> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Monday, 20 February 2023 16.41
>
> 20/02/2023 16:35, Bruce Richardson:
> > On Mon, Feb 20, 2023 at 04:30:20PM +0100, Thomas Monjalon wrote:
> > > 20/02/2023 16:07, Morten Brørup:
> > > > With the new viewing angle, the current define RTE
We were writing a value that should represent the number of items to be
processed by hardware. The value being written was off by 1 (N*4)+3;
The value should be (N*4) + 4 simplified to (N+1)*4
Fixes: 5dfa003db53f ("common/mlx5: fix post doorbell barrier")
Cc: sta...@dpdk.org
Signed-off-by: Gerry
Fix overflow of free QP mask.
Regex used 64 QPs and used a bitmask to select a free QP for use.
The bitmask in use was only 32 bits so did not allow half of the QPs
to be utilised.
Upgraded to 64 bit mask and using ffsll now instead of ffs.
Fixes: 270032608503 ("regex/mlx5: refactor HW queue objec
> -Original Message-
> From: fengchengwen
> Sent: Monday, February 20, 2023 2:58 PM
> To: Ashok Kaladi ; jer...@marvell.com;
> tho...@monjalon.net
> Cc: dev@dpdk.org; s.v.naga.haris...@intel.com; erik.g.carri...@intel.com;
> abhinandan.guj...@intel.com; sta...@dpdk.org; Ruifeng Wang
>
>
Add support for decompress LZ4 algorithm for mlx5 PMD.
Signed-off-by: Michael Baum
Acked-by: Matan Azrad
---
doc/guides/compressdevs/features/mlx5.ini | 18 ++-
doc/guides/compressdevs/mlx5.rst | 44 +-
doc/guides/rel_notes/release_23_03.rst| 4 +
drivers/compress/mlx5/mlx5
Add capabilities check for LZ4 decompression algorithm.
Signed-off-by: Michael Baum
Acked-by: Matan Azrad
---
drivers/common/mlx5/mlx5_devx_cmds.c | 6 ++
drivers/common/mlx5/mlx5_devx_cmds.h | 3 +++
drivers/common/mlx5/mlx5_prm.h | 16 ++--
3 files changed, 23 insertio
Separate the xform validations from create function into new function.
The new function checks caps directly from HCA attr structure, so all
capabilities in priv structure were removed.
Signed-off-by: Michael Baum
Acked-by: Matan Azrad
---
drivers/compress/mlx5/mlx5_compress.c | 75
The priv structure has variable named "min_block_size" coming from HCA
capabilities.
This field isn't used and copied into the priv structure for free.
This patch removes this field.
Signed-off-by: Michael Baum
Acked-by: Matan Azrad
---
drivers/compress/mlx5/mlx5_compress.c | 3 ---
1 file ch
Commit [1] add support in Bluefield-3, but Bluefield-3 has different
GGA opaque structure than Bluefield-2.
This patch updates the PRM structure to include both versions, and
culculate the relevant offset for each version in control path.
Commit [1] 559014f232b4 ("compress/mlx5: add Bluefield-3 d
The mlx5_compress_qp_setup() function creates QP for compress,
decompress and DMA. Thus, the MMO flag is turned on only when all
operations are supported.
However, since partial transformations have been allowed, it should be
turn on for part of them.
This patch removes the compress MMO support r
After de/compress dequeue, the output checksum is copied into the op
structure. The "output_checksum" field in op structure is "uint64_t"
type, and the 32-bit checksums (CRC32, Adler-32) are copied into the
lower 32 bits.
When both CRC32 and Adler-32 are configured, CRC32 is copied into the
lower
In xform creation, it first validate the xform according the
capabilities.
One of validations verifies that given "hash_algo" is
"RTE_COMP_HASH_ALGO_NONE" for both compress and decompress xform
objects.
However, the validation for decompress checks it again for compress
xform object.
This patch c
Add support for LZ4 decompression algorithm for mlx5 compress PMD.
Complete the support of compress PMD for Bluefield-3 smart-nic.
RFC:
https://patchwork.dpdk.org/project/dpdk/patch/20220410182622.8828-1-rzid...@nvidia.com/
v2:
- Rebase.
- Update PRM names.
