On 9/14/2021 7:02 AM, Min Hu (Connor) wrote:
> From: Huisong Li
>
> This patch adds dev_close() step to release network adapter resources
> when kni free.
>
> Signed-off-by: Huisong Li
> Signed-off-by: Min Hu (Connor)
> ---
> examples/kni/main.c | 1 +
> 1 file changed, 1 insertion(+)
>
> di
01/09/2021 10:19, Dmitry Kozlyuk:
> Dmitry Kozlyuk (2):
> net/mlx5: report error on indirect CT action destroy
> net/mlx5: fix flow indirect action reference counting
Applied in next-net-mlx, thanks.
01/09/2021 10:07, Dmitry Kozlyuk:
> Shared RSS resources were released before checking that the shared RSS
> has no more references. If it had, the destruction was aborted, leaving
> the shared RSS in an invalid state where it could no longer be used.
> Move reference counter check before resource
On 9/15/2021 9:16 AM, Min Hu (Connor) wrote:
<...>
>>> 在 2021/9/15 13:08, dapengx...@intel.com 写道:
From: Dapeng Yu
If the bond device was created by vdev mode, the kvlist was not free
after the bond device was closed.
This patch fixes it.
Fixes: 144dc4739
From: Dmitry Kozlyuk
Hugepage allocation from the system takes time, resulting in slow
startup or sporadic delays later. Most of the time spent in kernel
is zero-filling memory for security reasons, which may be irrelevant
in a controlled environment. The bottleneck is memory access speed,
so for
From: Dmitry Kozlyuk
get_hugepage_dir() searched for a hugetlbfs mount with a given page size
using handcraft parsing of /proc/mounts and mixing traversal logic with
selecting the needed entry. Separate code to enumerate hugetlbfs mounts
to eal_hugepage_mount_walk() taking a callback that can ins
From: Viacheslav Ovsiienko
The primary DPDK process launch might take a long time if initially
allocated memory is large. From practice allocation of 1 TB of memory
over 1 GB hugepages on Linux takes tens of seconds. Fast restart
is highly desired for some applications and launch delay presents
a
From: Dmitry Kozlyuk
Memory allocator performance is crucial to applications that deal
with large amount of memory or allocate frequently. DPDK allocator
performance is affected by EAL options, API used and, at least,
allocation size. New autotest is intended to be run with different
EAL options.
On 9/15/2021 6:47 AM, Alvin Zhang wrote:
> This patch defines new RSS offload types for IPv4 and
> L4(TCP/UDP/SCTP) checksum, which are required when users want
> to distribute packets based on the IPv4 or L4 checksum field.
>
> For example "flow create 0 ingress pattern eth / ipv4 / end
> actions
On Mon, Sep 20, 2021 at 9:41 PM David Christensen
wrote:
> >>> We do not have a ppc platform, so there is no adaptation.
> >>> doc/guides/prog_guide/asan.rst has stated that we currently only
> >>> support Linux x86_64. You can adapt according to the following documents,
> >> the main work is to m
On Mon, Sep 20, 2021 at 8:19 PM Harman Kalra wrote:
>
> Issue has been observed where fields of indirect buffers are
> accessed after being set free by the diver. Also fixing freeing
> of direct buffers to correct aura.
>
> Fixes: 5cbe184802aa ("net/octeontx: support fast mbuf free")
> Cc: sta...@
On Mon, Sep 20, 2021 at 12:42 AM Venkat Duvvuru
wrote:
>
> This patch set adds support for new offload features/enhancments for
> Thor adapters like VF representor support, new flow matches/actions
> & dynamic SRAM manager support.
>
> Farah Smith (4):
> net/bnxt: updates to TF core index table
> -Original Message-
> From: Bruce Richardson
> Sent: Monday, September 20, 2021 9:27 PM
> To: Harman Kalra
> Cc: dev@dpdk.org; ciara.po...@intel.com; Anatoly Burakov
>
> Subject: [EXT] Re: [PATCH] eal: add telemetry callbacks for memory info
>
> External Email
>
> --
On Fri, Sep 17, 2021 at 4:55 PM Ashwin Sekhar T K wrote:
>
> Network Pool accelerator (NPA) is part of ROC (Rest Of Chip). So
> NPA structures should be aligned to ROC Cache line size and not
> CPU cache line size.
