Re: [dpdk-dev] [PATCH v3 1/3] net/sfc: fix power of 2 round up when align has smaller type

On 7/24/2019 2:16 PM, Andrew Rybchenko wrote: > Substitute driver-defined P2ROUNDUP() h with EFX_P2ROUNDUP() > defined in libefx. > > Cast value and alignment to one specified type to guarantee result > correctness. > > Fixes: e1b944598579 ("net/sfc: build libefx") > Cc: sta...@dpdk.org > > Sign

[dpdk-dev] [PATCH] eal/interrupts: add function to allow interruptible epoll

The existing definition of rte_epoll_wait retries if interrupted by a signal. This behavior makes it hard to use rte_epoll_wait for applications that want to use signals do do things like exit polling loop and shutdown. Since changing existing semantic might break applications, add a new rte_epoll

Re: [dpdk-dev] [PATCH v1 4/4] net/i40e: replace license text with SPDX tag

Signed-off-by: Xiaolong Ye --- drivers/net/i40e/i40e_rxtx_vec_altivec.c | 35 ++-- drivers/net/i40e/i40e_rxtx_vec_avx2.c| 33 ++ 2 files changed, 5 insertions(+), 63 deletions(-) Reviewed-by: David Christensen

Re: [dpdk-dev] [PATCH v3 5/5] config: add WFE config entry for aarch64

Hi Stephen, > -Original Message- > From: Honnappa Nagarahalli > Sent: Wednesday, July 24, 2019 3:10 AM > To: Stephen Hemminger ; Gavin Hu (Arm > Technology China) > Cc: dev@dpdk.org; nd ; tho...@monjalon.net; > jer...@marvell.com; pbhagavat...@marvell.com; Honnappa Nagarahalli > ; nd > S

Re: [dpdk-dev] [PATCH v1] test/rcu: remove redundant header file

> > RCU library now includes stdbool.h header itself. > There is no need to include this header file in RCU user files. > > Signed-off-by: Ruifeng Wang > Reviewed-by: Gavin Hu > --- > app/test/test_rcu_qsbr.c | 1 - > 1 file changed, 1 deletion(-) > > diff --git a/app/test/test_rcu_qsbr.c b/a

Re: [dpdk-dev] [EXT] [PATCH v3 1/5] eal: add the APIs to wait until equal

Hi Jerin, > -Original Message- > From: Jerin Jacob Kollanukkaran > Sent: Wednesday, July 24, 2019 7:53 PM > To: Gavin Hu (Arm Technology China) ; dev@dpdk.org > Cc: nd ; tho...@monjalon.net; > step...@networkplumber.org; Pavan Nikhilesh Bhagavatula > ; Honnappa Nagarahalli > > Subject: RE

Re: [dpdk-dev] [PATCH 1/3] net/pcap: fix Rx with small buffers

On 7/24/2019 12:54 PM, David Marchand wrote: > If the pkt pool contains only buffers smaller than the default headroom, > then the driver will compute an invalid buffer size (negative value cast > to an uint16_t). > Rely on the mbuf api to check how much space is available in the mbuf. > > Fixes:

Re: [dpdk-dev] [PATCH 2/3] net/pcap: fix transmit return count in error conditions

On 7/24/2019 12:54 PM, David Marchand wrote: > When a packet cannot be transmitted, the driver is supposed to free this > packet and report it as handled. > This is to prevent the application from retrying to send the same packet > and ending up in a liveloop since the driver will never manage to s

Re: [dpdk-dev] [PATCH v3 1/3] net/sfc: fix power of 2 round up when align has smaller type

On 7/24/19 7:57 PM, Ferruh Yigit wrote: On 7/24/2019 2:16 PM, Andrew Rybchenko wrote: Substitute driver-defined P2ROUNDUP() h with EFX_P2ROUNDUP() defined in libefx. Cast value and alignment to one specified type to guarantee result correctness. Fixes: e1b944598579 ("net/sfc: build libefx") Cc

Re: [dpdk-dev] [PATCH 3/3] net/pcap: fix concurrent multiseg packet transmits

On 7/24/2019 12:54 PM, David Marchand wrote: > Two cores can send multi segment packets on two different pcap ports. > Because of this, we can't have one single buffer to linearize packets. +1, you are right. > > Use rte_pktmbuf_read() to copy the packet into a buffer on the stack > and remove e

Re: [dpdk-dev] [PATCH v3 0/3] net/sfc: fix power of 2 alignment macros

On 7/24/2019 2:16 PM, Andrew Rybchenko wrote: > P2ROUNDUP() and P2ALIGN() macros are buggy when alignment type is smaller > than type of the value to be aligned. > > IS_P2ALIGNED() has no the problem since it cast its arugments > to uintptr_t inside, but fixed anyway to follow the same approach as

