wei
> > Cc: dev@dpdk.org; Lu,
nnectX-5 must appear
somewhere (you can use past release notes as a template).
--
Adrien Mazarguil
6WIND
ibc. Doing so is
supported on the DPDK side since includes have been cleaned up.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 2 +-
drivers/net/mlx5/Makefile | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/mlx4/Makefile b/drivers/net/mlx4/Makef
keyword addresses it.
Fixes: 143c03925695 ("net/mlx: update C compliance standard")
Signed-off-by: Adrien Mazarguil
Tested-by: Yongseok Koh
---
Ferruh, squashing this commit with the previous one and discarding its
log is fine if you want. Thanks.
---
drivers/net/mlx5/mlx5_rxtx.h | 2
ot support this mode are automatically discarded.
>
> Signed-off-by: Vasily Philipov
Thanks for reordering the series. For this commit:
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
On Tue, Jul 04, 2017 at 11:22:49AM +, Vasily Philipov wrote:
> The user must request isolated mode before device configuration.
>
> Signed-off-by: Vasily Philipov
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
ing a flow in isolated mode
causes a crash due to RX QPs not being allocated. You should temporarily
make priv_flow_create_action_queue() call rxq_create_qp() when the target QP
does not exist.
Patch looks otherwise fine.
--
Adrien Mazarguil
6WIND
gned-off-by: Vasily Philipov
Looks good, assuming crash from patch 3/4 is addressed:
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
t; returning mbufs to the pool to avoid crashing.
>
> Fixes: 7fae69eeff13 ("mlx4: new poll mode driver")
> CC: sta...@dpdk.org
>
> Signed-off-by: Vasily Philipov
Thanks Vasily, except for the minor nit below,
Acked-by: Adrien Mazarguil
> ---
> drivers/net/m
ot support this mode are automatically discarded.
>
> Signed-off-by: Vasily Philipov
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
On Wed, Jul 05, 2017 at 11:14:09AM +0300, Vasily Philipov wrote:
> The user must request isolated mode before device configuration.
>
> Signed-off-by: Vasily Philipov
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
d
> in this case.
>
> Refactor RSS parent queue allocations (currently limited to a single
> parent) in preparation for flow API RSS action support.
>
> Signed-off-by: Vasily Philipov
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
gned-off-by: Vasily Philipov
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
_gre
> rte_flow_item_gre_mask = {
> * Matching accuracy level can be configure by threshold.
> * Driver can divide the range of threshold and map to different
> * accuracy levels that device support.
> + *
> + * Threshold 0 means perfect match(no fuzziness), while threshol
g
> ---
> v2:
> - minor fix
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
From: Gaetan Rivet
This allows PMDs and applications to save flow rules in their generic
format for later processing. This is useful when rules cannot be applied
immediately, such as when the device is not properly initialized.
Signed-off-by: Gaetan Rivet
Signed-off-by: Adrien Mazarguil
seems a mistake.
Why? It was done on purpose, I think exposing through Doxygen internal APIs
implemented by drivers must be a requirement, even if the exposed symbols
are not necessarily versioned.
--
Adrien Mazarguil
6WIND
ases since 17.02.
[1] http://dpdk.org/ml/archives/dev/2017-June/068605.html
Adrien Mazarguil (3):
cmdline: fix dynamic tokens initialization
cmdline: fix dynamic tokens interface
app/testpmd: fix token matching in flow command
app/test-pmd/cmdline_flow.c| 36 +++---
lib/libr
. Doing so is also
less expensive than a full memset() at each iteration.
Fixes: 4fffc05a2b2c ("cmdline: support dynamic tokens")
Cc: sta...@dpdk.org
Signed-off-by: Bernard Iremonger
Signed-off-by: Adrien Mazarguil
Acked-by: Olivier Matz
---
lib/librte_cmdline/cmdline_pa
9d2c ("app/testpmd: add queue actions to flow command")
Cc: sta...@dpdk.org
Signed-off-by: Adrien Mazarguil
---
app/test-pmd/cmdline_flow.c | 21 +
1 file changed, 17 insertions(+), 4 deletions(-)
diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow
.
Fixes: 4fffc05a2b2c ("cmdline: support dynamic tokens")
Fixes: 19c90af6285c ("app/testpmd: add flow command")
Cc: sta...@dpdk.org
Signed-off-by: Adrien Mazarguil
Acked-by: Olivier Matz
Cc: Bernard Iremonger
---
app/test-pmd/cmdline_flow.c| 15 ++--
lib/librte_cmd
ot support this mode are automatically discarded.