- Add a new fix at the beginning of
>-Original Message-
>From: lihuisong (C)
>Sent: Tuesday, February 21, 2023 2:38 PM
>To: He, ShiyangX ; dev@dpdk.org
>Cc: Zhou, YidingX ; sta...@dpdk.org; Singh, Aman
>Deep ; Zhang, Yuying
>; Burakov, Anatoly ;
>Li, Xiaoyun ; Alvin Zhang
>Subject: Re: [PATCH] app/testpmd: fix secondary p
The former logic of set IPv4/v6 DSCP flow action don't consider the
mask filed of control message passed to the firmware.
This caused the firmware skip the set action logic, and the offloaded
packets don't have the right IPv4/v6 DSCP as expected.
Fixes: 9c665d70f7a7 ("net/nfp: support IPv4 DSCP fl
The former logic of set TTL flow action don't consider the mask
filed of control message passed to the firmware.
This caused the firmware skip the set action logic, and the offloaded
packets don't have the right TTL as expected.
Fixes: ac12e126c482 ("net/nfp: support TTL flow action")
Cc: sta...@d
The former logic of set TP source/destination flow action don't
consider the mask filed of control message passed to the firmware.
This caused the firmware skip the set action logic, and the offloaded
packets don't have the right TP as expected.
Fixes: fc185097bbe6 ("net/nfp: support TP source flo
The former logic of set IPv6 source/destination address flow action don't
consider the mask filed of control message passed to the firmware.
This caused the firmware skip the set action logic, and the offloaded
packets don't have the right IPv6 address as expected.
Fixes: 596ae2217214 ("net/nfp: s
The former logic of set IPv4 source/destination address flow action don't
consider the mask filed of control message passed to the firmware.
This caused the firmware skip the set action logic, and the offloaded
packets don't have the right IPv4 address as expected.
Fixes: ab80975b0a14 ("net/nfp: s
The former logic of set MAC source/destination address flow action don't
consider the mask filed of control message passed to the firmware.
This caused the firmware skip the set action logic, and the offloaded
packets don't have the right MAC address as expected.
Fixes: 4f6983154570 ("net/nfp: sup
The former logic of set flow actions don't consider the mask filed
of control message passed to the firmware, and this caused the
firmware skip the set action logic wrongly.
This patch series aims to fix this problem.
Chaoyong He (6):
net/nfp: fix the set MAC flow action
net/nfp: fix the set I
在 2023/2/21 10:52, He, ShiyangX 写道:
-Original Message-
From: lihuisong (C)
Sent: Monday, February 20, 2023 8:46 PM
To: He, ShiyangX ; dev@dpdk.org
Cc: Zhou, YidingX ; sta...@dpdk.org; Singh, Aman
Deep ; Zhang, Yuying
; Burakov, Anatoly ;
Li, Xiaoyun ; Alvin Zhang
Subject: Re: [PATCH
From: James Hershaw
Due to changes in the firmware for NFPs, firmware will no longer write
the link speed of a port to the control BAR. In line with the behaviour
of the kernel NFP driver, this is now handled by the PMD by reading the
value provided by the NSP in the nfp_eth_table struct within t
Aligned with kernel driver, optimized for inner and outer VLAN handling
in DPDK, and implemented double vlan insertion and stripping support.
1.adjust vlan stripping
Remove the judgment on dvm, vlan stripping only operates inner vlan.
2.support QinQ stripping
This patch support ice outer vlan str
From: Long Wu
"NFP_NET_CFG_RSS_IPV4" and "NFP_NET_CFG_RSS_IPV6" represent that firmware
will calculate RSS base on l3 rather than udp or tcp information from
packets. So the logic of return RSS configuration in driver is wrong.
Modify code to return the right configuration.
Fixes: 934e4c60fbff (
From: Long Wu
Add support for RSS parsing SCTP packets, including ipv4/ipv6 SCTP
and ipv4/ipv6 tunnel SCTP packets.
Signed-off-by: Long Wu
Reviewed-by: Chaoyong He
Reviewed-by: Niklas Söderlund
---
drivers/net/nfp/nfp_common.c | 16 +++-
drivers/net/nfp/nfp_ctrl.h | 4
2
Add support for RSS parsing of SCTP packets and modify the logic to
return the right RSS configuration.
