Re: [PATCH v5 0/4] add pointer compression API

I have added macros to help find the parameters and I have added mempool functions that allow you to determine if you can use the mempool and what params it needs. The new mempool functions are mentioned in the docs for ptr compress. Please take a look at v11. I did not add a new make pkt buf API

[PATCH v11 0/6] add pointer compression API

This patchset is proposing adding a new header only library with utility functions that allow compression of arrays of pointers. Since this is a header only library a patch needed to be added to amend the build system to allow adding libraries without source files. When passing caches full of poi

[PATCH v11 1/6] lib: allow libraries with no sources

Allow header only libraries. Signed-off-by: Paul Szczepanek Reviewed-by: Honnappa Nagarahalli Reviewed-by: Jack Bond-Preston Acked-by: Bruce Richardson --- lib/meson.build | 16 1 file changed, 16 insertions(+) diff --git a/lib/meson.build b/lib/meson.build index 179a272932.

[PATCH v11 2/6] mempool: add functions to get extra mempool info

Add two functions: - rte_mempool_get_mem_range - get virtual memory range of the objects in the mempool, - rte_mempool_get_obj_alignment - get alignment of objects in the mempool. Add two tests that test these new functions. Signed-off-by: Paul Szczepanek Reviewed-by: Jack Bond-Preston Reviewed

[PATCH v11 3/6] ptr_compress: add pointer compression library

Add a new utility header for compressing pointers. The provided functions can store pointers as 32-bit or 16-bit offsets. The compression takes advantage of the fact that pointers are usually located in a limited memory region (like a mempool). We can compress them by converting them to offsets fr

[PATCH v11 4/6] test: add pointer compress tests to ring perf test

Add a test that runs a zero copy burst enqueue and dequeue on a ring of raw pointers and compressed pointers at different burst sizes to showcase performance benefits of newly added pointer compression APIs. Refactored threading code to pass more parameters to threads to reuse existing code. Added

[PATCH v11 5/6] docs: add pointer compression guide

Documentation added in the prog guide for the new utility functions for pointer compression showing example code and potential usecases. Signed-off-by: Paul Szczepanek Reviewed-by: Honnappa Nagarahalli Reviewed-by: Nathan Brown Reviewed-by: Jack Bond-Preston --- MAINTAINERS

[PATCH v11 6/6] test: add unit test for ptr compression

Test compresses and decompresses pointers with various combinations of memory regions and alignments and verifies the pointers are recovered correctly. Signed-off-by: Paul Szczepanek Reviewed-by: Honnappa Nagarahalli Reviewed-by: Nathan Brown Reviewed-by: Jack Bond-Preston --- MAINTAINERS

RE: [PATCH v5 0/4] add pointer compression API

> I have added macros to help find the parameters and I have added mempool > functions that allow you to determine if you can use the mempool and > what params it needs. The new mempool functions are mentioned in the > docs for ptr compress. > Please take a look at v11. Great, thanks. Will try t

Re: [PATCH 1/3] compress/uadk: Introduce UADK compression driver

On Fri, 24 May 2024 at 00:30, Konstantin Ananyev wrote: > > > > > -Original Message- > > From: Stephen Hemminger > > Sent: Thursday, May 23, 2024 5:21 PM > > To: Zhangfei Gao > > Cc: Akhil Goyal ; Fan Zhang ; > > Ashish Gupta ; > > dev@dpdk.org > > Subject: Re: [PATCH 1/3] compress/uadk

Re: [PATCH v2 22/27] net/ixgbe/base: add support for E610 Admin Command Interface

On Fri, May 03, 2024 at 02:57:53PM +0100, Anatoly Burakov wrote: > From: Piotr Kwapulinski > > Add low level support for Admin Command Interface (ACI). ACI is the > Firmware interface used by a driver to communicate with E610 adapter. Add > the following ACI features: > - data structures, macros,

Re: [PATCH v2 27/27] net/ixgbe/base: add various miscellaneous features

On Fri, May 03, 2024 at 02:57:58PM +0100, Anatoly Burakov wrote: > Add various features to the base driver that are not specifically > about packet I/O but are present in the shared code snapshot. > > Also, update documentation to reflect new base code snapshot version, > as well as document new h