>
> Signed-off-by: Vasily Philipov
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
; -Original Message-
> > From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
> > Sent: Thursday, June 1, 2017 4:14 PM
> > To: Dumitrescu, Cristian
> > Cc: dev@dpdk.org; tho...@monjalon.net;
> > jerin.ja...@caviumnetworks.com; hemant.agra...@nxp.com; Doherty,
and the RSS action. Both had to be handled through special
functions which I messed up.
Unfortunately I didn't catch it in testpmd because it doesn't do anything
with the copied data unlike the fail-safe PMD.
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
that the original author (now on Cc:) might not actually review,
> there is no value in keeping the Cc: in git commit history. If the
> original author performs a review, their Reviewed-by: or Acked-by: is
> stored in git history (same as now).
>
> Signed-off-by: Harry van Haa
gt; either "virtual machines" of "VMs", I think the
reader understands what they are at this point.
> + and NUMA node are pinned for the VM according to the above. Run
And you should remove "for the VM".
> +
> + .. code-block:: console
> +
> +lstopo-no-graphics
> +
> + to identify the NUMA node to which the PCIe adapter is connected.
> +
> +4. If more than one adapter is used, and root complex capabilities enables
> + to put both adapters on the same NUMA node without PCI bandwidth
> degredation,
degredation => degradation
> + it is recommended to locate both adapters on the same NUMA node.
> + This in order to forward packets from one to the other without
> + NUMA performance penalty.
> +
> +5. Disable pause frames
Missing period or colon.
> +
> + .. code-block:: console
> +
> +ethtool -A rx off tx off
> +
> +6. Verify IO non-posted prefetch is disabled by default. This can be checked
> + via the BIOS configuration. Please contact you server provider for more
> + information about the settings.
> +
> +.. hint::
> +
> +On Some machines, depends on the machine intergrator, it is
> beneficial
Some => some
intergrator => integrator
> +to set the PCI max read request parameter to 1K. This can be
> +done in the following way:
> +
> +To query the read request size use:
> +
> +.. code-block:: console
> +
> +setpci -s 68.w
> +
> +If the output is different than 3XXX, set it by:
> +
> +.. code-block:: console
> +
> +setpci -s 68.w=3XXX
> +
> +The XXX can be different on different systems. Make sure to configure
> +according to the setpci output.
> +
> Notes for testpmd
> -
>
> --
> 2.12.0
>
--
Adrien Mazarguil
6WIND
> dir = opendir(path);
> --
> 1.8.3.1
>
I think this patch is not necessary unless proved otherwise, have you
actually observed a crash addressed by it?
--
Adrien Mazarguil
6WIND
;
> }
> if (action->drop) {
> - qp = priv->flow_drop_queue->qp;
> + qp = priv->flow_drop_queue.qp;
> } else {
> int ret;
> unsigned int i;
> @@ -1307,7 +1289,7 @@ struct rte_flow *
> for (flow = LIST_FIRST(&priv->flows);
>flow;
>flow = LIST_NEXT(flow, next)) {
> - qp = flow->qp ? flow->qp : priv->flow_drop_queue->qp;
> + qp = flow->qp ? flow->qp : priv->flow_drop_queue.qp;
> flow->ibv_flow = ibv_create_flow(qp, flow->ibv_attr);
> if (!flow->ibv_flow) {
> DEBUG("Flow %p cannot be applied", (void *)flow);
> --
> 1.8.3.1
>
--
Adrien Mazarguil
6WIND
a larger MP cache
need to also increase CONFIG_RTE_LIBRTE_MLX4_TX_MP_CACHE like you did.
Because updating the default value cannot possibly satisfy all use cases,
I think it's better to leave it as is for the time being in order to not
affect existing applications.
[1] http://dpdk.org/ml/archives/dev/2017-July/071405.html
--
Adrien Mazarguil
6WIND
that depend on the application
or the host system, if for instance DPDK memory is spread across NUMA
nodes. Make sure it's not the case.
--
Adrien Mazarguil
6WIND
API change for detach flag
>
> app/test-pmd/testpmd.c | 9 ++---
> doc/guides/rel_notes/deprecation.rst | 6 ++
> drivers/net/mlx4/mlx4.c | 1 +
> drivers/net/mlx5/mlx5.c | 1 +
> lib/librte_ether/rte_ethdev.c| 11 +--
> 5 files changed, 15 insertions(+), 13 deletions(-)
>
> --
> 2.1.4
>
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
w->ibv_flow));
> + /* Current verbs does not allow to check real
> + * errors when the device was plugged out.
