[PATCH] compress/mlx5: fix double close for context device

From: Michael Baum The context of the device opens once in the common probe and closes with its removal. If the probe of one of the drivers fails, it releases its resources and then the common closes the context. But mistakenly in the compress probe, if ther isn't enough capabilities to support

[PATCH] net/mlx5: fix devargs validation for multiclass probe

From: Michael Baum The mlx5_args function reads the devargs and checks if they are valid for this driver and if not it returns an error. This was normal behavior as long as all the devargs come to this driver, but since it is possible to run several drivers together, the function may return an e

[PATCH] common/mlx5: fix error handling in multi-class probe

From: Michael Baum The common drivers_probe function calls in a loop to all probe functions for classes requested by the user. After it manages to probe them all, it updates this on the device in the "classes_loaded" field. If one of them fails, all those probed to it are remove using the driver

[PATCH v4] kni: allow configuring the kni thread granularity

The Kni kthreads seem to be re-scheduled at a granularity of roughly 1 millisecond right now, which seems to be insufficient for performing tests involving a lot of control plane traffic. Even if KNI_KTHREAD_RESCHEDULE_INTERVAL is set to 5 microseconds, it seems that the existing code cannot resch

Re: [PATCH v2] test/bpf_convert: do not require DNS lookup

On Sun, 2021-11-14 at 23:39 +, Luca Boccassi wrote: > On Wed, 10 Nov 2021 at 16:18, Stephen Hemminger > wrote: > > > > These tests were using strings with hostname (dpdk.org) and this makes > > test code do a DNS lookup. In container environment used for OpenSuse > > build, DNS is unavailable

release candidate 21.11-rc4

Hi all, A new DPDK release candidate is ready for testing: https://git.dpdk.org/dpdk/tag/?id=v21.11-rc4 There are 46 new patches in this snapshot, for last critical fixes and some documentation updates. Release notes: https://doc.dpdk.org/guides/rel_notes/release_21_11.html You

RE: [PATCH v3] ethdev: deprecate header fields and metadata flow actions

Hi > -Original Message- > From: Thomas Monjalon > Sent: Wednesday, November 24, 2021 7:33 PM > > 24/11/2021 17:37, Slava Ovsiienko: > > From: Ori Kam > > > From: Viacheslav Ovsiienko > > > > Action: ``OF_SET_MPLS_TTL`` > > > > ^^^ > > > > +This action is depre

Re: [PATCH v1] gpudev: return EINVAL if invalid input pointer for free and unregister

On Wed, Nov 24, 2021 at 09:24:42AM -0800, Tyler Retzlaff wrote: > On Fri, Nov 19, 2021 at 10:56:36AM +0100, Thomas Monjalon wrote: > > 19/11/2021 10:34, Ferruh Yigit: > > > >> + if (ptr == NULL) { > > > >> + rte_errno = EINVAL; > > > >> + return -rte_errno; > > > >> +

[PATCH] examples/ipsec-secgw: set L4 length for TSO packets

Using RTE_MBUF_F_TX_TCP_SEG requires L4 length to be set. Fixes: a7f32947a316 ("examples/ipsec-secgw: support TCP TSO") Signed-off-by: Radu Nicolau --- examples/ipsec-secgw/ipsec_process.c | 1 + 1 file changed, 1 insertion(+) diff --git a/examples/ipsec-secgw/ipsec_process.c b/examples/ipsec-

Re: [PATCH v3] ethdev: deprecate header fields and metadata flow actions

24/11/2021 17:37, Slava Ovsiienko: > From: Ori Kam > > From: Viacheslav Ovsiienko > > > Action: ``OF_SET_MPLS_TTL`` > > > ^^^ > > > +This action is deprecated. Consider `Action: MODIFY_FIELD`_. > > > > > > > Since no PMD support this action why set reference to MODIFY_F

Re: [PATCH v1] gpudev: return EINVAL if invalid input pointer for free and unregister

On Fri, Nov 19, 2021 at 10:56:36AM +0100, Thomas Monjalon wrote: > 19/11/2021 10:34, Ferruh Yigit: > > >> +if (ptr == NULL) { > > >> +rte_errno = EINVAL; > > >> +return -rte_errno; > > >> +} > > > > > > in general dpdk has real problems with how it i

[PATCH v3 1/2] doc: note KNI alternatives

Add more information on alternatives of KNI and the cons of KNI against these alternatives. Signed-off-by: Ferruh Yigit --- Cc: Olivier Matz Olivier Matz Cc: David Marchand David Marchand Cc: Stephen Hemminger Stephen Hemminger Cc: Elad Nachman Cc: Igor Ryzhov Cc: Dan Gora v3: * reference

[PATCH v3 2/2] doc: announce KNI deprecation

Announce the KNI kernel module move to out of dpdk repo and announce long term plan to deprecate the KNI. Signed-off-by: Ferruh Yigit --- Cc: Olivier Matz Olivier Matz Cc: David Marchand David Marchand Cc: Stephen Hemminger Stephen Hemminger Cc: Elad Nachman Cc: Igor Ryzhov Cc: Dan Gora Da

