On 20-05-2025 11:21, Gagandeep Singh wrote:
Due to race between KEY loading to CAAM’s internal memory
and initiating crypto operations can SEC errors in PDCP
AES algo combinations.
Please fix grammar.
To mitigate this, adding CALM instruction in SN 12bit case
and using older version of descri
On 20-05-2025 11:21, Gagandeep Singh wrote:
This patch adds NULL algo capabilities.
Do you also need to update documentation for the same?
Signed-off-by: Gagandeep Singh
---
drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h | 41 +++
1 file changed, 41 insertions(+)
diff --g
On 20-05-2025 11:21, Gagandeep Singh wrote:
update pdcp_insert_uplane_no_int_op() to support 12bit SN size
Signed-off-by: Gagandeep Singh
---
drivers/common/dpaax/caamflib/desc/pdcp.h | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/drivers/common/dpaax/caa
This workaround fixes the invalid key command SEC error.
Fixes: 6127fff842a7 ("common/dpaax: remove outdated caamflib code")
Cc: franck.lenorm...@nxp.com
Cc: sta...@dpdk.org
Signed-off-by: Gagandeep Singh
---
drivers/common/dpaax/caamflib/desc/pdcp.h | 16 ++--
1 file changed, 14 ins
From: Jun Yang
Make header room big enough for IPSec with TX dynamic confirm enabled.
Ingress minimum header room:
64(size of parser result) + 8(address of SEC context) * 2.
Egress minimum header room:
88(FAEAD offset) + 8(FAEAD size) + 8(address of SEC context) * 2 +
114(expansion).
Signed-off-
From: Jun Yang
For IVP mode of simple IPsec FD, store op context point just
before expansion and align with point size(8).
Otherwise, store op context point in offset to mbuf with
rte_mbuf_dynfield register.
Signed-off-by: Jun Yang
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 143
From: Jun Yang
For user to identify dpaa2 mempool by operation index.
Signed-off-by: Jun Yang
---
drivers/mempool/dpaa2/dpaa2_hw_mempool.c | 20 +++-
drivers/mempool/dpaa2/dpaa2_hw_mempool.h | 5 -
2 files changed, 23 insertions(+), 2 deletions(-)
diff --git a/drivers/mem
This patch adds NULL algo capabilities.
Signed-off-by: Gagandeep Singh
---
drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h | 41 +++
1 file changed, 41 insertions(+)
diff --git a/drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h
b/drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h
index 1c0bc3d6d
From: Vanshika Shukla
Fixes the uninitialized variable issue - reported by NXP internal
coverity.
Fixes: 1182b364312c ("crypto/dpaax_sec: set authdata in non-auth case")
Cc: g.si...@nxp.com
Cc: sta...@dpdk.org
Signed-off-by: Vanshika Shukla
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 1
Replacing existing device creation and deletion code in probe and
remove functions with rte_cryptodev_pmd_create() and
rte_cryptodev_pmd_destroy() APIs provided by the DPDK library,
simplifying drivers initialization and teardown.
Signed-off-by: Gagandeep Singh
---
drivers/crypto/dpaa2_sec/dpaa2
Due to race between KEY loading to CAAM’s internal memory
and initiating crypto operations can SEC errors in PDCP
AES algo combinations.
To mitigate this, adding CALM instruction in SN 12bit case
and using older version of descriptor for SN 18bit case.
Fixes: 6127fff842a7 ("common/dpaax: remove ou
The ixgbe macsec registers are not cleared on read. Reading the
registers and then accumulating the values will cause macsec stats
errors. Each time the data is queried via the testpmd command
“show port stats xstats all”, the macsec statistics are increased.
This commit will add macsec offset sta
The ixgbe macsec registers are not cleared on read. Reading the
registers and then accumulating the values will cause macsec stats
errors. Each time the data is queried via the testpmd command
“show port stats xstats all”, the macsec statistics are increased.
This commit will add macsec offset sta
From: Pavan Nikhilesh
Migrate all invocations of rte_event_dma_adapter_op
API to rte_dma_op.