Depends-on: series-27102 ("net/nfp: standardize the use of RSS-related")
Long Wu (2):
net/nfp: add support for RSS parsing of SCTP packets
net/nfp: fix getting wrong RSS configuration
dri
Added mempool test cases with zero-copy get and put API's
Signed-off-by: Kamalakshitha Aligeri
Reviewed-by: Ruifeng Wang
Reviewed-by: Feifei Wang
Acked-by: Morten Brørup
---
Depends-on: series-26984 ("[v9] mempool cache: add zero-copy get and put
functions")
app/test/test_mempool.c | 81 +++
Integrated zero-copy put API in mempool cache in i40e PMD.
On Ampere Altra server, l3fwd single core's performance improves by 5%
with the new API
Signed-off-by: Kamalakshitha Aligeri
Reviewed-by: Ruifeng Wang
Reviewed-by: Feifei Wang
Reviewed-by: Morten Brørup
---
Depends-on: series-26984 ("[
In i40e PMD, for fast_free path in Tx, freeing mbufs firstly need to do
an extra memcpy due to the special structure of ‘txep’.
Thus, replaced the put_bulk function with the zero-copy put API to
avoid the cost of this memcpy
On Ampere-altra server, for single-core DPDK l3fwd test, throughput
improv
From: Long Wu
The initial logic only support the single type metadata and this
commit add the support of chained type metadata. This commit also
make the relation between the RSS capability (v1/v2) and these
two types of metadata more clear.
Signed-off-by: Long Wu
Reviewed-by: Niklas Söderlund
From: Long Wu
The usage of RTE_ETH_MQ_RX_RSS and RTE_ETH_MQ_RX_RSS_FLAG are mixed in
nfp_net_configure(), use RTE_ETH_MQ_RX_RSS_FLAG uniformly.
Signed-off-by: Long Wu
Reviewed-by: Chaoyong He
Reviewed-by: Niklas Söderlund
---
drivers/net/nfp/nfp_common.c | 2 +-
1 file changed, 1 insertion(+
Select the appropriate RSS configuration in the initial process logic
and modify the logic of processing metadata based on RSS configuration
in the RX function.
---
v2:
* Correct spelling error
v3:
* Correct spelling error
---
Long Wu (2):
net/nfp: standardize the use of RSS-related codes
net/
From: Long Wu
The initial logic only support the single type metadata and this
commit add the support of chained type metadata. This commit also
make the relation between the RSS capability (v1/v2) and these
two types of metadata more clear.
Signed-off-by: Long Wu
Reviewed-by: Niklas Söderlund
From: Long Wu
The usage of RTE_ETH_MQ_RX_RSS and RTE_ETH_MQ_RX_RSS_FLAG are mixed in
nfp_net_configure(), use RTE_ETH_MQ_RX_RSS_FLAG uniformly.
Signed-off-by: Long Wu
Reviewed-by: Chaoyong He
Reviewed-by: Niklas Söderlund
---
drivers/net/nfp/nfp_common.c | 2 +-
1 file changed, 1 insertion(+
Select the appropriate RSS configuration in the initial process logic
and modify the logic of processing metadata based on RSS configuration
in the RX function.
---
v2:
* Correct spelling error
---
Long Wu (2):
net/nfp: standardize the use of RSS-related codes
net/nfp: modify RSS's processing
From: Long Wu
The initial logic only support the single type metadata and this
commit add the support of chained type metadata. This commit also
make the relation between the RSS capability (v1/v2) and these
two types of metadata more clear.
Signed-off-by: Long Wu
Reviewed-by: Niklas Söderlund
Select the appropriate RSS configuration in the initial process logic
and modify the logic of processing metadata based on RSS configuration
in the RX function.
Long Wu (2):
net/nfp: standardize the use of RSS-related codes
net/nfp: modify RSS's processing logic
drivers/net/nfp/nfp_common.c
From: Long Wu
The usage of RTE_ETH_MQ_RX_RSS and RTE_ETH_MQ_RX_RSS_FLAG are mixed in
nfp_net_configure(), use RTE_ETH_MQ_RX_RSS_FLAG uniformly.
Signed-off-by: Long Wu
Reviewed-by: Chaoyong He
Reviewed-by: Niklas Söderlund
---
drivers/net/nfp/nfp_common.c | 2 +-
1 file changed, 1 insertion(+
>-Original Message-
>From: lihuisong (C)
>Sent: Monday, February 20, 2023 8:46 PM
>To: He, ShiyangX ; dev@dpdk.org
>Cc: Zhou, YidingX ; sta...@dpdk.org; Singh, Aman
>Deep ; Zhang, Yuying
>; Burakov, Anatoly ;
>Li, Xiaoyun ; Alvin Zhang
>Subject: Re: [PATCH] app/testpmd: fix secondary pr
This patch add hw stats ierrors, when receive packets with
bad csum, ierrors value will increase.
