Re: [dpdk-dev] [PATCH 20.02] net/virtio-user: do not close tap when disabling queue pairs

2019-11-19 Thread Tiwei Bie
On Tue, Nov 19, 2019 at 08:36:38AM -0800, Stephen Hemminger wrote: > On Tue, 19 Nov 2019 14:29:48 +0800 > Tiwei Bie wrote: > > > -static int > > +int > > +vhost_kernel_tap_attach_queue(int fd) > > +{ > > + struct ifreq ifr; > > + > > + memset(&ifr, 0, sizeof(ifr)); > > + ifr.ifr_flags = IFF

[dpdk-dev] [PATCH] eal/linux: fix muti-secondary msl create issue

2019-11-19 Thread Li Han
when we run dpdk in docker, process pid may the same. add rte_rdtsc() to ensure all the names used in rte_fbarray_init are different. Signed-off-by: Li Han --- lib/librte_eal/linux/eal/eal_memalloc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/librte_eal/linux/eal

[dpdk-dev] [PATCH v2 0/6] octeontx: sync with latest SDK

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Sync octeontx mailbox with the latest version (10.1.2.x) of SDK available. Pavan Nikhilesh (6): octeontx: update mbox definition to version 1.1.3 net/octeontx: add application domain validation net/octeontx: cleanup redudant mbox structs mempool/octeontx: add applic

[dpdk-dev] [PATCH v2 1/6] octeontx: update mbox definition to version 1.1.3

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Sync mail box data structures to version 1.1.3. Add mail box version verification and defer initializing octeontx devices if mail box version mismatches. Update OCTEON TX limitaion with max mempool size used. Signed-off-by: Pavan Nikhilesh Reviewed-by: Jerin Jacob Kollanuk

[dpdk-dev] [PATCH v2 3/6] net/octeontx: cleanup redudant mbox structs

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Cleanup redudant mail box structures. Signed-off-by: Pavan Nikhilesh --- drivers/net/octeontx/base/octeontx_pkivf.c | 25 +-- drivers/net/octeontx/base/octeontx_pkivf.h | 242 +++-- 2 files changed, 43 insertions(+), 224 deletions(-) diff --git a/drivers

[dpdk-dev] [PATCH v2 5/6] event/octeontx: add appication domain validation

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Add applicaton domain validation for OCTEON TX TIM vfs aka Event timer. Signed-off-by: Pavan Nikhilesh --- drivers/event/octeontx/timvf_evdev.c | 12 ++--- drivers/event/octeontx/timvf_evdev.h | 8 +--- drivers/event/octeontx/timvf_probe.c | 65 ++

[dpdk-dev] [PATCH v2 4/6] mempool/octeontx: add application domain validation

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Add application domain validation for OCTEON TX FPA vfs. Signed-off-by: Pavan Nikhilesh --- drivers/mempool/octeontx/octeontx_fpavf.c | 86 +++ 1 file changed, 58 insertions(+), 28 deletions(-) diff --git a/drivers/mempool/octeontx/octeontx_fpavf.c b

[dpdk-dev] [PATCH v2 6/6] net/octeontx: make Rx queue offloads same as dev offloads

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Make Rx queue specific offloads same as device Rx offloads. Signed-off-by: Pavan Nikhilesh --- drivers/net/octeontx/octeontx_ethdev.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/octeontx/octeontx_ethdev.c b/drivers/net/octeontx/octeontx_ethdev.c ind

[dpdk-dev] [PATCH v2 2/6] net/octeontx: add application domain validation

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Add domain validation for PKI and PKO vfs Signed-off-by: Pavan Nikhilesh --- drivers/common/octeontx/octeontx_mbox.c | 15 - drivers/common/octeontx/octeontx_mbox.h | 6 +- .../octeontx/rte_common_octeontx_version.map | 1 + drivers/event/octeontx/ssovf

[dpdk-dev] [PATCH 1/2] doc: update bnxt feature list

2019-11-19 Thread Ajit Khaparde
Updating bnxt.ini file. Signed-off-by: Ajit Khaparde --- doc/guides/nics/features/bnxt.ini | 3 +++ 1 file changed, 3 insertions(+) diff --git a/doc/guides/nics/features/bnxt.ini b/doc/guides/nics/features/bnxt.ini index 9721dd61d..37a99e336 100644 --- a/doc/guides/nics/features/bnxt.ini +++ b

[dpdk-dev] [PATCH 2/2] net/bnxt: fix check for null spec and mask

2019-11-19 Thread Ajit Khaparde
bnxt_validate_and_parse_flow_type already has protocol specific NULL checks for rte_flow_item spec and mask and take actions accordingly. A check at the top of the loop is redundant and is preventing the protocol specific checks from being executed. Fixes: b7e01f386059 ("net/bnxt: handle cleanup i