Signed-off-by: Pavan Nikhilesh
Acked-by: Amit Prakash Shukla
---
app/test-eventdev/test_perf_common.c | 6 +-
app/test-eventdev/test_perf_common.h | 4 +-
app/test/test_event_dma_a
From: Pavan Nikhilesh
Implement DMA enqueue/dequeue operations when
application enables it via configuration.
Signed-off-by: Pavan Nikhilesh
---
drivers/dma/cnxk/cnxk_dmadev.c| 78
drivers/dma/cnxk/cnxk_dmadev.h| 7 ++
drivers/dma/cnxk/cnxk_dmadev_fp.c | 189 +++
From: Pavan Nikhilesh
This patch adds a option to measure performanc of
enq/deq operations in the benchmark app.
Signed-off-by: Pavan Nikhilesh
---
app/test-dma-perf/benchmark.c | 137 +++---
app/test-dma-perf/config.ini | 3 +
app/test-dma-perf/main.c | 13
From: Pavan Nikhilesh
Add enqueue dequeue operations tests.
Signed-off-by: Pavan Nikhilesh
---
app/test/test_dmadev.c | 160 +
app/test/test_dmadev_api.c | 76 --
2 files changed, 228 insertions(+), 8 deletions(-)
diff --git a/app/test/
From: Pavan Nikhilesh
Add enqueue/dequeue operations that use struct rte_dma_op
to communicate with the DMA device.
These operations need to be enabled at DMA device configuration
time by setting the flag rte_dma_conf::enable_enq_deq if the
device supports RTE_DMA_CAPA_OPS_ENQ_DEQ capability.
W
From: Pavan Nikhilesh
Introduce DMA enqueue/dequeue operations to the DMA device library.
Add configuration flags to rte_dma_config instead of boolean for
individual features.
The enqueue/dequeue operations allow applications to communicate with the
DMA device using the rte_dma_op structure, pr
Add jump action verification method and test case to Flow API test suite,
as well as a case for validating flows with different priority levels.
Signed-off-by: Dean Marx
---
dts/tests/TestSuite_flow.py | 187 +++-
1 file changed, 165 insertions(+), 22 deletions(-)
Add an RTE Flow API testing suite, which covers some basic
synchronous Flow API rules that should be supported across PMDs.
This suite will be added to over time, as the Flow API is too large
to cover all in one suite, and sending one monolithic series
would be impossible.
Signed-off-by: Dean Marx
> Hints API
>
> On Sat, May 17, 2025 at 5:18 PM Wathsala Vithanage
> wrote:
> > diff --git a/drivers/bus/pci/linux/pci.c b/drivers/bus/pci/linux/pci.c
> > index c20d159218..463c06ad64 100644
> > --- a/drivers/bus/pci/linux/pci.c
> > +++ b/drivers/bus/pci/linux/pci.c
> > @@ -814,3 +814,97 @@ rte_p
Linux uAPI changes
>
> Hi Wathsala,
>
> On Sat, May 17, 2025 at 5:18 PM Wathsala Vithanage
> wrote:
> >
> > This commit is a hack to prevent build failures the next commit in
> > this patch series causes due to missing vfio uapi definitions.
> > This commit should NEVER BE MERGED.
> > Next commi
From: Sunil Kumar Kori
hdr_lens is used to maintain header lengths after parsing packets.
When port receives different type of packets (say first is VXLAN
packet and second is GRE packet).
For first packet, L2/L3/L4 lengths are set for inner and outer header
alongwith tunnel_len.
Now for second
From: Sunil Kumar Kori
While parsing packet headers, offset must be added to get next
header but for geneve header parsing offset is overwritten.
Also inner_l2_len is not set in case of geneve packets.
Fixes: 64ed7f854cf4 ("net: add tunnel packet type parsing")
Signed-off-by: Sunil Kumar Kori
My bad, I got it confused with lcore topology patches.
I will look into this.
Thanks,
Pavan.
