[dpdk-dev] [PATCH RFC 2/2] net/ice: add support of VXLAN in switch filter

Add support of ipv4/ipv6 vxlan any pattern and outer/inner ipv6 vxlan pattern in switch filter. Pattern ipv4/ipv6_vxlan_any supports no input set. +-+---+ | Pattern | Input Set | +--

[dpdk-dev] [PATCH RFC 1/2] net/ice/base: add VXLAN support for switch filter

Add ipv4/ipv6 vxlan any support in switch filter. Add inner ipv6 and outer ipv6 support of vxlan protocol in switch filter. Signed-off-by: Yuying Zhang --- drivers/net/ice/base/ice_protocol_type.h | 6 + drivers/net/ice/base/ice_switch.c| 204 ++- drivers/net/ice/ba

Re: [dpdk-dev] [PATCH v4] net/bnxt: fix crash after port stop/start

On Mon, Aug 23, 2021 at 8:49 AM Somnath Kotur wrote: > > On chips like Thor, port stop/start sequence could result in a crash > in the application. This is because of false detection of a bad > opaque in the Rx completion and the subsequent kicking-in of the ring > reset code to recover from the c

Re: [dpdk-dev] [RFC 11/15] eventdev: reserve fields in timer object

Hi Erik, >Hi Pavan, > >One comment in-line: > >> -Original Message- >> From: pbhagavat...@marvell.com >> Sent: Monday, August 23, 2021 2:40 PM >> To: jer...@marvell.com; Carrillo, Erik G >> Cc: Ananyev, Konstantin ; >dev@dpdk.org; >> Pavan Nikhilesh >> Subject: [dpdk-dev] [RFC 11/15] ev

Re: [dpdk-dev] i40evf: potential segfault

> -Original Message- > From: dev On Behalf Of Stefan Baranoff > Sent: Saturday, August 21, 2021 2:18 AM > To: dev@dpdk.org > Subject: [dpdk-dev] i40evf: potential segfault > > Hi all! > > I was chasing a potential segfault and it appears, if I'm reading this driver > correctly, that in

Re: [dpdk-dev] [PATCH v2 8/8] examples: remove unnecessary include of atomic

> Hi Joyce, > > > -Original Message- > > From: Joyce Kong > > Sent: Monday, August 23, 2021 1:50 PM > > To: Chas Williams ; Min Hu (Connor) > > ; Ananyev, Konstantin > > ; Nicolau, Radu > > ; Akhil Goyal ; Doherty, > > Declan ; Sunil Kumar Kori > > ; Pavan Nikhilesh ; > > Richardson, Br

[dpdk-dev] [Bug 792] The dev_uev_parse function has a stack overflow bug.

https://bugs.dpdk.org/show_bug.cgi?id=792 Bug ID: 792 Summary: The dev_uev_parse function has a stack overflow bug. Product: DPDK Version: 21.08 Hardware: All OS: All Status: UNCONFIRMED Severity: normal

Re: [dpdk-dev] [RFC V1] ethdev: fix the issue that dev uninit may be called twice

Hi, Singh, Aman Deep Sorry, I missed your review. Thank you for your review.😁 在 2021/8/18 17:47, Singh, Aman Deep 写道: Hi Huison, On 8/2/2021 6:16 PM, Huisong Li wrote: Ethernet devices in DPDK can be released by rte_eth_dev_close() and rte_dev_remove(). However, these two APIs do not have ex

Re: [dpdk-dev] [PATCH] doc: abstract the behaviour of rte_ctrl_thread_create

> > > > > > 30/07/2021 23:44, Honnappa Nagarahalli: > > > > The current expected behaviour of the function > > > > rte_ctrl_thread_create is rigid which makes the implementation of the > function complex. > > > > Make the expected behaviour abstract to allow for simplified > > > > implementation.