> + */
> + ibv_destroy_flow(flow->ibv_flow);
> flow->ibv_flow = NULL;
> DEBUG("Flow %p removed", (void *)flow);
> }
> --
> 1.8.3.1
>
This approach looks way too intrusive. How about making the claim_zero()
definition not fail but still complain when compiled against a broken Verbs
version instead?
#include "mlx4_autoconf.h"
[...]
#ifndef HAVE_BROKEN_VERBS
#define claim_zero(...) assert((__VA_ARGS__) == 0)
#else /* HAVE_BROKEN_VERBS */
#define claim_zero(...) \
(void)(((__VA_ARGS__) == 0) || \
DEBUG("Assertion `" # __VA_ARGS__ "' failed (IGNORED)"))
#endif /* HAVE_BROKEN_VERBS */
You could use auto-config-h.sh to generate the HAVE_BROKEN_VERBS definition
in mlx4_autoconf.h (see mlx4 Makefile) based on some symbol, macro or type
that only exists or doesn't exist yet in problematic releases for instance.
--
Adrien Mazarguil
6WIND
Hi Matan,
On Mon, Jul 31, 2017 at 04:56:33PM +, Matan Azrad wrote:
> Hi Adrien
>
> > -Original Message-
> > From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
> > Sent: Monday, July 31, 2017 5:17 PM
> > To: Matan Azrad
> > Cc: dev@dp
support")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Nelio Laranjeiro
> Acked-by: Yongseok Koh
Acked-by: Adrien Mazarguil
Shouldn't this fix target an even older commit?
cf37ca9563d5 ("mlx5: support MTU configuration")
--
Adrien Mazarguil
6WIND
Hi Matan,
(snipping a bit of unnecessary context)
On Tue, Aug 01, 2017 at 10:12:29AM +, Matan Azrad wrote:
[...]
> > From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
[...]
> > On Mon, Jul 31, 2017 at 04:56:33PM +, Matan Azrad wrote:
[...]
> > > >
, as in mlx5.
This rework targets DPDK 17.11.
Adrien Mazarguil (48):
net/mlx4: add consistency to copyright notices
net/mlx4: remove limitation on number of instances
net/mlx4: check max number of ports dynamically
net/mlx4: remove useless compilation checks
net/mlx4: remove secondary
owner.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/mlx4.rst | 2 +-
drivers/net/mlx4/Makefile| 4 ++--
drivers/net/mlx4/mlx4.c | 4 ++--
drivers/net/mlx4/mlx4.h | 4 ++--
drivers/net/mlx4/mlx4_flow.c | 2 +-
drivers/net/mlx4/mlx4_flow.h | 2 +-
6 files changed, 9
an Ethernet device driver instead of a PCI driver spawning Ethernet
devices.
Getting rid of it simplifies device initialization.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 57 +++-
1 file changed, 3 insertions(+), 54 deletions(-)
diff
Use maximum number reported by hardware capabilities as replacement for the
static check on MLX4_PMD_MAX_PHYS_PORTS.
Cc: Gaëtan Rivet
Cc: Allain Legacy
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 43 +--
drivers/net/mlx4/mlx4.h | 3
to remove a few useless data path debugging messages
behind compilation checks on never defined macros.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile| 12
drivers/net/mlx4/mlx4.c | 35 ---
drivers/net/mlx4/mlx4.h | 2
Current implementation is partial (Tx only), not convenient to use and
not of primary concern.
Remove this feature before refactoring the PMD.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/features/mlx4.ini | 1 -
doc/guides/nics/mlx4.rst | 2 -
drivers/net/mlx4/mlx4.c
Less code makes refactoring easier. No impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 17 +
drivers/net/mlx4/mlx4.h | 12
2 files changed, 1 insertion(+), 28 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net
Software counters are mandatory since hardware counters are not
implemented.
Signed-off-by: Adrien Mazarguil
---
config/common_base| 1 -
doc/guides/nics/mlx4.rst | 6 --
drivers/net/mlx4/Makefile | 4
drivers/net/mlx4/mlx4.c | 37
This option both sets the maximum number of segments for Rx/Tx packets and
whether scattered mode is supported at all. This commit removes the latter
as well as configuration file exposure since the most appropriate value
should be decided at run-time.
Signed-off-by: Adrien Mazarguil
---
config
This should be a run-time parameter.
