Thanks for your reply .
The issue coming in function "vmxnet3_dev_tx_queue_setup"
below if condition is true because "tx_conf->txq_flags is coming as 0"
if ((tx_conf->txq_flags & ETH_TXQ_FLAGS_NOXSUMSCTP) !=
ETH_TXQ_FLAGS_NOXSUMSCTP) {
PMD_INIT_LOG(ERR, "SCTP checksum offload
> -Original Message-
> From: Ferruh Yigit [mailto:ferruh.yi...@intel.com]
> Sent: Saturday, October 06, 2018 3:50 AM
> To: Bruce Richardson ; John McNamara
> ; Marko Kovacevic
>
> Cc: dev@dpdk.org; Ferruh Yigit ; Juhamatti
> Kuusisaari ; Juhamatti Kuusisaari
>
> Subject: [PATCH v8] net
> > > On 07/10/2018, 06:03, "Jerin Jacob"
> wrote:
> > >
> > > In arm64 case, it will have ATOMIC_RELAXED followed by asm
> > volatile
> ("":::"memory") of rte_pause().
> > > I would n't have any issue, if the generated code code is same or
> better than the exiting case.
The new ethdev iterator lacks the support of filtering
by representor port id or by MAC address.
This series is not fully tested yet.
Thomas Monjalon (4):
kvargs: support list value
mk: remove broken check
ethdev: move representor parsing functions
ethdev: support representor id for itera
If a value contains a comma, rte_kvargs_tokenize() will split here.
In order to support list syntax [a,b] as value, an extra parsing of
the square brackets is added.
Signed-off-by: Thomas Monjalon
---
lib/librte_kvargs/rte_kvargs.c | 10 ++
test/test/test_kvargs.c| 17 +++
Next patch does not pass the experimental check
because it has an (unrelated) experimental function in his context.
Cc: Neil Horman
Signed-off-by: Thomas Monjalon
---
mk/internal/rte.compile-pre.mk | 1 -
1 file changed, 1 deletion(-)
diff --git a/mk/internal/rte.compile-pre.mk b/mk/internal/
The representor id is added in rte_eth_dev_data in order to be able
to match a port with its representor id in devargs.
Signed-off-by: Thomas Monjalon
---
drivers/net/i40e/i40e_vf_representor.c | 1 +
drivers/net/ixgbe/ixgbe_vf_representor.c | 1 +
drivers/net/mlx5/mlx5.c |
The functions for representor devargs parsing were static
in the file rte_ethdev.c.
In order to reuse them in the file rte_class_eth.c,
they are moved to the files ethdev_private.c/.h.
Signed-off-by: Thomas Monjalon
---
lib/librte_ethdev/ethdev_private.c | 82
lib/li
Any other error messages? There's really only one path out of
slave_configure() that doesn't emit another error message and
that's slave_configure_slow_queue. You need to set dedicated_queues to
be enabled to see that happen.
On 10/07/18 01:12, chetan bhasin wrote:
Can anybody suggest? Stuc
If no rte_device is given in the iterator,
eth_dev_match() is looking at all ports without any restriction,
except the ethdev kvargs filter.
It allows to iterate with a devargs filter referencing only
some ethdev parameters. The format (from the new devargs syntax) is:
class=eth,paramY=Y
The provided example of doxygen header is about a deprecated function.
It is replaced by rte_spinlock_trylock() which is small and
good enough for the purpose.
Signed-off-by: Thomas Monjalon
---
doc/guides/contributing/documentation.rst | 15 +--
1 file changed, 5 insertions(+), 10 d
These hotplug functions were deprecated and have some new replacements.
As announced earlier, the oldest ones are now removed.
Signed-off-by: Thomas Monjalon
---
doc/guides/rel_notes/deprecation.rst| 5 ---
doc/guides/rel_notes/release_18_11.rst | 6 +++
lib/librte_eal/common/eal_common_d
The hotplug attach/detach features are implemented in EAL layer.
There is a new ethdev iterator to retrieve ports from ethdev layer.
As announced earlier, the (buggy) ethdev functions are now removed.
Signed-off-by: Thomas Monjalon
---
app/test-pmd/testpmd.c| 17 ++-
..