Re: [dpdk-dev] [RFC 11/15] eventdev: reserve fields in timer object

Hi Pavan, One comment in-line: > -Original Message- > From: pbhagavat...@marvell.com > Sent: Monday, August 23, 2021 2:40 PM > To: jer...@marvell.com; Carrillo, Erik G > Cc: Ananyev, Konstantin ; dev@dpdk.org; > Pavan Nikhilesh > Subject: [dpdk-dev] [RFC 11/15] eventdev: reserve fields

Re: [dpdk-dev] [PATCH v14 9/9] Add unit tests for thread API

2021-08-19 14:31 (UTC-0700), Narcisa Ana Maria Vasile: > From: Narcisa Vasile > > As a new API for threading is introduced, > a set of unit tests have been added to test the new interface. There are three substantial issues with this tests: 1. They use pthread API to verify the results, but the

[dpdk-dev] [PATCH v3 3/3] common/octeontx2: enable build only on 64-bit Linux

From: Pavan Nikhilesh Due to Linux kernel dependency, only enable build for 64-bit Linux. Signed-off-by: Pavan Nikhilesh Acked-by: Jerin Jacob --- drivers/common/octeontx2/meson.build | 10 ++ drivers/event/octeontx2/meson.build | 4 ++-- drivers/mempool/octeontx2/meson.build | 9

[dpdk-dev] [PATCH v3 2/3] common/octeontx: enable build only on 64-bit Linux

From: Pavan Nikhilesh Due to Linux kernel dependency, only enable build for 64-bit Linux. Signed-off-by: Pavan Nikhilesh Acked-by: Jerin Jacob --- drivers/common/octeontx/meson.build | 4 ++-- drivers/compress/octeontx/meson.build | 6 ++ drivers/crypto/octeontx/meson.build | 7 ++

[dpdk-dev] [PATCH v3 1/3] net/thunderx: enable build only on 64-bit Linux

From: Pavan Nikhilesh Due to Linux kernel dependency, only enable build for 64-bit Linux. Signed-off-by: Pavan Nikhilesh Acked-by: Jerin Jacob --- drivers/net/thunderx/meson.build | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/thunderx/meson.build b/driver

[dpdk-dev] [PATCH v4] net/bnxt: fix crash after port stop/start

On chips like Thor, port stop/start sequence could result in a crash in the application. This is because of false detection of a bad opaque in the Rx completion and the subsequent kicking-in of the ring reset code to recover from the condition. The root cause being that the port stop/start would re

[dpdk-dev] [PATCH v4] net/bnxt: fix crash after port stop/start

On chips like Thor, port stop/start sequence could result in a crash in the application. This is because of false detection of a bad opaque in the Rx completion and the subsequent kicking-in of the ring reset code to recover from the condition. The root cause being that the port stop/start would re

[dpdk-dev] [RFC 15/15] eventdev: promote trace variables to stable

From: Pavan Nikhilesh Promote rte_trace global variables to stable i.e. remove them from experimental section of version map. Signed-off-by: Pavan Nikhilesh --- lib/eventdev/version.map | 78 ++-- 1 file changed, 35 insertions(+), 43 deletions(-) diff --git

[dpdk-dev] [RFC 14/15] eventdev: make trace APIs internal

From: Pavan Nikhilesh Slowpath trace APIs are only used in rte_eventdev.c so make them as internal. Signed-off-by: Pavan Nikhilesh --- lib/eventdev/{rte_eventdev_trace.h => eventdev_trace.h} | 0 lib/eventdev/eventdev_trace_points.c| 2 +- lib/eventdev/meson.build

[dpdk-dev] [RFC 13/15] eventdev: promote event vector API to stable

From: Pavan Nikhilesh Promote event vector configuration APIs to stable. Signed-off-by: Pavan Nikhilesh --- lib/eventdev/rte_event_eth_rx_adapter.h | 2 -- lib/eventdev/rte_eventdev.h | 1 - lib/eventdev/version.map| 6 +++--- 3 files changed, 3 insertions(+), 6 del

[dpdk-dev] [RFC 12/15] eventdev: move timer adapters memory to hugepage

From: Pavan Nikhilesh Move memory used by timer adapters to hugepage. Allocate memory on the first adapter create or lookup to address both primary and secondary process usecases. Signed-off-by: Pavan Nikhilesh --- lib/eventdev/rte_event_timer_adapter.c | 24 +++- 1 file ch

[dpdk-dev] [RFC 11/15] eventdev: reserve fields in timer object

From: Pavan Nikhilesh Reserve fields in rte_event_timer data structure to address future use cases. Also, remove volatile from rte_event_timer. Signed-off-by: Pavan Nikhilesh --- lib/eventdev/rte_event_timer_adapter.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/lib/e

[dpdk-dev] [RFC 10/15] eventdev: remove rte prefix for internal structs

From: Pavan Nikhilesh Remove rte_ prefix from rte_eth_event_enqueue_buffer, rte_event_eth_rx_adapter and rte_event_crypto_adapter as they are only used in rte_event_eth_rx_adapter.c and rte_event_crypto_adapter.c Signed-off-by: Pavan Nikhilesh --- lib/eventdev/rte_event_crypto_adapter.c | 66