>
> Non alignment of NPA stack to ROC cache line will result in
> undefined runtime
On 18/09/2021 02:33, Zhang, AlvinX wrote:
-Original Message-
From: Kevin Traynor
Sent: Saturday, September 18, 2021 1:25 AM
To: Zhang, AlvinX ; Zhang, Qi Z
; Guo, Junfeng
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH] net/ice: add ability to reduce the Rx latency
On 14/09/2021 02:31,
> -Original Message-
> From: David Marchand
> Sent: Monday, September 20, 2021 15:25
> To: Yunjian Wang ; Wang, Haiyue
> ; Yigit, Ferruh
>
> Cc: dev ; Xing, Beilei ; Yang, Qiming
> ;
> Zhang, Qi Z ; dingxiaoxi...@huawei.com; Renata Saiakhova
>
> Subject: Re: [dpdk-dev] [PATCH v2 0/4] d
this patch adds unit test for rte_event_eth_rx_adapter_create_with_params
api and validate all possible input combinations.
Signed-off-by: Naga Harish K S V
---
app/test/test_event_eth_rx_adapter.c | 53 +---
1 file changed, 49 insertions(+), 4 deletions(-)
diff --git a/
To configure per queue event buffer size, applications sets
``rte_event_eth_rx_adapter_params::use_queue_event_buf`` flag
as true and is passed to ``rte_event_eth_rx_adapter_create_with_params``
api.
The per queue event buffer size is populated in
``rte_event_eth_rx_adapter_queue_conf::event_buf_
Currently Rx event buffer is static array with a default size
of 192(6*BATCH_SIZE).
``rte_event_eth_rx_adapter_create_with_params`` api is added which takes
``struct rte_event_eth_rx_adapter_params`` to configure event
buffer size in addition other params . The event buffer is
allocated from heap
this patch implement the per queue event buffer after
required validations.
Signed-off-by: Naga Harish K S V
---
lib/eventdev/rte_event_eth_rx_adapter.c | 188 ++--
1 file changed, 139 insertions(+), 49 deletions(-)
diff --git a/lib/eventdev/rte_event_eth_rx_adapter.c
b/lib
this patch adds unit tests for per rx queue event buffer
Signed-off-by: Naga Harish K S V
---
app/test/test_event_eth_rx_adapter.c | 86
1 file changed, 86 insertions(+)
diff --git a/app/test/test_event_eth_rx_adapter.c
b/app/test/test_event_eth_rx_adapter.c
index
On 9/17/2021 11:57 AM, Morten Brørup wrote:
>> From: Lance Richardson [mailto:lance.richard...@broadcom.com]
>> Sent: Thursday, 16 September 2021 18.11
>>
>> Minor corrections and improvements to documentation
>> for checksum APIs.
>>
>> Fixes: 6006818cfb26 ("net: new checksum functions")
>> Fixes:
this patch adds unit test for rte_event_eth_rx_adapter_create_with_params
api and validate all possible input combinations.
Signed-off-by: Naga Harish K S V
---
app/test/test_event_eth_rx_adapter.c | 53 +---
1 file changed, 49 insertions(+), 4 deletions(-)
diff --git a/
To configure per queue event buffer size, applications sets
``rte_event_eth_rx_adapter_params::use_queue_event_buf`` flag
as true and is passed to ``rte_event_eth_rx_adapter_create_with_params``
api.
The per queue event buffer size is populated in
``rte_event_eth_rx_adapter_queue_conf::event_buf_
Currently Rx event buffer is static array with a default size
of 192(6*BATCH_SIZE).
``rte_event_eth_rx_adapter_create_with_params`` api is added which takes
``struct rte_event_eth_rx_adapter_params`` to configure event
buffer size in addition other params . The event buffer is
allocated from heap
this patch implement the per queue event buffer after
required validations.
