From: Natalie Samsonov
Since in DPDK 17.11 port type was changed from uint8_t to uint16_t
the MBUF_INVALID_PORT value became 0x but in mrvl_tx_pkt_burst()
when trying to lookup bpool using mbuf port, we check if the port
is invalid according to value 0xff. This causes segmentation fault.
Sol
26/01/2018 08:31, Lu, Wenzhuo:
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> > On Thu, 25 Jan 2018 10:04:11 +0100
> > Thomas Monjalon wrote:
> > > 25/01/2018 02:11, Lu, Wenzhuo:
> > > > > --- a/app/test-pmd/testpmd.c
> > > > > +++ b/app/test-pmd/testpmd.c
> > > > > @@ -305,9 +305
26/01/2018 08:30, Lu, Wenzhuo:
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > 25/01/2018 02:11, Lu, Wenzhuo:
> > > > --- a/app/test-pmd/testpmd.c
> > > > +++ b/app/test-pmd/testpmd.c
> > > > @@ -305,9 +305,7 @@ struct fwd_engine * fwd_engines[] = {
> > > > */
> > > > struct rte_eth_rxm
> -Original Message-
> From: Yang, Zhiyong
> Sent: Thursday, January 25, 2018 3:55 PM
> To: dev@dpdk.org
> Cc: Xing, Beilei ; Zhang, Qi Z ;
> Zhang, Helin ; Yang, Zhiyong
>
> Subject: [PATCH] net/i40e: remove unnecessary void pointer cast
>
> void pointer can be assigned to any type poi
26/01/2018 06:59, Andrew Rybchenko:
> On 01/26/2018 12:38 AM, Thomas Monjalon wrote:
> > 25/01/2018 18:00, Andrew Rybchenko:
> >> Unfortunately we're a bit late with dynamic logging implementation.
> >> So, it can wait for 18.05 release cycle if required.
> >>
> >> The series adds EXPERIMENTAL EAL
26/01/2018 06:11, Hemant Agrawal:
>
> > -Original Message-
> > From: Thomas Monjalon [mailto:tho...@monjalon.net]
> >
> > 22/01/2018 14:51, Hemant Agrawal:
> > > From: Pavan Nikhilesh
> > >
> > > Set the mempool preferred by the ethernet devices as default mbuf
> > > mempool before creat
> -Original Message-
> From: Zhao1, Wei
> Sent: Friday, January 26, 2018 11:08 AM
> To: dev@dpdk.org
> Cc: Zhang, Qi Z ; sta...@dpdk.org; Zhao1, Wei
>
> Subject: [PATCH v2] net/i40e: fix issue for RSS flow API
>
> This patch fix issue check from DPDK coverity issues.
>
> Fixes: ecad87d
Hi Stephen,
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Friday, January 26, 2018 12:02 AM
> To: Thomas Monjalon
> Cc: Lu, Wenzhuo ; Moti Haimovsky
> ; dev@dpdk.org; shah...@mellanox.com; Yigit,
> Ferruh
> Subject: Re: [dpdk-dev] [PATCH] app
Hi Thomas,
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Thursday, January 25, 2018 5:04 PM
> To: Lu, Wenzhuo
> Cc: Moti Haimovsky ; dev@dpdk.org;
> shah...@mellanox.com; Yigit, Ferruh
> Subject: Re: [PATCH] app/testpmd: do not enable Rx offloads by de
-Original Message-
> Date: Fri, 26 Jan 2018 12:10:16 +0530
> From: Pavan Nikhilesh
> To: jerin.ja...@caviumnetworks.com, santosh.shu...@caviumnetworks.com,
> ferruh.yi...@intel.com
> Cc: dev@dpdk.org, Pavan Nikhilesh
> Subject: [dpdk-dev] [PATCH v2] net/octeontx: use stub for building in
On 01/25/2018 09:42 PM, Stephen Hemminger wrote:
On Thu, 25 Jan 2018 17:00:43 +
Andrew Rybchenko wrote:
diff --git a/drivers/net/sfc/efsys.h b/drivers/net/sfc/efsys.h
index c7a54c3..8dd225e 100644
--- a/drivers/net/sfc/efsys.h
+++ b/drivers/net/sfc/efsys.h
@@ -26,6 +26,7 @@
#include
Use stub for building clang < 4.0 as it doesn't support the following
arm machine directives:
.cpu
Signed-off-by: Pavan Nikhilesh
---
v2 Changes:
- use stub for older clang compilers instead of disabling PMD
drivers/net/octeontx/Makefile | 8
drivers/net/octeontx/base/oct
On 01/26/2018 05:01 AM, Stephen Hemminger wrote:
Use the new API rte_eth_linkstatus_set to handle link status update.