[dpdk-dev] [RFC 09/15] eventdev: hide timer adapter pmd file

From: Pavan Nikhilesh Hide rte_event_timer_adapter_pmd.h file as it is an internal file. Remove rte_ prefix from rte_event_timer_adapter_ops structure. Signed-off-by: Pavan Nikhilesh --- drivers/event/cnxk/cnxk_tim_evdev.c | 5 ++-- drivers/event/cnxk/cnxk_tim_evdev.h | 2

[dpdk-dev] [RFC 08/15] eventdev: hide event device related structures

From: Pavan Nikhilesh Move rte_eventdev, rte_eventdev_data structures to eventdev_pmd.h. Signed-off-by: Pavan Nikhilesh --- drivers/event/dlb2/dlb2_inline_fns.h | 2 + drivers/event/dsw/dsw_evdev.h | 2 + drivers/event/octeontx/timvf_worker.h | 2 + drivers/net/octeontx/octeo

[dpdk-dev] [RFC 06/15] eventdev: use new API for inline functions

From: Pavan Nikhilesh Use new driver interface for the fastpath enqueue/dequeue inline functions. Signed-off-by: Pavan Nikhilesh --- lib/eventdev/rte_event_crypto_adapter.h | 13 +- lib/eventdev/rte_event_eth_tx_adapter.h | 22 ++--- lib/eventdev/rte_eventdev.h | 61 +++

[dpdk-dev] [RFC 05/15] eventdev: add helper functions for new driver API

From: Pavan Nikhilesh Add helper functions and macros to help drivers to transition to new fastpath interface. Signed-off-by: Pavan Nikhilesh --- lib/eventdev/eventdev_pmd.h | 396 lib/eventdev/rte_eventdev.c | 174 lib/eventdev/version.map

[dpdk-dev] [RFC 04/15] eventdev: move inline APIs into separate structure

From: Pavan Nikhilesh Move fastpath inline function pointers from rte_eventdev into a separate structure accessed via a flat array. The intension is to make rte_eventdev and related structures private to avoid future API/ABI breakages.` Signed-off-by: Pavan Nikhilesh --- lib/eventdev/eventdev_

[dpdk-dev] [RFC 03/15] eventdev: move eventdevs globals to hugepage mem

From: Pavan Nikhilesh Move the global struct eventdevs to hugepage memory, allocate memory on the first pmd allocate request for both primary and secondary process. Signed-off-by: Pavan Nikhilesh --- lib/eventdev/rte_eventdev.c | 30 -- 1 file changed, 20 insertions

[dpdk-dev] [RFC 02/15] eventdev: separate internal structures

From: Pavan Nikhilesh Create rte_eventdev_core.h and move all the internal data structures to this file. These structures are mostly used by drivers, but they need to be in the public header file as they are accessed by datapath inline functions for performance reasons. The accessibility of these

[dpdk-dev] [RFC 01/15] eventdev: make driver interface as internal

From: Pavan Nikhilesh Mark all the driver specific functions as internal, remove `rte` prefix from `struct rte_eventdev_ops`. Remove experimental tag from internal functions. Remove `eventdev_pmd.h` from non-internal header files. Signed-off-by: Pavan Nikhilesh --- drivers/event/cnxk/cn10k_eve

Re: [dpdk-dev] [PATCH v8] eal: remove sys/queue.h from public headers.

2021-08-23 13:03 (UTC+), William Tu: [...] > /* This macro permits both remove and free var within the loop safely.*/ > -#ifndef TAILQ_FOREACH_SAFE > -#define TAILQ_FOREACH_SAFE(var, head, field, tvar) \ > - for ((var) = TAILQ_FIRST((head)); \ > - (v

[dpdk-dev] [PATCH v2] maintainers: update for net mlx4/mlx5

For net/mlx4: - removing Shahaf Shuler - adding Viacheslav Ovsiienko For net/mlx5: - removing Shahaf Shuler Signed-off-by: Viacheslav Ovsiienko Acked-by: Matan Azrad --- MAINTAINERS | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS inde

