://www.openfabrics.org/
[2]
http://www.mellanox.com/page/products_dyn?product_family=26=linux_sw_drivers
v2:
- Include minor bugfix for VLAN filtering.
- Add maintainers entry.
- Add documentation.
Adrien Mazarguil (4):
scripts: add auto-config-h.sh
mlx4: new poll mode driver
maintainers: claim
This script looks for types, macros and functions in header files using
compilation options found in the environment (CC, CFLAGS, CPPFLAGS) to
define feature macros in a generated header.
Useful in combination with external headers that do not provide such macros.
Signed-off-by: Adrien Mazarguil
userland at the moment partly because /sys (sysfs) support is
required.
Signed-off-by: Adrien Mazarguil
Signed-off-by: Olga Shern
---
config/common_bsdapp | 11 +
config/common_linuxapp | 11 +
lib/Makefile |1 +
lib/librte_pmd_mlx4/Makefile | 119 ++
lib
Signed-off-by: Adrien Mazarguil
---
MAINTAINERS | 4
1 file changed, 4 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 7ac6d59..b4328d6 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -217,6 +217,10 @@ Intel fm10k
M: Jing Chen
F: lib/librte_pmd_fm10k/
+Mellanox mlx4
+M: Adrien
This documentation covers implementation details, features and limitations,
configuration, prerequisites and provides a usage example.
Signed-off-by: Adrien Mazarguil
---
doc/guides/prog_guide/index.rst | 1 +
doc/guides/prog_guide/mlx4_poll_mode_drv.rst | 327
.
- Remove unwanted executable bits.
- Fix coding style and typos found by checkpatch.
- Add shared library compilation support.
Adrien Mazarguil (3):
scripts: check features to generate configuration header
mlx4: new poll mode driver
doc: add librte_pmd_mlx4 documentation
MAINTAINERS
This script looks for types, macros and functions in header files using
compilation options found in the environment (CC, CFLAGS, CPPFLAGS) to
define feature macros in a generated header.
Useful in combination with external headers that do not provide such macros.
Signed-off-by: Adrien Mazarguil
userland at the moment partly because /sys (sysfs) support is
required.
Also claim responsibility in the MAINTAINERS file.
Signed-off-by: Adrien Mazarguil
Signed-off-by: Olga Shern
---
MAINTAINERS |4 +
config/common_bsdapp | 11
This documentation covers implementation details, features and limitations,
configuration, prerequisites and provides a usage example.
Signed-off-by: Adrien Mazarguil
---
MAINTAINERS | 1 +
doc/guides/prog_guide/index.rst | 1 +
doc/guides
ion.
In short, unload mlx4_core, mlx4_en, mlx4_ib and ib_uverbs, install this
package, run depmod and load them again. Since they have similar names, the
updated kernel modules should be found in a "extra" or "updates"
subdirectory of /lib/modules and come first when queried by modinfo.
[1]
http://www.mellanox.com/page/products_dyn?product_family=26=linux_sw_drivers
--
Adrien Mazarguil
6WIND
ce "mlx4_1" (VF:
false)
PMD: librte_pmd_mlx4: cannot use device, are drivers up to date?
EAL: No probed ethernet devices
[...]
Signed-off-by: Adrien Mazarguil
---
lib/librte_pmd_mlx4/mlx4.c | 14 --
1 file changed, 12 insertions(+), 2 deletions(-)
diff --git a/lib/librte_p
types")
>
> Signed-off-by: Thomas Monjalon
Makes sense.
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
This patchset adds compatibility with the upcoming Mellanox OFED 3.0
release (new kernel drivers and userland support libraries), which supports
new features such as L3/L4 checksum validation offloads and addresses
several bugs and limitations at the same time.
Adrien Mazarguil (3):
mlx4: add
Since Mellanox OFED 3.0 and Linux 3.15, interface port numbers are stored
in dev_port instead of dev_id sysfs files.