Signed-off-by: Naga Harish K S V
---
lib/eventdev/rte_event_eth_rx_adapter.c | 188 ++--
1 file changed, 139 insertions(+), 49 deletions(-)
diff --git a/lib/eventdev/rte_event_eth_rx_adapter.c
b/lib
this patch adds unit tests for per rx queue event buffer
Signed-off-by: Naga Harish K S V
---
app/test/test_event_eth_rx_adapter.c | 86
1 file changed, 86 insertions(+)
diff --git a/app/test/test_event_eth_rx_adapter.c
b/app/test/test_event_eth_rx_adapter.c
index
> -Original Message-
> From: Renata Saiakhova
> Sent: Wednesday, September 15, 2021 22:52
> To: Wang, Haiyue
> Cc: dev@dpdk.org; Renata Saiakhova
> Subject: [PATCH v2 1/1] net/e1000: igbvf VLAN offload implementation
>
> igbvf_vlan_offload_config and igbvf_vlan_offload_set primal
> impl
Hi Konstantin
On 9/21/2021 12:58 AM, Akhil Goyal wrote:
From: Gagandeep Singh
The current crypto raw data vectors is extended to support
rte_security usecases, where we need total data length to know
how much additional memory space is available in buffer other
than data length so that driver/
> Hi Konstantin
>
> On 9/21/2021 12:58 AM, Akhil Goyal wrote:
> >>> From: Gagandeep Singh
> >>>
> >>> The current crypto raw data vectors is extended to support
> >>> rte_security usecases, where we need total data length to know
> >>> how much additional memory space is available in buffer othe
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit
> Sent: Tuesday, 21 September 2021 11.28
>
> On 9/17/2021 11:57 AM, Morten Brørup wrote:
> >> From: Lance Richardson [mailto:lance.richard...@broadcom.com]
> >> Sent: Thursday, 16 September 2021 18.11
> >>
> >> Minor corrections a
On Tue, Sep 21, 2021 at 1:59 PM David Marchand
wrote:
>
> On Mon, Sep 20, 2021 at 9:41 PM David Christensen
> wrote:
> > >>> We do not have a ppc platform, so there is no adaptation.
> > >>> doc/guides/prog_guide/asan.rst has stated that we currently only
> > >>> support Linux x86_64. You can ada
This patch series enables telemetry for cnxk in the following:
- NPA LF
- Mempool driver
- NIX LF
- Ethdev driver
Depends-on: series-18612 ("net/cnxk: support for inline ipsec")
v8:
- removed lib telemetry and crypto patches for seperate series.
Gowrishankar Muthukrishnan (4):
common/cnxk
Add telemetry endpoints to npa.
Signed-off-by: Gowrishankar Muthukrishnan
---
drivers/common/cnxk/cnxk_telemetry.h | 26 +++
drivers/common/cnxk/cnxk_telemetry_npa.c | 224 +++
drivers/common/cnxk/meson.build | 6 +-
drivers/common/cnxk/roc_platform.h |
Adding telemetry end points to cnxk mempool driver.
Signed-off-by: Gowrishankar Muthukrishnan
---
drivers/mempool/cnxk/cnxk_mempool_telemetry.c | 100 ++
drivers/mempool/cnxk/meson.build | 1 +
2 files changed, 101 insertions(+)
create mode 100644 drivers/mempool/
Add telemetry endpoints to nix.
Signed-off-by: Gowrishankar Muthukrishnan
---
drivers/common/cnxk/cnxk_telemetry_nix.c | 849 +++
drivers/common/cnxk/meson.build | 3 +-
drivers/common/cnxk/roc_nix.c| 3 +
drivers/common/cnxk/roc_nix_priv.h |
Add telemetry endpoint to ethdev.
Signed-off-by: Gowrishankar Muthukrishnan
---
drivers/net/cnxk/cnxk_ethdev_telemetry.c | 129 +++
drivers/net/cnxk/meson.build | 1 +
2 files changed, 130 insertions(+)
create mode 100644 drivers/net/cnxk/cnxk_ethdev_telemetry.