Re: [PATCH v3] kni: allow configuring the kni thread granularity

Hi Ferruh, > As I tested both with KNI sample app and KNI PMD, change looks good, > practically we can't change the scheduler delay with existing code but > this patch enables it and lets configure performance/CPU usage trade of. > > Only possible change is to remove 'RTE_KNI_PREEMPT_DEFAULT' ma

[PATCH] doc/power: add info on scale mode reaction time

When using PMD Power Management, scale mode reacts slower than monitor mode and pause mode. Add note in user guide to this effect. Signed-off-by: David Hunt --- doc/guides/prog_guide/power_man.rst | 4 +++- doc/guides/sample_app_ug/l3_forward_power_man.rst | 5 +++-- 2 files change

Re: [PATCH v3] app/testpmd: fix hex string parser input length

On 11/24/2021 12:33 PM, Gregory Etelson wrote: Current hex string parser assumes input has even characters number. The parser fails input string with odd length. The patch parses hex strings with even and odd length. Parse result of an input with odd length will match result of even length input

Re: [dpdk-dev] [PATCH] doc: fix memif driver acronyms

On 11/19/2021 4:59 AM, jer...@marvell.com wrote: From: Jerin Jacob The commit d250589d5702 ("net/memif: replace master/slave arguments") replaced master/slave terms to server/client terms. Fix the documentation to reflect the same. Fixes: d250589d5702 ("net/memif: replace master/slave argument

RE: [PATCH v3] ethdev: deprecate header fields and metadata flow actions

Hi, Ori > -Original Message- > From: Ori Kam > Sent: Wednesday, November 24, 2021 18:21 > To: Slava Ovsiienko ; dev@dpdk.org > Cc: ferruh.yi...@intel.com; NBU-Contact-Thomas Monjalon (EXTERNAL) > > Subject: RE: [PATCH v3] ethdev: deprecate header fields and metadata flow > actions > > H

RE: [PATCH] net/mlx5: fix shared Rx queue segment configuration match

Hi, > -Original Message- > From: Slava Ovsiienko > Sent: Wednesday, November 24, 2021 4:33 PM > To: dev@dpdk.org > Cc: ferruh.yi...@intel.com; Xueming(Steven) Li ; > Raslan Darawsheh ; Matan Azrad > > Subject: [PATCH] net/mlx5: fix shared Rx queue segment configuration > match > > While

RE: [PATCH] net/mlx5: fix GENEVE and VXLAN-GPE item matching

Hi, > -Original Message- > From: Alexander Kozyrev > Sent: Wednesday, November 24, 2021 3:26 PM > To: dev@dpdk.org > Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava > Ovsiienko ; Matan Azrad ; > Gregory Etelson > Subject: [PATCH] net/mlx5: fix GENEVE and VXLAN-GPE item matching > > GENEVE

Re: [dpdk-dev] [PATCH] doc: remove repeated repeated words

On Fri, Nov 5, 2021 at 10:11 PM Thomas Monjalon wrote: > > The script devtools/check-dup-words.sh can detect some words > which are duplicated. Updated as: Some duplicate words were detected with a script. > > Fixes: fdec9301f52d ("doc: add flow classify guides") > Fixes: 4dc6d8e63c16 ("doc: add

RE: [PATCH v3] ethdev: deprecate header fields and metadata flow actions

Hi Slava, > -Original Message- > From: Viacheslav Ovsiienko > Sent: Wednesday, November 24, 2021 5:38 PM > To: dev@dpdk.org > Cc: ferruh.yi...@intel.com; NBU-Contact-Thomas Monjalon (EXTERNAL) > > Subject: [PATCH v3] ethdev: deprecate header fields and metadata flow actions > > The gen

Re: [kmods PATCH v2] windows/netuio: add Intel device ID

Hi, Can we merge this patch? Thank you! William On Tue, Oct 19, 2021 at 12:02 PM William Tu wrote: > > The patch adds three Intel device IDs, > I350 (0x1521), 82574L (0x10D3), and 82599 (0x10ED). > > Signed-off-by: William Tu > Acked-by: Dmitry Kozlyuk > Acked-by: Pallavi Kadam > --- > v2: > *

Re: [dpdk-dev] [PATCH] doc: remove repeated repeated words

On Sat, Nov 6, 2021 at 6:54 PM Thomas Monjalon wrote: > > 06/11/2021 10:01, David Marchand: > > On Fri, Nov 5, 2021 at 10:11 PM Thomas Monjalon wrote: > > > --- a/doc/guides/prog_guide/flow_classify_lib.rst > > > +++ b/doc/guides/prog_guide/flow_classify_lib.rst > > > @@ -366,7 +366,7 @@ Packet M