A virtual device can be matched with following syntax:
bus=vdev,name=X
Signed-off-by: Thomas Monjalon
---
drivers/bus/vdev/vdev_params.c | 19 +--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/drivers/bus/vdev/vdev_params.c b/drivers/bus/vdev/vdev_params.
The iterator will return the ethdev port ids matching a devargs string.
It is recommended to use the macro RTE_ETH_FOREACH_MATCHING_DEV()
for usage convenience.
The class string is prefixed with '+' in order to skip the validation
of the parameter keys. It is tolerated for the compatibility with
t
The functions for EAL attach/detach had already some replacements,
so they are removed.
The functions for ethdev attach/detach are removed and replaced
thanks to a new ethdev iterator working with devargs.
rte_eth_dev_attach(devargs, &port_id)
is replaced by:
rte_dev_probe(devargs);
RT
> -Original Message-
> From: Dumitrescu, Cristian
> Sent: Tuesday, October 09, 2018 0:00
> To: Xu, Rosen ; dev@dpdk.org
> Cc: Lu, Wenzhuo ; Singh, Jasvinder
> ; jerin.ja...@caviumnetworks.com; Yigit,
> Ferruh
> Subject: RE: [PATCH] lib/librte_ethdev: expand queue threshold size of RED
> pa
08/10/2018 14:52, Ferruh Yigit:
> On 10/8/2018 5:32 AM, Xiao Wang wrote:
> > The typedef of "__virtio16" is introduced into Linux kernel in v3.19.
> > To prevent build error on old kernel, this patch replaces the
> > "__virtio" usage with "uint16_t".
> >
> > Fixes: d7fe5a2861e7 ("net/ifc: support
Hi Everyone,
I would like to request comments on DFS as I presented at the DPDK summit in
Ireland.
As we know DPDK can be difficult to manage at run time and as DPDK becomes more
dynamic we need to address how to configure and monitor DPDK and DPDK-based
applications. A possible solution is t
All other visible functions in failsafe driver have 'failsafe_'
prefix.
Signed-off-by: Stephen Hemminger
---
drivers/net/failsafe/failsafe_ether.c | 2 +-
drivers/net/failsafe/failsafe_private.h | 4 ++--
drivers/net/failsafe/failsafe_rxtx.c| 2 +-
3 files changed, 4 insertions(+), 4 delet
On Mon, Oct 08, 2018 at 07:22:03PM +0800, Xiaoyu Min wrote:
> On 18-10-02 04:19:00, Yongseok Koh wrote:
> > On Sun, Sep 30, 2018 at 03:21:04PM +0800, Xiaoyu Min wrote:
> > > On 18-09-29 07:03:33, Yongseok Koh wrote:
> > > > On Tue, Sep 25, 2018 at 07:51:06PM +0800, Xiaoyu Min wrote:
> > > > > Offlo
On Mon, 8 Oct 2018 17:11:44 +0800
Phil Yang wrote:
> diff --git a/lib/librte_kni/rte_kni_fifo.h b/lib/librte_kni/rte_kni_fifo.h
> index ac26a8c..70ac14e 100644
> --- a/lib/librte_kni/rte_kni_fifo.h
> +++ b/lib/librte_kni/rte_kni_fifo.h
> @@ -28,8 +28,9 @@ kni_fifo_put(struct rte_kni_fifo *fifo,
On Sun, 7 Oct 2018 11:32:39 +0200
Thomas Monjalon wrote:
> This is a follow-up of an idea presented at Dublin
> during the "hotplug talk".
>
> Instead of changing the existing hotplug functions, as in the RFC,
> some new experimental functions are added.
> The old functions lose their experimen
On Mon, Oct 08, 2018 at 06:57:49AM -0700, Ori Kam wrote:
>
>
> > -Original Message-
> > From: dev On Behalf Of Dekel Peled
> > Sent: Monday, October 8, 2018 4:35 PM
> > To: Yongseok Koh ; Shahaf Shuler
> >
> > Cc: dev@dpdk.org; Ori Kam
> > Subject: [dpdk-dev] [PATCH] net/mlx5: fix calc
On Sun, Oct 07, 2018 at 05:25:08PM +0300, Dekel Peled wrote:
> This patch adds support for Direct Verbs decap operations, L2 and L3.
