On Tue, Aug 22, 2023 at 10:46 PM Ashwin Sekhar T K wrote:
>
> From: Sibaranjan Pattnayak
>
> Support for CNF10KA SoC by adding its PCI device ID.
>
> Signed-off-by: Sibaranjan Pattnayak
> Signed-off-by: Ashwin Sekhar T K
Updated .mailmap
diff --git a/.mailmap b/.mailmap
index 864d33ee46..59ea
For konstantin
> -Original Message-
> From: Feifei Wang
> Sent: Tuesday, August 22, 2023 3:27 PM
> To: Yuying Zhang ; Beilei Xing
>
> Cc: dev@dpdk.org; nd ; Feifei Wang
> ; Honnappa Nagarahalli
> ; Ruifeng Wang
>
> Subject: [PATCH v11 2/4] net/i40e: implement mbufs recycle mode
>
> Def
> -Original Message-
> From: Kevin Traynor
> Sent: Thursday, August 3, 2023 12:31 AM
> To: sta...@dpdk.org
> Cc: dev@dpdk.org; Abhishek Marathe ; Ali
> Alnubani ; Walker, Benjamin
> ; David Christensen ;
> Hemant Agrawal ; Stokes, Ian
> ; Jerin Jacob ; Mcnamara, John
> ; Ju-Hyoung Lee ; Ke
On Tue, Aug 22, 2023 at 6:40 PM wrote:
>
> From: Pavan Nikhilesh
>
> Add support to get the remaining ticks to expire for a
> given event timer.
>
> Signed-off-by: Pavan Nikhilesh
> ---
>
>
> +* **Updated Marvell cnxk eventdev driver.**
> +
> + Added support to get the remaining
On Tue, Aug 22, 2023 at 10:40 PM Ashwin Sekhar T K wrote:
>
> Currently mempool_cnxk driver uses asynchronous allocation
> for all pools. This asynchronous allocation can result in
> local caching of additional 512 objects on a single core
> even when cache is disabled. This will eventually lead t
For Konstantin
> -Original Message-
> From: Feifei Wang
> Sent: Tuesday, August 22, 2023 3:27 PM
> To: tho...@monjalon.net; Ferruh Yigit ; Andrew
> Rybchenko
> Cc: dev@dpdk.org; nd ; Feifei Wang
> ; Honnappa Nagarahalli
> ; Ruifeng Wang
> ; Morten Brørup
> Subject: [PATCH v11 1/4] ethde
> -Original Message-
> From: Stephen Hemminger
> Sent: Tuesday, August 22, 2023 10:03 PM
> To: Feifei Wang
> Cc: tho...@monjalon.net; Ferruh Yigit ; Andrew
> Rybchenko ; dev@dpdk.org; nd
> ; Honnappa Nagarahalli ;
> Ruifeng Wang ; Morten Brørup
>
> Subject: Re: [PATCH v11 1/4] ethdev:
> -Original Message-
> From: Qiao, Wenjing
> Sent: Friday, August 11, 2023 6:00 PM
> To: Zhang, Yuying ; Xing, Beilei
>
> Cc: dev@dpdk.org; Liu, Mingxia ; Qiao, Wenjing
>
> Subject: [PATCH v2 1/4] net/cpfl: parse flow parser file in devargs
>
> Add devargs "flow_parser" for rte_flow
> -Original Message-
> From: Stephen Hemminger
> Sent: Tuesday, August 22, 2023 9:59 PM
> To: Feifei Wang
> Cc: dev@dpdk.org; nd
> Subject: Re: [PATCH v11 0/4] Recycle mbufs from Tx queue into Rx queue
>
> On Tue, 22 Aug 2023 15:27:06 +0800
> Feifei Wang wrote:
>
> > Currently, t
On Wed, Aug 23, 2023 at 10:19:39AM +0100, Ferruh Yigit wrote:
> On 8/22/2023 11:30 PM, Konstantin Ananyev wrote:
> > 18/08/2023 14:48, Bruce Richardson пишет:
> >> On Fri, Aug 18, 2023 at 02:25:14PM +0100, Ferruh Yigit wrote:
> >>> On 8/17/2023 3:18 PM, Konstantin Ananyev wrote:
>
> >> Cau
Get rid of all the complicated code which copies data on the stack:
- allocate a new segment from the same mempool than the original mbuf,
- copy headers data in this segment,
- chain the new segment in place of headers of the original mbuf,
- use existing helpers for computing IP and TCP/UDP check
Checking that one of RTE_MBUF_F_TX_IPV4 or RTE_MBUF_F_TX_IP_CKSUM is
present is not compliant with the offloading API which specifies that IP
checksum requires RTE_MBUF_F_TX_IP_CKSUM.