Signed-off-by: Stephen Hemminger
---
drivers/net/sfc/sfc_ethdev.c | 24 +---
drivers/net/sfc/sfc_ev.c | 20 ++--
2 files changed, 7 i
On 01/26/2018 12:38 AM, Thomas Monjalon wrote:
25/01/2018 18:00, Andrew Rybchenko:
Unfortunately we're a bit late with dynamic logging implementation.
So, it can wait for 18.05 release cycle if required.
The series adds EXPERIMENTAL EAL feature which removes dependency
on EAL arguments processi
On Thu, Jan 25, 2018 at 01:58:32PM +, Ferruh Yigit wrote:
> On 1/23/2018 4:22 PM, Pavan Nikhilesh wrote:
> > Disable for clang < 4.0 as it doesn't support the following
> > arm machine directives:
> >
> > .cpu
> >
> > This also disables event octeontx as it depends on net octeontx.
> >
> > Sign
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
>
> 22/01/2018 14:51, Hemant Agrawal:
> > From: Pavan Nikhilesh
> >
> > Set the mempool preferred by the ethernet devices as default mbuf
> > mempool before creating the pktpool.
>
> The patch is just adding a lo
Hi Thomas,
>
> 22/01/2018 14:51, Hemant Agrawal:
> > This patch add support for various mempool ops config helper APIs.
> >
> > 1.User defined mempool ops
> > 2.Platform detected HW mempool ops (active).
> > 3.Best selection of mempool ops by looking into user defined,
> > platform registered
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
>
> 22/01/2018 14:51, Hemant Agrawal:
> > This patch change the logic to maintain the value of user defined and
> > compile time i.e. RTE_MBUF_DEFAULT_MEMPOOL_OPS.
> >
> > The pktmbuf_create_pool is updated to reflec
This commit provides a set of tests for verifying the correctness and
performance of both unsigned 32 and 64bit reciprocal based division.
Signed-off-by: Pavan Nikhilesh
---
test/test/Makefile| 2 +
test/test/test_reciprocal_division.c | 167 +++
In some use cases of integer division, denominator remains constant and
numerator varies. It is possible to optimize division for such specific
scenarios.
The librte_sched uses rte_reciprocal to optimize division so, moving it to
eal/common would allow other libraries and applications to use it.
Currently, rte_reciprocal only supports unsigned 32bit divisors. This
commit adds support for unsigned 64bit divisors.
Signed-off-by: Pavan Nikhilesh
---
lib/librte_eal/common/include/rte_reciprocal.h | 49 ++
lib/librte_eal/common/rte_reciprocal.c | 92 ++
> -Original Message-
> From: Yong Wang [mailto:wang.yon...@zte.com.cn]
> Sent: Thursday, January 25, 2018 5:01 PM
> To: Xing, Beilei
> Cc: dev@dpdk.org; Yong Wang
> Subject: [PATCH v4] net/i40e: add null point check and fix mem leak
>
> There are several func calls to rte_zmalloc() which
This patch aim to add a general uevent mechanism in eal device layer,
to enable all linux kernel object uevent monitoring, user could use these
APIs to monitor and read out the device status info that sent from the
kernel side, then corresponding to handle it, such as when detect hotplug
uevent typ
use testpmd for example, to show app how to request and use
uevent monitoring to handle the hot removal event and the
hot insertion event.