RE: [dpdk-dev] [RFC PATCH] ethdev: support priority based flow control

> -Original Message- > From: Jerin Jacob > Sent: Wednesday, November 24, 2021 12:48 PM > > On Wed, Nov 24, 2021 at 3:01 PM Ori Kam wrote: > > > > Hi Jerin, > > > > > -Original Message- > > > From: Jerin Jacob > > > Sent: Tuesday, November 23, 2021 12:58 PM > > > To: Ori Kam >

Re: [PATCH v3] ethdev: deprecate header fields and metadata flow actions

24/11/2021 16:37, Viacheslav Ovsiienko: > The generic RTE_FLOW_ACTION_TYPE_MODIFY_FIELD action was > introduced by [1]. This action provides an unified way > to perform various arithmetic and transfer operations over > packet network header fields and packet metadata. > > [1] commit 641dbe4fb053 (

Re: [EXT] Re: [dpdk-dev] [PATCH] doc: deprecation notice to remove octeontx2 drivers

24/11/2021 16:08, Thomas Monjalon: > 12/11/2021 12:24, Akhil Goyal: > > > On 11/9/2021 3:51 PM, jer...@marvell.com wrote: > > > > From: Jerin Jacob > > > > > > > > In the view of enabling unified driver for octeontx2(cn9k)/ > > > > octeontx3(cn10k), removing drivers/octeontx2 drivers and > > > > re

Re: [PATCH v1] doc: update release notes for 21.11

On Mon, Nov 22, 2021 at 6:05 PM Ajit Khaparde wrote: > > Fix grammar, spelling and formatting of DPDK 21.11 release notes. > > > > Signed-off-by: John McNamara > Acked-by: Ajit Khaparde Applied, thanks. -- David Marchand

[PATCH v3] ethdev: deprecate header fields and metadata flow actions

The generic RTE_FLOW_ACTION_TYPE_MODIFY_FIELD action was introduced by [1]. This action provides an unified way to perform various arithmetic and transfer operations over packet network header fields and packet metadata. [1] commit 641dbe4fb053 ("net/mlx5: support modify field flow action") On ot

Re: [PATCH V1] doc: add tested Intel platforms with Intel NICs

On Fri, Nov 19, 2021 at 11:35 AM Lin, Xueqin wrote: > > Add tested Intel platforms with Intel NICs to v21.11 release note. > > > > Signed-off-by: Yan Xia > Acked-by: Xueqin Lin Applied, thanks. -- David Marchand

Re: [PATCH v1] doc: update release notes for 21.11

On Mon, Nov 22, 2021 at 6:05 PM Mcnamara, John wrote: > > Thomas, > > Coul you also remove the boilerplate comments and the empty Known Issue > section from the final version. I'll handle it. Thanks John. -- David Marchand

Re: [dpdk-dev] [PATCH] doc: deprecation notice to remove octeontx2 drivers

On Tue, Nov 9, 2021 at 4:55 PM wrote: > > From: Jerin Jacob > > In the view of enabling unified driver for octeontx2(cn9k)/ > octeontx3(cn10k), removing drivers/octeontx2 drivers and > replace with drivers/cnxk/ which supports both octeontx2(cn9k) > and octeontx3(cn10k) SoCs. > > This deprecation

Re: [EXT] Re: [dpdk-dev] [PATCH] doc: deprecation notice to remove octeontx2 drivers

12/11/2021 12:24, Akhil Goyal: > > On 11/9/2021 3:51 PM, jer...@marvell.com wrote: > > > From: Jerin Jacob > > > > > > In the view of enabling unified driver for octeontx2(cn9k)/ > > > octeontx3(cn10k), removing drivers/octeontx2 drivers and > > > replace with drivers/cnxk/ which supports both octe

Re: [PATCH] net/memif: allow stopping and closing device

On 11/18/2021 5:33 PM, Junxiao Shi wrote: Bugzilla ID: 888 Fixes: febc855b358e ("ethdev: forbid closing started device") Signed-off-by: Junxiao Shi Reviewed-by: Ferruh Yigit Applied to dpdk-next-net/main, thanks.

Re: [PATCH v2] ethdev: deprecate header fields and metadata flow actions

+Cc some maintainers 24/11/2021 09:06, Viacheslav Ovsiienko: > There generic RTE_FLOW_ACTION_TYPE_MODIFY_FIELD action was s/There/The/ > introduced by [1]. This action provides the unified way s/the/an/ > to perform various arithmetic and transfer operations over > packet network header fields

Re: [PATCH] net/memif: allow stopping and closing device

On 11/24/2021 2:01 PM, yoursunny wrote: Hi Junxiao, comment moved down, please don't top post, it makes following the discussion very hard in the archive. Yours, Junxiao On Wed, Nov 24, 2021, 06:02 Ferruh Yigit mailto:ferruh.yi...@intel.com>> wrote: On 11/18/2021 5:33 PM, Junxiao Shi wrot