On the other hand, RTE_MBUF_F_TX_IP_CKSUM is invalid for IPv6 packets,
so we can simply check for RTE_MBUF_F_TX_IP_
The L4 checksum offloading API does not require l4_len to be set.
Make the driver discover the L4 headers size by itself.
Fixes: 6546e76056e3 ("net/tap: calculate checksums of multi segs packets")
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
Tested-by: Ales Musil
---
.mailmap
On Wed, Aug 23, 2023 at 4:45 PM Amit Prakash Shukla
wrote:
>
> roc_dpi was using vfid as part of name for memzone allocation.
> This led to memzone allocation failure in case of multiple
> physical functions. vfid is not unique by itself since multiple
> physical functions can have the same virtua
> It seems the issue arose due to changes in the DPDK read-write lock
> implementation. Following these changes, the RW-lock no longer supports
> recursion, implying that a single thread shouldn't obtain a read lock if it
> already possesses one. The problem arises during initialization: the
> rt
On Wed, Aug 23, 2023 at 11:47 AM Gujjar, Abhinandan S
wrote:
>
> It is not about supporting other PMDs.
> This is about keeping the test case simple with minimal conf overhead so that
> it is easy for testing and maintaining.
> From the code, I see that few PMDs supports NULL too without much ch
https://bugs.dpdk.org/show_bug.cgi?id=1277
Bug ID: 1277
Summary: memory_hotplug_lock deadlock during initialization
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: no
> Validate copied memory to ensure DMA copy did not fail.
>
> Fixes: 623dc9364dc ("app/dma-perf: introduce DMA performance test")
>
> Signed-off-by: Gowrishankar Muthukrishnan
> ---
> app/test-dma-perf/benchmark.c | 23 +--
> app/test-dma-perf/main.c | 16 +++---
From: Vamsi Attunuru
last index value might lost the order when dma stats are reset
in between copy operations. Patch adds a variable to track the
completed count, that can be used to compute the last index, also
patch adds misc other changes.
Signed-off-by: Vamsi Attunuru
---
v2:
- Fix for bug
From: Vamsi Attunuru
Adds a check to avoid tail wrap when completion desc ring
is full. Also patch increase max desc size to 2048.
Fixes: b56f1e2dad38 ("dma/cnxk: add channel operations")
Fixes: 3340c3e22783 ("dma/cnxk: add scatter-gather copy")
Fixes: 681851b347ad ("dma/cnxk: support CN10K DMA
Code changes to realign dpi private structure based on vchan.
Changeset also resets DMA dev stats while starting dma device.
Signed-off-by: Amit Prakash Shukla
---
v2:
- Fix for bugs observed in v1.
- Squashed few commits.
v3:
- Resolved review suggestions.
- Code improvement.
v4:
- Resolved ch
To support multiple directions in same queue make use of multiple vchan
per queue. Each vchan can be configured in some direction and used.
Signed-off-by: Amit Prakash Shukla
Signed-off-by: Radha Mohan Chintakuntla
---
v2:
- Fix for bugs observed in v1.
- Squashed few commits.
v3:
- Resolved re
Valid function pointers are set for DMA device operations
i.e. cn10k_dmadev_ops are used for all cn10k devices.
Signed-off-by: Radha Mohan Chintakuntla
Signed-off-by: Amit Prakash Shukla
---
v2:
- Fix for bugs observed in v1.
- Squashed few commits.
v3:
- Resolved review suggestions.
- Code imp
Use pfid and vfid of src_port for incoming DMA transfers and dst_port
for outgoing DMA transfers.
Signed-off-by: Radha Mohan Chintakuntla
Signed-off-by: Amit Prakash Shukla
---
v2:
- Fix for bugs observed in v1.
- Squashed few commits.
v3:
- Resolved review suggestions.
- Code improvement.
v4:
On chunk buffer alloc failure, ENOMEM is returned. As per DMA spec
ENOSPC shall be returned on failure to allocate memory. This
changeset fixes the same.
Fixes: b56f1e2dad38 ("dma/cnxk: add channel operations")
Cc: sta...@dpdk.org
Signed-off-by: Amit Prakash Shukla
---
v2:
- Fix for bugs observe
Completion buffer was a static array per dma device. This would
consume memory for max number of descriptor supported by device
which might be more than configured by application. The patchset
allocates the memory for completion buffer based on the number
of descriptor configured by application.
F
Multiple call to configure, setup queues without stopping the device
would leak the ring descriptor and hardware queue memory. This patch
adds flags support to prevent configuring without stopping the
device.