Signed-off-by: Mingxia Liu
---
drivers/net/idpf/idpf_ethdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/idpf/idpf_ethdev.c b/drivers/net/idpf/idpf_ethdev.c
index 38cbbf369d..5b
> -邮件原件-
> 发件人: David Marchand
> 发送时间: Monday, February 20, 2023 8:07 PM
> 收件人: Feifei Wang
> 抄送: dev@dpdk.org; david.h...@intel.com; step...@networkplumber.org;
> tho...@monjalon.net; nd
> 主题: Re: [PATCH v6 0/2] Enable PMD power management on Arm
>
> On Mon, Feb 20, 2023 at 9:51 AM F
Add support for programming PMU counters and reading their values
in runtime bypassing kernel completely.
This is especially useful in cases where CPU cores are isolated
i.e run dedicated tasks. In such cases one cannot use standard
perf utility without sacrificing latency and performance.
Si
Thanks, I'll check it out.
On Mon, Feb 20, 2023 at 3:04 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> On Mon, 20 Feb 2023 12:44:06 -0800
> Bili Dong wrote:
>
> > Hi Cristian,
> >
> > I agree the 64-bit version could enable better performance, and I will do
> > it in the next versio
diff --git a/lib/pmu/rte_pmu.h b/lib/pmu/rte_pmu.h new file
mode
100644 index 00..6b664c3336
--- /dev/null
+++ b/lib/pmu/rte_pmu.h
@@ -0,0 +1,212 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Marvell */
+
+#ifndef _RTE_PMU_H_
+#define _RTE_PMU_H_
+
+/**
+ * @file
+
> -Original Message-
> From: Radu Nicolau
> Sent: Friday, February 17, 2023 7:53 PM
> To: dev@dpdk.org
> Cc: Nicolau, Radu ; Wu, Jingjing
> ; Xing, Beilei
> Subject: [PATCH] net/iavf: remove contrains for AEAD IV and AAD size
>
> Because of incosistencies between IPsec library and sam
> -Original Message-
> From: Zhang, Ke1X
> Sent: Monday, February 13, 2023 3:03 PM
> To: Zhang, Qi Z ; Yang, Qiming
> ; dev@dpdk.org
> Cc: Zhang, Ke1X
> Subject: [PATCH v2] net/ice: fix ice dcf contrl thread crash
>
> The control thread accesses the hardware resources after the resour
The code for setting algorithm for hash is not at all perf sensitive,
and doing it inline has a couple of problems. First, it means that if
multiple files include the header, then the initialization gets done
multiple times. But also, it makes it harder to fix usage of RTE_LOG().
Despite what the
Use dynamic type for hash and add subtypes for crc and gfni.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/hash/rte_cuckoo_hash.c | 5 +
lib/hash/rte_fbk_hash.c | 5 +
lib/hash/rte_hash_crc.c |
Having stubs in header file makes it harder to update
RTE_LOG(). Also modify to only print warning once.
Signed-off-by: Stephen Hemminger
---
lib/hash/meson.build | 8 +++-
lib/hash/rte_thash_gfni.c | 43 +++
lib/hash/rte_thash_gfni.h | 28 +-
Use a dynamically allocated logtype.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/pipeline/rte_pipeline.c | 3 +++
3 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/lib/eal/common/eal_common_log.c b/lib/e
Instead of using static type PIPELINE for logging in test application
use stderr instead. If not testing RTE_LOG() better to not use
it since log also goes to syslog.
Signed-off-by: Stephen Hemminger
---
app/test/test_table_acl.c | 50 --
app/test/test_table
Use logtype suffixes for various components of this library.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/table/rte_table_acl.c | 3 +++
lib/table/rte_table_array.c | 3 +++
lib/table/rte_table_hash_cuck
Looks like some code got copy/paste in to the IPSEC gateway
example from another place. Shouldn't be using RTE_LOGTYPE_PORT
here.
Fixes: ec17993a145a ("examples/ipsec-secgw: support security offload")
Cc: akhil.go...@nxp.com
Signed-off-by: Stephen Hemminger
---
examples/ipsec-secgw/sa.c | 6 +++-
Split up the single static RTE_LOGTYPE_PORT into separate
sub types for each component: port.ethdev, port.evendev, ...