As per an known HW issue RVUM interrupts may get dropped, If an RVUM
interrupt event occurs when PCCPF_XXX_MSIX_CAP_HDR[MSIXEN]=0 then no
interrupt is triggered, which is expected. But after MSIXEN is set to
1, subsequently if same interrupts event occurs again, still no
interrupt will be triggered
An issue exists on some HW revisions whereby if a CQ overflows
NIX may have undefined behavior, e.g. free incorrect buffers.
Implementing a workaround for this known HW issue.
Signed-off-by: Harman Kalra
---
drivers/common/cnxk/roc_nix_priv.h | 3 ++-
drivers/common/cnxk/roc_nix_queue.c | 18 +
Fix json output buffer size for a single largest value.
Fixes: 52af6ccb2b39 ("telemetry: add utility functions for creating JSON")
v2:
- split from series 18768 ("cnxk: enable telemetry endpoints").
Signed-off-by: Gowrishankar Muthukrishnan
---
lib/telemetry/telemetry_json.h | 7 +--
1 fi
Anatoly, Jeff, I would appreciate review.
Thanks,
Maciej
-Original Message-
From: David Marchand
Sent: Friday, September 3, 2021 2:01 PM
To: Szwed, Maciej ; Burakov, Anatoly
; Jeff Guo
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] |FAILURE| pw97277 [PATCH] [v4] eal: read data buffer on
RT
On Tue, Sep 21, 2021 at 4:23 PM Gowrishankar Muthukrishnan
wrote:
>
> Add telemetry endpoint to ethdev.
>
> Signed-off-by: Gowrishankar Muthukrishnan
> ---
> drivers/net/cnxk/cnxk_ethdev_telemetry.c | 129 +++
> drivers/net/cnxk/meson.build | 1 +
> 2 files chan
Add telemetry endpoints to cryptodev.
Signed-off-by: Gowrishankar Muthukrishnan
---
Depends-on: series-19052 ("cnxk: enable telemetry endpoints for mempool and
ethdev")
Depends-on: patch-19054 ("telemetry: fix json output buffer size")
v2:
- split from series 18768 ("cnxk: enable telemetry end
If the user requests to use an lcore above 128 using -l,
the eal will exit with "EAL: invalid core list syntax" and
very little else useful information.
THis patch adds some extra information suggesting to use --lcores
so that physical cores above RTE_MAX_LCORE (default 128) can be
used. This is a
If the user requests to use an lcore above 128 using -c,
the eal will exit with "EAL: invalid core list syntax" and
very little else useful information.
This patch adds some extra information suggesting to use --lcores
so that physical cores above RTE_MAX_LCORE (default 128) can be
used. This is a
On Tue, Sep 21, 2021 at 04:57:52PM +0530, Jerin Jacob wrote:
> On Tue, Sep 21, 2021 at 4:23 PM Gowrishankar Muthukrishnan
> wrote:
> >
> > Add telemetry endpoint to ethdev.
> >
> > Signed-off-by: Gowrishankar Muthukrishnan
> > ---
> > drivers/net/cnxk/cnxk_ethdev_telemetry.c | 129 ++
On Tue, Sep 21, 2021 at 12:50:14PM +0100, David Hunt wrote:
> If the user requests to use an lcore above 128 using -l,
> the eal will exit with "EAL: invalid core list syntax" and
> very little else useful information.
>
> THis patch adds some extra information suggesting to use --lcores
> so that
On Tue, Sep 21, 2021 at 12:50:15PM +0100, David Hunt wrote:
> If the user requests to use an lcore above 128 using -c,
> the eal will exit with "EAL: invalid core list syntax" and
> very little else useful information.
>
> This patch adds some extra information suggesting to use --lcores
> so that
On 21/9/2021 12:57 PM, Bruce Richardson wrote:
On Tue, Sep 21, 2021 at 12:50:14PM +0100, David Hunt wrote:
If the user requests to use an lcore above 128 using -l,
the eal will exit with "EAL: invalid core list syntax" and
very little else useful information.