Fixes: b56f1e2dad38 ("dma/cnxk: add channel operations")
Cc: sta...@dpdk.org
Signed-off-
When a device is not set to a ready state, on exiting the application
proper cleanup is not done. This causes the application to fail on
trying to run next time.
Setting the device to ready state on successful probe fixes the issue.
Fixes: 53f6d7328bf4 ("dma/cnxk: create and initialize device on
Adds support for the burst capacity. Call to the function return
number of vacant space in descriptor ring for the current burst.
Signed-off-by: Amit Prakash Shukla
---
v2:
- Fix for bugs observed in v1.
- Squashed few commits.
v3:
- Resolved review suggestions.
- Code improvement.
v4:
- Resolv
roc_dpi was using vfid as part of name for memzone allocation.
This led to memzone allocation failure in case of multiple
physical functions. vfid is not unique by itself since multiple
physical functions can have the same virtual function indices.
So use complete DBDF as part of memzone name to ma
> Currently, the TEST_CASE_NAMED_ST macro is implemented in a way that is
> identical to TEST_CASE_NAMED_WITH_DATA. In particular, it expects the
> test function to have a data argument, which is why a bunch of tests have an
> unused data arguments.
>
> Fix this by adjusting the macro to not expec
Acked-by: Hemant Agrawal
> Currently, the TEST_CASE_NAMED_ST macro is implemented in a way that
> is identical to TEST_CASE_NAMED_WITH_DATA. In particular, it expects
> the test function to have a data argument, which is why a bunch of tests
> have an unused data arguments.
>
> Fix this by adjusting the macro to not expec
On 23/08/2023 10:55, Mcnamara, John wrote:
-Original Message-
From: Kevin Traynor
Sent: Monday, August 21, 2023 1:38 PM
To: Mcnamara, John ; sta...@dpdk.org; Peng, Yuan
; Chen, Zhaoyan ; Xu, HailinX
Cc: dev@dpdk.org; Walker, Benjamin ; Stokes,
Ian ; Luca Boccassi ; Thomas
Monjalon
S
Currently, the TEST_CASE_NAMED_ST macro is implemented in a way that
is identical to TEST_CASE_NAMED_WITH_DATA. In particular, it expects
the test function to have a data argument, which is why a bunch of tests
have an unused data arguments.
Fix this by adjusting the macro to not expect a test fun
> -Original Message-
> From: Kevin Traynor
> Sent: Monday, August 21, 2023 1:38 PM
> To: Mcnamara, John ; sta...@dpdk.org; Peng, Yuan
> ; Chen, Zhaoyan ; Xu, HailinX
>
> Cc: dev@dpdk.org; Walker, Benjamin ; Stokes,
> Ian ; Luca Boccassi ; Thomas
> Monjalon
> Subject: Re: 21.11.5 patche
On 8/22/2023 11:30 PM, Konstantin Ananyev wrote:
> 18/08/2023 14:48, Bruce Richardson пишет:
>> On Fri, Aug 18, 2023 at 02:25:14PM +0100, Ferruh Yigit wrote:
>>> On 8/17/2023 3:18 PM, Konstantin Ananyev wrote:
>> Caution: This message originated from an External Source. Use
>> proper c
> -Original Message-
> From: Aakash Sasidharan
> Sent: Wednesday, August 23, 2023 1:19 PM
> To: Gujjar, Abhinandan S
> Cc: Jerin Jacob Kollanukkaran ; Shijith Thotton
> ; Anoob Joseph ;
> dev@dpdk.org
> Subject: RE: [PATCH 2/3] test/event_crypto_adapter: skip unsupported test
>
> Hi A
> -Original Message-
> From: David Marchand
> Sent: Wednesday, August 23, 2023 2:29 PM
> To: dev@dpdk.org
> Cc: echau...@redhat.com; m...@redhat.com; Zhang, Qi Z
> ; sta...@dpdk.org; Wu, Jingjing
> ; Xing, Beilei ; Sinha, Abhijit
> ; Doherty, Declan ;
> Nicolau, Radu
> Subject: [PATCH
Hi Abhinandan,
Please see inline.
> -Original Message-
> From: Aakash Sasidharan
> Sent: Wednesday, August 23, 2023 1:18 PM
> To: 'Gujjar, Abhinandan S'
> Cc: Jerin Jacob Kollanukkaran ; Shijith Thotton
> ; Anoob Joseph ;
> dev@dpdk.org
> Subject: RE: [PATCH 2/3] test/event_crypto_adapt
On Mon, Jul 31, 2023 at 4:48 PM David Marchand
wrote:
>
> This series is one implementation to try and please users who want to
> select more easily which parts of DPDK are built.
