The selection between scalar mode and vector mode was removed for
ixgbe driver in DPDK 20.02. The code base still has code
for scalar mode but it cannot be selected at compile time. The
vector mode is adequate for applications which require high
throughput which is the case most of the time, but
On Mon, May 24, 2021 at 11:31 AM Thierry Herbelot
wrote:
>
> Use stats only after it was checked not to be NULL.
>
> Fixes: af1ae8b6a32c9 ("graph: implement stats")
> Cc: sta...@dpdk.org
> Cc: Jerin Jacob
> Cc: Kiran Kumar K
>
> Signed-off-by: Thierry Herbelot
> --
> V2: reword patch title
> --
> -Original Message-
> From: Chengwen Feng
> Sent: Monday, May 24, 2021 9:23 PM
> To: tho...@monjalon.net; ferruh.yi...@intel.com
> Cc: dev@dpdk.org; jer...@marvell.com; Ruifeng Wang
> ; vikto...@rehivetech.com;
> bruce.richard...@intel.com; Honnappa Nagarahalli
> ; jerinjac...@gmail.com;
> -Original Message-
> From: Chengwen Feng
> Sent: Friday, May 21, 2021 9:53 AM
> To: tho...@monjalon.net; ferruh.yi...@intel.com; Ruifeng Wang
>
> Cc: dev@dpdk.org; bruce.richard...@intel.com;
> vladimir.medved...@intel.com; vikto...@rehivetech.com;
> jer...@marvell.com
> Subject: [PATCH
The tests 'alarm_autotest' and 'cycles_autotest' rely on the underlying
system having very accurate and precise timing. On systems where the timing
isn't as rigid, or the load is particularly high, these tests are unreliable
since the wake latency from the scheduler can be high enough to miss the
Make the definition of the table used to map hardware packet type
information to DPDK packet type more generic.
Add macro definitions for constants used in creating table
indices, use these to eliminate raw constants in code.
Add build-time assertions to validate ptype mapping constants.
Signed-
Implement AVX2 vector PMD.
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Khaparde
---
doc/guides/nics/bnxt.rst | 57 ++-
drivers/net/bnxt/bnxt_ethdev.c| 119 +++--
drivers/net/bnxt/bnxt_rxr.c | 4 +-
drivers/net/bnxt/bnxt_rxr.h | 11 +-
drivers/n
The burst receive function should return all packets currently
present in the receive ring up to the requested burst size,
update vector mode receive functions accordingly.
Fixes: 398358341419 ("net/bnxt: support NEON")
Fixes: bc4a000f2f53 ("net/bnxt: implement SSE vector mode")
Cc: sta...@dpdk.or
Vector mode updates for the bnxt PMD.
Lance Richardson (3):
net/bnxt: refactor HW ptype mapping table
net/bnxt: fix Rx burst size constraint
net/bnxt: add AVX2 vector PMD
doc/guides/nics/bnxt.rst | 57 ++-
drivers/net/bnxt/bnxt_ethdev.c| 119 +++--
drivers/net/bnxt/bn
I am looking at the source for rte_memcpy (this is a discussion only for
x86-64)
For one of the cases, when aligned correctly, it uses
/**
* Copy 64 bytes from one location to another,
* locations should not overlap.
*/
static __rte_always_inline void
rte_mov64(uint8_t *dst, const uint8_t *src
Add in missing rte_mbuf size in mempool allocation for out-of-place op.
Fixes: bf9d6702eca9 ("app/crypto-perf: use single mempool")
Cc: pablo.de.lara.gua...@intel.com
Signed-off-by: Kai Ji
Acked-by: Fan Zhang
Acked-by: Pablo de Lara
---
v2:
-- commit message update
app/test-crypto-pe
On Mon, 24 May 2021 11:58:19 +0100
"Liguzinski, WojciechX" wrote:
> DPDK sched library is equipped with mechanism that secures it from the
> bufferbloat problem
> which is a situation when excess buffers in the network cause high latency
> and latency
> variation. Currently, it supports RED fo
Currently, the soc_kunpeng930 declares '-march=armv8.2-a+crypto+sve',
but some compiler doesn't recognize the march because it doesn't
support sve.
To solve this bug we use the following scheme:
1. Define 'march_base' tuple which defines support march, it should
arrange from lower to higher.
e.g.
Currently, the SVE code is compiled only when -march supports SVE
(e.g. '-march=armv8.2a+sve'), there maybe some problem[1] with this
approach.