Signed-off-by: Jeff Guo
---
v13->v12:
refine some code style
---
app/test-pmd/testpmd.c | 169 +
app/test-pmd
In order to handle the uevent which have been detected from the kernel
side, add uevent process function, let hot plug event to be example to
show uevent mechanism how to pass the uevent and process the uevent.
About uevent passing and processing, add below functions in linux eal
dev layer. FreeBS
We need the synchronous way for multi-process communication,
i.e., blockingly waiting for reply message when we send a request
to the peer process.
We add two APIs rte_eal_mp_request() and rte_eal_mp_reply() for
such use case. By invoking rte_eal_mp_request(), a request message
is sent out, and th
Previouly, there are three channels for multi-process
(i.e., primary/secondary) communication.
1. Config-file based channel, in which, the primary process writes
info into a pre-defined config file, and the secondary process
reads the info out.
2. vfio submodule has its own channel ba
v5->v6:
- Correct the API name issue in rte_eal_version.map.
v3->v5:
- Drop the patch 3 on vfio communication (postponed).
- Change names from rte_eal_mp_* -> rte_mp_* as suggested by Thomas.
- Add nb_sent and nb_received in struct rte_mp_reply.
- Standardize the return val of sendmsg, r
Hi, zhang qi
V2 for this patch.
Thanks.
https://dpdk.org/dev/patchwork/patch/34533/
> -Original Message-
> From: Zhang, Qi Z
> Sent: Thursday, January 25, 2018 1:40 PM
> To: Zhao1, Wei ; dev@dpdk.org
> Subject: RE: [PATCH] net/i40e: fix issues for RSS flow API
>
> Hi Wei:
>
>
> > --
Hi Yong,
> -Original Message-
> From: Yong Wang [mailto:wang.yon...@zte.com.cn]
> Sent: Tuesday, January 23, 2018 2:45 AM
> To: Xing, Beilei ; Lu, Wenzhuo ;
> Trahe, Fiona
>
> Cc: dev@dpdk.org; Yong Wang
> Subject: [PATCH v8 3/3] crypto/qat: add null point check and fix mem leak
>
> Th
This patch fix issue check from DPDK coverity issues.
Fixes: ecad87d22383e ("net/i40e: move RSS to flow API")
Coverity issue: 257020 257024 257037
Signed-off-by: Wei Zhao
---
v2:
-fix an error introduced in v1 patch.
---
drivers/net/i40e/i40e_ethdev.c | 2 +-
drivers/net/i40e/i40e_flow.c |
Hi Olivier,
> -Original Message-
> From: Olivier Matz [mailto:olivier.m...@6wind.com]
> Sent: Thursday, January 25, 2018 6:40 PM
> To: Lu, Wenzhuo
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [RFC] lib/librte_ether: add a return value for MAC set
> ops
>
> Hi Wenzhuo,
>
> On Thu, Jan 2
This driver was not doing atomic update of link status information.
And the return value was different than others.
The hardware also does not do autonegotiation (at least on Linux).
Signed-off-by: Stephen Hemminger
---
drivers/net/enic/enic_ethdev.c | 5 ++---
drivers/net/enic/enic_main.c |
Yet another driver which was not returing correct value on
link change.
Since this driver can't be built on x86 could not even
do a compile test.
Signed-off-by: Stephen Hemminger
---
drivers/net/szedata2/rte_eth_szedata2.c | 11 ---
1 file changed, 4 insertions(+), 7 deletions(-)
diff
Use the new link update API, and cleanup the logic in the the
link update routine.
Tested-by: Shijith Thotton
Signed-off-by: Stephen Hemminger
---
drivers/net/liquidio/lio_ethdev.c | 44 ---
1 file changed, 4 insertions(+), 40 deletions(-)
diff --git a/drive
Use new helper function.
Signed-off-by: Stephen Hemminger
---
drivers/net/thunderx/nicvf_ethdev.c | 45 +++--
1 file changed, 18 insertions(+), 27 deletions(-)
diff --git a/drivers/net/thunderx/nicvf_ethdev.c
b/drivers/net/thunderx/nicvf_ethdev.c
index d34938c64
Common function matches this drivers usage.