[dpdk-dev] [PATCH v3] net/bnxt: fix crash after port stop/start

In bnxt_init_one_rx_ring(), reset rx_next_cons to 0, so that there is no scope for any false mismatch errors being flagged on traffic resumption. Typically, this would show up in a port stop/start sequence as then the HW would start with fresh values, but this driver internal variable would still

Re: [dpdk-dev] [PATCH] net/virtio: wait device ready in device reset

> -Original Message- > From: Andrew Rybchenko > Sent: Monday, August 23, 2021 5:57 PM > To: Xueming(Steven) Li > Cc: dev@dpdk.org; Maxime Coquelin ; Chenbo Xia > > Subject: Re: [dpdk-dev] [PATCH] net/virtio: wait device ready in device reset > > On 8/23/21 9:39 AM, Xueming Li wrote:

[dpdk-dev] [PATCH] ring: fix comment for __rte_ring_move_cons_head()

Change "enqueue" to "dequeue" because the __rte_ring_move_cons_head() function is updating the consumer head for dequeue. Fixes: 0dfc98c507b1 ("ring: separate out head index manipulation") Cc: bruce.richard...@intel.com Cc: sta...@dpdk.org Signed-off-by: Cian Ferriter --- lib/ring/rte_ring_c11_

Re: [dpdk-dev] [PATCH RESEND v4 2/2] drivers/raw: remove octeontx2-ep driver

On Mon, Aug 23, 2021 at 8:51 PM Radha Mohan Chintakuntla wrote: > > Removing the rawdev based octeontx2-ep driver as the dependent > common/octeontx2 will soon be going away. Moreover this driver is no > longer required as the net/octeontx_ep driver is sufficient. > > Signed-off-by: Radha Mohan Ch

Re: [dpdk-dev] 20.11.3 patches review and test

> -Original Message- > From: dev On Behalf Of luca.bocca...@gmail.com > Sent: Monday, August 9, 2021 4:44 PM > To: sta...@dpdk.org > Cc: dev@dpdk.org; Abhishek Marathe ; > Akhil Goyal ; Ali Alnubani ; > Walker, Benjamin ; David Christensen > ; Govindharajan, Hariprasad > ; Hemant Agrawal >

Re: [dpdk-dev] [PATCH 1/2] lib/eal: add amd epyc2 memcpy routine to eal

On Mon, Aug 23, 2021 at 2:14 PM Aman Kumar wrote: > > This patch provides rte_memcpy* calls optimized for > AMD EPYC Gen2 platforms. This option is disabled by > default and can be enabled by defining 'rte_memcpy_amdepyc2' Generic options should support all the architectures. Another more scalabl

[dpdk-dev] [PATCH RESEND v4 2/2] drivers/raw: remove octeontx2-ep driver

Removing the rawdev based octeontx2-ep driver as the dependent common/octeontx2 will soon be going away. Moreover this driver is no longer required as the net/octeontx_ep driver is sufficient. Signed-off-by: Radha Mohan Chintakuntla --- Changes from v3: Fixed patch application failure due to conf

[dpdk-dev] [PATCH RESEND v4 1/2] drivers/raw: remove octeontx2-dma driver

Removing the rawdev based octeontx2-dma driver as the dependent common/octeontx2 will be soon be going away. Also a new DMA driver will be coming in this place once the rte_dmadev library is in. Signed-off-by: Radha Mohan Chintakuntla --- Changes from v3: Fixed patch application failure due to co

Re: [dpdk-dev] [PATCH v2] doc: fix CI typo warnings

Ray Kinsella writes: > Fix documentation typos that are generating spurious CI warnings. > > Signed-off-by: Ray Kinsella > --- > v2: > * Fix some additional typos pointed out by Aaron Conole > > doc/guides/rel_notes/deprecation.rst | 4 ++-- > doc/guides/sample_app_ug/ioat.rst

[dpdk-dev] [PATCH v2] eal: remove the deprecated whitelist/blacklist and master/slave API's

New API's for these were added in 20.11 and the old API was retained but marked deprecated. Since 21.11 is the next LTS, it is time to remove the deprecated ones. Signed-off-by: Stephen Hemminger --- v2 remove the short 'w' option as well doc/guides/rel_notes/release_21_11.rst | 7 +++ l

[dpdk-dev] [PATCH v7 2/2] ethdev: use ext hdr for gtp psc item

This updates the gtp_psc item to use the net hdr definition of the gtp_psc to be based on RFC 38415-g30 Signed-off-by: Raslan Darawsheh --- app/test-pmd/cmdline_flow.c | 23 +++-- doc/guides/testpmd_app_ug/testpmd_funcs.rst | 2 +- drivers/net/iavf/iavf_fdir.c