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/port/rte_port_ethdev.c | 3 +++
lib/port/rte_port_ev
Also can remove unused RTE_LOGTYPE_METER
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 2 --
lib/eal/include/rte_log.h | 4 ++--
lib/sched/rte_pie.c | 1 +
lib/sched/rte_sched.c | 5 +
lib/sched/rte_sched_log.h | 4
5 files chan
Even though KNI will eventually disappear, fix the
logtype now.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/kni/rte_kni.c | 3 +++
3 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/lib/eal/comm
Split lpm and lpm6 into separate log types since they
are in different files and user may want to change log
levels for IPv4 vs IPv6.
For rib and fib libraries give them own types as well.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h
Convert from RTE_LOGTYPE_MEMPOOL to logtype_mempool.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/mempool/rte_mempool.c | 2 ++
lib/mempool/rte_mempool.h | 8
lib/mempool/version.map | 3 +++
The logtype for ring only used in library.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/ring/rte_ring.c | 3 +++
3 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/lib/eal/common/eal_common_log.c b
Use dynamic log type for power library.
Also replace use of RTE_LOGTYPE_USER1 with lib.power.guest.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/power/guest_channel.c | 3 ++-
lib/power/power_common.c| 2 ++
Convert to using a dynamic logtype for the application.
Signed-off-by: Stephen Hemminger
---
examples/l3fwd-power/main.c | 15 ---
1 file changed, 8 insertions(+), 7 deletions(-)
diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.c
index f57c54d2b57c..76b890b76c88
Don't use static logtype in sample application.
Signed-off-by: Stephen Hemminger
---
examples/distributor/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/examples/distributor/main.c b/examples/distributor/main.c
index 21304d661873..542f76cf9664 100644
--- a/examples/di
Introduce a new dynamic logtype for mbuf related messages.
Since this is used in multiple files put one macro in mbuf_log.h
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/mbuf/mbuf_log.h | 4
lib/mbuf/rte_m
Get rid of RTE_LOGTYPE_ACL and RTE_LOGTYPE_MALLOC.
For ACL library use a dynamic type.
The one message using RTE_LOGTYPE_MALLOC should have been
under the ACL logtype anyway.
The test code should not have been using fixed log type
so just change that to stderr.
Signed-off-by: Stephen Hemminger
-
Replace all uses of the global logtype with a dynamic log type.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
lib/efd/rte_efd.c | 4
3 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/lib/eal/com
The timer code does not use rte_log.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/lib/eal/common/eal_common_log.c b/lib/eal/common/eal_common_log.c
index 894701e8c
The message that used this was replaced in previous patch.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_log.c | 1 -
lib/eal/include/rte_log.h | 2 +-
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/lib/eal/common/eal_common_log.c b/lib/eal/common/eal_common
If a large packet is passed into GSO routines of unknown protocol
then library would log a message.
Better to tell the application instead of logging.
Fixes: 119583797b6a ("gso: support TCP/IPv4 GSO")
Reviewed-by: Jiayu Hu
Signed-off-by: Stephen Hemminger
---
lib/gso/rte_gso.c | 5 ++---
lib/gs
This patchset removes the main uses of static LOGTYPE's in DPDK
libraries. It starts with the easy one and goes on to the more complex ones.
There are several options on how to treat the old static types:
- leave them there
- mark the definitions as deprecated
- remove them
On 2/20/2023 10:09 PM, Ed Czeck wrote:
> Hi Ferruh,
> We had looked at providing this patch as an LTS backport, but there are
> some ties to FPGA firmware which complicate its application. As this is
> a uncommon net/ark feature we support users on an as-needed basis.
Got it, you don't want this
On 2/20/2023 10:04 PM, Ed Czeck wrote:
> Hi ferruh,
> We have limited support for secondary processes. This patch simply
> avoids corrupting the FPGA state if a secondary process attaches.
> Improved support for secondary processes is on our list, but we need a
> strong customer driver for this fe
On 2/20/2023 9:11 PM, Levend Sayar wrote:
> rx no_mbufs stats counter update is added for another error case.