>
> It introduces a change in behavior for enabling deprecated libraries:
> this series is aimed at the next release b
Hi Abhinandan,
Please see inline.
> -Original Message-
> From: Gujjar, Abhinandan S
> Sent: Wednesday, August 23, 2023 10:04 AM
> To: Aakash Sasidharan
> Cc: Jerin Jacob Kollanukkaran ; Shijith Thotton
> ; Anoob Joseph ;
> dev@dpdk.org
> Subject: [EXT] RE: [PATCH 2/3] test/event_crypto_
Cipher length alignment checked for 3DES-CBC and AES-CBC to avoid slice
hang error in QAT CPM1.8
Signed-off-by: Sivaramakrishnan VenkatX
---
drivers/crypto/qat/dev/qat_crypto_pmd_gens.h | 21
1 file changed, 21 insertions(+)
diff --git a/drivers/crypto/qat/dev/qat_crypto_pm
Hello Bruce,
On Wed, Aug 16, 2023 at 5:35 PM Bruce Richardson
wrote:
>
> This patchset contains changes to DPDK to make the build more configurable
> overall. The changes can be grouped into two areas:
>
> * Firstly, there are changes to make the build of the unit tests more
> flexible and main
This patch enables the ipsec protocol offload to copy DF and diffserv
Signed-off-by: Hemant Agrawal
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 31 +
drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h | 2 ++
2 files changed, 21 insertions(+), 12 deletions(-)
diff --git a/dr
LX216x suppots ESN.
Also enable to correctly print the SEC era.
Signed-off-by: Hemant Agrawal
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 2 +-
drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/crypto/dpaa2_sec/dpaa2_s
This patch adds supports for UDP encapsulation in NAT-T for
IPSEC security protocol offload case.
Signed-off-by: Hemant Agrawal
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 101 ++--
drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h | 3 +
2 files changed, 75 insertions(+), 29 d
LX216x can support upto 1024 anti replay window size.
Signed-off-by: Hemant Agrawal
---
drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h
b/drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h
index 0
This patch adds support for enabling capability to copy
dscp and df bits from inner to outer header and vice-versa.
Signed-off-by: Hemant Agrawal
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 18 ++
drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h | 10 --
2 files changed,
From: Apeksha Gupta
The macro DPAA2_SET_FLE_OFFSET(fle, offset) only works for masking the
offset upto with 12 bits. When the offset value is more that 12 bits,
this macro may over writing the FMT/SL/F bits which are beyond the
offset bits.
The FLE_ADDR is modified to FLE_ADDR + OFFSET, and the F
From: Gagandeep Singh
The descriptors refers to auth data as well, so initializing it
properly for the non-auth cases.
Signed-off-by: Gagandeep Singh
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 16
drivers/crypto/dpaa_sec/dpaa_sec.c | 5 +
2 files changed,
Adding NULL cipher and auth in capabilities.
Signed-off-by: Hemant Agrawal
---
drivers/crypto/dpaa_sec/dpaa_sec.h | 42 +-
1 file changed, 41 insertions(+), 1 deletion(-)
diff --git a/drivers/crypto/dpaa_sec/dpaa_sec.h
b/drivers/crypto/dpaa_sec/dpaa_sec.h
index 412a
IPSEC proto offload support NULL in combo cases, thus adding
NULL cipher and auth in security capabilities. Non-supported
cases are already protected in the code.
Signed-off-by: Hemant Agrawal
---
drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h | 43 +--
1 file changed, 41 insertio
From: Vanshika Shukla
This patch initializes the pdcp alg to null.
Signed-off-by: Vanshika Shukla
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c
b/drivers/crypto/dpaa2_sec/dpaa2_
From: Gagandeep Singh
In case of protocol based offload, it is better to wait before the
share descriptor complete the execution. Simultaneous sharing may
cause issues.
Signed-off-by: Gagandeep Singh
---
drivers/common/dpaax/caamflib/desc/pdcp.h | 82 +++
1 file changed, 41
From: Gagandeep Singh
If all the keys are inlined, the descriptor would
be 32 + 20 = 52 which is the size of the CURRENT shared
descriptor created.
So 32 * CAAM_CMD_SZ is the value that must be passed to
rta_inline_query() for its "sd_base_len" parameter and
drivers are using IPSEC_AUTH_VAR_AES_
This series include misc enhancements in dpaax_sec drivers.
- improving the IPsec protocol offload features
- enhancing PDCP protocol processing
- code optimization and cleanup
Apeksha Gupta (1):
crypto/dpaa2_sec: enhance dpaa FD FL FMT offset set
Gagandeep Singh (3):
common/dpaax: update IP
57 matches
Mail list logo