[dpdk-dev] [PATCH v7 0/2] update gtp psc to use net hdr

This series add header definition for gtp_psc in the net library, and update the relevant rte_flow_item gtp_psc to use this new header definition. Raslan Darawsheh (2): net: add new ext hdr for gtp psc ethdev: use ext hdr for gtp psc item app/test-pmd/cmdline_flow.c | 23 +++-

[dpdk-dev] [PATCH v7 1/2] net: add new ext hdr for gtp psc

Define new rte header for gtp PDU session container based on RFC 38415-g30 Signed-off-by: Raslan Darawsheh --- lib/net/rte_gtp.h | 78 +++ 1 file changed, 78 insertions(+) diff --git a/lib/net/rte_gtp.h b/lib/net/rte_gtp.h index 6a6f9b238d..9f6deb9c7e

[dpdk-dev] [PATCH v2] eal: read data buffer on RTE_INTR_HANDLE_VFIO_REQ

We should call read() on RTE_INTR_HANDLE_VFIO_REQ event to confirm that event. Signed-off-by: Maciej Szwed --- lib/eal/linux/eal_interrupts.c | 11 +-- 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/lib/eal/linux/eal_interrupts.c b/lib/eal/linux/eal_interrupts.c index 22b3b7

Re: [dpdk-dev] [PATCH v16 7/9] dma/skeleton: add test cases

On Mon, Aug 23, 2021 at 11:31:32AM +0800, Chengwen Feng wrote: > Patch introduces dmadev unit testcase for validation against the > skeleton dmadev PMD implementation. > > Test cases are added along with the skeleton driver implementation. > It can be enabled by using vdev argument to any DPDK bin

[dpdk-dev] [PATCH] eal: read data buffer on RTE_INTR_HANDLE_VFIO_REQ

We should call read() on RTE_INTR_HANDLE_VFIO_REQ event to confirm that event. Signed-off-by: Maciej Szwed maciej.sz...@intel.com --- lib/eal/linux/eal_interrupts.c | 11 +-- 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/lib/eal/linux/eal_inte

[dpdk-dev] [PATCH v8] eal: remove sys/queue.h from public headers.

Currently there are some public headers that include 'sys/queue.h', which is not POSIX, but usually provided by the Linux/BSD system library. (Not in POSIX.1, POSIX.1-2001, or POSIX.1-2008. Present on the BSDs.) The file is missing on Windows. During the Windows build, DPDK uses a bundled copy, so

[dpdk-dev] [PATCH 1/1] net/mlx5: add stub functions to null drv ops

There are several functions implementation that queries the drv type to understand which fops function to use. In case the type is DV, the fucntion gets the concrete DV function and calls it. In case it’s not, the function returns an error. The current implementation is not flexible enough and wil

[dpdk-dev] [PATCH] crypto/openssl: fix CCM processing 0 length source

When given a source length 0 for CCM, the encryption and decryption functions did not call the EVP_ENCRYPTUPDATE/EVP_DECRYPTUPDATE functions with a src and dst, causing some FIPS validation failures for testcases with PLen=0: process_openssl_auth_encryption_ccm() line 1131: Process openssl auth en

Re: [dpdk-dev] [PATCH v4 2/2] drivers/raw: remove octeontx2-ep driver

On Mon, Aug 23, 2021 at 6:02 PM David Marchand wrote: > > On Mon, Aug 23, 2021 at 2:19 PM Radha Mohan wrote: > > I received this report that this patch failed to apply > > http://mails.dpdk.org/archives/test-report/2021-August/211611.html > > The two patches of this series are seen as two separat

Re: [dpdk-dev] [PATCH v1] net/ice: fix the reversed priority of DCF switch rule

Thomas Monjalon writes: > 13/08/2021 12:45, Ferruh Yigit: >> On 8/2/2021 8:25 AM, Wenjun Wu wrote: >> > This patch fixes the reversed priority of DCF switch rule. Priority 0 and 1 >> > are supported, and priority 0 should be the highest priority. >> > >> > Fixes: 2321e34c23b3 ("net/ice: support

Re: [dpdk-dev] [PATCH] build: fix install from arbitrary directory for meson 0.55

Good catch - thanks Dmitry. Ack-by: Nick Connolly

Re: [dpdk-dev] [PATCH v7] eal: remove sys/queue.h from public headers.