THis patch adds some extra inform
On Tue, Sep 21, 2021 at 12:53:53PM +0100, Bruce Richardson wrote:
> On Tue, Sep 21, 2021 at 04:57:52PM +0530, Jerin Jacob wrote:
> > On Tue, Sep 21, 2021 at 4:23 PM Gowrishankar Muthukrishnan
> > wrote:
> > >
> > > Add telemetry endpoint to ethdev.
> > >
> > > Signed-off-by: Gowrishankar Muthukris
> -Original Message-
> From: Zhang, Qi Z
> Sent: Friday, September 17, 2021 10:43 PM
> To: Yang, Qiming
> Cc: Guo, Junfeng ; dev@dpdk.org; Zhang, Qi Z
>
> Subject: [PATCH v2 02/20] net/ice/base: init imem table for parser
>
> Parse DDP section ICE_SID_RXPARSER_IMEM into an array of s
Add the parser module that can parse on a raw packet then figure
out the low-level metadata to program the hardware packet process
pipeline for flow offloading(Switch/FDIR/RSS). This is the pre-step
to enable a protocol-agnostic flow offloading solution for ice devices
that leverage Intel DDP techn
Add new parser module which can parse a packet in binary
and generate information like ptype, protocol/offset pairs
and flags which can be used to feed the FXP profile creation
directly.
The patch added skeleton of the parser instance create and
destroy APIs:
ice_parser_create
ice_parser_destroy
Parse DDP section ICE_SID_RXPARSER_IMEM into an array of
struct ice_imem_item.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_imem.c| 244 +
drivers/net/ice/base/ice_imem.h| 109 +++
drivers/net/ice/base/ice_parser.c
Parse DDP section ICE_SID_RXPARSER_METADATA_INIT into an array of
struct ice_metainit_item.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_metainit.c| 143 +
drivers/net/ice/base/ice_metainit.h| 46
drivers/net/ice/base/ice_p
On 21/9/2021 1:04 PM, David Hunt wrote:
On 21/9/2021 12:57 PM, Bruce Richardson wrote:
On Tue, Sep 21, 2021 at 12:50:14PM +0100, David Hunt wrote:
If the user requests to use an lcore above 128 using -l,
the eal will exit with "EAL: invalid core list syntax" and
very little else useful infor
Parse DDP section ICE_SID_RXPARSER_CAM or ICE_SID_RXPARSER_PG_SPILL
into an array of struct ice_pg_cam_item.
Parse DDP section ICE_SID_RXPARSER_NOMATCH_CAM or
ICE_SID_RXPARSER_NOMATCH_SPILL into an array of struct ice_pg_nm_cam_item.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net
Parse DDP section ICE_SID_RXPARSER_CAM into an array of
ice_bst_tcam_item.
Parse DDP section ICE_SID_LBL_RXPARSER_TMEM into an array of
ice_lbl_item.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_bst_tcam.c| 241 +
drivers/net/ice/base/ic
Parse DDP section ICE_SID_RXPARSER_MARKER_PTYPE into an array of
ice_ptype_mk_tcam_item.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_parser.c | 11 ++
drivers/net/ice/base/ice_parser.h | 3 ++
drivers/net/ice/base/ice_ptype_mk.c | 54 +
Parse DDP section ICE_SID_RXPARSER_MARKER_GRP into an array of
ice_mk_grp_item.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_mk_grp.c | 55 +++
drivers/net/ice/base/ice_mk_grp.h | 15 +
drivers/net/ice/base/ice_parser.c | 11
Parse DDP section ICE_SID_RXPARSER_PROTO_GRP into an array of
ice_proto_grp_item.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_parser.c| 11 +++
drivers/net/ice/base/ice_parser.h| 3 +
drivers/net/ice/base/ice_proto_grp.c | 108 ++
Parse DDP section ICE_SID_RXPARSER_FLAG_REDIR into an array of
ice_flag_rd_item.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_flex_type.h | 2 ++
drivers/net/ice/base/ice_flg_rd.c| 53
drivers/net/ice/base/ice_flg_rd.h| 16
Parse below DDP section into struct ice_xlt_kb:
ICE_SID_XLT_KEY_BUILDER_SW
ICE_SID_XLT_KEY_BUILDER_FD
ICE_SID_XLT_KEY_BUILDER_RSS
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_flex_pipe.c | 2 +-
drivers/net/ice/base/ice_flex_pipe.h | 3 +
drivers/net/ice/base/i
Add parser runtime data struct ice_parser_rt.