Signed-off-by: Or Ami
Signed-off-by: Nitzan Weller
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 51 +
1 file
This API implements additional flags in work completions that are required
to support checksum offloads.
Signed-off-by: Gilad Berman
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 22 +++---
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/drivers
From: Olga Shern <ol...@mellanox.com>
HAVE_EXP_QUERY_DEVICE is used to check whether ibv_exp_query_device() can be
used. RSS and inline receive features depend on it.
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 4
drivers/net/mlx4/
From: Gilad Berman <gil...@mellanox.com>
Mellanox ConnectX-3 adapters can handle L3 (IPv4) and L4 (TCP, UDP, TCP6,
UDP6) RX checksums validation, with and without 802.1Q (VLAN) headers.
Signed-off-by: Gilad Berman
Signed-off-by: Adrien Mazarguil
Acked-by: Guillaume Gaudonville
---
d
Depending on adapters features and VXLAN support in the kernel, VXLAN frames
can be automatically recognized, in which case checksum validation occurs on
inner and outer L3 and L4.
Signed-off-by: Adrien Mazarguil
Acked-by: Guillaume Gaudonville
---
drivers/net/mlx4/mlx4.c | 34
From: Alex Rosenbaum <al...@mellanox.com>
Replace ibv_exp_poll_cq() with direct function call to improve performance.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 13 +++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff
From: Olga Shern <ol...@mellanox.com>
This feature is now also supported in VMs.
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/
From: Alex Rosenbaum <al...@mellanox.com>
This is done by storing the current index in the RX queue structure.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 24 ++--
1 file changed, 22 insertions(+), 2 deletions(-)
diff
From: Alex Rosenbaum <al...@mellanox.com>
Make rxq_setup_qp() handle inline support like rxq_setup_qp_rss() instead of
having two separate functions.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.
From: Olga Shern <ol...@mellanox.com>
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index f7186fa..cc16e8c 100644
--- a/drivers/net/mlx4/
ed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 44 +---
1 file changed, 25 insertions(+), 19 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index cc16e8c..f9faeb0 100644
--- a/drivers/net/mlx
From: Alex Rosenbaum <al...@mellanox.com>
This commit makes librte_pmd_mlx4 support both the extended Verbs API from
upstream and the original experimental Verbs API.
Signed-off-by: Olga Shern
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefil
From: Or Ami <o...@mellanox.com>
The number of descriptors must be a multiple of MLX4_PMD_SGE_WR_N.
Signed-off-by: Or Ami
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drive
From: Or Ami <o...@mellanox.com>
Starting from MLNX_OFED 3.0 FW 2.34.5000 when working with optimized
steering mode (-7) QPs can be attached to the port's MAC, therefore no need
for the check.
Signed-off-by: Or Ami
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drive
ecifications per configured MAC address.
Signed-off-by: Olga Shern
Signed-off-by: Or Ami
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 174
1 file changed, 115 insertions(+), 59 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/
ed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 34 ++
1 file changed, 18 insertions(+), 16 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index 4c4f693..04cc5e1 100644
--- a/drivers/net/mlx4/mlx4.c
+++ b/drivers/net/mlx4/
D(wr_id).offset);
> struct rte_mbuf *rep;
>
> /* Sanity checks. */
> @@ -2891,7 +2895,8 @@ rxq_rehash(struct rte_eth_dev *dev, struct rxq *rxq)
> for (i = 0; (i != elemof(*elts)); ++i) {
> struct rxq_elt *elt = &(
t that, as a PMD's Makefile is obviously not
the right place to add a -pedantic parameter anyway.
However outside of PMD usage, I think public API headers (I'm not talking
about the entire DPDK code base, just headers) should handle all kind of
warnings a user application might throw at it for its own
Following John McNamara's changes we've noticed that mlx4 release notes were
missing from DPDK 2.0.0. This patchset adds them retroactively and also
provides mlx4 release notes for DPDK 2.1.0.