Hi Dmitry, Thanks for your feedback. On Thu, Aug 19, 2021 at 4:29 PM Dmitry Kozlyuk wrote: > > 2021-08-18 23:26 (UTC+), William Tu: > [...] > > diff --git a/lib/eal/linux/include/rte_os.h b/lib/eal/linux/include/rte_os.h > > index 1618b4df22..ce5b0aed52 100644 > > --- a/lib/eal/linux/include/

Re: [dpdk-dev] [PATCH v4 2/2] drivers/raw: remove octeontx2-ep driver

On Mon, Aug 23, 2021 at 2:19 PM Radha Mohan wrote: > I received this report that this patch failed to apply > http://mails.dpdk.org/archives/test-report/2021-August/211611.html The two patches of this series are seen as two separate series. http://patchwork.dpdk.org/project/dpdk/list/?submitter=2

Re: [dpdk-dev] [PATCH v4 2/2] drivers/raw: remove octeontx2-ep driver

On Thu, Aug 19, 2021 at 1:29 PM Radha Mohan Chintakuntla wrote: > > Removing the rawdev based octeontx2-ep driver as the dependent > common/octeontx2 will soon be going away. Moreover this driver is no > longer required as the net/octeontx_ep driver is sufficient. > > Signed-off-by: Radha Mohan Ch

[dpdk-dev] [PATCH] sched: support 64 bits greatest common divisor

This patch changes inline function that compute the greatest common divisor to 64 bits. Signed-off-by: Xueming Li --- lib/sched/rte_sched_common.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/sched/rte_sched_common.h b/lib/sched/rte_sched_common.h index 96706df7b

Re: [dpdk-dev] [PATCH v2 8/8] examples: remove unnecessary include of atomic

Hi Joyce, > -Original Message- > From: Joyce Kong > Sent: Monday, August 23, 2021 1:50 PM > To: Chas Williams ; Min Hu (Connor) ; > Ananyev, Konstantin ; Nicolau, Radu > ; Akhil Goyal ; Doherty, Declan > ; Sunil Kumar Kori ; Pavan > Nikhilesh ; Richardson, Bruce > ; Hunt, David ; Burakov,

Re: [dpdk-dev] [PATCH] app/testpmd: configurable number of flows in flowgen

Hi > -Original Message- > From: Zhihong Wang > Sent: Thursday, August 19, 2021 20:36 > To: dev@dpdk.org; Yigit, Ferruh ; Li, Xiaoyun > ; Singh, Aman Deep ; > irussk...@marvell.com; cchempara...@tilera.com > Cc: Zhihong Wang > Subject: [PATCH] app/testpmd: configurable number of flows in

[dpdk-dev] [PATCH] net/virtio: wait device ready in device reset

According to virtio spec, the device MUST reset when 0 is written to device_status, and present a 0 in device_status once that is done. This patch adds the missing part of waiting status 0 in reset function. Signed-off-by: Xueming Li --- drivers/net/virtio/virtio.c | 7 +-- 1 file changed,

Re: [dpdk-dev] [PATCH] net/bnxt: fix to reset Rx next consumer index

Thank you Ferruh, have sent V2 with updated commit log, please see if it sounds good now? -Som On Fri, Aug 20, 2021 at 6:17 PM Ferruh Yigit wrote: > On 8/10/2021 7:07 AM, Somnath Kotur wrote: > > In bnxt_init_one_rx_ring(), reset this variable internal to the driver > > ring to 0, so that there

[dpdk-dev] [PATCH v2] net/bnxt: fix to reset Rx next consumer index

In bnxt_init_one_rx_ring(), reset this variable internal to the driver ring to 0, so that there is no scope for any false mismatch errors being flagged on traffic resumption. Typically, this would show up in a port stop/start sequence as then the HW would start afresh, but this driver internal var

Re: [dpdk-dev] [PATCH] RFC: ethdev: add reassembly offload

On 8/23/21 1:02 PM, Akhil Goyal wrote: > Reassembly is a costly operation if it is done in > software, however, if it is offloaded to HW, it can > considerably save application cycles. > The operation becomes even more costlier if IP fragmants > are encrypted. > > To resolve above two issues, a ne

Re: [dpdk-dev] [RFC v2] eal: simplify the implementation of rte_ctrl_thread_create