Add below APIs for parser runtime preparation:
ice_parser_rt_reset
ice_parser_rt_pkt_buf_set
Add below API skeleton for parser runtime execution:
ice_parser_rt_execute
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_pars
Add internal helper function ice_bst_tcam_match to perform ternary
match on boost TCAM.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_bst_tcam.c | 22 +++
drivers/net/ice/base/ice_bst_tcam.h | 3 ++
drivers/net/ice/base/ice_parser.h | 1 +
drivers/ne
Add below two internal helper functions for parse graph key matching
in cam table:
ice_pg_cam_match
ice_pg_nm_cam_match
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_pg_cam.c | 76 +++
drivers/net/ice/base/ice_pg_cam.h | 6 +++
2 files
Add internal helper function ice_ptype_mk_tcam_match for ptype markers
matching in tcam table.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_ptype_mk.c | 22 ++
drivers/net/ice/base/ice_ptype_mk.h | 3 +++
2 files changed, 25 insertions(+)
diff
Add internal helper function ice_flg_redirect to redirect parser flags
to packet flags.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_flg_rd.c | 23 +++
drivers/net/ice/base/ice_flg_rd.h | 1 +
2 files changed, 24 insertions(+)
diff --git a/dri
Add internal helper function ice_xlt_kb_flg_get to aggregate 64 bit
packet flag into 16 bit key builder flags.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_xlt_kb.c | 27 +++
drivers/net/ice/base/ice_xlt_kb.h | 1 +
2 files changed, 28 inse
Implement function ice_parser_rt_execute which perform the main
loop of the parser.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_parser_rt.c | 782 ++-
drivers/net/ice/base/ice_parser_rt.h | 27 +-
2 files changed, 803 insertions(+), 6 dele
Add API ice_parser_dvm_set to support turn on/off parser's
double vlan mode.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_bst_tcam.c | 28
drivers/net/ice/base/ice_bst_tcam.h | 4
drivers/net/ice/base/ice_parser.c | 27 +
UDP tunnel can be added/deleted for vxlan, geneve, ecpri through
below APIs:
ice_parser_vxlan_tunnel_set
ice_parser_geneve_tunnel_set
ice_parser_ecpri_tunnel_set
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_parser.c | 75 +++
drivers/net
Add API ice_parser_profile_init to init a parser profile base on
a parser result and a mask buffer. The ice_parser_profile can feed to
low level FXP engine to create HW profile / field vector directly.
Signed-off-by: Qi Zhang
Acked-by: Junfeng Guo
---
drivers/net/ice/base/ice_parser.c | 112 +++
On Tue, Sep 21, 2021 at 02:16:35PM +0100, David Hunt wrote:
>On 21/9/2021 1:04 PM, David Hunt wrote:
>
> On 21/9/2021 12:57 PM, Bruce Richardson wrote:
>
> On Tue, Sep 21, 2021 at 12:50:14PM +0100, David Hunt wrote:
>
> If the user requests to use an lcore above 128 using -l,
This patch fixes a possible buffer overrun problem in crypto perf test.
Previously when user configured aad size is over 12 bytes the copy of template
aad will cause a buffer overrun.
The problem is fixed by only copy up to 12 bytes of aad template.
