Add support for the 6th Gen Intel Xeon Scalable processors. Note that NTB
devices within the 6th Gen Intel Xeon compliant to PCIe 5.0 spec.
Signed-off-by: Junfeng Guo
---
Depends-on: series-30460 \
("raw/ntb: refactor to support NTB on 5th Gen Intel Xeon")
---
doc/guides/rawdevs/ntb.rst
Adds support to configure writing result to offset of the DMA
response event.
Signed-off-by: Amit Prakash Shukla
---
drivers/common/cnxk/roc_dpi.c | 6 +-
drivers/common/cnxk/roc_dpi.h | 2 +-
drivers/common/cnxk/roc_dpi_priv.h | 4
drivers/common/cnxk/roc_idev.c |
Added cnxk driver support for dma event enqueue and dequeue.
Also added changes for work queue entry completion status.
Signed-off-by: Amit Prakash Shukla
---
doc/guides/eventdevs/cnxk.rst| 5 +
drivers/dma/cnxk/cnxk_dma_event_dp.h | 21 +++
drivers/dma/cnxk/cnxk_dmadev.c | 2
Added support of dma driver callback assignment to eventdev
enqueue and dequeue. The change also defines dma adapter
capabilities function.
Depends-on: patch-134955 ("lib/dmadev: get DMA device using device ID")
Signed-off-by: Amit Prakash Shukla
---
drivers/event/cnxk/cn10k_eventdev.c | 7
Hi,
> -Original Message-
> From: Luca Boccassi
> Sent: 12/7/2023 17:50
> To: dev
> Cc: Xueming(Steven) Li ; Kevin Traynor
> ; Mcnamara, John ; NBU-
> Contact-Thomas Monjalon (EXTERNAL) ; David
> Marchand
> Subject: Re: Next year of LTSes
>
> On Thu, 9 Nov 2023 at 10:30, Luca Boccassi
On 12/8/2023 7:24 AM, Jie Hai wrote:
> On 2023/12/8 1:42, Ferruh Yigit wrote:
>> On 12/7/2023 1:57 PM, Ferruh Yigit wrote:
>>> On 12/7/2023 1:42 AM, Jie Hai wrote:
Replace the use of gcc builtin __atomic_xxx intrinsics with
corresponding rte_atomic_xxx optional stdatomic API.
Si
On 12/8/2023 6:55 AM, Jie Hai wrote:
> This patchset refactors mailbox codes.
> We will send a patch fix for all __atomic_xxx, so this
> patchset still use __atomic_xxx.
>
> --
> v4:
> 1. use __atomic_xxx instead of rte_atomic_XXX.
> 2. use '__rte_packed' instead of '#pragma pack()'.
> v3:
> 1. fi
On 12/8/2023 7:44 AM, Jie Hai wrote:
> This patchset fixes some bugs on reset, and the replaces
> of gcc builtin __atomic_xxx with rte_atomic_xxx is removed
> and will send later separately.
>
> Dengdui Huang (3):
> net/hns3: fix VF multiple count on one reset
> net/hns3: fix disable command w
packet_ordering example works in two modes (--disable-reorder):
- When reorder is enabled: rx_thread - N*worker_thread - send_thread
- When reorder is disabled: rx_thread - N*worker_thread - tx_thread
N parallel worker_thread(s) generate out-of-order packets.
When reorder is enabled, send_thre
On 12/8/2023 2:38 AM, Liangxing Wang wrote:
> The refcnt update of stored mbufs in memif driver is redundant since
> those mbufs are only freed in eth_memif_tx_zc(). No other place
> can free those stored mbufs quietly. So remove the redundant mbuf
> refcnt update in dpdk memif driver to avoid extr
Getting readable and consistent logs is important when running a DPDK
application, especially when troubleshooting.
A common issue with logs is when a DPDK change do not add (or on the
contrary add too many \n) in the format string.
This issue would only get noticed when actually hitting this log
This macro is not used.
Fixes: 769b2de7fb52 ("hash: implement RCU resources reclamation")
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
---
lib/hash/rte_cuckoo_hash.h | 11 ---
1 file changed, 11 deletions(-)
diff --git a/lib/hash/rte_cuckoo_hash.h b/lib/hash/rte_cuckoo_hash.h
inde
This library logtype was not initialized so its logs would end up under
the 0 logtype, iow, RTE_LOGTYPE_EAL.
Fixes: b25246beaefc ("regexdev: add core functions")
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
---
lib/regexdev/rte_regexdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(
No printf!