Signed-off-by: Adrien Mazarguil
---
config/common_base| 1 -
drivers/net/mlx4/Makefile | 4
drivers/net/mlx4/mlx4.c | 6 --
drivers/net/mlx4/mlx4.h | 4
4 files changed, 15 deletions(-)
diff --git a/config/common_base b/config
This is done in preparation for a major refactoring.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/features/mlx4.ini | 2 -
doc/guides/nics/mlx4.rst | 2 -
drivers/net/mlx4/mlx4.c | 311 -
drivers/net/mlx4/mlx4.h | 4 -
4
This is done in preparation for a major refactoring.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/features/mlx4.ini | 1 -
doc/guides/nics/mlx4.rst | 1 -
drivers/net/mlx4/mlx4.c | 206 +++--
drivers/net/mlx4/mlx4.h | 13
Only the default port MAC address remains and is not configurable.
This is done in preparation for a major refactoring.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/features/mlx4.ini | 2 -
doc/guides/nics/mlx4.rst | 1 -
drivers/net/mlx4/mlx4.c | 322
.
Removing support for the RSS flow rule action is the first step toward this
goal.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 6 +-
drivers/net/mlx4/mlx4.h | 5 -
drivers/net/mlx4/mlx4_flow.c | 206 +++---
drivers/net/mlx4
This reverts commit ff00a0dc5600dbb0a29e4aa7fa4b078f98c7a360.
Support for several RSS parent queues was necessary to implement the RSS
flow rule action, dropped in a prior commit.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 332
specific flow rules.
There is no need for "ignored" Rx queues anymore since their number is no
longer limited to powers of two.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/features/mlx4.ini | 1 -
doc/guides/nics/mlx4.rst | 13 --
drivers/net/mlx4/mlx4.c
it belongs.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 215 +++
drivers/net/mlx4/mlx4.h | 2 +-
2 files changed, 57 insertions(+), 160 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index dd42c96..c11e789
The Verbs API used to implement Tx and Rx checksum offloads is deprecated.
Support for these will be added back after refactoring the PMD.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/features/mlx4.ini | 4 --
doc/guides/nics/mlx4.rst | 2 -
drivers/net/mlx4/mlx4.c
The Verbs API used to implement packet type recognition is deprecated.
Support will be added back after refactoring the PMD.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/features/mlx4.ini | 1 -
drivers/net/mlx4/mlx4.c | 70 +-
2 files changed
The Verbs API used to implement inline receive is deprecated.
Support will be added back after refactoring the PMD.
Signed-off-by: Moti Haimovsky
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 52
drivers/net/mlx4/mlx4.h | 1 -
2
The Verbs API used to implement Tx and Rx burst functions is deprecated.
Drop scatter/gather support to ease refactoring while maintaining basic
single-segment Rx/Tx functionality in the meantime.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/features/mlx4.ini | 1 -
doc/guides/nics
The Verbs API used to set QP attributes is deprecated. Revert to the
standard API since it actually supports the remaining ones.
Signed-off-by: Moti Haimovsky
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 25 +++--
1 file changed, 11 insertions(+), 14
This reverts commit 8b3ffe95e75d6d305992505005cbb95969874a15.
Multicast loopback prevention is not part of the standard Verbs interface.
Remove it temporarily.
Signed-off-by: Moti Haimovsky
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 6 +-
drivers/net/mlx4/mlx4.c | 7
allocation
interface.
Signed-off-by: Moti Haimovsky
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 88 ---
drivers/net/mlx4/mlx4.h | 2 -
drivers/net/mlx4/mlx4_flow.c | 30 +
3 files changed, 20 insertions(+), 100 deletions
data path implementation. This one focuses on maintaining basic
functionality in the meantime.
Signed-off-by: Moti Haimovsky
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 120 ++-
drivers/net/mlx4/mlx4.h | 4 +-
2 files changed, 52 inserti
data path implementation. This one focuses on maintaining basic
functionality in the meantime.
Signed-off-by: Moti Haimovsky
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 127 +++
drivers/net/mlx4/mlx4.h | 2 -
2 files changed, 33 insertions(+), 96
).
Standardize on negative errno values and rely on rte_errno instead of
errno in all functions.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 485 ---
1 file changed, 274 insertions(+), 211 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b
Returning a different value when the current link status differs from the
previous one was probably useful at some point in the past but is now
meaningless; this value is ignored both internally (mlx4 PMD) and
externally (ethdev wrapper).
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4
. This adds unnecessary complexity.