The solution:
a. If the minimum instruction set support SVE then compiles it.
b. Else if the compiler support SVE then compiles it.
c. Otherwise don't com
This patch set contains two bugfixes for Kunpeng SVE compile.
---
v8:
* rm config/arm/meson.build from 2/2 patch which add in v7
v7:
* reuse CC_SVE_ACLE_SUPPORT macro in 2/2 patch
v6:
* fix compile error with gcc8.3 and thunder2 of 2/2 patch
v5:
* make sure 'march_feature' parsed depend on 'march_
Fix in v7, thanks
On 2021/5/24 18:03, Ruifeng Wang wrote:
>> -Original Message-
>> From: fengchengwen
>> Sent: Monday, May 24, 2021 4:44 PM
>> To: Ruifeng Wang ; tho...@monjalon.net;
>> ferruh.yi...@intel.com
>> Cc: dev@dpdk.org; jer...@marvell.com; vikto...@rehivetech.com;
>> bruce.richa
Currently, the SVE code is compiled only when -march supports SVE
(e.g. '-march=armv8.2a+sve'), there maybe some problem[1] with this
approach.
The solution:
a. If the minimum instruction set support SVE then compiles it.
b. Else if the compiler support SVE then compiles it.
c. Otherwise don't com
Currently, the soc_kunpeng930 declares '-march=armv8.2-a+crypto+sve',
but some compiler doesn't recognize the march because it doesn't
support sve.
To solve this bug we use the following scheme:
1. Define 'march_base' tuple which defines support march, it should
arrange from lower to higher.
e.g.
This patch set contains two bugfixes for Kunpeng SVE compile.
---
v7:
* reuse CC_SVE_ACLE_SUPPORT macro in 2/2 patch
v6:
* fix compile error with gcc8.3 and thunder2 of 2/2 patch
v5:
* make sure 'march_feature' parsed depend on 'march_base'
* only the kunpeng930 use the new option
v4:
* fix compil
From: Pavan Nikhilesh
Add Rx event vector fastpath to convert HW defined metadata into
rte_mbuf and rte_event_vector.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn10k_worker.h | 50 +++
1 file changed, 50 insertions(+)
diff --git a/drivers/event/cnxk/cn1
From: Pavan Nikhilesh
Add event vector support for cnxk event Rx adapter, add control path
APIs to get vector limits and ability to configure event vectorization
on a given Rx queue.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn10k_eventdev.c | 106 ++-
driv
From: Pavan Nikhilesh
Add support for event eth Tx adapter fastpath operations.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn10k_eventdev.c | 35
drivers/event/cnxk/cn10k_worker.c | 32 +++
drivers/event/cnxk/cn10k_worker.h | 67 ++
drive
From: Pavan Nikhilesh
Add support for event eth Tx adapter.
Signed-off-by: Pavan Nikhilesh
---
doc/guides/eventdevs/cnxk.rst| 4 +-
drivers/event/cnxk/cn10k_eventdev.c | 90 +
drivers/event/cnxk/cn9k_eventdev.c | 117 +++
drivers/ev
From: Pavan Nikhilesh
Add support for event eth Rx adapter fastpath operations.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn10k_eventdev.c | 115 -
drivers/event/cnxk/cn10k_worker.c | 164 +
drivers/event/cnxk/cn10k_worker.h | 91 +--
drivers/event/cnxk
From: Pavan Nikhilesh
Add support for event eth Rx adapter.
Signed-off-by: Pavan Nikhilesh
---
This patch set depends on
http://patches.dpdk.org/project/dpdk/list/?series=15515
doc/guides/eventdevs/cnxk.rst| 4 +
drivers/event/cnxk/cn10k_eventdev.c | 76 +++
driv
Hi, all,
any comments for this set?
在 2021/4/23 19:01, Min Hu (Connor) 写道:
This patch set contains two bugfixes for sched.
Huisong Li (2):
lib/sched: fix return value judgment
lib/sched: optimize exception handling code
lib/sched/rte_sched.c | 60 +++---
Hi, Thierry,
This bug has been fixed in my earlier patch, please refer to:
https://patches.dpdk.org/project/dpdk/patch/1619175672-20016-2-git-send-email-humi...@huawei.com/
Thanks.
在 2021/5/24 19:30, Thierry Herbelot 写道:
Check port->subport_profiles after it was allocated.
Fix
Hi, Thierry,
This bug has been fixed in my earlier patch, please refer to:
https://patches.dpdk.org/project/dpdk/patch/1618839289-33224-5-git-send-email-humi...@huawei.com/
Thanks.