Signed-off-by: Stephen Hemminger
---
drivers/net/octeontx/octeontx_ethdev.c | 16 +---
1 file changed, 1 insertion(+), 15 deletions(-)
diff --git a/drivers/net/octeontx/octeontx_ethdev.c
b/drivers/net/octeontx/octeontx_ethdev.c
index adca
Use new rte_linkstatus update API
Signed-off-by: Stephen Hemminger
---
drivers/net/i40e/i40e_ethdev.c| 46 ++-
drivers/net/i40e/i40e_ethdev_vf.c | 19 ++--
2 files changed, 9 insertions(+), 56 deletions(-)
diff --git a/drivers/net/i40e/i40e_et
Use the new API rte_eth_linkstatus_set to handle link status update.
Signed-off-by: Stephen Hemminger
---
drivers/net/sfc/sfc_ethdev.c | 24 +---
drivers/net/sfc/sfc_ev.c | 20 ++--
2 files changed, 7 insertions(+), 37 deletions(-)
diff --git a/drivers/ne
Use new rte_eth_linkstatus_get/set helper function.
Signed-off-by: Stephen Hemminger
---
drivers/net/nfp/nfp_net.c | 72 +--
1 file changed, 7 insertions(+), 65 deletions(-)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 0044
Use new rte_eth_linkstatus_get/set API.
Signed-off-by: Stephen Hemminger
---
drivers/net/e1000/em_ethdev.c | 70 +++--
drivers/net/e1000/igb_ethdev.c | 71 +++---
2 files changed, 10 insertions(+), 131 deletions(-)
diff --
Use the new helper functions from eth_dev for
handling atomic link_info update.
Signed-off-by: Stephen Hemminger
---
drivers/net/ixgbe/ixgbe_ethdev.c | 90 ++--
1 file changed, 13 insertions(+), 77 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/
Use new helper function to update the link status.
Signed-off-by: Stephen Hemminger
---
drivers/net/dpaa2/dpaa2_ethdev.c | 78 ++--
1 file changed, 10 insertions(+), 68 deletions(-)
diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c b/drivers/net/dpaa2/dpaa2_ethde
Many drivers are all doing copy/paste of the same code to atomically
update the link status. Reduce duplication, and allow for future
changes by having common function for this.
Signed-off-by: Stephen Hemminger
---
lib/librte_ether/rte_ethdev.c| 22 +++--
lib/librte_ether/rte_eth
Use the new comon code in ethdev to handle link status update.
Signed-off-by: Stephen Hemminger
---
drivers/net/virtio/virtio_ethdev.c | 52 --
1 file changed, 5 insertions(+), 47 deletions(-)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/vir
To handle atomic update of link status (64 bit), every driver
was doing its own version using cmpset.
Atomic exchange is a useful primitive in its own right;
therefore make it a EAL routine.
Signed-off-by: Stephen Hemminger
---
.../common/include/arch/ppc_64/rte_atomic.h| 21 +-
.../
Use new rte_eth_link_update helper.
Also remove no longer necessary include of rte_atomic.h
Signed-off-by: Stephen Hemminger
---
drivers/net/vmxnet3/vmxnet3_ethdev.c | 63 ++--
1 file changed, 3 insertions(+), 60 deletions(-)
diff --git a/drivers/net/vmxnet3/vmxn
While writing hyper-v driver, noticed a lot of unnecessary
duplication of code in drivers for handling the eth_dev link status
information. While consolidating this, it also became obvious that
some drivers behave differently for no good reason.
It also was a good chance to introduce atomic exchan
> -Original Message-
> From: Matan Azrad [mailto:ma...@mellanox.com]
> Sent: Thursday, January 25, 2018 4:30 PM
> To: Lu, Wenzhuo ; Thomas Monjalon
> ; Gaetan Rivet ; Wu,
> Jingjing
> Cc: dev@dpdk.org; Neil Horman ; Richardson,
> Bruce ; Ananyev, Konstantin
>
> Subject: RE: [dpdk-dev] [PA
On Thu, 11 Jan 2018 17:01:10 +
Ferruh Yigit wrote:
> On 1/8/2018 5:45 PM, Stephen Hemminger wrote:
> > To handle atomic update of link status (64 bit), every driver
> > was doing its own version using cmpset.