Re: [EXTERNAL] [PATCH 2/3] compress/uadk: support basic operations

Hi, Akhil Thanks for your time. On Thu, 23 May 2024 at 23:38, Akhil Goyal wrote: > > Hi Zhangfei, > > Overall, a well written driver. > Please see below comment. > > > +static int > > +uadk_compress_pmd_config(struct rte_compressdev *dev, > > + struct rte_compressdev_config

RE: [EXTERNAL] [PATCH 2/3] compress/uadk: support basic operations

> > > +static int > > > +uadk_compress_pmd_config(struct rte_compressdev *dev, > > > + struct rte_compressdev_config *config) > > > +{ > > > + char mp_name[RTE_MEMPOOL_NAMESIZE]; > > > + struct uadk_compress_priv *priv; > > > + struct rte_mempool *mp; > > > + in

RE: [PATCH v11 2/6] mempool: add functions to get extra mempool info

> From: Paul Szczepanek [mailto:paul.szczepa...@arm.com] > Sent: Friday, 24 May 2024 10.37 > > +size_t rte_mempool_get_obj_alignment(struct rte_mempool *mp) > +{ > + if (mp == NULL) > + return 0; > + > + if (mp->flags & RTE_MEMPOOL_F_NO_CACHE_ALIGN) > + return sizeo

RE: [PATCH v11 3/6] ptr_compress: add pointer compression library

> +#define BITS_REQUIRED_TO_STORE_VALUE(x) \ > + ((x) == 0 ? 1 : (sizeof(size_t) * CHAR_BIT - __builtin_clzl((size_t)x))) > + > +#define BIT_SHIFT_FROM_ALIGNMENT(x) ((x) == 0 ? 0 : __builtin_ctzl(x)) The two built-ins used are not available with MSVC. Please see if you can find similar functio

Re: [PATCH] dma/cnxk: add higher chunk size support

On Wed, May 22, 2024 at 8:15 PM wrote: > > From: Pavan Nikhilesh > > Add support to configure higher chunk size by using the new > OPEN_V2 mailbox, this improves performance as the number of > mempool allocs are reduced. > > Signed-off-by: Pavan Nikhilesh > Signed-off-by: Amit Prakash Shukla >

Re: [PATCH v3 1/7] dma/odm: add framework for ODM DMA device

On Fri, Apr 19, 2024 at 12:13 PM Anoob Joseph wrote: > > Add framework for Odyssey ODM DMA device. > > Signed-off-by: Anoob Joseph > Signed-off-by: Gowrishankar Muthukrishnan > Signed-off-by: Vidya Sagar Velumuri > --- > MAINTAINERS | 6 +++ > drivers/dma/meson.build |

Re: [PATCH v3 2/7] dma/odm: add hardware defines

On Fri, Apr 19, 2024 at 12:22 PM Anoob Joseph wrote: > > Add ODM registers and structures. Add mailbox structs as well. > > Signed-off-by: Anoob Joseph > Signed-off-by: Gowrishankar Muthukrishnan > Signed-off-by: Vidya Sagar Velumuri > --- > drivers/dma/odm/odm.h | 116 +++

[PATCH] doc: remove empty section from testpmd guide

When removing the documentation for the obsolete flow filtering, the related section was kept almost empty. It can be safely removed. Fixes: 5007ac13189d ("ethdev: remove deprecated Flow Director configuration") Cc: sta...@dpdk.org Signed-off-by: Thomas Monjalon --- doc/guides/testpmd_app_ug/te

Re: [PATCH v3 4/7] dma/odm: add device ops

On Fri, Apr 19, 2024 at 3:22 PM Anoob Joseph wrote: > > From: Gowrishankar Muthukrishnan > > Add DMA device control ops. > > Signed-off-by: Anoob Joseph > Signed-off-by: Gowrishankar Muthukrishnan > Signed-off-by: Vidya Sagar Velumuri > --- > drivers/dma/odm/odm.c| 144 +++