>
> Signed-off-by: Dekel Peled
Same comment for the order of 'dev' arg.
> Conflicts:
> drivers/net/mlx5/mlx5_flow_dv.c
What is this? Conflicts?
> ---
> d
Only send notification to mpu when crossing 64 index boundary.
Signed-off-by: Ed Czeck
---
drivers/net/ark/ark_ethdev_rx.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ark/ark_ethdev_rx.c b/drivers/net/ark/ark_ethdev_rx.c
index 77149f6..e019b9d 100644
--
Support for version 2 of DDM hardware.
Add configurable setup for using DDM user meta data via
CONFIG_RTE_LIBRTE_ARK_DDM_USER_META
defaults to 0 for the meta data compatible with v1 hardware.
Set to 1 use uudata64.
Signed-off-by: Ed Czeck
---
config/common_base | 1 +
drivers/net/a
Fail with rte_exit() when the lack of mbufs is not recoverable.
Signed-off-by: Ed Czeck
---
drivers/net/ark/ark_ethdev_rx.c | 52 -
1 file changed, 46 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ark/ark_ethdev_rx.c b/drivers/net/ark/ark_ethdev_
On Sun, Oct 07, 2018 at 05:25:07PM +0300, Dekel Peled wrote:
> This patch adds support for Direct Verbs encap operations, L2 and L3.
>
> Signed-off-by: Dekel Peled
> ---
For flow_dv_create_encap(), flow_dv_create_encap_l3() and
flow_dv_create_action(), please make the 'dev' come first in the arg
On Sun, Oct 07, 2018 at 05:25:06PM +0300, Dekel Peled wrote:
> This patch adds the required definitions for DV encap/decap actions.
> It also adds usage of the new actions definition in validation
> function of existing drop operation.
>
> Signed-off-by: Dekel Peled
> ---
You may have to rebase
08/10/2018 17:59, Trahe, Fiona:
> Hi Akhil, Joseph, Thomas,
> Just spotted this now.
> See below.
>
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > 24/09/2018 13:36, Joseph, Anoob:
> > > Hi Fiona,
> > >
> > > Can you please comment on this?
> > >
> > > We are adding all capabilities of oc
On Mon, Oct 8, 2018 at 9:03 PM Yongseok Koh wrote:
> 1) VLAN modify isn't supported by driver.
>
>
OVS doesn't support it also @ the DP level, modify is translated to DP
pop + push
How it is done on OVS/AVS-DPDK?
> 2) FW syndrome (0xA9C090):
> set_flow_table_entry: push vlan action ft
On Sun, Oct 07, 2018 at 05:25:05PM +0300, Dekel Peled wrote:
> This patch adds glue functions for operations:
> - Create packet reformat (encap/decap) flow action.
> - Destroy flow action.
>
> Signed-off-by: Dekel Peled
> ---
I'm not still able to compile this patch on my host, which is RH7.4 wi
On Sun, Oct 07, 2018 at 05:01:05PM +0300, Dekel Peled wrote:
> This patch complements [1], adding to MLX5 PMD the option to set
> flow rule for egress traffic.
>
> [1] "net/mlx5: support metadata as flow rule criteria"
> http://mails.dpdk.org/archives/dev/2018-September/113275.html
>
> Signed
rte_flow has 'group' attribute and 'jump' action in order to support
multiple groups. This feature is known as multi-table support ('chain' in
linux TC flower) in general because a group means a table of flows. Example
commands are:
flow create 0 transfer priority 1 ingress
pa
Fixes: 4f07e13d6af5 ("net/mlx5: split flow validation to dedicated function")
Fixes: f7adfffa3de1 ("net/mlx5: add Direct Verbs validation function")
Fixes: edcdef4e5fe4 ("net/mlx5: add Linux TC flower driver for E-Switch flow")
Cc: or...@mellanox.com
Signed-off-by: Yongseok Koh
---
drivers/net/m
1) VLAN modify isn't supported by driver.
2) FW syndrome (0xA9C090):
set_flow_table_entry: push vlan action fte in fdb can ONLY be
forward to the uplink.