When a dedicated log helper exists, use it.
And no usurpation please: a library should log under its logtype
(see the eventdev rx adapter update for example).
Note: the RTE_ETH_VALID_PORTID_OR_GOTO_ERR_RET macro is renamed for
consistency with the rest of eventdev (private) macros.
Cc:
Fix places where two newline characters may be logged.
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
---
Changes since RFC v1:
- split fixes on direct calls to printf or RTE_LOG in a previous patch,
---
drivers/crypto/ipsec_mb/ipsec_mb_ops.c | 2 +-
lib/bbdev/rte_bbdev.c
Fix places leading to a log message not terminated with a newline.
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
---
lib/eal/common/eal_common_options.c | 2 +-
lib/eal/linux/eal_hugepage_info.c | 2 +-
lib/eal/linux/eal_interrupts.c | 2 +-
lib/ethdev/ethdev_pci.h |
An error log message does not need to paraphrase the DPDK documentation.
Signed-off-by: David Marchand
---
lib/eal/linux/eal_timer.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/lib/eal/linux/eal_timer.c b/lib/eal/linux/eal_timer.c
index 3a30284e3a..df9ad61ae9 100644
There is no other log level than debug, simplify this helper.
Signed-off-by: David Marchand
---
lib/bpf/bpf_validate.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/lib/bpf/bpf_validate.c b/lib/bpf/bpf_validate.c
index 95b9ef99ef..f246b3c5eb 100644
--- a/lib/bpf/
Those error log messages don't need to span on multiple lines.
Signed-off-by: David Marchand
---
lib/acl/tb_mem.c| 2 +-
lib/bpf/bpf_stub.c | 4 ++--
lib/eal/windows/eal_hugepages.c | 4 ++--
lib/ethdev/rte_ethdev.c | 14 +++---
4 files changed, 12
Signed-off-by: David Marchand
---
lib/rcu/rte_rcu_qsbr.c | 62 --
lib/rcu/rte_rcu_qsbr.h | 1 +
2 files changed, 24 insertions(+), 39 deletions(-)
diff --git a/lib/rcu/rte_rcu_qsbr.c b/lib/rcu/rte_rcu_qsbr.c
index 41a44be4b9..5b6530788a 100644
--- a/lib/r
gcc builtin __builtin_strchr can be used as a static assertion to check
whether passed format strings contain a \n.
This can be useful to detect double \n in log messages.
Signed-off-by: David Marchand
---
Changes since RFC v1:
- added a check in checkpatches.sh,
---
devtools/checkpatches.sh |
Add the device name as a prefix of logs associated to madvise() calls.
Signed-off-by: David Marchand
---
lib/vhost/iotlb.c | 18 +-
lib/vhost/vhost.h | 2 +-
lib/vhost/vhost_user.c | 26 +-
3 files changed, 23 insertions(+), 23 deletions(-)
dif
Use RTE_LOG_LINE in existing macros that append a \n.
Signed-off-by: David Marchand
---
Changes since RFC v1:
- converted all logging helpers in lib/,
---
lib/bbdev/rte_bbdev.c | 5 +++--
lib/bpf/bpf_impl.h | 2 +-
lib/cfgfile/rte_cfgfile.c
Signed-off-by: David Young
Thanks!
David Young
Professional Copywriter/Technical Writer
Young Copy
+1 (678) 500-9550
https://www.youngcopy.com
On Thu, Dec 7, 2023 at 9:11 PM David Young wrote:
> ---
> doc/guides/prog_guide/bpf_lib.rst | 4 +-
> doc/guides/prog_guide/dmadev.rst
32nd bit of the eimc register is reserved according to the datasheet
Signed-off-by: Simon Ellmann
---
drivers/net/ixgbe/base/ixgbe_type.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ixgbe/base/ixgbe_type.h
b/drivers/net/ixgbe/base/ixgbe_type.h
index 1094df589
On 12/5/2023 2:54 AM, Chaoyong He wrote:
> Add the very basic rte_flow support for corenic firmware.
>
> ---
> v2:
> * Update the 'nfp.ini' document.
> * Rebase to the latest main branch.
> ---
>
> Chaoyong He (11):
> net/nfp: move some source files
> net/nfp: add the structures and functions
On Fri, 8 Dec 2023 15:59:35 +0100
David Marchand wrote:
> This macro is not used.