Adrien Mazarguil (2):
doc: add missing release notes for mlx4 availability in DPDK 2.0.0
doc: add
Signed-off-by: Adrien Mazarguil
---
doc/guides/rel_notes/release_2_0.rst | 6 ++
1 file changed, 6 insertions(+)
diff --git a/doc/guides/rel_notes/release_2_0.rst
b/doc/guides/rel_notes/release_2_0.rst
index 4341a0c..62737ae 100644
--- a/doc/guides/rel_notes/release_2_0.rst
+++ b/doc
Signed-off-by: Adrien Mazarguil
---
doc/guides/rel_notes/release_2_1.rst | 20
1 file changed, 20 insertions(+)
diff --git a/doc/guides/rel_notes/release_2_1.rst
b/doc/guides/rel_notes/release_2_1.rst
index c39418c..00edd4b 100644
--- a/doc/guides/rel_notes/release_2_1.rst
n commitlog since Thomas does not like "safer"
> argument :-)
Obvious issue that did not show up in our tests under Debian.
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
>
> Signed-off-by: Thomas Monjalon
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
rte_cpuflags.h.
Fixes: ba7468997ea6 ("spinlock: add HTM lock elision for x86")
Signed-off-by: Adrien Mazarguil
---
lib/librte_eal/common/include/generic/rte_cpuflags.h | 4
1 file changed, 4 insertions(+)
diff --git a/lib/librte_eal/common/include/generic/rte_cpuflags.h
b/lib/
when allocating mbufs.
- Several API changes following the release of Mellanox OFED 3.0.
- Performance improvements made possible by the new API.
- Add TX checksum offloads.
- Update documentation to reflect the changes.
Adrien Mazarguil (6):
mlx4: fix possible crash on scattered mbuf
When failing to allocate a segment, mlx4_rx_burst_sp() may call
rte_pktmbuf_free() on an incomplete scattered mbuf whose next pointer in the
last segment is not set.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git
Since Mellanox OFED 3.0 and Linux 3.15, interface port numbers are stored
in dev_port instead of dev_id sysfs files.
Signed-off-by: Or Ami
Signed-off-by: Nitzan Weller
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 51 +
1 file
From: Olga Shern <ol...@mellanox.com>
HAVE_EXP_QUERY_DEVICE is used to check whether ibv_exp_query_device() can be
used. RSS and inline receive features depend on it.
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 4
drivers/net/mlx4/
From: Alex Rosenbaum <al...@mellanox.com>
This is done by storing the current index in the RX queue structure.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 24 ++--
1 file changed, 22 insertions(+), 2 deletions(-)
diff
From: Alex Rosenbaum <al...@mellanox.com>
Make rxq_setup_qp() handle inline support like rxq_setup_qp_rss() instead of
having two separate functions.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.
From: Olga Shern <ol...@mellanox.com>
Although using the PMD from a forked process is still unsupported, this
commit makes Verbs safe enough for applications to call fork() for other
purposes.
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4
ed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 44 +---
1 file changed, 25 insertions(+), 19 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index ed68beb..02dd894 100644
--- a/drivers/net/mlx
From: Alex Rosenbaum <al...@mellanox.com>
This commit drops "exp" from related function and type names to stop using
the experimental API.
Signed-off-by: Olga Shern
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/
From: Or Ami <o...@mellanox.com>
The number of descriptors must be a multiple of MLX4_PMD_SGE_WR_N.
Signed-off-by: Or Ami
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drive
From: Or Ami <o...@mellanox.com>
Starting from MLNX_OFED 3.0 FW 2.34.5000 when working with optimized
steering mode (-7) QPs can be attached to the port's MAC, therefore no need
for the check.
Signed-off-by: Or Ami
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drive
ecifications per configured MAC address.