Re: [dpdk-dev] [PATCH v3 1/3] net/sfc: fix power of 2 round up when align has smaller type

On 7/24/2019 7:41 PM, Andrew Rybchenko wrote: > On 7/24/19 7:57 PM, Ferruh Yigit wrote: >> On 7/24/2019 2:16 PM, Andrew Rybchenko wrote: >>> Substitute driver-defined P2ROUNDUP() h with EFX_P2ROUNDUP() >>> defined in libefx. >>> >>> Cast value and alignment to one specified type to guarantee result

Re: [dpdk-dev] [PATCH v1 4/4] net/i40e: replace license text with SPDX tag

On 7/24/2019 6:34 PM, David Christensen wrote: >> Signed-off-by: Xiaolong Ye >> --- >> drivers/net/i40e/i40e_rxtx_vec_altivec.c | 35 ++-- >> drivers/net/i40e/i40e_rxtx_vec_avx2.c| 33 ++ >> 2 files changed, 5 insertions(+), 63 deletions(-) > > Revi

Re: [dpdk-dev] [PATCH v3] net/mlx5: fix doorbell release on Rx queue release

השג את Outlook עבור Android מאת: Dekel Peled נשלח: יום רביעי, 24 ביולי, 12:05 נושא: [dpdk-dev] [PATCH v3] net/mlx5: fix doorbell release on Rx queuerelease אל: Yongseok Koh, Slava Ovsiienko, Shahaf Shuler עותק: Ori Kam, dev@dpdk.org Function mlx5_rxq_release() calls m

Re: [dpdk-dev] [PATCH v3] net/mlx5: fix Rx queue release of resources

השג את Outlook עבור Android מאת: Dekel Peled נשלח: יום רביעי, 24 ביולי, 12:05 נושא: [dpdk-dev] [PATCH v3] net/mlx5: fix Rx queue release of resources אל: Yongseok Koh, Slava Ovsiienko, Shahaf Shuler עותק: Ori Kam, dev@dpdk.org Function rxq_release_rq_resources() relea

[dpdk-dev] [PATCH 0/2] mk: fix build with shared library

If RTE_BUILD_SHARED_LIB is enabled, then there a couple of places in the mk scripts where constructors are not called but need to be. Stephen Hemminger (2): mk: fix missing pci bus with shared library build mk: fix missing mempool driver registration mk/rte.app.mk | 9 + 1 file chang

[dpdk-dev] [PATCH 1/2] mk: fix missing pci bus with shared library build

If DPDK is built as a shared library, then any application linked with rte.app.mk will not find any buses (including PCI devices). This is because the linker will not call the initializers of the bus libraries since they are not directly called by application or EAL. Bugzilla ID: 333 Fixes: c7529

[dpdk-dev] [PATCH 2/2] mk: fix missing mempool driver registration

Since mempool driver is now registered during constructor, the mempool driver must always be linked in during shared library build. Otherwise, application will fail because of missing default mbuf pool operations. Fixes: c6b3570f9e14 ("mk: link app with ring mempool by default") Cc: stable.dpdk.or

Re: [dpdk-dev] [PATCH v2] Revert "eal: fix parsing option --telemetry"

24/07/2019 18:28, Morrissey, Sean: > > Subject: Re: [PATCH v2] Revert "eal: fix parsing option --telemetry" > > 24/07/2019 17:20, Sean Morrissey: > > This reverts commit debacba0297fbe214b4185a9791e6a9fdf6642ba. > > > > Reverting this patch as it currently breaks the initialization of > > telem

[dpdk-dev] [v3] net/ice: enable multi-process support

Add multiple processes support for ice, secondary processes will share memory and configuration with primary process, do not need further initialization for secondary processes. Cc: sta...@dpdk.org Signed-off-by: Xiao Zhang --- v3 Update the title and commit message. v2 Remove limitation for sec

[dpdk-dev] [PATCH] net/enic: retain previous message logging

Prior to fix, RTE_LOGTYPE_INFO messages would display in testpmd by default. After the fix, using dynamic logging, only NOTICE level and higher were displayed by default and INFO level were not. Change the messages to NOTICE level so they continue to display. DTS uses testpmd and parses messages a

Re: [dpdk-dev] [v3] net/ice: enable multi-process support

On 07/25, Xiao Zhang wrote: >Add multiple processes support for ice, secondary processes will share >memory and configuration with primary process, do not need further >initialization for secondary processes. > >Cc: sta...@dpdk.org This is a new feature, it's no need to cc sta...@dpdk.org. Thanks