Leave this role to wrapper functions in ethdev.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 90 +++
drivers/net/mlx4/mlx4.h | 4 --
drivers/net/mlx4/mlx4_flow.c | 15 +--
3 files changed, 6
This addresses badly formatted comments and needless empty lines before
refactoring functions into different files.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 89 ---
drivers/net/mlx4/mlx4_flow.c | 1 -
2 files changed, 36 insertions
Wrapper functions whose main purpose was to take a lock on the private
structure are no longer needed since this lock does not exist anymore.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 61 --
drivers/net/mlx4/mlx4_flow.c | 106
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 11 ++-
drivers/net/mlx4/mlx4.h | 3 ---
2 files changed, 6 insertions(+), 8 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index 0f1169c..f4dc67f 100644
--- a/drivers/net/mlx4/mlx4.c
+++ b
These were originally used for compatibility between DPDK releases when
this PMD was built out of tree.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 29 ++---
1 file changed, 10 insertions(+), 19 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b
-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 32
drivers/net/mlx4/mlx4.h | 3 +--
2 files changed, 21 insertions(+), 14 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index 07a47ea..7fc9b4c 100644
--- a/drivers/net/mlx4/mlx4.c
interrupts (priv->intr_conf).
This complexity can be avoided entirely since the PMD does not disable
interrupts configuration parameters in case of error anymore.
With this commit, only two functions are necessary to toggle interrupts
(including Rx) during start/stop cycles.
Signed-off-by: Adr
emoves the unnecessary dev argument which can
be retrieved through the private structure where needed. Documentation is
updated accordingly.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 145 ---
1 file changed, 55 insertions(+), 90 deletions(-)
The new definitions also rely on the existing DPDK logging subsystem
instead of using fprintf() directly.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 2 +-
drivers/net/mlx4/mlx4.h | 46 ---
drivers/net/mlx4/mlx4_flow.c | 1 +
drivers/net/mlx4
Make clear this field is related to interrupt handling.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 14 +++---
drivers/net/mlx4/mlx4.h | 6 +++---
2 files changed, 10 insertions(+), 10 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index
: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 1 +
drivers/net/mlx4/mlx4.c | 63 ++--
drivers/net/mlx4/mlx4.h | 4 +++
drivers/net/mlx4/mlx4_utils.c | 66 ++
drivers/net/mlx4/mlx4_utils.h | 4 +++
5 files
This commit groups all data plane functions (Rx/Tx) into a separate file
and adjusts header files accordingly.
Private functions are now prefixed with "mlx4_" to prevent them from
conflicting with their mlx5 PMD counterparts at link time.
No impact on functionality.
Signed-off-
Private functions are now prefixed with "mlx4_" to prevent them from
conflicting with their mlx5 PMD counterparts at link time.
No impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 1 +
drivers/net/mlx4/mlx4.c
Except for a minor documentation update on internal structure definitions
to make them more Doxygen-friendly, there is no impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 13 +---
drivers/net/mlx4/mlx4.h | 69 +---
drivers/net
Private functions are now prefixed with "mlx4_" to prevent them from
conflicting with their mlx5 PMD counterparts at link time.
No impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile| 1 +
drivers/net/mlx4/mlx4.c
Private functions are now prefixed with "mlx4_" to prevent them from
conflicting with their mlx5 PMD counterparts at link time.
No impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile| 1 +
drivers/net/mlx4/mlx4.c
Private functions are now prefixed with "mlx4_" to prevent them from
conflicting with their mlx5 PMD counterparts at link time.
No impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile| 1 +
drivers/net/mlx4/mlx4.c
Only the common filter control operation callback needs to be exposed.
No impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 48 +-
drivers/net/mlx4/mlx4_flow.c | 72 ---
drivers/net/mlx4
No impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 1 +
drivers/net/mlx4/mlx4.c| 115 -
drivers/net/mlx4/mlx4.h| 8 +-
drivers/net/mlx4/mlx4_mr.c | 183
4 files changed, 188
While internal static functions do not cause link time conflicts, this
differentiates them from their mlx5 PMD counterparts while debugging.
No impact on functionality.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 4 ++--
drivers/net/mlx4/mlx4_flow.c | 30
Add missing includes and sort them, then update/remove comments around them
for consistency.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 40 ---
drivers/net/mlx4/mlx4.h | 3 +--
drivers/net/mlx4/mlx4_flow.c | 5 +
drivers/net
On Tue, Aug 01, 2017 at 05:35:30PM +, Legacy, Allain wrote:
> > -Original Message-
> > From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
> > Sent: Tuesday, August 01, 2017 12:54 PM
> <...>
> > @@ -5946,12 +5949,11 @@ mlx4_arg_parse(const cha
mmunity or could
> > > you just share it with me via an e-mail ??