Signed-off-by: Olga Shern
Signed-off-by: Or Ami
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 174
1 file changed, 115 insertions(+), 59 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/
ed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 34 ++
1 file changed, 18 insertions(+), 16 deletions(-)
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index 37aca55..cdc679a 100644
--- a/drivers/net/mlx4/mlx4.c
+++ b/drivers/net/mlx4/
From: Alex Rosenbaum <al...@mellanox.com>
This commit replaces the CQ polling and QP posting functions
(mlx4_rx_burst() only) with a new low level interface to improve
performance.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Gilad Berman
Signed-off-by: Adrien Mazarguil
---
drivers/ne
From: Alex Rosenbaum <al...@mellanox.com>
Instead of requesting a completion event for each TX burst, request it on a
fixed schedule once every MLX4_PMD_TX_PER_COMP_REQ (currently 64) packets to
improve performance.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drive
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 4 --
drivers/net/mlx4/mlx4.c | 167 +++---
2 files changed, 85 insertions(+), 86 deletions(-)
diff --git a/drivers/net/mlx4/Makefile b/drivers/net/mlx4/Makefile
index ce1f2b0..fd74
This commit makes scattered TX support entirely optional by moving it to a
separate function that is only available when MLX4_PMD_SGE_WR_N > 1.
Improves performance when scattered support is not needed.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx
g SGEs. For scattered mbufs, SGEs are
allocated on the stack and passed to send_pending_sg_list().
Signed-off-by: Alex Rosenbaum
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 244 +---
1 file changed, 84 insertions(+), 160 deletions(-)
diff
From: Gilad Berman <gil...@mellanox.com>
Mellanox ConnectX-3 adapters can handle L3 (IPv4) and L4 (TCP, UDP, TCP6,
UDP6) RX checksums validation and TX checksums generation, with and without
802.1Q (VLAN) headers.
Signed-off-by: Gilad Berman
Signed-off-by: Olga Shern
Signed-off-by:
Depending on adapters features and VXLAN support in the kernel, VXLAN frames
can be automatically recognized, in which case checksum validation and
generation occurs on inner and outer L3 and L4.
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/mlx4.c | 49
From: Alex Rosenbaum <al...@mellanox.com>
RDs are a new feature of MOFED 3.0 that makes Verbs aware of how CQ and QP
resources are being used for internal performance tuning.
Signed-off-by: Alex Rosenbaum
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/
From: Olga Shern <ol...@mellanox.com>
Multicast loopback must be disabled on PF devices to prevent the adapter
from sending frames back. Required with MOFED 3.0.
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 5 +
drivers/net/mlx4/mlx4.c
.
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/mlx4.rst | 29 ++---
1 file changed, 6 insertions(+), 23 deletions(-)
diff --git a/doc/guides/nics/mlx4.rst b/doc/guides/nics/mlx4.rst
index ac2dd56..c33aa38 100644
--- a/doc/guides/nics/mlx4.rst
+++ b/doc/guides/nics/mlx4
t;-pedantic"
> #endif
>
> -/* Verbs header. */
> -/* ISO C doesn't support unnamed structs/unions, disabling -pedantic. */
> -#ifdef PEDANTIC
> -#pragma GCC diagnostic ignored "-pedantic"
> -#endif
> -
> -#include
> -
> -#ifdef PEDANTIC
> -#pragma GCC diagnostic error "-pedantic"
> -#endif
> -
> /* Generated configuration header. */
> #include "mlx4_autoconf.h"
>
> --
> 2.2.2
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
NE_RECV_SZ
> + enum IBV_EXP_DEVICE_ATTR_INLINE_RECV_SZ $(AUTOCONF_OUTPUT)
> $Q sh -- '$<' '$@' \
> SEND_RAW_WR_SUPPORT \
> infiniband/verbs.h \
> - type 'struct ibv_send_wr_raw'
> + type 'struct ibv_send_wr_raw
steering API), remove it
entirely.
