RE: [dpdk-dev] [PATCH v2] doc: define qualification criteria for external library

> From: Jerin Jacob [mailto:jerinjac...@gmail.com] > Sent: Sunday, 19 November 2023 08.08 > > On Fri, Nov 17, 2023 at 1:57 PM Morten Brørup > wrote: > > > > > From: Jerin Jacob [mailto:jerinjac...@gmail.com] > > > Sent: Friday, 17 November 2023 05.34 > > > > > > On Thu, Sep 28, 2023 at 11:10 AM

[PATCH] doc/mlx4: set limitation for RSS

This adds a limitation explanation on RSS queue usage for MLX4 pmd. MLX4 pmd requires a power of two queues to do RSS but, the user can still open more queues which can be utilized through flow API. Signed-off-by: Raslan Darawsheh --- doc/guides/nics/mlx4.rst | 12 1 file changed,

The problem with priv_data in crypto and security API's

In the patchset, that removes the GNU style zero length arrays, there is a problem caused by the use of these in the cryptodev header files. For both rte_session and rte_cryptodev, the current convention is to do: struct rte_security_session { RTE_MARKER cacheline0; uint64_t opaqu

[Bug 366] i40e PMD returns 0 for secondary invoking rx_burst on queue 0, when Primary dies

https://bugs.dpdk.org/show_bug.cgi?id=366 dengkaiwen (kaiwenx.d...@intel.com) changed: What|Removed |Added Ever confirmed|1 |0 Status|RESOLVE

RE: [PATCH v2] net/txgbe: fix out of bound access

On Friday, November 17, 2023 6:12 PM, ferruh.yi...@amd.com wrote: > Reported by SuSe CI [1] by GCC [2], possibly false positive. Error: > > In function 'txgbe_host_interface_command', > inlined from 'txgbe_host_interface_command' > at ../drivers/net/txgbe/base/txgbe_mng.c:104:1,

Re: [PATCH v7 1/4] app/dma-perf: add skip support

Hi Gowrishankar and Amit, On 2023/11/17 20:15, Gowrishankar Muthukrishnan wrote: > From: Amit Prakash Shukla > > Add support to skip running a dma-perf test-case. > > Signed-off-by: Amit Prakash Shukla > Acked-by: Anoob Joseph > --- > app/test-dma-perf/config.ini | 2 ++ > app/test-dma-perf

RE: [EXT] The problem with priv_data in crypto and security API's

Hi Stephen, Since the issue is only with cnxk drivers, may I propose a change in driver to mitigate this issue? Flex array that is proposed is already used by asymmetric sessions. We have addressed the issue you highlighted with below scheme. #define ASYM_SESS_SIZE sizeof(struct rte_cryptodev_

Re: [PATCH 3/4] drivers: add extension keyword to statement expressions

On 2023/11/16 7:02 AM, Tyler Retzlaff wrote: Add __extension__ keyword to gcc statement expression extensions. Signed-off-by: Tyler Retzlaff --- drivers/bus/fslmc/mc/fsl_mc_sys.h| 6 ++--- drivers/common/cnxk/roc_io.h | 6 ++--- drivers/common/cnxk/ro

[PATCH 2/3] doc/hns3: update the features for hns3

Update the features for hns3. Signed-off-by: Huisong Li --- doc/guides/nics/hns3.rst | 10 +- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/doc/guides/nics/hns3.rst b/doc/guides/nics/hns3.rst index 7eac1b3c92..11d0c04884 100644 --- a/doc/guides/nics/hns3.rst +++ b/doc/guid

[PATCH 0/3] doc/hns3: fix hns3 doc

This series fix some feature description and update features for hns3. Hope this series merge into 23.11. Huisong Li (3): doc/hns3: fix compilation option about max queue number doc/hns3: update the features for hns3 doc/hns3: fix the usage description for RSS flow doc/guides/nics/hns3.rs

[PATCH 1/3] doc/hns3: fix compilation option about max queue number

This patch fixes the description of compilation option about setting max queue number. The maximum queue number of HIP09 and HIP10 is determined by the RTE_LIBRTE_HNS3_MAX_TQP_NUM_PER_PF and the value the firmware report. Fixes: 21938cf4e43e ("doc: add build config option in hns3 guide") Cc: sta..

[PATCH 3/3] doc/hns3: fix the usage description for RSS flow

The hns3 driver supports for creating rule base on input tuple, hash key, queues and hash algorithm. But hash key, queues and hash algorithm are the global configuration for hardware which will affect other rules. The rule just setting input tuple is completely independent. Fixes: 63a0f65c9572 ("d