> > Atomic exchange is a useful primitive in its own right;
> > therefore make it a
Register for async events from the FW.
New events we are registering for include Link speed config changes,
PF driver unload and VF config change. Also log a message when the
async event arrives on the completion ring.
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_cpr.c | 11 ++
While using RSS, the pool count should be 1.
Fixes: 8103a57ab432a ("net/bnxt: handle Rx multi queue creation properly")
Cc: sta...@dpdk.org
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_rxq.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/bnxt/bnxt_rxq.c
When the driver is loaded on a 100G NIC, the port speed is not
displayed correctly. Parse the 100G speed before displaying it.
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_hwrm.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_
Currently this is implemented entirely in the PMD as there is no explicit
support in the HW. Re-program the RSS Table without this queue on stop
and add it back to the table on start.
Signed-off-by: Somnath Kotur
Signed-off-by: Ajit Khaparde
--
v1->v2: address review comments
---
drivers/net/bn
In certain cases the MAC address of a port could be all zeros.
Catch it early, log a message and fail the initiaization.
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_ethdev.c | 10 ++
drivers/net/bnxt/bnxt_filter.c | 2 +-
drivers/net/bnxt/bnxt_filter.h | 1 +
3 files changed
Please apply this patchset.
I have incorporated most of the comments that we got in v1.
The switch to SPDX tags is in the works and will be submitted separately.
Ajit Khaparde (6):
net/bnxt: fix size of tx ring in HW
net/bnxt: use driver specific dynamic log type
net/bnxt: register for more
During Tx ring allocation, the actual ring size configured in the HW
ends up being twice the number of txd parameter specified to the driver.
The power of 2 ring size wrongly adds a +1 while sending the ring
create command to the FW.
Fixes: 6eb3cc2294fd ("net/bnxt: add initial Tx code")
Cc: sta...
This patch implements driver specific log type doing away with
usage of RTE_LOG() for logging.
Signed-off-by: Ajit Khaparde
--
v1 -> v2: address review comments
---
drivers/net/bnxt/bnxt.h | 8 ++
drivers/net/bnxt/bnxt_cpr.c | 10 +-
drivers/net/bnxt/bnxt_ethdev.c | 241 ++
Ping
19/01/2018 00:49, Thomas Monjalon:
> 04/12/2017 14:22, Pavan Nikhilesh:
> > +/*
> > + * BSD LICENSE
> > + *
> > + * Copyright (C) Cavium, Inc. 2017.
> > + *
> > + * Redistribution and use in source and binary forms, with or without
> > + * modification, are permitted provided that the
22/01/2018 14:17, Olivier Matz:
> On Sun, Jan 21, 2018 at 05:05:10PM +, Andrew Rybchenko wrote:
> > From: Ivan Malov
> >
> > Fixes: a5279180f510 ("eal: change several log levels matching a regexp")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: Ivan Malov
> > Signed-off-by: Andrew Rybchenko
22/01/2018 15:49, Ferruh Yigit:
> On 1/22/2018 2:14 PM, Jasvinder Singh wrote:
> > Free allocated memory of the rule if not added to the table.
> >
> > Coverity issue: 257032
> > Fixes: 50bdac5916d9 ("flow_classify: remove table id parameter from API")
> >
> > Signed-off-by: Jasvinder Singh
>
>
24/01/2018 12:55, Remy Horton:
> Done quick smoke test and it all seems fine :)
>
> On 23/01/2018 15:43, Andriy Berestovskyy wrote:
> [..]
> > Fixes: e70a61ad50ab ("keepalive: export states")
> > Cc: remy.hor...@intel.com
> > Signed-off-by: Andriy Berestovskyy
>
> Acked-by: Remy Horton
Applied
23/01/2018 14:59, Shreyansh Jain:
> Defining a very generic super-set of device type and its device operations
> that
> can be exposed such that any new/upcoming/experimental device can be layered
> over it. 'rawdevice' semantic in this patchset represents a device that
> doesn't
> have any flavo
22/01/2018 14:14, Olivier Matz:
> On Sat, Jan 20, 2018 at 11:26:31AM +0800, Xueming Li wrote:
> > The result buffer was not initialized before parsing, inducing garbage
> > in unused fields or padding of the parsed structure.