[PATCH] net/mlx5: fix shared Rx queue segment configuration match

While joining the shared Rx queue to the existing queue group the queue configurations is checked to be the same as it was specified in the first group queue creation - all shared queues should be created with identical configurations. During the Rx queue creation the buffer split segment configur

Re: [PATCH v2] config: sort Meson options by categories

On Wed, Nov 24, 2021 at 01:56:42PM +0100, Thomas Monjalon wrote: > Options used to be sorted alphabetically. > It looks easier to read when major options are first, > then path tuning, libs options, and drivers options. > > Signed-off-by: Thomas Monjalon > --- > v2: add blank lines for readabilit

Re: [PATCH v2 1/2] doc: note KNI alternatives

On 11/24/2021 1:58 PM, Thomas Monjalon wrote: 23/11/2021 13:08, Ferruh Yigit: --- a/doc/guides/nics/tap.rst +++ b/doc/guides/nics/tap.rst @@ -1,6 +1,8 @@ .. SPDX-License-Identifier: BSD-3-Clause Copyright(c) 2016 Intel Corporation. +.. _TunTap_PMD: No need for such anchor at the be

Re: [PATCH v3] mbuf: fix dump of dynamic fields and flags

24/11/2021 13:57, Olivier Matz: > From: Alexander Bechikov > > The dump of dynamic fields and flags fails if the shm is already > allocated. Add a check to fix the issue. > > Cc: sta...@dpdk.org > Fixes: d4902ed31c63 ("mbuf: check shared memory before dumping dynamic space") > > Signed-off-by:

Re: [PATCH] net/memif: allow stopping and closing device

Hi Ferruh You have to rely on user to call stop before calling close/remove. This is mandated in ethdev library, as implemented in: febc855b358e ("ethdev: forbid closing started device") Yours, Junxiao On Wed, Nov 24, 2021, 06:02 Ferruh Yigit wrote: > On 11/18/2021 5:33 PM, Junxiao Shi wrote:

Re: [PATCH v2 2/2] doc: announce KNI deprecation

23/11/2021 13:08, Ferruh Yigit: > --- a/doc/guides/rel_notes/deprecation.rst > +++ b/doc/guides/rel_notes/deprecation.rst > +* kni: KNI kernel module will be moved to `dpdk-kmods > `_ > + repository by the `DPDK technical board decision > +

Re: [PATCH v2 1/2] doc: note KNI alternatives

23/11/2021 13:08, Ferruh Yigit: > --- a/doc/guides/nics/tap.rst > +++ b/doc/guides/nics/tap.rst > @@ -1,6 +1,8 @@ > .. SPDX-License-Identifier: BSD-3-Clause > Copyright(c) 2016 Intel Corporation. > > +.. _TunTap_PMD: No need for such anchor at the beginning of a doc. You can reference a d

Re: [PATCH v2] examples/ptpclient: fix delay request message

On Mon, Nov 22, 2021 at 8:31 AM wrote: > The size of delay request message sent out by the DPDK > ptpclient application was observed to have extra length > than expected. Due to this, bad messages were observed > on the master side and delay response was not received. > This patch fixes this bug.

Re: [PATCH v2] kni: restrict bifurcated device support

23/11/2021 17:46, Ferruh Yigit: > To enable bifurcated device support, rtnl_lock is released before calling > userspace callbacks and asynchronous requests are enabled. > > But these changes caused more issues, like bug #809, #816. To reduce the > scope of the problems, the bifurcated device suppo

RE: [PATCH] net/mlx5: fix GENEVE and VXLAN-GPE item matching

> -Original Message- > From: Alexander Kozyrev > Sent: Wednesday, November 24, 2021 15:26 > To: dev@dpdk.org > Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava > Ovsiienko ; Matan Azrad ; > Gregory Etelson > Subject: [PATCH] net/mlx5: fix GENEVE and VXLAN-GPE item matching > > GENEVE and VX

Re: [PATCH v2] examples/ptpclient: fix delay request message

On Tue, Nov 23, 2021 at 6:40 PM Nipun Gupta wrote: > > Acked-by: Nipun Gupta Please don't top post. Can you also be careful with your ack tag? It is missing a space between name and mail address. > > > -Original Message- > > From: vanshika.shu...@nxp.com > > Sent: 22 November 2021 13:01

Re: [PATCH v4] bus: fix device iterator match from arguments

On Wed, Nov 24, 2021 at 2:00 PM Xueming(Steven) Li wrote: > On Wed, 2021-11-24 at 13:45 +0100, Olivier Matz wrote: > > From: Xueming Li > > > > Device iterator RTE_DEV_FOREACH() failed to return devices from > > classifier like "class=vdpa", because matching name from empty kvargs > > returns no

[PATCH] net/mlx5: fix GENEVE and VXLAN-GPE item matching

GENEVE and VXLAN-GPE item matching is done similarly to GRE matching. Users can skip the specification of the protocol type and expect that this type is deducted from the inner header type automatically. But the inner header type may not be specified in order to match all the protocol types. In thi