Fixes: 761a321acf91 ("event/cnxk: support vecto
> -Original Message-
> From: Zhang, Qi Z
> Sent: Tuesday, September 21, 2021 9:20 PM
> To: Yang, Qiming
> Cc: Guo, Junfeng ; dev@dpdk.org; Zhang, Qi Z
>
> Subject: [PATCH v3 00/20] ice/base: add parser module
>
> Add the parser module that can parse on a raw packet then figure out th
>
> Not all net PMD's/HW can parse packet and identify L2 header and
> L3 header locations on Tx. This is inline with other Tx offloads
> requirements such as L3 checksum, L4 checksum offload, etc,
> where mbuf.l2_len, mbuf.l3_len etc, needs to be set for HW to be
> able to generate checksum. Sinc
On Tue, Sep 21, 2021 at 1:50 PM David Hunt wrote:
> static int
> eal_parse_coremask(const char *coremask, int *cores)
> {
> @@ -839,54 +880,89 @@ eal_parse_service_corelist(const char *corelist)
> static int
> eal_parse_corelist(const char *corelist, int *cores)
> {
> - unsigned count
Hi, Maxime
Very nice catch, thank you for the patch.
Minor typo in the commit message "ithat" -> "that".
Besides this:
Acked-by: Viacheslav Ovsiienko
> -Original Message-
> From: Maxime Coquelin
> Sent: Friday, September 10, 2021 12:18
> To: dev@dpdk.org; chenbo@intel.com; amore..
Hi, Winxu
Thank you for the patch.
> If the mlx5_flow_tunnel_validate validate the flow tunnel rule failed, the
Typo? "validate" -> "validating" ?
> err_msg is empty in the rte_flow_error.
Sorry, what do you mean "empty"? It is NULL, not pointing to empty string "".
Patch looks OK for me, could
Hi Jerin,
Please see the replies inline.
> -Original Message-
> From: Jerin Jacob
> Sent: Monday, September 20, 2021 11:50 AM
> To: Naga Harish K, S V
> Cc: Jerin Jacob ; Jayatheerthan, Jay
> ; dpdk-dev ; Kundapura,
> Ganapati
> Subject: Re: [dpdk-dev] [PATCH v1 1/5] eventdev: rx_adap
Hi Jerin,
> -Original Message-
> From: Jerin Jacob
> Sent: Monday, September 20, 2021 11:53 AM
> To: Naga Harish K, S V ; Pavan Nikhilesh
> ; mattias.ronnblom
>
> Cc: Jerin Jacob ; Jayatheerthan, Jay
> ; dpdk-dev
> Subject: Re: [dpdk-dev] [PATCH v1 3/5] eventdev:rx_adapter:add per queue
On 9/17/2021 1:31 PM, Elad Nachman wrote:
> Fix lack of multiple KNI requests handling support by introducing a
> rotating ring mechanism for the sync buffer.
>
Thanks Elad for the patch.
I have missed that 'rte_kni_handle_request()' can be called by multiple cores,
at least kni sample app does
Hi Jerin,
> > > > > From: Kevin Laatz <[2]kevin.la...@intel.com>
> > > > > Add a burst capacity check API to the dmadev library. This API is
> > > > > useful to
> > > > > applications which need to how many descriptors can be enqueued
> in
> > > > > the
> > > > > cur
11/09/2021 01:07, Dmitry Kozlyuk:
> Add the current status and the actual roadmap, remove completed entries.
> More detailed plan in the mailing list:
> https://mails.dpdk.org/archives/dev/2021-August/217463.html
>
> Signed-off-by: Dmitry Kozlyuk
> ---
> v2: improve wording, fix typos, remove imp
>This patch fixes a possible buffer overrun problem in crypto perf test.
>Previously when user configured aad size is over 12 bytes the copy of
>template aad will cause a buffer overrun.
>The problem is fixed by only copy up to 12 bytes of aad template.
>
>Fixes: 761a321acf91 ("event/cnxk: support
On Tue, Sep 21, 2021 at 7:46 PM Pai G, Sunil wrote:
>
> Hi Jerin,
Hi Sunil,
>
> > > > > > From: Kevin Laatz <[2]kevin.la...@intel.com>
> > > > > > Add a burst capacity check API to the dmadev library. This API
> > > > > > is
> > > > > > useful to
> > > > > > applications wh
On 21/9/2021 2:51 PM, David Marchand wrote:
On Tue, Sep 21, 2021 at 1:50 PM David Hunt wrote:
static int
eal_parse_coremask(const char *coremask, int *cores)
{
@@ -839,54 +880,89 @@ eal_parse_service_corelist(const char *corelist)
static int
eal_parse_corelist(const char *corelist,
On 9/15/2021 11:40 AM, Andrew Rybchenko wrote:
> From: Ivan Ilchenko
>
> Add missing initialisation of the accumulative SW xstat to
> zero since it is sum of per-queue xstats.