Re: [PATCH v2 1/1] net/cnxk: fix RSS config via ethdev configure API

On Thu, May 23, 2024 at 8:23 PM wrote: > > From: Sunil Kumar Kori > > Currently user passed RSS configuration is ignored via > rte_eth_dev_configure() API. Instead default RSS setup is done by > driver. > > Adding handling for user passed RSS configuration too via > rte_eth_dev_configure(). > > F

[PATCH] net/mlx5: remove redundant macro

The macro MLX5_BITSHIFT() is not used anymore, and is redundant with RTE_BIT64(), so it can be removed. Signed-off-by: Thomas Monjalon --- drivers/net/mlx5/mlx5_utils.h | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/net/mlx5/mlx5_utils.h b/drivers/net/mlx5/mlx5_utils.h index d86a8

Re: [PATCH v2 0/2] dts: update mypy and clean up

Disregard that last comment, it was a mistake on my end. Updating the dependencies in my container instance fixed the problem. On Tue, May 21, 2024 at 10:59 AM Luca Vizzarro wrote: > > On 20/05/2024 16:37, Nicholas Pratte wrote: > > Ran into no issues with this when testing your port stats and in

Re: [EXTERNAL] [PATCH 2/3] compress/uadk: support basic operations

On Fri, 24 May 2024 at 18:20, Akhil Goyal wrote: > > > > > +static int > > > > +uadk_compress_pmd_config(struct rte_compressdev *dev, > > > > + struct rte_compressdev_config *config) > > > > +{ > > > > + char mp_name[RTE_MEMPOOL_NAMESIZE]; > > > > + struct uadk_compres

Re: [PATCH v2 1/2] dts: update mypy static checker

On Tue, May 14, 2024 at 7:34 AM Luca Vizzarro wrote: > diff --git a/dts/framework/runner.py b/dts/framework/runner.py > index db8e3ba96b..d74f1871db 100644 > --- a/dts/framework/runner.py > +++ b/dts/framework/runner.py > @@ -23,7 +23,7 @@ > import re > import sys > from pathlib import Path > -

RE: [EXTERNAL] [PATCH 2/3] compress/uadk: support basic operations

> > > > > > One more question, > > > rte_compressdev_pmd_init_params does not have .max_nb_queue_pairs as > > > rte_cryptodev_pmd_init_params. > > > So dpdk-test-compress-perf will use 128 queues by default, except > > > adding -l 1,2. > > > Is this expected? > > > > > rte_compressdev_pmd_init_par

RE: [PATCH] doc: add firmware update links in mlx5 guide

> +The firmware can be updated with `mlxup > +`_. > +The latest firmwares can be downloaded at > +https://network.nvidia.com/support/firmware/firmware-downloads/ > + What do you think about moving these lines from "Upstream dependencies" t

RE: [PATCH] net/mlx5: remove redundant macro

> -Original Message- > From: Thomas Monjalon > Sent: Friday, May 24, 2024 15:43 > To: dev@dpdk.org > Cc: Dariusz Sosnowski ; Slava Ovsiienko > ; Ori Kam ; Suanming Mou > ; Matan Azrad > Subject: [PATCH] net/mlx5: remove redundant macro > > The macro MLX5_BITSHIFT() is not used anymore, a

Re: [PATCH] doc: add firmware update links in mlx5 guide

24/05/2024 17:36, Dariusz Sosnowski: > > +The firmware can be updated with `mlxup > > +`_. > > +The latest firmwares can be downloaded at > > +https://network.nvidia.com/support/firmware/firmware-downloads/ > > + > > What do you think abou

RE: [PATCH v1 2/3] app/testpmd: add VXLAN last reserved modification command

> -Original Message- > From: rongwei liu > Sent: Thursday, May 23, 2024 09:29 > To: dev@dpdk.org; Matan Azrad ; Slava Ovsiienko > ; Ori Kam ; Suanming Mou > ; NBU-Contact-Thomas Monjalon (EXTERNAL) > > Cc: Dariusz Sosnowski ; Aman Singh > ; Yuying Zhang > Subject: [PATCH v1 2/3] app/test