Re: [PATCH] bus/auxiliary: fix device iteration

On Mon, Nov 22, 2021 at 10:43 AM Xueming(Steven) Li wrote: > On Mon, 2021-11-22 at 10:25 +0100, David Marchand wrote: > > rte_kvargs_parse() expects a NULL terminated keys array. > > > > Fixes: 1afce3086cf4 ("bus/auxiliary: introduce auxiliary bus") > > > > Signed-off-by: David Marchand > Acked b

Re: [PATCH v2] doc: add examples.dox path to STRIP_FROM_PATH

On Tue, Nov 23, 2021 at 4:17 PM Luca Boccassi wrote: > > On Tue, 2021-11-23 at 15:48 +0100, Timothy Redaelli wrote: > > examples.dox is built inside builddir/doc/api and so doxygen generates > > some dir_HASH.html that includes the builddir name and this may prevent > > DPDK documentation to be co

RE: [EXT] Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

Regards Sunil Kumar Kori >-Original Message- >From: Sunil Kumar Kori >Sent: Wednesday, November 24, 2021 6:17 PM >To: Thomas Monjalon >Cc: Jerin Jacob Kollanukkaran ; dev@dpdk.org; >nikhil@intel.com; Pavan Nikhilesh Bhagavatula >; hemant.agra...@nxp.com; >nipun.gu...@nxp.com; harry.v

RE: [PATCH] net/mlx5: fix GRE item matching

Hi, > -Original Message- > From: Alexander Kozyrev > Sent: Wednesday, November 24, 2021 6:33 AM > To: dev@dpdk.org > Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava > Ovsiienko ; Matan Azrad ; > Gregory Etelson > Subject: [PATCH] net/mlx5: fix GRE item matching > > GRE protocol type is i

RE: [PATCH v3 0/2] net/mlx5: fix indirect RSS reference counting

Hi, > -Original Message- > From: Dmitry Kozlyuk > Sent: Wednesday, November 24, 2021 11:40 AM > To: dev@dpdk.org > Cc: Raslan Darawsheh > Subject: [PATCH v3 0/2] net/mlx5: fix indirect RSS reference counting > > v3: > * No changes, slipped when sending v2. > v2: > * Fix build with G

[PATCH v9 1/1] doc: add eventdevs feature matrices

From: Sunil Kumar Kori Patch adds feature matrices for event dev, Rx/Tx, Crypto and Timer adapters. Along with, .ini files are added for all supported PMDs. Signed-off-by: Sunil Kumar Kori --- v9: - Squah doc patches into a single patch. - Add .ini file for dlb2 driver also - Fix minor revi

Re: [PATCH] examples/power: revert "add wakeup log"

24/11/2021 13:17, Pattan, Reshma: > > From: David Hunt > > This reverts commit 931e3a994597acb9ffa827376206e8aa324476b7. > > > > Wakeup happens quite often (several hundred times a second) in l3fwd-power > > example app in PMD power management mode, so this message is appearing > > too often to b

Re: [PATCH v4] bus: fix device iterator match from arguments

On Wed, 2021-11-24 at 13:45 +0100, Olivier Matz wrote: > From: Xueming Li > > Device iterator RTE_DEV_FOREACH() failed to return devices from > classifier like "class=vdpa", because matching name from empty kvargs > returns no result. If device name not specified in kvargs, the function > should

Minutes of Technical Board Meeting, 2021-11-17

Members Attending - - Aaron - Bruce - Ferruh - Honnappa - Jerin - Kevin - Konstantin - Maxime - Olivier (Chair) - Stephen - Thomas NOTE: The technical board meetings every second Wednesday at https://meet.jit.si/DPDK at 3 pm UTC. Meetings are public, and DPDK community members are

[PATCH v3] mbuf: fix dump of dynamic fields and flags

From: Alexander Bechikov The dump of dynamic fields and flags fails if the shm is already allocated. Add a check to fix the issue. Cc: sta...@dpdk.org Fixes: d4902ed31c63 ("mbuf: check shared memory before dumping dynamic space") Signed-off-by: Alexander Bechikov Acked-by: Olivier Matz --- Hi

[PATCH v2] config: sort Meson options by categories

Options used to be sorted alphabetically. It looks easier to read when major options are first, then path tuning, libs options, and drivers options. Signed-off-by: Thomas Monjalon --- v2: add blank lines for readability --- meson_options.txt | 99 ++-

RE: [EXT] Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

Regards Sunil Kumar Kori >-Original Message- >From: Thomas Monjalon >Sent: Wednesday, November 24, 2021 6:12 PM >To: Sunil Kumar Kori >Cc: Jerin Jacob Kollanukkaran ; dev@dpdk.org; >nikhil@intel.com; Pavan Nikhilesh Bhagavatula >; hemant.agra...@nxp.com; >nipun.gu...@nxp.com; harr