>
> Fixes: fdd7719eb3c ("net/sfc: add xstats for Rx/Tx doorbells")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Ivan Ilchen
Hi Jerin,
> > > > The fact is that it's very hard for apps to calculate the
> > > > available space of a
> > > DMA ring.
> > >
> > > Yes, I agree.
> > >
> > > My question is more why to calculate the space per burst and
> > > introduce yet another fastpath API.
> > > For example, the applicatio
Hi Anoob,
Few minor comments, Please see inline.
Apart from that,
Acked-by: Akhil Goyal
>
Update title as
Test/crypto: add lookaside IPsec cases.
> +static int
> +security_proto_supported(enum rte_security_session_action_type action,
> + enum rte_security_session_protocol proto);
> +
> +sta
The previous commit 18effad9cfa7 ("stack: reload head when pop fails")
only changed C11 implementation, not generic implementation.
List head must be loaded right before continue (when failed to find the
new head). Without this, one thread might keep trying and failing to pop
items without ever lo
> Subject: [PATCH v4 2/5] test/crypto: add combined mode tests
Title: test/crypto: add combined mode IPSec cases
>
In description explain the meaning of combined mode.
Also mention that the encap and decap are done one after the other,
Hence known test vectors are not required.
Apart from that
On Fri, Sep 17, 2021 at 03:42:18PM +, Conor Walsh wrote:
When a suitable device is found during the PCI probe, create a dmadev
instance for each channel. Internal structures and HW definitions required
for device creation are also included.
Signed-off-by: Conor Walsh
Reviewed-by: Kevin L
On Fri, Sep 17, 2021 at 03:42:27PM +, Conor Walsh wrote:
Move Intel IOAT devices on Ice Lake systems from Misc to DMA devices.
Signed-off-by: Conor Walsh
Reviewed-by: Kevin Laatz
---
usertools/dpdk-devbind.py | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/us
> Subject: [PATCH v4 4/5] test/crypto: add IV gen tests
>
> From: Tejasree Kondoj
>
> Add test cases to verify IV generated by PMD.
Title and description of the patch do not specify that
This is for lookaside IPSec.
Title: test/crypto: add IV gen cases for IPsec
Description:
Added cases to ver
> Subject: [PATCH v4 3/5] test/crypto: add lookaside IPsec ICV corrupt test case
>
> From: Tejasree Kondoj
>
> Adding lookaside IPsec ICV corrupt test case.
Please elaborate the test case a bit more.
There is no documentation for the test cases, we can
Atleast add appropriate description.
Apart
On Fri, Sep 17, 2021 at 03:42:17PM +, Conor Walsh wrote:
Add the basic device probe/remove skeleton code and initial documentation
for new IOAT DMA driver. Maintainers update is also included in this
patch.
Signed-off-by: Conor Walsh
Reviewed-by: Kevin Laatz
---
MAINTAINERS
On Tue, Sep 21, 2021 at 2:52 PM Naga Harish K S V
wrote:
>
> To configure per queue event buffer size, applications sets
> ``rte_event_eth_rx_adapter_params::use_queue_event_buf`` flag
> as true and is passed to ``rte_event_eth_rx_adapter_create_with_params``
> api.
>
> The per queue event buffer
On 9/18/2021 1:35 PM, Xueming Li wrote:
> Some drivers don't need Rx and Tx queue release callback, make them
> optional. Clean up empty queue release callbacks for some drivers.
>
> Signed-off-by: Xueming Li
> Reviewed-by: Andrew Rybchenko
Acked-by: Ferruh Yigit
1 - 100 of 166 matches
Mail list logo