在 2021/5/24 16:57, Thierry Herbelot 写道:
Check that nb_jobs is not zero before using it for a divi
X550 NIC can support runtime tx/rx queue setup.
Add capacity in dev_capa and queue offload capacity.
Signed-off-by: Wu Jianyue
---
drivers/net/ixgbe/ixgbe_ethdev.c | 22 ++
drivers/net/ixgbe/ixgbe_rxtx.c | 21 +++--
2 files changed, 41 insertions(+), 2 delet
Take into account VLAN presence fields in items ETH and VLAN.
Provided that the item ETH does not match on the EtherType,
the pattern behaviour will be as follows:
- ETH (mask->has_vlan = 0) | IPv4= match both tagged and untagged;
- ETH (mask->has_vlan = 1) | IPv4= match as per sp
Introduce necessary infrastructure for these fields to
be set, validated and compared during class comparison.
Enumeration and mappings envisaged are MCDI-compatible.
Signed-off-by: Ivan Malov
Reviewed-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Acked-by: Ray Kinsella
---
drivers/common/sf
Changes in v2:
* [1/3] no longer triggers spelling warnings
Changes in v3:
* Attempt to correct mailing order of the patches
Changes in v4:
* Yet another attempt to correct mailing order of the patches
Andrew Rybchenko (1):
common/sfc_efx/base: update MCDI headers
Ivan Malov (2):
common/sfc
> -Original Message-
> From: Thierry Herbelot
> Sent: Monday, May 24, 2021 12:30 PM
> To: dev@dpdk.org
> Cc: Thierry Herbelot ; Thomas Monjalon
> ; sta...@dpdk.org; Dumitrescu, Cristian
> ; Singh, Jasvinder
>
> Subject: [PATCH v2] sched: check newly allocated pointer
>
> Check port->s
> -Original Message-
> From: Thierry Herbelot
> Sent: Monday, May 24, 2021 12:29 PM
> To: dev@dpdk.org
> Cc: Thierry Herbelot ; Thomas Monjalon
> ; sta...@dpdk.org; Dumitrescu, Cristian
>
> Subject: [PATCH v2] table: fix typo
>
> Check Bucket key for all possible indices.
>
> Fixes:
Check port->subport_profiles after it was allocated.
Fixes: 0ea4c6afcaf14 ("sched: add subport profile table")
Cc: sta...@dpdk.org
Cc: Cristian Dumitrescu
Cc: Jasvinder Singh
Signed-off-by: Thierry Herbelot
--
V2: reword patch title
---
lib/sched/rte_sched.c | 2 +-
1 file changed, 1 insertio
Check Bucket key for all possible indices.
Fixes: d0a00966618ba ("table: add exact match SWX table")
Cc: sta...@dpdk.org
Cc: Cristian Dumitrescu
Signed-off-by: Thierry Herbelot
--
V2: reword patch title
---
lib/table/rte_swx_table_em.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
dif
> -Original Message-
> From: Thierry Herbelot
> Sent: Monday, May 24, 2021 9:58 AM
> To: dev@dpdk.org
> Cc: Thierry Herbelot ; Thomas Monjalon
> ; sta...@dpdk.org; Dumitrescu, Cristian
> ; Singh, Jasvinder
>
> Subject: [PATCH] rte_sched: check newly allocated pointer
>
> Check port->s
> -Original Message-
> From: Thierry Herbelot
> Sent: Monday, May 24, 2021 9:54 AM
> To: dev@dpdk.org
> Cc: Thierry Herbelot ; Thomas Monjalon
> ; sta...@dpdk.org; Dumitrescu, Cristian
>
> Subject: [PATCH] rte_swx: fix typo
>
> Check Bucket key for all possible indices.
>
> Fixes: d0
Do not return "GOOD" flag if the status was bad.
Fixes: 9e6e7f479ae9b ("net/sfc: support Rx checksum offload for EF100")
Cc: sta...@dpdk.org
Cc: Andrew Rybchenko
Signed-off-by: Thierry Herbelot
---
drivers/net/sfc/sfc_ef100_rx.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
Take into account VLAN presence fields in items ETH and VLAN.
Provided that the item ETH does not match on the EtherType,
the pattern behaviour will be as follows:
- ETH (mask->has_vlan = 0) | IPv4= match both tagged and untagged;
- ETH (mask->has_vlan = 1) | IPv4= match as per sp
Introduce necessary infrastructure for these fields to
be set, validated and compared during class comparison.
Enumeration and mappings envisaged are MCDI-compatible.