Re: [dpdk-dev] [PATCH v1 0/2] declare crypto asym xform immutable

Hi Ayuj, I believe there are couple of issues with this patch. Are these experimental APIs? I believe they were made stable this release and I'm not sure if it is a right practice to edit an API without deprecation notice after it is made stable. Especially now that RC2 is done. @Akhil, what i

[dpdk-dev] [PATCH] net/iavf: fix endless loop

Change loop index from uint16_t to uint32_t since max index 65535 could be exceeded when ring size is 2k+. Fixes: 69dd4c3d0898 ("net/avf: enable queue and device") Cc: sta...@dpdk.org Reported-by: Lei Yao Signed-off-by: Qi Zhang --- drivers/net/iavf/iavf_rxtx.c | 3 ++- 1 file changed, 2 inser

[dpdk-dev] [PATCH] net/iavf: fix endless loop

Change loop index from uint16_t to uint32_t since max index 65535 could be exceeded when ring size is 2k+. Fixes: 69dd4c3d0898 ("net/avf: enable queue and device") Cc: sta...@dpdk.org Reported-by: Lei Yao Signed-off-by: Qi Zhang --- drivers/net/iavf/iavf_rxtx.c | 3 ++- 1 file changed, 2 inser

[dpdk-dev] [PATCH v2] net/iavf: fix endless loop

Change loop index from uint16_t to uint32_t since max index 65535 could be exceeded when ring size is 2k+. Fixes: 69dd4c3d0898 ("net/avf: enable queue and device") Cc: sta...@dpdk.org Reported-by: Lei Yao Signed-off-by: Qi Zhang --- v2: - add missing fix in reset_tx_queue. drivers/net/iavf/ia

Re: [dpdk-dev] [v3] net/ice: enable multi-process support

> -Original Message- > From: Zhang, Xiao > Sent: Thursday, July 25, 2019 7:19 PM > To: dev@dpdk.org > Cc: Yang, Qiming ; Zhang, Qi Z > ; Zhang, Xiao ; > sta...@dpdk.org > Subject: [v3] net/ice: enable multi-process support > > Add multiple processes support for ice, secondary processes

[dpdk-dev] [PATCH v2 0/4] bnxt patches

This set contains bug fixes for commits that made it into 19.08, so it's imperative that these go into 19.08 as well Please apply Kalesh AP (3): net/bnxt: fix to use chimp hwrm channel for few commands net/bnxt: fix endianness issue net/bnxt: reduce verbosity of a message Santoshkumar Kara

[dpdk-dev] [PATCH 1/4] net/bnxt: fix extended port counter statistics

From: Santoshkumar Karanappa Rastapur We were trying to fill in more rx extended stats than the size allocated for stats causing segfault. Fixed this by adding an explicit check. Rearranged the code to return statistic values in xstats_get as per the names returned in xstats_get_names. Fixes: f5

[dpdk-dev] [PATCH 4/4] net/bnxt: reduce verbosity of a message

From: Kalesh AP Change verbosity of a message to DEBUG from ERROR. This is just debug message. Fixes: 3e92fd4e4ec0c7b ("net/bnxt: use dynamic log type") Signed-off-by: Kalesh AP Reviewed-by: Somnath Kotur Signed-off-by: Somnath Kotur --- drivers/net/bnxt/bnxt_hwrm.c | 2 +- 1 file changed,

[dpdk-dev] [PATCH 3/4] net/bnxt: fix endianness issue

From: Kalesh AP Use rte_cpu_to_le_16/32 while parsing the hwrm command response. Fixes: 11e5e19695c7bfb3 ("net/bnxt: support redirecting tunnel packets to VF") Signed-off-by: Kalesh AP Reviewed-by: Somnath Kotur Signed-off-by: Somnath Kotur --- drivers/net/bnxt/bnxt_hwrm.c | 4 ++-- 1 file

[dpdk-dev] [PATCH 2/4] net/bnxt: fix to use chimp hwrm channel for few commands

From: Kalesh AP Modified to send the tunnel redirect commands to Chimp as Kong does not support these commands. Fixes: 11e5e19695c7bfb3 ("net/bnxt: support redirecting tunnel packets to VF") Signed-off-by: Kalesh AP Reviewed-by: Somnath Kotur Signed-off-by: Somnath Kotur --- drivers/net/bnx

Re: [dpdk-dev] [PATCH 0/5] bnxt patches

Ferruh, Please ignore this set as i have sent a v2 with just 4 of the patches below...the 1st patch in this set got pulled in to Lance's latest patch ( (net/bnxt: use dedicated cpr for async events) Thanks and apologies for any confusion -Som On Wed, Jul 24, 2019 at 11:37 AM Somnath Ko