> -Original Message-
> From: Pavan Nikhilesh Bhagavatula
> Sent: Monday, May 19, 2025 7:32 PM
> To: Morten Brørup ; Varghese, Vipin
>
> Cc: dev
> Subject: RE: rte_lcore_has_role() return value
Hi Morten,
Looks like this is meant for Vipin.
+Vipin
Thanks,
Pavan.
> Pavan,
>
> The rte_lcore_has_role() documentation says it returns bool.
>
> Its implementation returns a true value (-1) if given a non-compliant
> lcore_id,
> e.g. LCORE_ID_ANY.
>
> I think it should behave like rte_lc
Updates CPT parse header HW structure for O20.
Signed-off-by: Rahul Bhansali
---
drivers/common/cnxk/hw/cpt.h| 12 +++-
drivers/common/cnxk/roc_cpt_debug.c | 8
2 files changed, 11 insertions(+), 9 deletions(-)
diff --git a/drivers/common/cnxk/hw/cpt.h b/drivers/common
From: Monendra Singh Kushwaha
This patch enables setting CPT result address offset relative
to wqe address.
Signed-off-by: Monendra Singh Kushwaha
---
Depends-on: series-34970 ("[v2,1/3] common/cnxk: update steer rule mbox for
cn20k")
drivers/common/cnxk/roc_mbox.h | 4
drivers/
Update frag offset calculation based on CSR change.
Signed-off-by: Rahul Bhansali
---
drivers/net/cnxk/cn20k_rx.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/cnxk/cn20k_rx.h b/drivers/net/cnxk/cn20k_rx.h
index 09042e384d..5682631815 100644
--- a/drivers/net/cn
Update bufsize in lookup-mem by default to maintain
data order.
Signed-off-by: Rahul Bhansali
---
drivers/net/cnxk/cn20k_ethdev.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/cnxk/cn20k_ethdev.c b/drivers/net/cnxk/cn20k_ethdev.c
index 740fdb7f76..159ce39e97 1
Inbound CPT LF ID update for Rx inline config.
Signed-off-by: Rahul Bhansali
---
drivers/common/cnxk/roc_nix_inl.c | 10 +-
drivers/common/cnxk/roc_nix_inl_dev.c | 4 +---
drivers/common/cnxk/roc_nix_inl_priv.h | 2 +-
3 files changed, 11 insertions(+), 5 deletions(-)
diff --git
This configures the CPT result address offset from WQE pointer
in RX_INLINE_CFG0. CPT result address offset is a signed
number in multiple of 16 bytes and configured to the
corresponding reserved space in packet meta area itself.
Signed-off-by: Rahul Bhansali
---
drivers/common/cnxk/roc_nix_inl.
> -Original Message-
> From: Minggang(Gavin) Li
> Sent: Friday, May 16, 2025 10:11 AM
> To: Matan Azrad ; Slava Ovsiienko
> ; Ori Kam ; NBU-Contact-
> Thomas Monjalon (EXTERNAL) ; Dariusz Sosnowski
> ; Bing Zhao ; Suanming Mou
> ; NBU-Contact-N?lio Laranjeiro (EXTERNAL)
> ; Yongseok Koh
>
Hi Yangming,
How did you configure the queue pairs differently between the primary and
secondary processes?
The application must call rte_cryptodev_queue_pair_setup() with a unique qp_id
for each process. This implies that each process should receive distinct queue
pair configurations at runti
Pavan,
The rte_lcore_has_role() documentation says it returns bool.
Its implementation returns a true value (-1) if given a non-compliant lcore_id,
e.g. LCORE_ID_ANY.
I think it should behave like rte_lcore_is_enabled(), i.e. return false if
given a non-compliant lcore_id, e.g. LCORE_ID_ANY.
Hi Wathsala,
On Sat, May 17, 2025 at 5:18 PM Wathsala Vithanage
wrote:
>
> This commit is a hack to prevent build failures the next commit in this
> patch series causes due to missing vfio uapi definitions.
> This commit should NEVER BE MERGED.
> Next commit in this patch series depends on additi
37 matches
Mail list logo