Signed-off-by: Ivan Malov
Reviewed-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Acked-by: Ray Kinsella
---
drivers/common/sf
Changes in v2:
* [1/3] no longer triggers spelling warnings
Changes in v3:
* Attempt to correct mailing order of the patches
Andrew Rybchenko (1):
common/sfc_efx/base: update MCDI headers
Ivan Malov (2):
common/sfc_efx/base: add MAE VLAN presence match bits
net/sfc: support matching on VLA
Signed-off-by: Liguzinski, WojciechX
---
examples/ip_pipeline/tmgr.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/examples/ip_pipeline/tmgr.c b/examples/ip_pipeline/tmgr.c
index e4e364cbc0..406184e760 100644
--- a/examples/ip_pipeline/tmgr.c
+++ b/examples/ip_pipeline/t
patch add support enable pie or red by
parsing config file.
Signed-off-by: Liguzinski, WojciechX
---
config/rte_config.h | 1 -
examples/qos_sched/app_thread.c | 1 -
examples/qos_sched/cfg_file.c | 82 ++---
examples/qos_sched/init.c | 5 +-
examples/qos_sched
Implement pie based congestion management based on rfc8033
Signed-off-by: Liguzinski, WojciechX
---
drivers/net/softnic/rte_eth_softnic_tm.c | 4 +-
lib/sched/meson.build| 10 +-
lib/sched/rte_sched.c| 220 +--
lib/sched/rte_sched.h
DPDK sched library is equipped with mechanism that secures it from the
bufferbloat problem
which is a situation when excess buffers in the network cause high latency and
latency
variation. Currently, it supports RED for queue congestion control (which is
designed
to control the queue length bu
> -Original Message-
> From: Thierry Herbelot
> Sent: Monday, May 24, 2021 3:08 PM
> To: dev@dpdk.org
> Cc: Thierry Herbelot ; Thomas Monjalon
> ; sta...@dpdk.org; Gagandeep Singh
>
> Subject: [PATCH v2] net/pfe: no need to check dev to be NULL Pointer
>
> librte_ethdev library is alr
> -Original Message-
> From: fengchengwen
> Sent: Monday, May 24, 2021 4:44 PM
> To: Ruifeng Wang ; tho...@monjalon.net;
> ferruh.yi...@intel.com
> Cc: dev@dpdk.org; jer...@marvell.com; vikto...@rehivetech.com;
> bruce.richard...@intel.com; Honnappa Nagarahalli
> ; jerinjac...@gmail.com;
>
librte_ethdev library is already doing some checks on dev before calling
the link update.
Fixes: acd4818ea2a45 ("net/pfe: add link status update")
Cc: sta...@dpdk.org
Cc: Gagandeep Singh
Signed-off-by: Thierry Herbelot
--
V2: rework after noting dev cannot be NULL
---
drivers/net/pfe/pfe_ethde
Use stats only after it was checked not to be NULL.
Fixes: af1ae8b6a32c9 ("graph: implement stats")
Cc: sta...@dpdk.org
Cc: Jerin Jacob
Cc: Kiran Kumar K
Signed-off-by: Thierry Herbelot
--
V2: reword patch title
---
lib/graph/graph_stats.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
> -Original Message-
> From: Thierry Herbelot
> Sent: Monday, May 24, 2021 2:29 PM
> To: dev@dpdk.org
> Cc: Thierry Herbelot ; Thomas Monjalon
> ; sta...@dpdk.org; Gagandeep Singh
>
> Subject: [PATCH] net/pfe: do not use a possibly NULL Pointer
>
> Use dev only after it was checked no
On Mon, May 24, 2021 at 2:28 PM Thierry Herbelot
wrote:
>
> Use stats only after it was checked not to be NULL.
Please change the subject to graph: ..
With the above fix:
Acked-by: Jerin Jacob
>
> Fixes: af1ae8b6a32c9 ("graph: implement stats")
> Cc: sta...@dpdk.org
> Cc: Jerin Jacob
> C
Use vc_req only after it was checked not to be NULL.
Fixes: 2d962bb736521 ("vhost/crypto: fix possible TOCTOU attack")
Cc: sta...@dpdk.org
Cc: Maxime Coquelin
Cc: Chenbo Xia
Signed-off-by: Thierry Herbelot
--
V2: fix copy/paste typo
---
lib/vhost/vhost_crypto.c | 6 --
1 file changed, 4 i
The flag use_avx2 and use_avx512 are defined as local variables, they
will not be aware by the secondary process, then wrong data path is
selected. Fix the issue by moving them into struct ice_adapter.