[PATCH v4] bus: fix device iterator match from arguments

From: Xueming Li Device iterator RTE_DEV_FOREACH() failed to return devices from classifier like "class=vdpa", because matching name from empty kvargs returns no result. If device name not specified in kvargs, the function should iterate all devices. This patch allows empty devargs or devargs wi

Re: [EXT] Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

24/11/2021 13:35, Sunil Kumar Kori: > Hi Thomas, > > Just wanted to confirm that all documentation patches should go as single > patch or I can send it as a series of patches. > I am ready with dlb2 changes also. Once you confirm I will send patches > accordingly. Please confirm. There is no s

Re: [dpdk-dev] [PATCH] config/x86: add support for AMD platform

Ping techboard for comments 18/11/2021 15:05, Thomas Monjalon: > 18/11/2021 14:52, Bruce Richardson: > > On Thu, Nov 18, 2021 at 01:25:38PM +0100, Thomas Monjalon wrote: > > > I request a techboard decision for this patch. > > > > > > > > > 02/11/2021 20:04, Thomas Monjalon: > > > > 02/11/2021 1

RE: [EXT] Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

Hi Thomas, Just wanted to confirm that all documentation patches should go as single patch or I can send it as a series of patches. I am ready with dlb2 changes also. Once you confirm I will send patches accordingly. Please confirm. Regards Sunil Kumar Kori >-Original Message- >From:

[PATCH v3] app/testpmd: fix hex string parser input length

Current hex string parser assumes input has even characters number. The parser fails input string with odd length. The patch parses hex strings with even and odd length. Parse result of an input with odd length will match result of even length input, that has `0` as MSB, following by the original

Re: [dpdk-dev] [PATCH] doc: cleanup flow mark Rx offload deprecation notice

On 11/22/2021 4:10 AM, Jerin Jacob wrote: On Fri, Nov 19, 2021 at 11:23 PM Ferruh Yigit wrote: On 11/5/2021 4:03 PM, Ferruh Yigit wrote: On 11/2/2021 12:50 PM, Andrew Rybchenko wrote: The problem is solved using Rx metadata delivery negotiation API [1]. [1] commit f6d8a6d3fad7 ("ethdev: neg

RE: [PATCH] examples/power: revert "add wakeup log"

> -Original Message- > From: David Hunt > This reverts commit 931e3a994597acb9ffa827376206e8aa324476b7. > > Wakeup happens quite often (several hundred times a second) in l3fwd-power > example app in PMD power management mode, so this message is appearing > too often to be useful. This

Re: [PATCH v3] bus: fix device iterator match from arguments

On Wed, Nov 24, 2021 at 11:43:29AM +, Xueming(Steven) Li wrote: > On Wed, 2021-11-24 at 19:30 +0800, Xueming Li wrote: > > On Wed, 2021-11-24 at 12:02 +0100, Olivier Matz wrote: > > > From: Xueming Li > > > > > > Device iterator RTE_DEV_FOREACH() failed to return devices from > > > classifier

[PATCH v1 1/1] doc: update eventdev feature matrix for dlb2

From: Sunil Kumar Kori Patch updates eventdev and adapters feature matrices for dlb2 driver. Signed-off-by: Sunil Kumar Kori --- doc/guides/eventdevs/features/default.ini | 46 +++ doc/guides/eventdevs/features/dlb2.ini| 21 +++ 2 files changed, 44 insertions(+)

Re: [pull-request] next-crypto 21.11 rc4

24/11/2021 09:35, Akhil Goyal: > http://dpdk.org/git/next/dpdk-next-crypto Pulled, thanks.

Re: [PATCH v3] bus: fix device iterator match from arguments

On Wed, 2021-11-24 at 19:30 +0800, Xueming Li wrote: > On Wed, 2021-11-24 at 12:02 +0100, Olivier Matz wrote: > > From: Xueming Li > > > > Device iterator RTE_DEV_FOREACH() failed to return devices from > > classifier like "class=vdpa", because matching name from empty kvargs > > returns no resul

Re: [PATCH v3] bus: fix device iterator match from arguments

On Wed, 2021-11-24 at 12:02 +0100, Olivier Matz wrote: > From: Xueming Li > > Device iterator RTE_DEV_FOREACH() failed to return devices from > classifier like "class=vdpa", because matching name from empty kvargs > returns no result. If device name not specified in kvargs, the function > should

Re: [EXT] Re: [PATCH v8 10/10] devtools: check event device doc tables

24/11/2021 12:16, Sunil Kumar Kori: > Hi Thomas, > > It will take some time to handle all the comments for this patch. So I would > request you to defer this patch for next release and take other patches in > series. > Also I will send one patch to add feature matrix for dlb2 platform and fix

Re: [PATCH v2] kvargs: fix device iterator match from arguments

On Wed, 2021-11-24 at 12:07 +0100, Olivier Matz wrote: > Hi Xueming, > > On Wed, Nov 24, 2021 at 06:17:40PM +0800, Xueming Li wrote: > > Device iterator RTE_DEV_FOREACH() failed to return devices from > > classifier like "class=vdpa", because matching name from empty kvargs > > returns no result.