3) FW syndrome (0x294609):
set_flow_table_entry: modify/pop/push actions in fdb flow table are
supported only w
Fixes: 4f07e13d6af5 ("net/mlx5: split flow validation to dedicated function")
Fixes: f7adfffa3de1 ("net/mlx5: add Direct Verbs validation function")
Fixes: edcdef4e5fe4 ("net/mlx5: add Linux TC flower driver for E-Switch flow")
Cc: Ori Kam
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_f
This can cause crash by null pointer reference.
Fixes: 4f07e13d6af5 ("net/mlx5: split flow validation to dedicated function")
Cc: Ori Kam
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_flow.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/drivers/net/mlx5/mlx5
MLX5_ACTION_* -> MLX5_FLOW_ACTION_*
MLX5_VXLAN* -> MLX5_UDP_PORT_VXLAN*
Fixes: 0f8775dd8f1c ("net/mlx5: add support for multiple flow drivers")
Fixes: edcdef4e5fe4 ("net/mlx5: add Linux TC flower driver for E-Switch flow")
Cc: Ori Kam
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_flow.
Fixes: 0f8775dd8f1c ("net/mlx5: add support for multiple flow drivers")
Cc: Ori Kam
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_flow.c | 9 +
drivers/net/mlx5/mlx5_flow.h | 9 -
drivers/net/mlx5/mlx5_flow_verbs.c | 7 ---
3 files changed, 13 insertions(
Fixes: 5c83c536783c ("net/mlx5: add Direct Verbs final functions")
Cc: Ori Kam
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_flow.h| 10 --
drivers/net/mlx5/mlx5_flow_dv.c | 2 +-
2 files changed, 1 insertion(+), 11 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow.h b/
Minor fixes accumulated since the following two patchsets.
net/mlx5: add Direct Verbs flow driver support [1]
net/mlx5: migrate Linux TC flower driver to new flow engine
[1] http://patches.dpdk.org/cover/45248
[2] http://patches.dpdk.org/cover/44897
Yongseok Koh (7):
net/mlx5:
This patch add AES-CMAC support. CMAC is a keyed hash function
that is based on a symmetric key block cipher. It is One-Key
CBC MAC improvement over XCBC-MAC. RFC 4493. NIST SP 800-38B.
Signed-off-by: Tomasz Cel
Signed-off-by: Arek Kusztal
---
drivers/crypto/qat/qat_sym_session.c | 190
This patch add support for AES-CMAC, moreover update QAT documentation
and enable tests for AES-CMAC.
v2:
* deleted unnecessary white signs from documentation part
v3:
* added AES-CMAC to the capabilities list
* updated release note for 18.11
* updated qat.rst file
Tomasz Cel (3):
crypto/qat:
Enable tests for AES-CMAC authentication algorithm.
Signed-off-by: Tomasz Cel
---
test/test/test_cryptodev_hash_test_vectors.h | 12
1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/test/test/test_cryptodev_hash_test_vectors.h
b/test/test/test_cryptodev_hash_test_vecto
Update the QAT documentation to show that it supports AES-CMAC.
Signed-off-by: Tomasz Cel
---
doc/guides/cryptodevs/features/qat.ini| 1 +
doc/guides/cryptodevs/qat.rst | 1 +
doc/guides/rel_notes/release_18_11.rst| 6 ++
drivers/crypto/qat/qat_sym_capabilities.h | 20
On 10/08/2018 06:25 PM, Ilya Maximets wrote:
Have you missed this mail while preparing v4?
Yes I did... Thanks for the heads-up!
Maxime
Best regards, Ilya Maximets.
On 05.10.2018 17:24, Ilya Maximets wrote:
I think it'll be nice to have a warning in case zero-copy
and postcopy enabled at
Have you missed this mail while preparing v4?
Best regards, Ilya Maximets.
On 05.10.2018 17:24, Ilya Maximets wrote:
> I think it'll be nice to have a warning in case zero-copy
> and postcopy enabled at the same time. This warning will
> clarify for users why postcopy is not enabled.
> What do yo
On 08.10.2018 18:25, Maxime Coquelin wrote:
> This patch opens a userfaultfd and sends it back to Qemu's
> VHOST_USER_POSTCOPY_ADVISE request.
>
> Signed-off-by: Dr. David Alan Gilbert
> Signed-off-by: Maxime Coquelin
> ---
> lib/librte_vhost/vhost.h | 2 ++
> lib/librte_vhost/vhost_user.