Fixes: ae60d3c9b227 ("net/ice: support Rx AVX2 vector")
Fixes: 2d5f6953d56d ("net/ice: support v
Use out only after it was checked not to be NULL.
Fixes: 4d07cbefe3ba0 ("app/testpmd: add commands for conntrack")
Cc: sta...@dpdk.org
Cc: Ori Kam
Signed-off-by: Thierry Herbelot
---
app/test-pmd/cmdline_flow.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/test-pmd/cm
Use m only after it was checked not to be NULL.
Fixes: 202d375c60bc1 ("app/test: add cryptodev unit and performance tests")
Cc: sta...@dpdk.org
Cc: Declan Doherty
Signed-off-by: Thierry Herbelot
---
app/test/test_cryptodev.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --gi
Check that pointers are valid before using them.
Fixes: 7bc8e9a227ccb ("net/bnxt: support async link notification")
Cc: sta...@dpdk.org
Cc: Ajit Khaparde
Cc: Somnath Kotur
Signed-off-by: Thierry Herbelot
---
drivers/net/bnxt/bnxt_irq.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
mz is known to be NULL: do not use it to print a memzone name.
Fixes: 242e18c056890 ("net/octeontx_ep: add Rx queue setup and release")
Cc: sta...@dpdk.org
Cc: Nalla Pradeep
Cc: Radha Mohan Chintakuntla
Cc: Veerasenareddy Burru
Signed-off-by: Thierry Herbelot
---
drivers/net/octeontx_ep/otx_
Use dev only after it was checked not to be NULL.
Fixes: acd4818ea2a45 ("net/pfe: add link status update")
Cc: sta...@dpdk.org
Cc: Gagandeep Singh
Signed-off-by: Thierry Herbelot
---
drivers/net/pfe/pfe_ethdev.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/
Use stats only after it was checked not to be NULL.
Fixes: af1ae8b6a32c9 ("graph: implement stats")
Cc: sta...@dpdk.org
Cc: Jerin Jacob
Cc: Kiran Kumar K
Signed-off-by: Thierry Herbelot
---
lib/graph/graph_stats.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/graph/g
Use vc_req only after it was checked not to be NULL.
Fixes: 2d962bb736521 ("vhost/crypto: fix possible TOCTOU attack")
Cc: sta...@dpdk.org
Cc: Maxime Coquelin
Cc: Chenbo Xia
Signed-off-by: Thierry Herbelot
---
lib/vhost/vhost_crypto.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-
Check port->subport_profiles after it was allocated.
Fixes: 0ea4c6afcaf14 ("sched: add subport profile table")
Cc: sta...@dpdk.org
Cc: Cristian Dumitrescu
Cc: Jasvinder Singh
Signed-off-by: Thierry Herbelot
---
lib/sched/rte_sched.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
Check that nb_jobs is not zero before using it for a division.
Fixes: f5cffb7eb7fb6 ("app/regex: read data file once at startup")
Cc: sta...@dpdk.org
Cc: Ori Kam
Signed-off-by: Thierry Herbelot
---
app/test-regex/main.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/app/test-regex/main.
Check Bucket key for all possible indices.
Fixes: d0a00966618ba ("table: add exact match SWX table")
Cc: sta...@dpdk.org
Cc: Cristian Dumitrescu
Signed-off-by: Thierry Herbelot
---
lib/table/rte_swx_table_em.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/table/rte_sw
I am looking at the source for rte_memcpy (this is a discussion only for
x64-64)
For one of the cases, when aligned correctly, it uses
/**
* Copy 64 bytes from one location to another,
* locations should not overlap.
*/
static __rte_always_inline void
rte_mov64(uint8_t *dst, const uint8_t *src
On 2021/5/24 13:38, Ruifeng Wang wrote:
>> -Original Message-
>> From: fengchengwen
>> Sent: Friday, May 21, 2021 2:53 PM
>> To: Ruifeng Wang ; tho...@monjalon.net;
>> ferruh.yi...@intel.com
>> Cc: dev@dpdk.org; jer...@marvell.com; vikto...@rehivetech.com;
>> bruce.richard...@intel.com;
Hi Thomas,
Sure, no problem. Thanks.
Congrats on 21.05 release!
Regards,
Fan
> -Original Message-
> From: Thomas Monjalon
> Sent: Friday, May 21, 2021 4:37 PM
> To: Zhang, Roy Fan ; Ji, Kai ;
> De Lara Guarch, Pablo
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [dpdk-dev v1] cperf: fix
67 matches
Mail list logo