>
> Fixes: 769b2de7fb52 ("hash: implement RCU resources reclamation")
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
Reviewed-by: Stephen Hemminger
On Fri, 8 Dec 2023 15:59:38 +0100
David Marchand wrote:
> Fix places leading to a log message not terminated with a newline.
>
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
Maybe a coccinelle fixup script would help in future.
Acked-by: Stephen Hemminger
On Fri, 8 Dec 2023 15:59:39 +0100
David Marchand wrote:
> Fix places where two newline characters may be logged.
>
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
> Changes since RFC v1:
> - split fixes on direct calls to printf or RTE_LOG in a previous patch,
>
> ---
> driver
On Fri, 8 Dec 2023 15:59:41 +0100
David Marchand wrote:
> There is no other log level than debug, simplify this helper.
>
> Signed-off-by: David Marchand
Acked-by: Stephen Hemminger
On Fri, 8 Dec 2023 15:59:42 +0100
David Marchand wrote:
> Those error log messages don't need to span on multiple lines.
>
> Signed-off-by: David Marchand
> ---
> lib/acl/tb_mem.c| 2 +-
> lib/bpf/bpf_stub.c | 4 ++--
> lib/eal/windows/eal_hugepages.c | 4 ++--
On Fri, 8 Dec 2023 15:59:44 +0100
David Marchand wrote:
> Add the device name as a prefix of logs associated to madvise() calls.
>
> Signed-off-by: David Marchand
> ---
> lib/vhost/iotlb.c | 18 +-
> lib/vhost/vhost.h | 2 +-
> lib/vhost/vhost_user.c | 26 ++
On Fri, 8 Dec 2023 15:59:46 +0100
David Marchand wrote:
> Convert many libraries that call RTE_LOG(... "\n", ...) to RTE_LOG_LINE.
>
> Note:
> - for acl and sched libraries that still has some debug multilines
> messages, a direct call to RTE_LOG is used: this will make it easier to
> notic
On Mon, 4 Dec 2023 18:41:18 +
Maryam Tahhan wrote:
> Hi Shibin
>
> I'm not really sure what you are suggesting, is to make an assumption on
> the path part where the socket resides (aka hard code it) and then try
> to build the full UDS path in DPDK?
>
> Yes the plugin is using constants
In the current implementation when packets is received with
special flag(s) set, only that packet is delivered out of order.
There could be already coalesced packets in the GRO table
beloging to the same flow but not handled.
This fix makes sure that the entire segment is delivered with the
special
In the current implementation when packets is received with
special flag(s) set, only that packet is delivered out of order.
There could be already coalesced packets in the GRO table
beloging to the same flow but not handled.
This fix makes sure that the entire segment is delivered with the
special
In the current implementation when packets is received with
special flag(s) set, only that packet is delivered out of order.
There could be already coalesced packets in the GRO table
beloging to the same flow but not handled.
This fix makes sure that the entire segment is delivered with the
special
In the current implementation when a packet is received with
special TCP flag(s) set, only that packet is delivered out of order.
There could be already coalesced packets in the GRO table
belonging to the same flow but not delivered.
This fix makes sure that the entire segment is delivered with the
> -Original Message-
> From: David Marchand
> Sent: Friday, December 8, 2023 9:00 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; ferruh.yi...@amd.com;
> bruce.richard...@intel.com; step...@networkplumber.org;
> m...@smartsharesystems.com; Honnappa Nagarahalli
>
> Subject: [RFC v2 09/
Hi Michael,
> Subject: [PATCH v3 2/2] app/testpmd: add random item support
>
> Add support for random item, usage example:
>
> pattern random spec value 0x1 mask value 0x3 / eth / end
>
> Flow rule with above pattern matching 25% of the traffic, it hits only when
> random value suffix i
Hi Michael,
> +Item: ``RANDOM``
> +
> +
> +Matches a random value.
> +
> +The rundom number is generated by PMD,
s/rundom/random
I'm not sure that mentioning PMD here is fully correct, because in my opinion
it implies that SW generates it.
HW, SW and system clock were mentioned a
> + RTE_FLOW_FIELD_RANDOM /**< Random value. */
I think that this new modify field type should be mentioned in release notes as
well.