> >
> > Initialize the result buffer each time before parsing.
> >
> >
22/01/2018 14:13, Olivier Matz:
> On Sat, Jan 20, 2018 at 02:16:10AM +0800, Xueming Li wrote:
> > When using dynamic tokens, the result buffer contains pointers to some
> > location inside the result buffer. When the content of the temporary
> > buffer is copied in the final one, these pointers sti
24/01/2018 18:02, Harry van Haaren:
> This commit ensures that if that if we run out of memory
> during the initialization of the service library, that the
> first allocated memory is correctly freed instead of leaked.
>
> Fixes: 21698354c832 ("service: introduce service cores concept")
> Cc: sta.
11/01/2018 19:20, Vipin Varghese:
> This patch fixes the memory leak created by rte_service_init. When
> secondary application which shares the huge page from primary, is
> executed multiple times memory is initialized but not freed on exit
>
> The rte_service_finalize checks if the service is ini
22/01/2018 14:51, Hemant Agrawal:
> From: Pavan Nikhilesh
>
> Set the mempool preferred by the ethernet devices as default mbuf
> mempool before creating the pktpool.
The patch is just adding a log.
Something is not consistent.
22/01/2018 14:51, Hemant Agrawal:
> This patch change the logic to maintain the value of
> user defined and compile time i.e. RTE_MBUF_DEFAULT_MEMPOOL_OPS.
>
> The pktmbuf_create_pool is updated to reflect the same.
I cannot understand this text.
22/01/2018 14:51, Hemant Agrawal:
> This patch add support for various mempool ops config helper APIs.
>
> 1.User defined mempool ops
> 2.Platform detected HW mempool ops (active).
> 3.Best selection of mempool ops by looking into user defined,
> platform registered and compile time configured.
22/01/2018 14:07, Olivier Matz:
> On Fri, Jan 19, 2018 at 06:18:13PM +0800, Zhiyong Yang wrote:
> > It is unnecessary to cast from void * to struct rte_mbuf *,
> > the change can make code clearer.
> >
> > Signed-off-by: Zhiyong Yang
>
> Acked-by: Olivier Matz
Applied, thanks
25/01/2018 18:00, Andrew Rybchenko:
> Unfortunately we're a bit late with dynamic logging implementation.
> So, it can wait for 18.05 release cycle if required.
>
> The series adds EXPERIMENTAL EAL feature which removes dependency
> on EAL arguments processing and log types registration. It stores
On Thu, Jan 25, 2018 at 11:12:25AM +, Bruce Richardson wrote:
> Rather than having to work off files all the time, allow stdin and stdout
> to be used as the source and destination for pmdinfogen. This will allow
> other possible usages from scripts, e.g. taking files from ar archive and
> buil
25/01/2018 20:21, Jianfeng Tan:
> v3->v5:
> - Drop the patch 3 on vfio communication (postponed).
> - Change names from rte_eal_mp_* -> rte_mp_* as suggested by Thomas.
You did not update the .map file for this change.
Polling a new packet is basically sensing the generation bit in a
completion entry. For some processors not having strongly-ordered memory
model, there has to be a memory barrier between reading the generation bit
and other fields of the entry in order to guarantee data is not stale.
Fixes: 570acd
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_rxtx.c | 4 ++--
drivers/net/mlx5/mlx5_rxtx.h | 2 +-
drivers/net/mlx5/mlx5_rxtx_vec.h | 2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
index 3b8f71c28..
Signed-off-by: Yongseok Koh
---
lib/librte_eal/common/include/arch/ppc_64/rte_atomic.h | 4
1 file changed, 4 insertions(+)
diff --git a/lib/librte_eal/common/include/arch/ppc_64/rte_atomic.h
b/lib/librte_eal/common/include/arch/ppc_64/rte_atomic.h
index 150810cdb..f38618f90 100644
--- a/l
As rte_write64() has an IO barrier, there's no need to have a barrier
before the call.