Re: [dpdk-dev] [EXT] [PATCH v2] crypto/openssl: support SG for inplace buffers

2019-11-19 Thread Anoob Joseph
Hi Akhil, A minor question. Otherwise, looks good. Thanks, Anoob > -Original Message- > From: Akhil Goyal > Sent: Monday, November 18, 2019 7:06 PM > To: dev@dpdk.org > Cc: declan.dohe...@intel.com; Anoob Joseph ; > konstantin.anan...@intel.com; Jerin Jacob Kollanukkaran > ; Akhil Goyal

Re: [dpdk-dev] [PATCH] crypto/armv8: enable meson build

2019-11-19 Thread Honnappa Nagarahalli
> > Hi Dharmik, > > As per the recent communication with Honnappa on a separate mail chain, > ARM has agreed to Host the armv8_crypto library. If it is happening in 19.11 To the wider audience, yes Arm will host a crypto library which is an existing repo on Arm github. We will pull the Marvell

[dpdk-dev] [PATCH 2/5] event/octeontx2: use opposite bucket to store current chunk

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Since TIM buckets are always aligned to 32B and our cache line size being 128B, we will always have a cache miss when reading current_chunk pointer. Avoid the cache miss by storing the current_chunk pointer in the bucket opposite to the current bucket. Signed-off-by: Pavan

[dpdk-dev] [PATCH 3/5] event/octeontx2: improve chunk pool performance

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Enable mempool cache for internal mempool to improve alloc performance. Signed-off-by: Pavan Nikhilesh --- drivers/event/octeontx2/otx2_tim_evdev.c | 4 ++-- drivers/event/octeontx2/otx2_tim_worker.h | 15 ++- 2 files changed, 16 insertions(+), 3 deletions(-

[dpdk-dev] [PATCH 1/5] event/octeontx2: fix TIM HW race condition

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Fix HW race condition observed when timeout resolution is low (<5us). When HW traverses a given TIM bucket it will clear chunk_remainder, but since SW always decreases the chunk_remainder at the start of the arm routine it might cause a race where SW updates chunk_remainder

[dpdk-dev] [PATCH 5/5] event/octeontx2: update start timestamp periodically

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Update start timestamp periodically to prevent drift. Signed-off-by: Pavan Nikhilesh --- drivers/event/octeontx2/otx2_tim_evdev.c | 28 +++ drivers/event/octeontx2/otx2_tim_evdev.h | 7 -- drivers/event/octeontx2/otx2_tim_worker.c | 19 +

[dpdk-dev] [PATCH 4/5] event/octeontx2: update SSO buffers based on timer count

2019-11-19 Thread pbhagavatula
From: Pavan Nikhilesh Update SSO internal XAQ buffers based on number of timers in event timer adapter. Signed-off-by: Pavan Nikhilesh --- drivers/event/octeontx2/otx2_evdev.h | 6 +- drivers/event/octeontx2/otx2_evdev_adptr.c | 84 +- drivers/event/octeontx2/otx2_ti

Re: [dpdk-dev] [PATCH 1/2] doc: update bnxt feature list

2019-11-19 Thread Ajit Khaparde
On Tue, Nov 19, 2019 at 8:29 PM Ajit Khaparde wrote: > Updating bnxt.ini file. > > Signed-off-by: Ajit Khaparde > Patch applied to dpdk-next-net-brcm. > --- > doc/guides/nics/features/bnxt.ini | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/doc/guides/nics/features/bnxt.ini > b/do

Re: [dpdk-dev] [PATCH 2/2] net/bnxt: fix check for null spec and mask

2019-11-19 Thread Ajit Khaparde
On Tue, Nov 19, 2019 at 8:29 PM Ajit Khaparde wrote: > bnxt_validate_and_parse_flow_type already has protocol specific NULL > checks for rte_flow_item spec and mask and take actions accordingly. > A check at the top of the loop is redundant and is preventing the > protocol specific checks from be

[dpdk-dev] Minutes of Technical Board Meeting, 2019-11-06

2019-11-19 Thread Jerin Jacob Kollanukkaran
Minutes of Technical Board Meeting, 2019-11-06 Members Attending - -Bruce -Ferruh -Hemant -Honnappa -Jerin (Chair) -Kevin -Konstantin -Maxime -Olivier -Thomas NOTE: The technical board meetings every second Wednesday on IRC channel #dpdk-board, at 3pm UTC. Meetings are public a

[dpdk-dev] [PATCH 0/2] fix FDIR support for non-extension-header GTPU

2019-11-19 Thread Yahui Cao
This patch series fixes FDIR support only for GTP-U with extension header by adding support for GTP-U without extension header. - Patch 1 fixes ice flow GTPU with extension header definition - Patch 2 adds FDIR support for non-extension-header GTPU -- Yahui Cao (2): net/ice: fix GTPU and GTPU