The following changes since commit 77b5311d0ece90b1c197c1dc0f38bc6dbdcc28a8:
event/dpaa: support select based event (2018-10-04 16:38:54 +0200)
are available in the git repository at:
http://dpdk.org/git/next/dpdk-next-qos
for you to fetch changes up to 367ff7d2d7cc204c0b80eb4b6e4e3c19f51c
Introduced DEV_TX_OFFLOAD_OUTER_UDP_CKSUM offload flags and
PKT_TX_OUTER_UDP_CKSUM mbuf ol_flags to enable Tx outer UDP
checksum offload.
To use hardware Tx outer UDP checksum offload, the user needs to,
- enable following in mbuf:
a) fill outer_l2_len and outer_l3_len in mbuf
b) set the PKT_TX_O
Added outer-udp Tx HW checksum support for csum forward engine
if device supports DEV_TX_OFFLOAD_OUTER_UDP_CKSUM.
Signed-off-by: Jerin Jacob
Acked-by: Bernard Iremonger
---
v3:
- Added Acked-by: Bernard Iremonger
v2:
- Added outer-udp support for csum engine
---
app/test-pmd/cmdline.c
Collect and prints the statistics for PKT_RX_EL4_CKSUM_BAD
errors.
Signed-off-by: Jerin Jacob
Reviewed-by: Andrew Rybchenko
Acked-by: Bernard Iremonger
---
v4:
- Added Acked-by: Bernard Iremonger
- s/PKT_RX_EL4_CKSUM_BAD/PKT_RX_OUTER_L4_CKSUM_BAD
v3:
- Added Reviewed-by: Andrew Rybchenko
-
Introduced DEV_RX_OFFLOAD_OUTER_UDP_CKSUM Rx offload flag and
PKT_RX_OUTER_L4_CKSUM_* mbuf ol_flags to detect outer UDP checksum
status.
- To use hardware Rx outer UDP checksum offload, the user needs to
configure DEV_RX_OFFLOAD_OUTER_UDP_CKSUM offload flags in slowpath.
- Driver updates checksum
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Jasvinder Singh
> Sent: Wednesday, September 12, 2018 10:02 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo ; Wu, Jingjing
> ; Iremonger, Bernard
>
> Subject: [dpdk-dev] [PATCH] app/testpmd: fix metering and policing cl
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Krzysztof Kanas
> Sent: Friday, August 17, 2018 12:39 PM
> To: dev@dpdk.org
> Cc: krzysztof.ka...@caviumnetworks.com;
> jerin.jacobkollanukka...@cavium.com; nithin.dabilpu...@cavium.com
> Subject: [dpdk-dev] [PAT
Hi Akhil, Joseph, Thomas,
Just spotted this now.
See below.
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Monday, October 1, 2018 11:05 AM
> To: Joseph, Anoob
> Cc: dev@dpdk.org; Trahe, Fiona ; Akhil Goyal
> ; Anoob
> Joseph ; De Lara Guarch, Pablo
>
> -Original Message-
> From: Xu, Rosen
> Sent: Thursday, August 16, 2018 2:14 AM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian ; Lu, Wenzhuo
> ; Singh, Jasvinder ;
> jerin.ja...@caviumnetworks.com; Xu, Rosen ; Yigit,
> Ferruh
> Subject: [PATCH] lib/librte_ethdev: expand queue threshold
> -Original Message-
> From: Dumitrescu, Cristian
> Sent: Monday, September 24, 2018 6:12 PM
> To: Vivek Sharma
> Cc: Sharma, Vivek ; sta...@dpdk.org;
> dev@dpdk.org
> Subject: RE: [PATCH v1] eal: use correct data type for slab operations
>
>
>
> > -Original Message-
> > From:
> -Original Message-
> From: Singh, Jasvinder
> Sent: Thursday, September 27, 2018 6:11 PM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian
> Subject: [PATCH] librte_sched: allocate memory on the given socket id
>
> Replace rte_zmalloc() with rte_zmalloc_socket() to allocate
> memory on t
On 08/10/2018, 16:46, "Ola Liljedahl" wrote:
On 08/10/2018, 16:44, "Bruce Richardson" wrote:
On Mon, Oct 08, 2018 at 09:22:05AM +, Ola Liljedahl wrote:
> "* multi-producer safe lock-free ring buffer enqueue"
> The comment is also wrong.