RE: [EXT] Re: [PATCH v8 10/10] devtools: check event device doc tables

Hi Thomas, It will take some time to handle all the comments for this patch. So I would request you to defer this patch for next release and take other patches in series. Also I will send one patch to add feature matrix for dlb2 platform and fix minor review comments given by You. Regards Su

RE: [EXT] Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

Regards Sunil Kumar Kori >-Original Message- >From: Thomas Monjalon >Sent: Wednesday, November 24, 2021 4:18 PM >To: Sunil Kumar Kori >Cc: Jerin Jacob Kollanukkaran ; nikhil@intel.com; >Pavan Nikhilesh Bhagavatula ; >hemant.agra...@nxp.com; nipun.gu...@nxp.com; >harry.van.haa...@i

Re: [PATCH v2] kvargs: fix device iterator match from arguments

Hi Xueming, On Wed, Nov 24, 2021 at 06:17:40PM +0800, Xueming Li wrote: > Device iterator RTE_DEV_FOREACH() failed to return devices from > classifier like "class=vdpa", because matching name from empty kvargs > returns no result. If device name not specified in kvargs, the function > should itera

Re: [PATCH v2] gpu/cuda: set rte_errno

24/11/2021 19:34, eagost...@nvidia.com: > From: Elena Agostini > > Set correct rte_errno variable in gpu/cuda and return > -rte_errno in case of error. > > rte_errno values are compliant with the gpudev library > documentation. > > Fixes: 1306a73b1958 ("gpu/cuda: introduce CUDA driver") > > Si

[PATCH v3] bus: fix device iterator match from arguments

From: Xueming Li Device iterator RTE_DEV_FOREACH() failed to return devices from classifier like "class=vdpa", because matching name from empty kvargs returns no result. If device name not specified in kvargs, the function should iterate all devices. This patch allows empty devargs or devargs wi

Re: [PATCH] net/memif: allow stopping and closing device

On 11/18/2021 5:33 PM, Junxiao Shi wrote: Bugzilla ID: 888 Fixes: febc855b358e ("ethdev: forbid closing started device") Signed-off-by: Junxiao Shi Thanks Junxiao, +1 to this fix, cc'ed memif maintainer Jakub. --- drivers/net/memif/rte_eth_memif.c | 11 --- 1 file changed, 8 inse

Re: [PATCH v8 10/10] devtools: check event device doc tables

23/11/2021 12:07, sk...@marvell.com: > --- a/devtools/check-doc-vs-code.sh > +++ b/devtools/check-doc-vs-code.sh > +all_event_drivers() > +{ > + find $rootdir/drivers/event -mindepth 1 -maxdepth 1 -type d | > + sed 's,.*/,,' | > + sort > +} > + > +check_event_dev() # > +{ > + code=

Re: [dpdk-dev] [RFC PATCH] ethdev: support priority based flow control

On Wed, Nov 24, 2021 at 3:01 PM Ori Kam wrote: > > Hi Jerin, > > > -Original Message- > > From: Jerin Jacob > > Sent: Tuesday, November 23, 2021 12:58 PM > > To: Ori Kam > > > > On Sun, Nov 21, 2021 at 3:20 PM Ori Kam wrote: > > > > > > Hi Jerin, > > > > Hi Ori, > > > > > > > > Sorry fo

Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

23/11/2021 12:07, sk...@marvell.com: > +[Scheduling Features] > +queue_qos= > +event_qos= > +distributed_sched= > +queue_all_types = > +burst_mode = > +implicit_release_disable = > +nonseq_mode = > +runtime_port_link

Re: [PATCH] doc: remove custom kernel patch link in ENA's guide

On 11/23/2021 7:23 AM, Michal Krawczyk wrote: ENAv2 device requires write combining support which isn't supported by the upstream vfio-pci. amzn-driver repository provided non-upstream patch to enable this feature and it was linked directly by the ENA PMD guide. To avoid custom kernel patch link

Re: [EXT] Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

24/11/2021 11:30, Sunil Kumar Kori: > From: Thomas Monjalon > >23/11/2021 12:07, sk...@marvell.com: > >> .gitignore | 5 ++ > >> doc/guides/conf.py | 25 ++ > >> doc/guides/eventdevs/features/cnxk.ini | 14 ++ > >> doc/gu

RE: [EXT] Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

>-Original Message- >From: Thomas Monjalon >Sent: Wednesday, November 24, 2021 3:34 PM >To: Jerin Jacob Kollanukkaran ; Sunil Kumar Kori > >Cc: nikhil@intel.com; Pavan Nikhilesh Bhagavatula >; hemant.agra...@nxp.com; >nipun.gu...@nxp.com; harry.van.haa...@intel.com; >mattias.ronnb...@e