[dpdk-dev] [PATCH 1/2] net/ice: fix GTPU and GTPU extension header mix

2019-11-19 Thread Yahui Cao
GTP-U packet with extension header is GTP-U packet with GTP_PSC header inserted between GTP-U header and inner payload. Fixes: 3e4eab9c2192 ("net/ice: add pattern manifest") Cc: ying.a.w...@intel.com Signed-off-by: Yahui Cao --- drivers/net/ice/ice_generic_flow.c | 14 ++--- drivers/net

[dpdk-dev] [PATCH 2/2] net/ice: fix FDIR support for GTPU without extension header

2019-11-19 Thread Yahui Cao
Add FDIR support for normal GTP-U packet, which doesn't contain GTP_PSC extension header. Fixes: efc16c621415 ("net/ice: support flow director GTPU tunnel") Cc: yahui@intel.com Signed-off-by: Yahui Cao --- drivers/net/ice/ice_fdir_filter.c | 10 -- 1 file changed, 8 insertions(+), 2

Re: [dpdk-dev] [EXT] [PATCH v2] crypto/openssl: support SG for inplace buffers

2019-11-19 Thread Akhil Goyal
> > static inline int > > process_openssl_decryption_update(struct rte_mbuf *mbuf_src, int offset, > > - uint8_t **dst, int srclen, EVP_CIPHER_CTX *ctx) > > + uint8_t **dst, int srclen, EVP_CIPHER_CTX *ctx, uint8_t inplace) > > { > > struct rte_mbuf *m; > > int dstlen

[dpdk-dev] [PATCH v3] crypto/openssl: support SG for inplace buffers

2019-11-19 Thread Akhil Goyal
As per current support, Scatter Gather is only supported for out of place input and output buffers. This patch add support for Scatter Gather for inplace buffers. Signed-off-by: Akhil Goyal --- changes in v3: - fixed hardcode in decryption path changes in v2: - fixed typo - enabled corresponding

Re: [dpdk-dev] [PATCH v3] crypto/openssl: support SG for inplace buffers

2019-11-19 Thread Anoob Joseph
> -Original Message- > From: dev On Behalf Of Akhil Goyal > Sent: Wednesday, November 20, 2019 11:29 AM > To: dev@dpdk.org > Cc: declan.dohe...@intel.com; Anoob Joseph ; > konstantin.anan...@intel.com; Jerin Jacob Kollanukkaran > ; Akhil Goyal > Subject: [dpdk-dev] [PATCH v3] crypto/opens

Re: [dpdk-dev] [PATCH] mbuf: extend pktmbuf pool private structure

2019-11-19 Thread Shahaf Shuler
Wednesday, November 20, 2019 1:51 AM, Stephen Hemminger: > Subject: Re: [dpdk-dev] [PATCH] mbuf: extend pktmbuf pool private > structure > > On Tue, 19 Nov 2019 23:30:15 +0100 > Thomas Monjalon wrote: > > > 19/11/2019 17:25, Stephen Hemminger: > > > On Tue, 19 Nov 2019 15:23:50 + > > > Shaha

[dpdk-dev] [PATCH v2 0/2] fix FDIR support for GTPU

2019-11-19 Thread Yahui Cao
This patch series enables FDIR support for both GTP-U and GTP-U with extension header. - Patch 1 renames GTPU with extension header related variables - Patch 2 adds FDIR support for non-extension-header GTPU -- Yahui Cao (2): net/ice: fix GTPU extension header renaming net/ice: fix FDIR sup

[dpdk-dev] [PATCH v2 2/2] net/ice: fix FDIR support for GTPU without extension header

2019-11-19 Thread Yahui Cao
Add FDIR support for normal GTP-U packet, which doesn't contain GTP_PSC extension header. Fixes: efc16c621415 ("net/ice: support flow director GTPU tunnel") Cc: yahui@intel.com Signed-off-by: Yahui Cao --- drivers/net/ice/ice_fdir_filter.c | 9 +++-- drivers/net/ice/ice_generic_flow.c

[dpdk-dev] [PATCH v2 1/2] net/ice: fix GTPU extension header renaming

2019-11-19 Thread Yahui Cao
GTP-U packet with extension header is GTP-U packet with GTP_PSC header inserted between GTP-U header and inner payload. Fixes: 3e4eab9c2192 ("net/ice: add pattern manifest") Cc: ying.a.w...@intel.com Signed-off-by: Yahui Cao --- drivers/net/ice/ice_fdir_filter.c | 5 +++-- drivers/net/ice/ice

Re: [dpdk-dev] [PATCH] eal/linux: fix muti-secondary msl create issue

2019-11-19 Thread David Marchand
On Wed, Nov 20, 2019 at 3:48 AM Li Han wrote: > > when we run dpdk in docker, process pid may the same. > add rte_rdtsc() to ensure all the names used in rte_fbarray_init > are different. There is already a fix being prepared by Yasufumi Ogawa. https://patchwork.dpdk.org/patch/62972/ -- David

<    1   2