Signed-off-by: Olga Shern
Signed-off-by: Adrien Mazarguil
---
config/common_bsdapp | 1 -
config/common_linuxapp | 1 -
doc/guides/prog_guide/mlx4_poll_mode_drv.rst | 11 -
lib/librte_pmd_mlx4/Makefile
+), 3 deletions(-)
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
c |2 +-
> lib/librte_malloc/malloc_heap.h |2 +-
> lib/librte_malloc/rte_malloc.c |4 ++--
> lib/librte_malloc/rte_malloc.h |2 +-
> 5 files changed, 8 insertions(+), 8 deletions(-)
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
-
> lib/librte_eal/linuxapp/eal/eal.c |4
> 2 files changed, 16 insertions(+), 1 deletions(-)
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
> + m = __rte_mbuf_raw_alloc(mp);
> __rte_mbuf_sanity_check(m, RTE_MBUF_PKT, 1);
> return m;
> }
> --
> 1.7.10.4
Looks good.
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
-by: Adrien Mazarguil
---
doc/guides/nics/mlx4.rst | 85
1 file changed, 78 insertions(+), 7 deletions(-)
diff --git a/doc/guides/nics/mlx4.rst b/doc/guides/nics/mlx4.rst
index b26c219..93a7f57 100644
--- a/doc/guides/nics/mlx4.rst
+++ b/doc/guides
lies on it.
> I would be happy to contribute to the help page with end-to-end
> instructions for using the Mellanox with testpmd.
I've just submitted a patch to provide more details about the installation
process, you should have a look.
--
Adrien Mazarguil
6WIND
{} while (0)
How about defining it as (void)0 instead of an empty do/while block?
Doing so will prevent warnings if this macro happens to be used in an
expression. RTE_LOG() supports it.
> +#endif
>
> /*
> * The set of PCI devices this driver supports
> --
> 2.2.2
--
Adrien Mazarguil
6WIND
On Thu, Apr 09, 2015 at 09:28:53AM -0700, Stephen Hemminger wrote:
> On Thu, 9 Apr 2015 10:32:24 +0200
> Adrien Mazarguil wrote:
>
> > >
> > > +#ifdef RTE_LIBRTE_ENIC_DEBUG
> > > #define ENICPMD_FUNC_TRACE() \
> > > RTE_LOG(DEBUG, PMD,
This field is not supposed to contain the RX queue index. Applications can
rely on it to determine the port a given mbuf comes from.
Signed-off-by: Adrien Mazarguil
---
lib/librte_pmd_enic/enic.h| 1 +
lib/librte_pmd_enic/enic_ethdev.c | 1 +
lib/librte_pmd_enic/enic_main.c | 4
documentation will be added in V2.
[1] https://www.openfabrics.org/
[2]
http://www.mellanox.com/page/products_dyn?product_family=26=linux_sw_drivers
Adrien Mazarguil (2):
scripts: add auto-config-h.sh
mlx4: new poll mode driver
config/common_bsdapp | 11 +
config/common_linuxapp
This script looks for types, macros and functions in header files using
compilation options found in the environment (CC, CFLAGS, CPPFLAGS) to
define feature macros in a generated header.
Useful in combination with external headers that do not provide such macros.
Signed-off-by: Adrien Mazarguil
userland at the moment partly because /sys (sysfs) support is
required.
Signed-off-by: Adrien Mazarguil
Signed-off-by: Olga Shern
---
config/common_bsdapp | 11 +
config/common_linuxapp | 11 +
lib/Makefile |1 +
lib/librte_pmd_mlx4/Makefile | 119 ++
lib
On Thu, Jan 29, 2015 at 03:37:39PM +, Stephen Hemminger wrote:
> On Thu, 29 Jan 2015 16:20:46 +0100
> Adrien Mazarguil wrote:
>
> > This PMD manages all variants of Mellanox ConnectX-3 (EN 40, EN 10, Pro EN
> > 40) as well as their virtual functions in SR-IOV con
rt from Christian Ehrhardt's)?
Yes, since you're suggesting to prefix driver names, shall "librte_pmd_mlx5"
really become "net_librte_pmd_mlx5" or shortened to "net_mlx5" instead?
What about using a '/' separator instead of '_'?
Will this impact directories as well ("net/mlx5" -> "net/net_mlx5")?