> -Original Message-
> From: Pattan, Reshma
> Sent: Monday, October 8, 2018 12:19 PM
> To: dev@dpdk.org; Dumitrescu, Cristian ;
> Singh, Jasvinder
> Cc: Pattan, Reshma
> Subject: [PATCH v4] net/softnic: add flow flush API
>
> Add rte flow flush api for flushing
> all the flows of the
> -Original Message-
> From: Pattan, Reshma
> Sent: Friday, September 28, 2018 5:19 PM
> To: dev@dpdk.org; Dumitrescu, Cristian ;
> Singh, Jasvinder
> Cc: Pattan, Reshma
> Subject: [PATCH] examples/ip_pipeline: fix ipv6 address endianness
>
> Fix ipv6 endianness from big endian to cpu
Postcopy live-migration feature require the application to
not populate the guest memory. As the vhost library cannot
prevent the application to that (e.g. preventing the
application to call mlockall()), the feature is disabled by
default.
The application should only enable the feature if it does
> -Original Message-
> From: Pattan, Reshma
> Sent: Friday, September 28, 2018 5:19 PM
> To: dev@dpdk.org; Dumitrescu, Cristian ;
> Singh, Jasvinder
> Cc: Pattan, Reshma
> Subject: [PATCH] net/softnic: fix ipv6 address from big endian to cpu order
>
> Fix ipv6 endianness from big endi
Introduce a new postcopy-support parameter to Vhost PMD that
passes the RTE_VHOST_USER_POSTCOPY_SUPPORT flag at vhost
device register time.
Flag should only be set if application does not prefault guest
memory using, for example, mlockall() syscall.
Default value is 0, meaning that postcopy suppo
Enable postcopy protocol feature except if dequeue
zero-copy is enabled. In this case, guest memory requires
to be populated, which is not compatible with userfaultfd.
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
---
lib/librte_vhost/vhost_user.c | 7 +++
lib/librte_
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
---
lib/librte_vhost/vhost_user.c | 49 ---
1 file changed, 46 insertions(+), 3 deletions(-)
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index 9a50b962b..75e15
The VHOST_USER_SET_MEM_TABLE payload is copied when handled,
whereas it could directly be referenced.
This is not very important, but next, we'll need to update the
payload and send it back to Qemu.
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
Acked-by: Ilya Maximets
--
The master sends this message before stopping handling
userfaults, so that the backend closes the userfaultfd.
The master waits for the slave to acknowledge the request
with an empty 64bits payload for synchronization purpose.
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
---
lib/librte_vhost/vhost_user.c | 26 ++
1 file changed, 26 insertions(+)
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index 7a79145c2..46c97836a 100644
--- a/lib/librte_
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
---
lib/librte_vhost/vhost.h | 1 +
lib/librte_vhost/vhost_user.c | 21 +
lib/librte_vhost/vhost_user.h | 3 ++-
3 files changed, 24 insertions(+), 1 deletion(-)
diff --git a/lib/librte_vhost/vhost.h
This patch opens a userfaultfd and sends it back to Qemu's
VHOST_USER_POSTCOPY_ADVISE request.
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
---
lib/librte_vhost/vhost.h | 2 ++
lib/librte_vhost/vhost_user.c | 49 +++
lib/librte_vhost
This is not used for now, but will be needed for the
special handling of VHOST_USER_SET_MEM_TABLE message
once postcopy will be supported.
Signed-off-by: Maxime Coquelin
---
lib/librte_vhost/vhost_user.c | 71 +++
1 file changed, 47 insertions(+), 24 deletions(-)
Postcopy live-migration features relies on userfaultfd,
which was only introduced in kernel v4.3.
This patch introduces a new define to allow building vhost
library on kernels not supporting userfaultfd.
With legacy build system, user has to explicitly set
CONFIG_RTE_LIBRTE_VHOST_POSTCOPY to 'y'.
Passing userfault fds to Qemu will be required for postcopy
live-migration feature.