>
> Fixes: 4f6b1dd8240c ("net/gve: support basic statistics")
> Cc: junfeng@intel.com
>
> Signed-off-by: Levend Sayar
> ---
> drivers/net/gve/gve_rx.c | 5 -
> 1 file changed, 4
On 2/20/2023 9:11 PM, Levend Sayar wrote:
> Google Virtual NIC rx/tx queue stats are added as extended stats.
>
> Signed-off-by: Levend Sayar
Thank you for the update, mainly looks good to me, there area a few
minor questions/comments below.
<...>
> +static const struct gve_xstats_name_offset t
Allows the creation of multiple ports from one ark device via
the use of ark pmd extension, though the splitting of queues
Add unique dev_private data for each port.
Signed-off-by: Ed Czeck
---
v3:
* Backport to LTS version is not recommended since firmware
changes are required and limited
* re
From: John Miller
disable fpga device configuration for secondary processes.
This is minimum support to avoid unneeded fpga resets
Signed-off-by: John Miller
Signed-off-by: Ed Czeck
---
v2:
* Use standard logging
v3:
* clarify limits on secondary process support
---
drivers/net/ark/ark_ethdev
Hi Ferruh,
We had looked at providing this patch as an LTS backport, but there are
some ties to FPGA firmware which complicate its application. As this is a
uncommon net/ark feature we support users on an as-needed basis.
Thanks.
Ed.
On Mon, Feb 20, 2023 at 9:54 AM Ferruh Yigit wrote:
> On 2/17
Hi ferruh,
We have limited support for secondary processes. This patch simply avoids
corrupting the FPGA state if a secondary process attaches.
Improved support for secondary processes is on our list, but we need a
strong customer driver for this feature.
An update patch is following soon.
Thanks
Google Virtual NIC rx/tx queue stats are added as extended stats.
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_ethdev.c | 143 +++
drivers/net/gve/gve_ethdev.h | 27 +--
drivers/net/gve/gve_rx.c | 10 +--
drivers/net/gve/gve_tx.c | 11 +--
4
rx no_mbufs stats counter update is added for another error case.
Fixes: 4f6b1dd8240c ("net/gve: support basic statistics")
Cc: junfeng@intel.com
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_rx.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/gve/
Google Virtual NIC rx/tx queue stats are added as extended stats.
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_ethdev.c | 143 +++
drivers/net/gve/gve_ethdev.h | 27 +--
drivers/net/gve/gve_rx.c | 10 +--
drivers/net/gve/gve_tx.c | 11 +--
4
Hi Cristian,
I agree the 64-bit version could enable better performance, and I will do
it in the next version.
For endianness, see my comments below (inline):
On Mon, Feb 20, 2023 at 12:19 PM Dumitrescu, Cristian <
cristian.dumitre...@intel.com> wrote:
> HI Bili,
>
> Comments inline below:
>
>
>-Original Message-
>From: Konstantin Ananyev
>Sent: Monday, February 20, 2023 6:21 PM
>To: Tomasz Duszynski ; Konstantin Ananyev
>;
>dev@dpdk.org
>Subject: RE: [EXT] Re: [PATCH v11 1/4] lib: add generic support for reading
>PMU events
>
>
>> >> >> >> diff --git a/lib/pmu/rte_pmu.h b/l
On 2/15/23 18:09, Hernan Vargas wrote:
Add a call to queue start and queue stop specifically for testing the
bbdev API.
Signed-off-by: Hernan Vargas
---
app/test-bbdev/test_bbdev_perf.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/app/test-bbdev/test_bbdev_per
HI Bili,
Comments inline below:
> diff --git a/lib/hash/rte_hash_xor.h b/lib/hash/rte_hash_xor.h
> new file mode 100644
> index 00..7004f83ec2
> --- /dev/null
> +++ b/lib/hash/rte_hash_xor.h
> @@ -0,0 +1,65 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2023 Intel Cor
Got it. Will fix. Thanks!
On Mon, Feb 20, 2023 at 12:10 PM Medvedkin, Vladimir <
vladimir.medved...@intel.com> wrote:
> Hi Bill,
>
> On 15/02/2023 11:06, Bili Dong wrote:
> > An XOR32 hash is needed in the Software Switch (SWX) Pipeline for its
> > use case in P4. We implement it in this patch so
On 2/15/23 18:09, Hernan Vargas wrote:
No functional impact.