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_rxq.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/mlx5/mlx5_rxq.c b/drivers/net/mlx5/mlx5_rxq.c
index 3c716b960..ca6cd3ae6 100644
--- a/
Cc: Thomas Speier
Signed-off-by: Yongseok Koh
Acked-by: Thomas Speier
Acked-by: Jianbo Liu
---
lib/librte_eal/common/include/arch/arm/rte_atomic_64.h | 4
1 file changed, 4 insertions(+)
diff --git a/lib/librte_eal/common/include/arch/arm/rte_atomic_64.h
b/lib/librte_eal/common/include
Signed-off-by: Yongseok Koh
Acked-by: Jianbo Liu
---
lib/librte_eal/common/include/arch/arm/rte_atomic_32.h | 4
1 file changed, 4 insertions(+)
diff --git a/lib/librte_eal/common/include/arch/arm/rte_atomic_32.h
b/lib/librte_eal/common/include/arch/arm/rte_atomic_32.h
index 14c048640..d2
This patchset is to introduce coherent I/O memory barriers, which could be more
efficient for coherent memory between I/O device and CPU, especially for ARMv8.
v4:
* rename barriers to "coherent I/O memory barrier".
* Make groups for various barriers in Doxygen doc.
v3:
* add more detailed commen
Signed-off-by: Yongseok Koh
---
lib/librte_eal/common/include/generic/rte_atomic.h | 12
1 file changed, 12 insertions(+)
diff --git a/lib/librte_eal/common/include/generic/rte_atomic.h
b/lib/librte_eal/common/include/generic/rte_atomic.h
index 3ba7245a3..58c40489b 100644
--- a/lib
This commit introduces rte_cio_wmb() and rte_cio_rmb(), in order to
guarantee the ordering of coherent shared memory between the CPU and a DMA
capable device.
Signed-off-by: Yongseok Koh
Acked-by: Andrew Rybchenko
---
lib/librte_eal/common/include/generic/rte_atomic.h | 39 +
Signed-off-by: Yongseok Koh
---
lib/librte_eal/common/include/arch/x86/rte_atomic.h | 4
1 file changed, 4 insertions(+)
diff --git a/lib/librte_eal/common/include/arch/x86/rte_atomic.h
b/lib/librte_eal/common/include/arch/x86/rte_atomic.h
index 36cfabc38..8fb796c63 100644
--- a/lib/librte
We need the synchronous way for multi-process communication,
i.e., blockingly waiting for reply message when we send a request
to the peer process.
We add two APIs rte_eal_mp_request() and rte_eal_mp_reply() for
such use case. By invoking rte_eal_mp_request(), a request message
is sent out, and th
v3->v5:
- Drop the patch 3 on vfio communication (postponed).
- Change names from rte_eal_mp_* -> rte_mp_* as suggested by Thomas.
- Add nb_sent and nb_received in struct rte_mp_reply.
- Standardize the return val of sendmsg, request, reply: 0 on sucess,
(-1) on failure.
- If we found
Previouly, there are three channels for multi-process
(i.e., primary/secondary) communication.
1. Config-file based channel, in which, the primary process writes
info into a pre-defined config file, and the secondary process
reads the info out.
2. vfio submodule has its own channel ba
Apology, please ignore this version which is not correct. Will send out
a new version.
On 1/26/2018 3:14 AM, Jianfeng Tan wrote:
v3->v4:
- Drop the patch 3 on vfio communication (postponed).
- Change names from rte_eal_mp_* -> rte_mp_* as suggested by Thomas.
- Add nb_sent and nb_rece
Previously, vfio uses its own private channel for the secondary
process to get container fd and group fd from the primary process.
This patch changes to use the generic mp channel.
Test:
1. Bind two NICs to vfio-pci.
2. Start the primary and secondary process.
$ (symmetric_mp) -c 2 -- -p
We need the synchronous way for multi-process communication,
i.e., blockingly waiting for reply message when we send a request
to the peer process.