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
---
lib/librte_vhost/vhost_user.c | 27 +++
1 file changed, 15 insertions(+), 12 deletions(-)
diff --git a/lib/librte
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
Acked-by: Ilya Maximets
---
lib/librte_vhost/rte_vhost.h | 4
1 file changed, 4 insertions(+)
diff --git a/lib/librte_vhost/rte_vhost.h b/lib/librte_vhost/rte_vhost.h
index b02673d4a..b3cc6990d 100644
--- a/lib/librte_vh
As soons as some anciliarry datai (fds) are received, it is copied
without checking its length.
This patch adds adds the number of fds received to the message,
which is set in read_vhost_message().
This is preliminary work to support sending fds to Qemu.
Signed-off-by: Dr. David Alan Gilbert
Si
When the memory table gets updated, the rings addresses need
to be translated again. If it fails, we need to exit cleanly
by unmapping memory regions.
Fixes: d5022533c20a ("vhost: retranslate vring addr when memory table changes")
Cc: sta...@dpdk.org
Signed-off-by: Maxime Coquelin
Acked-by: Ilya
For messages that require a reply, a second ack should not be
sent when reply-ack protocol feature is negotiated, even if
the corresponding flag is set in the message.
The code is compliant with the spec but it isn't clear it is,
so this patch adds a comment to make it explicit.
Suggested-by: Ily
QEMU doesn't expect any payload for the reply of
VHOST_USER_SET_LOG_BASE request, so don't send any.
Note that the Vhost-user specification isn't clear about
it and would need to be fixed.
Fixes: 54f9e32305d4 ("vhost: handle dirty pages logging request")
Cc: sta...@dpdk.org
Reported-by: Ilya Maxi
VHOST_USER_GET_PROTOCOL_FEATURES, VHOST_USER_GET_VRING_BASE
and VHOST_USER_SET_LOG_BASE require replies, so their handlers
should return VH_RESULT_REPLY, not VH_RESULT_OK.
Fixes: 0bff510b5ea6 ("vhost: unify message handling function signature")
Signed-off-by: Maxime Coquelin
Acked-by: Ilya Maxim
Return of message handling has now changed to an enum that can
take non-negative value that is not zero in case a reply is
needed. But the code checking the variable afterwards has not
been updated, leading to success messages handling being
treated as errors.
Fixes: 2f270595c05d ("vhost: rework m
In this v4:
- Fix VHOST_USER_MAX in patch 12 (Ilya)
- Don't close postcopy_fd in set_mem_table error path (Ilya)
- Close postcopy_fd in vhost_backend_cleanup() (Ilya)
- Re-init postcopy_listening in vhost_backend_cleanup() (Ilya)
- Close FDs in read_fd_message() error path (Ilya)
- Fix various typo
Thanks, PSB
From: Andrew Rybchenko
Sent: Friday, October 5, 2018 4:40 PM
To: Ferruh Yigit ; Dekel Peled ;
wenzhuo...@intel.com; jingjing...@intel.com; bernard.iremon...@intel.com;
dev@dpdk.org; olivier.m...@6wind.com; Adrien Mazarguil
; Thomas Monjalon
Cc: Shahaf Shuler ; Ori Kam
Subject: R
> -Original Message-
> From: Dumitrescu, Cristian
> Sent: Monday, October 8, 2018 11:36 AM
> To: Pattan, Reshma ; dev@dpdk.org; Singh,
> Jasvinder
> Subject: RE: [PATCH v3] net/softnic: add flow flush API
>
>
>
> > -Original Message-
> > From: Pattan, Reshma
> > Sent: Monday,
On 08/10/2018, 16:44, "Bruce Richardson" wrote:
On Mon, Oct 08, 2018 at 09:22:05AM +, Ola Liljedahl wrote:
> "* multi-producer safe lock-free ring buffer enqueue"
> The comment is also wrong. This design is not lock-free, how could it be
when there is spinning
> (wai
On Mon, Oct 08, 2018 at 09:22:05AM +, Ola Liljedahl wrote:
> "* multi-producer safe lock-free ring buffer enqueue"
> The comment is also wrong. This design is not lock-free, how could it be when
> there is spinning
> (waiting) for other threads in the code? If a thread must wait for other
>
Hi Thomas,
On 08-10-2018 19:07, Thomas Monjalon wrote:
> External Email
>
> 08/10/2018 15:07, Joseph, Anoob:
>> On 08-10-2018 17:57, Thomas Monjalon wrote:
>>> 05/10/2018 14:58, Anoob Joseph:
+Cavium OCTEON TX
+M: Anoob Joseph
+F: drivers/common/cpt/
>>> What is the real wording fo
On Mon, 2018-10-08 at 08:43 +, Zhao1, Wei wrote:
> Hi, Luca Boccassi
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Luca Boccassi
> > Sent: Saturday, July 28, 2018 1:26 AM
> > To: dev@dpdk.org
> > Cc: Lu, Wenzhuo ; Luca Boccassi
> > ; sta...@dpdk.org
> -Original Message-
> From: dev On Behalf Of Dekel Peled
> Sent: Monday, October 8, 2018 4:35 PM
> To: Yongseok Koh ; Shahaf Shuler
>
> Cc: dev@dpdk.org; Ori Kam
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix calculation of struct size in bytes
>
> Update definition of MLX5_ST_SZ_DB to
Update definition of MLX5_ST_SZ_DB to calculate struct size correctly.