Signed-off-by: Hernan Vargas
---
app/test-bbdev/test_bbdev_perf.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/app/test-bbdev/test_bbdev_perf.c b/app/test-bbdev/test_bbdev_perf.c
index 8c4b82efca..057fb5ac9e 100644
---
On 2/15/23 18:09, Hernan Vargas wrote:
Enhancement to enable early termination for validation tests if the
device supports it.
Signed-off-by: Hernan Vargas
---
app/test-bbdev/test_bbdev_perf.c | 30 +++---
1 file changed, 19 insertions(+), 11 deletions(-)
diff --g
Hi Bill,
On 15/02/2023 11:06, Bili Dong wrote:
An XOR32 hash is needed in the Software Switch (SWX) Pipeline for its
use case in P4. We implement it in this patch so it could be easily
registered in the pipeline later.
Signed-off-by: Bili Dong
---
+static inline uint32_t
+rte_hash_xor(const vo
In the flow template API, non-masked action must set its configuration
to NULL.
If flow action defined a configuration structure, then
testpmd *always* provided the action with configuration buffer.
Testpmd also implicitly assigned 0 to all action configuration
members that were not mentioned in a
I think the endianness conversions are necessary, otherwise, the hash
function return value will be host endianness dependent.
For example, what should rte_hash_xor32(/*data*/ &{0x00, 0x01, 0x02, 0x03},
/*data_len*/ 4, /*init_val*/ 0x0) return? With the current implementation,
it returns 0x0001020
There already is a logtype in this driver, use it!
Fixes: b35848bc01f6 ("crypto/ipsec_mb: add multi-process IPC request handler")
Cc: kai...@intel.com
Signed-off-by: Stephen Hemminger
---
drivers/crypto/ipsec_mb/ipsec_mb_ops.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a
Fix instances of USER1 logtype in fdset and crypto
sections.
Acked-by: Chenbo Xia
Reviewed-by: Maxime Coquelin
Signed-off-by: Stephen Hemminger
---
lib/vhost/fd_man.c | 4 ++--
lib/vhost/vhost_crypto.c | 4
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/lib/vhost/fd_
Libraries should not reuse RTE_LOGTYPE_USER1 for their
logging. Instead they should register their own type.
Fixes: 5cd3cac9ed22 ("latency: added new library for latency stats")
Cc: reshma.pat...@intel.com
Signed-off-by: Stephen Hemminger
---
lib/latencystats/rte_latencystats.c | 3 ++-
1 file c
Libraries should not be reusing RTE_LOGTYPE_USER1 instead
of doing proper logtype registration.
Fixes: b70b56032bff ("reorder: new library")
Cc: sergio.gonzalez.mon...@intel.com
Signed-off-by: Stephen Hemminger
---
lib/reorder/rte_reorder.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletion
DPDK libraries should not be reusing RTE_LOGTYPE_USER1 in
lieu of doing proper logtype registration.
Acked-by: Konstantin Ananyev
Fixes: 416707812c03 ("ip_frag: refactor reassembly code into a proper library")
Signed-off-by: Stephen Hemminger
---
lib/ip_frag/ip_frag_common.h | 5 -
lib/
The DPDK libraries and drivers should not be using the USER1 logtype.
v6 - fix checkpatch warnings
Stephen Hemminger (5):
ip_frag: use a dynamic logtype
reorder: use a dynamic logtype
latencystats: use dynamic logtype
vhost: use logtype instead of RTE_LOGTYPE_USER1
ipsec: fix usage of R
Reduce code duplication by introducing a helper that takes care of
transmitting, retrying if enabled and incrementing tx counter.
inc_tx_burst_stats() is then unneeded and removed.
Signed-off-by: David Marchand
---
Changes since v1:
- changed Tx helper so it matches rte_eth_tx_burst() semantic,
-
Reduce code duplication by introducing a helper that takes care of
receiving packets and incrementing rx counter.
inc_rx_burst_stats() is then unneeded and removed.
Signed-off-by: David Marchand
Reviewed-by: Ferruh Yigit
---
Changes since v1:
- updated shared_rxq fwd engine,
- removed inc_rx_bur
Reduce code duplication by introducing a helper that takes care of
initialising the fs object.
While at it, remove unneeded initialisation of fwd_engine empty fields.
Signed-off-by: David Marchand
Reviewed-by: Ferruh Yigit
---
Changes since v1:
- updated ieee1588,
---
app/test-pmd/5tswap.c
1 - 100 of 221 matches
Mail list logo