--
Adrien Mazarguil
6WIND
v] [PATCH] doc: announce driver name changes
> >
> > On 7/22/2016 1:54 PM, Adrien Mazarguil wrote:
> > > Hi Pablo,
> > >
> > > On Fri, Jul 22, 2016 at 12:37:22PM +, De Lara Guarch, Pablo wrote:
> > >> Hi,
> > >>
> > >>
. Considering users are not expected to reset the
reference count to satisfy assert() and that raw functions are designed on
purpose without safety belts, remove these checks.
Signed-off-by: Adrien Mazarguil
---
lib/librte_mbuf/rte_mbuf.h | 2 --
1 file changed, 2 deletions(-)
diff --git a/lib
ake these PMDs slower and is not acceptable either.
What remains is the consistency issue, I think these statements were only
added to catch multiple frees, and those should be caught at a higher
level, where other consistency checks are also performed.
> > Signed-off-by: Adrien Mazarguil
&
r names when dealing with mbufs.
> Why instead you choose to change common functions and compromise
> librte_mbuf debug ability?
No, I'm fine with keeping the debug ability, however I did not find a way to
both keep it and fix the consistency issue without breaking something
(performance or r
s in received mbufs, which is
fine but not very useful.
That suggestion is also in line with the recent RX checksum discussion [1],
where basically HW not reporting something about a packet does not mean it
is the opposite.
[1] http://dpdk.org/ml/archives/dev/2016-May/039920.html
> By the way, I'm working on a software implementation that return a
> packet_type from a mbuf. I may need it for virtio offload, but before
> that, I think it could be useful for debug purposes. I'll submit a
> patchset for this in the coming days.
>
> Regards,
> Olivier
>
> PS: sorry, many questions to you these days... answer when you have
> the time ;)
--
Adrien Mazarguil
6WIND
Various BSD ioctl macros are not exposed anymore when _XOPEN_SOURCE is
defined, and linux/if.h now conflicts with net/if.h.
Suggested-by: Bruce Richardson
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 1 +
drivers/net/mlx4/mlx4.c| 7 +++
drivers/net/mlx5
Mellanox PMDs must be rebuilt if a Verbs update would cause the autoconf
file to differ.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 13 +++--
drivers/net/mlx5/Makefile | 13 +++--
2 files changed, 22 insertions(+), 4 deletions(-)
diff --git a/drivers/net
A missing define in the previous patch causes additional compilation
issues.
Fixes: 3025206f5060 ("mlx: fix compilation with recent Glibc")
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 1 +
drivers/net/mlx5/Makefile | 1 +
2 files changed, 2 insertions(+)
priv *priv, enum
> hash_rxq_flow_type type)
> int
> priv_rehash_flows(struct priv *priv)
> {
> - unsigned int i;
> + enum hash_rxq_flow_type i;
>
> - for (i = 0; (i != RTE_DIM((*priv->hash_rxqs)[0].special_flow)); ++i)
> + for (i = HASH_RXQ_FLOW_TYPE_PROMISC;
> + i != RTE_DIM((*priv->hash_rxqs)[0].special_flow); ++i)
Here also.
> if (!priv_allow_flow_type(priv, i)) {
> priv_special_flow_disable(priv, i);
> } else {
> --
> 2.5.5
>
Otherwise that fix looks fine, thanks.
--
Adrien Mazarguil
6WIND
;mlx5: handle link status interrupts")
> Fixes: 0d2186743d62 ("mlx5: manage all special flow types at once")
> Fixes: 612ad38209f7 ("mlx5: fix hash Rx queue type in RSS mode")
> Fixes: 083c2dd31776 ("mlx5: refactor special flows handling")
>
> Signed-off-by: Ferruh Yigit
Acked-by: Adrien Mazarguil
--
Adrien Mazarguil
6WIND
On Fri, Jun 17, 2016 at 05:29:53PM +0100, Ferruh Yigit wrote:
> On 6/14/2016 10:22 AM, Adrien Mazarguil wrote:
> > A missing define in the previous patch causes additional compilation
> > issues.