[PATCH v2] gpu/cuda: set rte_errno

From: Elena Agostini Set correct rte_errno variable in gpu/cuda and return -rte_errno in case of error. rte_errno values are compliant with the gpudev library documentation. Fixes: 1306a73b1958 ("gpu/cuda: introduce CUDA driver") Signed-off-by: Elena Agostini --- drivers/gpu/cuda/cuda.c | 18

Re: [PATCH] kvargs: fix device iterator match from arguments

On Tue, 2021-11-23 at 21:02 +0100, Olivier Matz wrote: > Hello Xueming, > > On Tue, Nov 23, 2021 at 12:49:32PM +, Xueming(Steven) Li wrote: > > On Tue, 2021-11-23 at 13:31 +0100, Olivier Matz wrote: > > > On Tue, Nov 23, 2021 at 11:25:22AM +, Xueming(Steven) Li wrote: > > > > On Tue, 2021-

[PATCH v2] kvargs: fix device iterator match from arguments

Device iterator RTE_DEV_FOREACH() failed to return devices from classifier like "class=vdpa", because matching name from empty kvargs returns no result. If device name not specified in kvargs, the function should iterate all devices. This patch allows empty devargs or devargs without name specifie

RE: [PATCH] net/mlx5: fix GRE item matching

> -Original Message- > From: Alexander Kozyrev > Sent: Wednesday, November 24, 2021 6:33 > To: dev@dpdk.org > Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava > Ovsiienko ; Matan Azrad ; > Gregory Etelson > Subject: [PATCH] net/mlx5: fix GRE item matching > > GRE protocol type is implicitly

Re: [PATCH v8 01/10] doc: add skeleton for eventdevs feature matrices

23/11/2021 12:07, sk...@marvell.com: > .gitignore | 5 ++ > doc/guides/conf.py | 25 ++ > doc/guides/eventdevs/features/cnxk.ini | 14 ++ > doc/guides/eventdevs/features/default.ini | 51 + > doc/guid

[PATCH v3 1/2] net/mlx5: fix indirect RSS creation when port is stopped

mlx5_ind_table_obj_setup() was incrementing RxQ reference counters even when the port was stopped, which prevented RxQ release and triggered an internal assertion. Only increment reference counter when the port is started. Fixes: ec4e11d41d12 ("net/mlx5: preserve indirect actions on restart") Sig

[PATCH v3 2/2] net/mlx5: fix RxQ reference counting for indirect RSS

mlx5_ind_table_obj_modify() was not changing the reference counters of neither the new set of RxQs, nor the old set of RxQs. On the other hand, creation of the RSS incremented the RxQ refcnt. If an RxQ was present in both the initial and the modified set, its reference counter was incremented one e

[PATCH v3 0/2] net/mlx5: fix indirect RSS reference counting

v3: * No changes, slipped when sending v2. v2: * Fix build with GCc on RHEL7 (CI). * Fix RxQ leak when an indirect RSS had been created before the start and then modified after the start (Matan). Dmitry Kozlyuk (2): net/mlx5: fix indirect RSS creation when port is stopped net/mlx5: f

[PATCH v2 1/2] net/mlx5: fix indirect RSS creation when port is stopped

mlx5_ind_table_obj_setup() was incrementing RxQ reference counters even when the port was stopped, which prevented RxQ release and triggered an internal assertion. Only increment reference counter when the port is started. Fixes: ec4e11d41d12 ("net/mlx5: preserve indirect actions on restart") Sig

[PATCH 2/2] net/mlx5: fix RxQ reference counting for indirect RSS

mlx5_ind_table_obj_modify() was not changing the reference counters of neither the new set of RxQs, nor the old set of RxQs. On the other hand, creation of the RSS incremented the RxQ refcnt. If an RxQ was present in both the initial and the modified set, its reference counter was incremented one e

[PATCH v2 0/2] net/mlx5: fix indirect RSS reference counting

v2: * Fix build with GCc on RHEL7 (CI). * Fix RxQ leak when an indirect RSS had been created before the start and then modified after the start (Matan). Dmitry Kozlyuk (2): net/mlx5: fix indirect RSS creation when port is stopped net/mlx5: fix RxQ reference counting for indirect RSS

[PATCH 1/2] net/mlx5: fix indirect RSS creation when port is stopped

mlx5_ind_table_obj_setup() was incrementing RxQ reference counters even when the port was stopped, which prevented RxQ release and triggered an internal assertion. Only increment reference counter when the port is started. Fixes: ec4e11d41d12 ("net/mlx5: preserve indirect actions on restart") Sig

[PATCH 0/2] next/mlx5: fix indirect RSS reference counting

Dmitry Kozlyuk (2): net/mlx5: fix indirect RSS creation when port is stopped net/mlx5: fix RxQ reference counting for indirect RSS drivers/net/mlx5/mlx5_rxq.c | 56 + 1 file changed, 32 insertions(+), 24 deletions(-) -- 2.25.1

  1   2   >