> > >
> > > [1] http://dpdk.org/ml/archives/users/2017-April/001786.html
> > >
> > > Thanks,
> > >
> > > Tetsuro
> > >
> > > # re-sending the e-mail because I failed to submit the e-mail to
> > > us...@dpdk.org.
> > >
> > > --
> > > Tetsuro Nakamura NTT Network Service
> > > Systems Laboratories
> > > TEL:0422 59 6914(National)/+81 422 59 6914(International) 3-9-11,
> > > Midori-Cho Musashino-Shi, Tokyo 180-8585 Japan
> > >
>
--
Adrien Mazarguil
6WIND
[-Werror,-Wc11-extensions]
Fixes: 5d109deffa87 ("ethdev: add traffic management API")
Cc: Cristian Dumitrescu
Signed-off-by: Adrien Mazarguil
---
lib/librte_ether/rte_tm.h | 5 +
1 file changed, 5 insertions(+)
diff --git a/lib/librte_ether/rte_tm.h b/lib/librte_ether/rte_tm.h
ind
AEAD parameters in crypto operation")
Cc: Pablo de Lara
Signed-off-by: Adrien Mazarguil
---
lib/librte_cryptodev/rte_crypto_sym.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_cryptodev/rte_crypto_sym.h
b/lib/librte_cryptodev/rte_crypto_sym.h
index 58a33b8..0ceaa91 100644
des of strtoul() is
useless, just that it seems overkill in this specific case. Using atoi()
without any error checking would have been perfectly fine as well.
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
would be more complicated than the current
> > approach of leaving the payload type field unspecified or set it to some
> > custom value that PMDs may or may not accept depending on their
> > capabilities.
>
> You're right. My comment was based on the misconception that it was mandatory
> to correctly specify ether_types / next_protocol_id / ...
Well thanks to that you've raised an interesting issue with the VLAN item
(TBH Wenzhuo and other people warned me about that, at the time I was
certain it would not be a problem.) I'll attempt to address it as soon as
possible.
Best regards,
--
Adrien Mazarguil
6WIND
ced.
>
> Fixes: c4da6caa426d ("mlx4: handle link status interrupts")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Gaetan Rivet
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
Hi Ferruh,
On Mon, Mar 20, 2017 at 01:24:36PM +, Ferruh Yigit wrote:
> On 3/16/2017 11:04 AM, Adrien Mazarguil wrote:
> > On Fri, Mar 03, 2017 at 04:40:06PM +0100, Gaetan Rivet wrote:
> >> Most ConnectX-3 adapters expose two physical ports on a single PCI bus
> >> a
{
> + /**
> + * Lable (20b), TC (3b), Bottom of Stack (1b), TTL (8b).
Typo, "Lable".
> + */
> + uint32_t label_tc_s_ttl;
> +};
> +
> +/** Default mask for RTE_FLOW_ITEM_TYPE_MPLS. */
> +static const struct rte_flow_item_mpls rte_flow_item_mpls_mask = {
> + .label_tc_s_ttl = 0xf,
> +};
This default mask is wrong, it has to be specified in network order (you can
include rte_byteorder.h if you need some #ifdef).
> +
> +/**
> + * RTE_FLOW_ITEM_TYPE_GRE.
> + *
> + * Matches a GRE header.
> + */
> +struct rte_flow_item_gre {
> + /**
> + * Checksum (1b), reserved 0 (12b), version (3b).
> + * Refer to RFC 2784.
> + */
> + uint16_t c_rsvd0_ver;
> + uint16_t protocol; /**< Protocol type. */
> +};
Default mask is missing, you must add one.
> +
> +/**
> * Matching pattern item definition.
> *
> * A pattern is formed by stacking items starting from the lowest protocol
> --
> 2.5.5
>
--
Adrien Mazarguil
6WIND
t does not allow different rule types.
If after initialization, matching TCP, UDP and ICMP is possible, creating a
UDP rule might subsequently prevent the creation of otherwise valid TCP and
ICMP rules. rte_flow_validate() should (but is not forced to) check for
that.
What do you think about keeping the defined error codes as is and merging
somehow my above statements in the documentation instead?
--
Adrien Mazarguil
6WIND
201 - 300 of 1899 matches
Mail list logo