Hi Honnappa, On Mon, Aug 02, 2021 at 12:16:52AM -0500, Honnappa Nagarahalli wrote: > The current described behaviour of rte_ctrl_thread_create is > rigid which makes the implementation of the function complex. > The behavior is abstracted to allow for simplified implementation. I agree that the b

Re: [dpdk-dev] 20.11.3 patches review and test

On Mon, 2021-08-23 at 10:37 +0800, Pei Zhang wrote: > Hello Luca, > > The testing with dpdk 20.11.3-rc1 from Red Hat looks good. We tested > below 16 scenarios and all got PASS on RHEL8: > > (1)Guest with device assignment(PF) throughput testing(1G hugepage > size): PASS > (2)Guest with device as

Re: [dpdk-dev] remove dpdk-pdump pcap file, disk space is not freed

On 8/23/2021 5:14 AM, Yan, Xiaoping (NSB - CN/Hangzhou) wrote: > Hi, > > Before I run pdump, 21MB is used in /tmp > tmpfs63G 21M 63G 1% /tmp > Then I run pdump with such command: > dpdk-pdump -c 0x4001000400100 -a :03:00.6 -a :03:06.1 --legacy-mem > --base-v

[dpdk-dev] [PATCH] RFC: ethdev: add reassembly offload

Reassembly is a costly operation if it is done in software, however, if it is offloaded to HW, it can considerably save application cycles. The operation becomes even more costlier if IP fragmants are encrypted. To resolve above two issues, a new offload DEV_RX_OFFLOAD_REASSEMBLY is introduced in

Re: [dpdk-dev] [PATCH v4] app/testpmd: add option to display extended statistics

On 8/21/21 4:09 AM, Ajit Khaparde wrote: > On Fri, Aug 20, 2021 at 6:55 AM Andrew Rybchenko > wrote: >> >> From: Ivan Ilchenko >> >> Add 'display-xstats' option for using in accompanying with Rx/Tx statistics >> (i.e. 'stats-period' option or 'show port stats' interactive command) to >> display s

Re: [dpdk-dev] [PATCH] net/virtio: wait device ready in device reset

On 8/23/21 9:39 AM, Xueming Li wrote: > According to virtio spec, the device MUST reset when 0 is written to > device_status, and present a 0 in device_status once that is done. > > This patch adds the missing part of waiting status 0 in reset function. > > Signed-off-by: Xueming Li > --- > dri

[dpdk-dev] [PATCH RFC 1/1] vhost: add DMADEV support for async datapath

This patch simplifies the async data path enablement for applications by allowing just the DMADEV ID to be passed as a parameter rather than having to implement its own logic to enable DMA offload. Remove the callbacks transfer_data and check_completed_copies and utilize the generic DMADEV API's t

[dpdk-dev] [PATCH RFC 0/1] vhost-add-DMADEV-support-for-async-datapath

Note to the reader: --- The intent of this patch is to explore possible different approaches of async implementations. Please consider this patch for discussions only and not for merge/upstream. This patch simplifies the vhost async datapath usability by utilizing the generic DMADE

Re: [dpdk-dev] [PATCH] doc: abstract the behaviour of rte_ctrl_thread_create

Hi Honnappa, Back from holidays, sorry for the late answer. On Mon, Aug 09, 2021 at 01:18:42PM +, Honnappa Nagarahalli wrote: > > > > > 30/07/2021 23:44, Honnappa Nagarahalli: > > > The current expected behaviour of the function rte_ctrl_thread_create > > > is rigid which makes the implemen

[dpdk-dev] [PATCH v5 2/2] net: added macro to extract MAC address bytes

Added macros to simplify print of MAC address. The six bytes of a MAC address are extracted in a macro here, to improve code readablity. Signed-off-by: Aman Singh Reviewed-by: Ferruh Yigit --- The change in the document will be done in seperate patch. To ensure document has direct reference of t

[dpdk-dev] [PATCH v5 1/2] net: added macro for MAC address print

Added macro to print six bytes of MAC address. The MAC addresses will be printed in upper case hexadecimal format. In case there is a specific check for lower case MAC address, the user may need to make a change in such test case after this patch. Signed-off-by: Aman Singh Reviewed-by: Ferruh Yig

[dpdk-dev] [PATCH v5 0/2] Use macro to print MAC address

Added macros to simplyfy print of MAC address. The six bytes of mac address is extracted using a macro to improve code readability. Aman Singh (2): net: added macro for MAC address print net: added macro to extract MAC address bytes app/pdump/main.c | 5 +--- ap