We add two APIs rte_eal_mp_request() and rte_eal_mp_reply() for
such use case. By invoking rte_eal_mp_request(), a request message
is sent out, and th
v3->v4:
- Drop the patch 3 on vfio communication (postponed).
- Change names from rte_eal_mp_* -> rte_mp_* as suggested by Thomas.
- Add nb_sent and nb_received in struct rte_mp_reply.
- Standardize the return val of sendmsg, request, reply: 0 on sucess,
(-1) on failure.
- If we found
> On Jan 22, 2018, at 8:35 PM, Jerin Jacob
> wrote:
>
> -Original Message-
>> Date: Mon, 22 Jan 2018 18:29:31 +
>> From: Yongseok Koh
>> To: Andrew Rybchenko , Thomas Monjalon
>> , "jianbo@arm.com" , Jerin
>> Jacob
>> CC: Adrien Mazarguil , Nélio Laranjeiro
>> , "bruce.richard
> On Jan 25, 2018, at 8:18 AM, Shahaf Shuler wrote:
>
> Since commit f81ec748434b ("net/mlx5: fix memory region lookup") the
> Memory Region (MR) are no longer ovetrlaps.
>
> Comparing the end address of the MR should be exclusive, otherwise two
> contiguous MRs may cause wrong matching.
>
> F
Hi Shally, Ahmed,
> -Original Message-
> From: Verma, Shally [mailto:shally.ve...@cavium.com]
> Sent: Thursday, January 25, 2018 10:25 AM
> To: Ahmed Mansour ; Trahe, Fiona
> ;
> dev@dpdk.org; Akhil Goyal
> Cc: Challa, Mahipal ; Athreya, Narayana Prasad
> ; De Lara Guarch, Pablo
> ;
>
On Thu, 25 Jan 2018 17:00:43 +
Andrew Rybchenko wrote:
> diff --git a/drivers/net/sfc/efsys.h b/drivers/net/sfc/efsys.h
> index c7a54c3..8dd225e 100644
> --- a/drivers/net/sfc/efsys.h
> +++ b/drivers/net/sfc/efsys.h
> @@ -26,6 +26,7 @@
> #include
>
> #include "sfc_debug.h"
> +#include "s
Hi All,
Sorry for the delay. Please see responses inline.
Ahmed
On 1/12/2018 8:50 AM, Verma, Shally wrote:
> Hi Fiona
>
>> -Original Message-
>> From: Trahe, Fiona [mailto:fiona.tr...@intel.com]
>> Sent: 12 January 2018 00:24
>> To: Verma, Shally ; Ahmed Mansour
>> ; dev@dpdk.org
>> Cc:
On 25-Jan-18 5:10 PM, Tan, Jianfeng wrote:
On 1/26/2018 12:22 AM, Burakov, Anatoly wrote:
On 25-Jan-18 3:03 PM, Ananyev, Konstantin wrote:
-Original Message-
From: Burakov, Anatoly
Sent: Thursday, January 25, 2018 1:10 PM
To: Ananyev, Konstantin ; Tan,
Jianfeng ; dev@dpdk.org
Cc:
25/01/2018 17:48, Bruce Richardson:
> On Wed, Jan 24, 2018 at 03:07:45PM +, Marko Kovacevic wrote:
> > Fix build issue with pdf guides. Some indentations in the bbdev test
> > application doc were causing build failures. Latex Log message:
> >
> > doc.log:! LaTeX Error: Too deeply nested.
> -Original Message-
> From: Van Haaren, Harry [mailto:harry.van.haa...@intel.com]
> Sent: Thursday, January 25, 2018 12:06 PM
> Subject: RE: [PATCH 06/18] net/avp: align dynamic log names with standard
>
> > From: Legacy, Allain [mailto:allain.leg...@windriver.com]
> > Sent: Thursday, Jan
Thursday, January 25, 2018 7:12 PM, Mordechay Haimovsky:
not support "don't care" bits in its flow spec mask, therefore
> we do cannot use The following rules used by mlx5 for promiscuous and
> allmulti:
> if (dev->data->promiscuous) {
> struct rte_flow_item_eth promisc = {
1 - 100 of 258 matches
Mail list logo