Re: [dpdk-dev] [PATCH v1 0/2] declare crypto asym xform immutable

Hi Anoob Seems your reply is not in plaintext which will make it difficult for inline response. So, please take care of that when you reply. Rest, please see my response inline. From: Anoob Joseph Sent: Thursday, July 25, 2019 9:46 AM To: Ayuj Verma ; akhil.go...@nxp.com Cc: arkadiuszx.kusz...@

Re: [dpdk-dev] [PATCH v1 0/2] declare crypto asym xform immutable

Hi Shally, Oh yeah! Should've noticed. My settings is to use the incoming mail's formatting. Mail Ayuj created the issue. Thanks, Anoob > -Original Message- > From: Shally Verma > Sent: Thursday, July 25, 2019 10:52 AM > To: Anoob Joseph ; Ayuj Verma > ; akhil.go...@nxp.com > Cc: arkadi

[dpdk-dev] [PATCH 0/2] add symmetric toeplitz hash support

From: Simei Su [PATCH 1/2] ethdev: add new hash function "Symmetric Toeplitz" supported by hardware. [PATCH 2/2] app/testpmd: add command line support for symmetric toeplitz hash configuration. Simei Su (2): ethdev: add symmetric toeplitz hash support app/testpmd: add symmetric toeplitz ha

[dpdk-dev] [PATCH 1/2] ethdev: add symmetric toeplitz hash support

From: Simei Su Currently, there are DEFAULT,TOEPLITZ and SIMPLE_XOR hash funtion. To support symmetric hash by rte_flow RSS action, this patch adds new hash function "Symmetric Toeplitz" which is supported by some hardware. Signed-off-by: Simei Su --- lib/librte_ethdev/rte_flow.h | 1 + 1 file

[dpdk-dev] [PATCH 2/2] app/testpmd: add symmetric toeplitz hash support

From: Simei Su This patch adds command line support for Symmetric Toeplitz hash configuration. Signed-off-by: Simei Su --- app/test-pmd/cmdline.c | 12 +--- app/test-pmd/cmdline_flow.c | 12 +++- doc/guides/testpmd_app_ug/testpmd_funcs.rst |

[dpdk-dev] [PATCH] net/octeontx2: fix handling indirect mbufs during TX

Multi segmented packet may be spliced with indirect mbufs also. Currently driver causes buffer leak for indirect mbufs as they were not being freed to packet pool. Patch fixes handling of indirect mbufs for following use cases: - packet contains all indirect mbufs only. - packet contains mixed mbu

[dpdk-dev] [PATCH] examples/ip_fragmentation: fix DEV_TX_OFFLOAD_MBUF_FAST_FREE flag

Application uses different pool to allocate direct and indirect mbufs which are further spliced together to consturct a fragmented packet and same is transmitted over the port which is configured with DEV_TX_OFFLOAD_MBUF_FAST_FREE enabled i.e. all segments must belong to the same pool. But construc

[dpdk-dev] [PATCH v2 1/1] examples/ip_frag: fix DEV_TX_OFFLOAD_MBUF_FAST_FREE flag

Application uses different pool to allocate direct and indirect mbufs which are further spliced together to consturct a fragmented packet and same is transmitted over the port which is configured with DEV_TX_OFFLOAD_MBUF_FAST_FREE enabled i.e. all segments must belong to the same pool. But construc

Re: [dpdk-dev] [Bug 335] Compilation fails when HEADROOM is 0

HI Matan, > > On Wed, 24 Jul 2019 15:37:40 + > > Matan Azrad wrote: > > > > > Hi Hemant, Sachin > > > > > > Looks like issue in dpaa. > > > Please check. > > > > > > > -Original Message- > > > > From: dev On Behalf Of bugzi...@dpdk.org > > > > Sent: Wednesday, July 24, 2019 6:35 PM >

Re: [dpdk-dev] [PATCH v2] net/iavf: fix endless loop

On 07/25, Qi Zhang wrote: >Change loop index from uint16_t to uint32_t since max >index 65535 could be exceeded when ring size is 2k+. > >Fixes: 69dd4c3d0898 ("net/avf: enable queue and device") >Cc: sta...@dpdk.org > >Reported-by: Lei Yao >Signed-off-by: Qi Zhang >--- >v2: >- add missing fix in

Re: [dpdk-dev] [PATCH v1 0/2] declare crypto asym xform immutable

Hi Shally, Please see inline. Thanks, Anoob > -Original Message- > From: Shally Verma > Sent: Thursday, July 25, 2019 10:52 AM > To: Anoob Joseph ; Ayuj Verma > ; akhil.go...@nxp.com > Cc: arkadiuszx.kusz...@intel.com; Sunila Sahu ; Kanaka > Durga Kotamarthy ; dev@dpdk.org; Fiona > Trahe

<    1   2