RE: [PATCH] doc: add firmware update links in mlx5 guide

> -Original Message- > From: Thomas Monjalon > Sent: Friday, May 24, 2024 17:43 > To: Dariusz Sosnowski > Cc: dev@dpdk.org; Slava Ovsiienko ; Ori Kam > ; Suanming Mou ; Matan Azrad > > Subject: Re: [PATCH] doc: add firmware update links in mlx5 guide > > 24/05/2024 17:36, Dariusz Sosnow

[PATCH v2] doc: remove empty section from testpmd guide

When removing the documentation for the obsolete flow filtering, the related section was kept almost empty. It can be safely removed. Fixes: 5007ac13189d ("ethdev: remove deprecated Flow Director configuration") Cc: sta...@dpdk.org Signed-off-by: Thomas Monjalon --- v2: remove a reference to the

Re: [PATCH v2 2/2] dts: clean up config types

Looks good to me. Reviewed-by: Patrick Robb Tested-by: Patrick Robb

Re: [PATCH v2 0/2] dts: update mypy and clean up

Series-reviewed-by: Patrick Robb As a reminder, once this is merged, UNH will add a CI job which runs mypy and reports its results to PW (only runs for patches with changes in dpdk/dts). Luca, I assume what you had in mind was we will run dts-check-format.sh, so including isort and pylama, as op

[PATCH v3] test/crypto: fix enqueue dequeue callback case

The enqueue/dequeue callback test cases were using the test_null_burst_operation() for doing enqueue/dequeue. But this function is only designed to be run for NULL PMD. Hence for other PMDs, the callback was not getting called. Now, separate processing thread is removed, instead NULL crypto operati

[RFC PATCH 0/1] Initial Implementation For Jumbo Frames

The following is a rough design for the implementation of the jumbo frames test suite in new DTS. The test suite uses the same testing methodology from the test suite implementation in old DTS. In doing so, much of the logic in this implementation has been stripped away for the sake of maintaining

[RFC PATCH 1/1] Initial Implementation For Jumbo Frames Test Suite

The following test suite reflects the fundamental outline for how the jumbo frames test suite may be designed. The test suite consists of five individual test cases, each of which assesses the behavior of packet transmissions for both 1518 byte and 9000 byte frames. The edge cases are ripped from

Re: [RFC PATCH 0/1] Initial Implementation For Jumbo Frames

This is being superseded, ignore this submission. Thanks. On Fri, May 24, 2024 at 2:07 PM Nicholas Pratte wrote: > > The following is a rough design for the implementation of the jumbo > frames test suite in new DTS. The test suite uses the same > testing methodology from the test suite implemen

[RFC PATCH 0/1] Initial Implementation For Jumbo Frames

The following is a rough design for the implementation of the jumbo frames test suite in new DTS. The test suite uses the same testing methodology from the test suite implementation in old DTS. In doing so, much of the logic in this implementation has been stripped away for the sake of maintaining

[RFC PATCH] Initial Implementation For Jumbo Frames Test Suite

The following test suite reflects the fundamental outline for how the jumbo frames test suite may be designed. The test suite consists of five individual test cases, each of which assesses the behavior of packet transmissions for both 1518 byte and 9000 byte frames. The edge cases are ripped direc

[PATCH] event/dsw: support explicit release only mode

Add the RTE_EVENT_DEV_CAP_IMPLICIT_RELEASE_DISABLE capability to the DSW event device. This feature may be used by an EAL thread to pull more work from the work scheduler, without giving up the option to forward events originating from a previous dequeue batch. This in turn allows an EAL thread to

Re: [RFC PATCH v2] dts: skip test cases based on capabilities

I think this implementation is great, and I plan on testing it properly with the jumbo frames suite that I am developing before giving the final review. The only input that I could reasonably give is a couple rewordings on the docstrings which I'll highlight below. On Thu, Apr 11, 2024 at 4:48 AM