[dpdk-dev] [PATCH 2/2] net/mlx5: optimize mprq memcpy for AMD EPYC2 platforms

add non temporal load and temporal store for mprq memcpy. define mlx5_ntload_tstore in meson build configuration to enable this optimization. This utilizes AMD EPYC2 optimized rte_memcpy* routines. Signed-off-by: Aman Kumar --- drivers/net/mlx5/meson.build | 1 + drivers/net/mlx5/mlx5.c |

[dpdk-dev] [PATCH 1/2] lib/eal: add amd epyc2 memcpy routine to eal

This patch provides rte_memcpy* calls optimized for AMD EPYC Gen2 platforms. This option is disabled by default and can be enabled by defining 'rte_memcpy_amdepyc2' in the meson build. Signed-off-by: Aman Kumar --- lib/eal/x86/include/meson.build | 1 + lib/eal/x86/include/rte_memcpy.h | 502

[dpdk-dev] [PATCH] maintainers: update for NFP

From: Heinrich Kuhn Some of Netronome's activities and people have moved over to Corigine, including NFP PMD maintenance and myself. Signed-off-by: Heinrich Kuhn --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 8013ba1f14..ba2c

[dpdk-dev] [PATCH] net/i40e: fix mac counting error

The i40evf_add_del_all_mac_addr function implements the mac batch delete or add function. The vsi mac_num variable should be determined according to the value of the variable add to add j or subtract j. Fixes: 185993420249 ("net/i40e: fix VF add/del MAC") Cc: sta...@dpdk.org Signed-off-by: Qiming

[dpdk-dev] [PATCH] net/nfp: fix minimum descriptor sizes

From: Heinrich Kuhn The NFP4000/6000 supports a minimum of 256 Tx/Rx descriptors and not 64. Before this patch when a DPDK application configured < 256 Tx descriptors the hardware read/write pointers would be unmasked and not wrapped at the expected size of the ring. The PMD logic to determine th

Re: [dpdk-dev] i40evf: potential segfault

On 8/20/2021 7:18 PM, Stefan Baranoff wrote: > Hi all! > > I was chasing a potential segfault and it appears, if I'm reading this > driver correctly, that in i40evf_init_vf() the value vf->adapter->eth_dev > is never set like pf->adapter->eth_dev is in eth_i40e_dev_init(). > > I believe this is l

[dpdk-dev] [PATCH 3/3] net/bnxt: fix VLAN indication in Rx mbuf

From: Kalesh AP Fix to ignore the RX_PKT_CMPL_FLAGS2_META_FORMAT_VLAN when the firmware doesn't support the VLAN strip offload. Fixes: 2eb53b134aae ("net/bnxt: add initial Rx code") Cc: sta...@dpdk.org Signed-off-by: Kalesh AP Reviewed-by: Somnath Kotur Reviewed-by: Lance Richardson Reviewed

[dpdk-dev] [PATCH 2/3] net/bnxt: check FW capability for VLAN offloads

From: Kalesh AP VLAN offload capability may be disabled in the FW. The driver should not attempt to override or utilize this feature in such scenarios since it will not work as expected. Fixes: 0a6d2a720078 ("net/bnxt: get device infos") Cc: sta...@dpdk.org Signed-off-by: Kalesh AP Reviewed-by

[dpdk-dev] [PATCH 1/3] net/bnxt: fix Rx mbuf VLAN in scalar Rx

From: Kalesh AP In the scalar Rx path, for the VLAN packet, TCI is not saved in the "mbuf->vlan_tci", however the STRIPPED offload flag is set along with PKT_RX_VLAN flag. Fixes: c1b33d40315f ("net/bnxt: use table based mbuf flags handling") Cc: sta...@dpdk.org Signed-off-by: Kalesh AP Reviewe

[dpdk-dev] [PATCH 0/3] bnxt fixes

From: Kalesh AP Please apply. Kalesh AP (3): net/bnxt: fix Rx mbuf VLAN in scalar Rx net/bnxt: check FW capability for VLAN offloads net/bnxt: fix VLAN indication in Rx mbuf drivers/net/bnxt/bnxt.h| 7 --- drivers/net/bnxt/bnxt_ethdev.c | 4 drivers/net/bnxt/bnxt_hwrm.