> >
> > Fixes: 3025206f5060 ("mlx: fix compilation with recent Glibc&
with older
versions.
Suggested-by: Bruce Richardson
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx4/Makefile | 2 ++
drivers/net/mlx4/mlx4.c| 7 +++
drivers/net/mlx5/Makefile | 2 ++
drivers/net/mlx5/mlx5.h| 1 -
drivers/net/mlx5/mlx5_ethdev.c | 7 +++
drivers/net
I'm dropping this patch for now. We'll update our last
mlx5 patchset to do it properly and keep these assert in place.
Thanks.
--
Adrien Mazarguil
6WIND
;
> > Changes in v3:
> > - Rebased patchset on top of next-net/rel_16_07.
> >
> > Changes in v2:
> > - Rebased patchset on top of dpdk/master.
> > - Fixed CQE size on Power8.
> > - Fixed mbuf assertion failure in debug mode.
> > - Fixed missing class_id field
On Wed, Jun 22, 2016 at 10:19:14AM +0100, Bruce Richardson wrote:
> On Wed, Jun 22, 2016 at 10:20:54AM +0200, Adrien Mazarguil wrote:
> > On Tue, Jun 21, 2016 at 05:42:29PM +0100, Ferruh Yigit wrote:
> > > On 6/21/2016 8:23 AM, Nelio Laranjeiro wrote:
> > > >
On Wed, Jun 22, 2016 at 11:30:05AM +0200, Adrien Mazarguil wrote:
> On Wed, Jun 22, 2016 at 10:19:14AM +0100, Bruce Richardson wrote:
> > On Wed, Jun 22, 2016 at 10:20:54AM +0200, Adrien Mazarguil wrote:
> > > On Tue, Jun 21, 2016 at 05:42:29PM +0100, Ferruh Yigit wrote:
> &g
On Mon, Jun 27, 2016 at 01:12:15PM +0100, Bruce Richardson wrote:
> On Fri, Jun 24, 2016 at 03:18:01PM +0200, Nelio Laranjeiro wrote:
> > From: Adrien Mazarguil
> >
> > Since commit "mlx5: resurrect Tx gather support", older GCC versions (such
> > as 4.
these modes.
- Updated comments about maximum number of MAC addresses and VLAN filters.
- Improved performance with better prefetching.
- Fixed deadlock in case of error during port start.
- Simplified VLAN filtering configuration storage using a basic list instead
of a table (with holes).
Adrien
In its current state, this driver implements the bare minimum to initialize
itself and Mellanox ConnectX-4 adapters without doing anything else
(no RX/TX for instance). It is disabled by default since it is based on the
mlx4 driver and also depends on libibverbs.
Signed-off-by: Adrien Mazarguil
RSS implementation with parent/child QPs comes from mlx4 and is temporary.
Signed-off-by: Adrien Mazarguil
Signed-off-by: Nelio Laranjeiro
---
config/common_bsdapp | 3 +
config/common_linuxapp| 3 +
drivers/net/mlx5/Makefile | 15 +
drivers/net/mlx5/mlx5.c
This commit adds support for MAC flow steering rules mandatory for the RX
path as well as the related callbacks to add/remove MAC addresses.
Signed-off-by: Adrien Mazarguil
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Didier Pallard
---
drivers/net/mlx5/mlx5.c | 4 +-
drivers/net
This commit adds the remaining missing callbacks to make mlx5 usable.
Like mlx4, device start and stop are implemented on top of MAC RX flows.
Signed-off-by: Adrien Mazarguil
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Francesco Santoro
Signed-off-by: Didier Pallard
---
drivers/net/mlx5
in a given mbuf, the
remaining segments are linearized in the last SGE entry.
Signed-off-by: Adrien Mazarguil
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxq.c | 175 +++-
drivers/net/mlx5/mlx5_rxtx.c | 376 +++
drivers/net
1 - 100 of 486 matches
Mail list logo