Best regards,
Dariusz Sosnowski
On 12/5/2023 2:54 AM, Chaoyong He wrote:
> diff --git a/drivers/net/nfp/nfp_net_cmsg.h b/drivers/net/nfp/nfp_net_cmsg.h
> new file mode 100644
> index 00..15e0bb60d8
> --- /dev/null
> +++ b/drivers/net/nfp/nfp_net_cmsg.h
> @@ -0,0 +1,37 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * C
> The HW adds a random number per each hash, this value can be used for
> statistic calculation over the packets, for example by setting one bit in the
> mask
> of that filed we will get half of the traffic in the flow, and so on with the
> rest of
> the mask.
Looks like there's a typo - s/filed/
> +- Match on random value:
> +
> + - Supported only with HW Steering enabled (``dv_flow_en`` = 2).
> + - Supported only in table with ``nb_flows=1``.
> + - NIC ingress flow in group 0 is not supported.
Shouldn't egress flow in group 0 be mentioned here as well?
> +* **Updated NVIDIA mlx5 net d
On Fri, Dec 08, 2023 at 03:59:35PM +0100, David Marchand wrote:
> This macro is not used.
>
> Fixes: 769b2de7fb52 ("hash: implement RCU resources reclamation")
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
Reviewed-by: Tyler Retzlaff
On Fri, Dec 08, 2023 at 03:59:36PM +0100, David Marchand wrote:
> This library logtype was not initialized so its logs would end up under
> the 0 logtype, iow, RTE_LOGTYPE_EAL.
>
> Fixes: b25246beaefc ("regexdev: add core functions")
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
On Fri, Dec 08, 2023 at 03:59:37PM +0100, David Marchand wrote:
> No printf!
> When a dedicated log helper exists, use it.
> And no usurpation please: a library should log under its logtype
> (see the eventdev rx adapter update for example).
>
> Note: the RTE_ETH_VALID_PORTID_OR_GOTO_ERR_RET macro
On Fri, Dec 08, 2023 at 03:59:38PM +0100, David Marchand wrote:
> Fix places leading to a log message not terminated with a newline.
>
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
Reviewed-by: Tyler Retzlaff
On Fri, Dec 08, 2023 at 03:59:40PM +0100, David Marchand wrote:
> An error log message does not need to paraphrase the DPDK documentation.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Tyler Retzlaff
On Fri, Dec 08, 2023 at 03:59:41PM +0100, David Marchand wrote:
> There is no other log level than debug, simplify this helper.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Tyler Retzlaff
On Fri, Dec 08, 2023 at 03:59:42PM +0100, David Marchand wrote:
> Those error log messages don't need to span on multiple lines.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Tyler Retzlaff
On Fri, Dec 08, 2023 at 03:59:43PM +0100, David Marchand wrote:
> Signed-off-by: David Marchand
> ---
Reviewed-by: Tyler Retzlaff
> lib/rcu/rte_rcu_qsbr.c | 62 --
> lib/rcu/rte_rcu_qsbr.h | 1 +
> 2 files changed, 24 insertions(+), 39 deletions(-)
>
>
I tested below 18 scenarios on RHEL9 and didn't find any new dpdk issues.
Guest with device assignment(PF) throughput testing(1G hugepage size): PASS
Guest with device assignment(PF) throughput testing(2M hugepage size) : PASS
Guest with device assignment(VF) throughput testing: PASS
PVP (host dpd
Hi Amit,
On 2023/12/8 15:55, Amit Prakash Shukla wrote:
> DMA library has a function to get DMA device based on device name but
> there is no function to get DMA device using device id.
>
> Added a function that lookup for the dma device using device id and
> returns the pointer to the same.
>
>
For lib/dmadev part
Reviewed-by: Chengwen Feng
On 2023/12/8 22:59, David Marchand wrote:
> Fix places where two newline characters may be logged.
>
> Cc: sta...@dpdk.org
>
> Signed-off-by: David Marchand
> ---
> Changes since RFC v1:
> - split fixes on direct calls to printf or RTE_LOG in a pr
For lib/dmadev part
Reviewed-by: Chengwen Feng
On 2023/12/8 22:59, David Marchand wrote:
> Use RTE_LOG_LINE in existing macros that append a \n.
>
> Signed-off-by: David Marchand
> ---
> Changes since RFC v1:
> - converted all logging helpers in lib/,
>
> ---
> lib/bbdev/rte_bbdev.c
Acked-by: Chengwen Feng
On 2023/12/8 22:59, David Marchand wrote:
> gcc builtin __builtin_strchr can be used as a static assertion to check
> whether passed format strings contain a \n.
> This can be useful to detect double \n in log messages.
>
> Signed-off-by: David Marchand
> ---
> Changes s
58 matches
Mail list logo