Fixes: 86a8d2b59f54 ("net/mlx5: add Direct Verbs prepare function")
Cc: or...@mellanox.com
Signed-off-by: Dekel Peled
---
drivers/net/mlx5/mlx5_prm.h |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/
08/10/2018 15:07, Joseph, Anoob:
> On 08-10-2018 17:57, Thomas Monjalon wrote:
> > 05/10/2018 14:58, Anoob Joseph:
> >> +Cavium OCTEON TX
> >> +M: Anoob Joseph
> >> +F: drivers/common/cpt/
> > What is the real wording for this device family?
> > Sometimes I read OcteonTX with lowercases and no spa
-Original Message-
> Date: Mon, 08 Oct 2018 15:03:49 +0200
> From: Thomas Monjalon
> To: Jerin Jacob
> Cc: Ferruh Yigit , "Ananyev, Konstantin"
> , Andrew Rybchenko
> , "Lu, Wenzhuo" , "Wu,
> Jingjing" , "Iremonger, Bernard"
> , "Mcnamara, John" ,
> "Kovacevic, Marko" , Olivier Matz
08/10/2018 14:58, Joseph, Anoob:
> On 08-10-2018 18:03, Thomas Monjalon wrote:
> > A lot of patches are split in 2 parts for common/cpt and crypto/octeontx.
> > I think it is useless most of the time.
> > You are allowed to update any related code in the common part
> > while implementing some feat
Hi Thomas,
On 08-10-2018 17:57, Thomas Monjalon wrote:
External Email
05/10/2018 14:58, Anoob Joseph:
+Cavium OCTEON TX
+M: Anoob Joseph
+F: drivers/common/cpt/
What is the real wording for this device family?
Sometimes I read OcteonTX with lowercases and no space,
sometimes OCTEONTX without
Unit testcases are added for metrics library
Added metrics unit test to autotest list
Updated meson build file
Updated MAINTAINERSHIP for metrics unit test
Signed-off-by: Hari Kumar Vemula
Reviewed-by: Reshma Pattan
Reviewed-by: Remy Horton
Acked-by: Remy Horton
---
v5: Rebased on new codebase
08/10/2018 14:25, Jerin Jacob:
> From: Ferruh Yigit
> > On 10/8/2018 12:55 PM, Jerin Jacob wrote:
> > > From: Ferruh Yigit
> > >> On 10/8/2018 10:37 AM, Jerin Jacob wrote:
> > >>> From: Thomas Monjalon
> > 08/10/2018 10:24, Jerin Jacob:
> > > From: Ferruh Yigit
> > >> On 10/6/2018
On 10/05/2018 02:34 PM, Ilya Maximets wrote:
On 04.10.2018 11:13, Maxime Coquelin wrote:
Signed-off-by: Dr. David Alan Gilbert
Signed-off-by: Maxime Coquelin
---
lib/librte_vhost/vhost_user.c | 33 -
1 file changed, 32 insertions(+), 1 deletion(-)
diff --
1 - 100 of 197 matches
Mail list logo
