Re: [dpdk-dev] [PATCH v2] vhost: fix null pointer checking

On 4/8/19 5:38 AM, Tiwei Bie wrote: On Thu, Apr 04, 2019 at 09:53:17AM +0100, Mohammad Abdul Awal wrote: Null value for parameters will cause segfault. Fixes: d7280c9fff ("vhost: support selective datapath") Fixes: 72e8543093df ("vhost: add API to get MTU value") Fixes: a277c71598 ("vhost: r

Re: [dpdk-dev] [PATCH] raw/ifpga: fix fd leak in rte fpga do pr

Hi, > -Original Message- > From: Pei, Andy > Sent: Wednesday, April 17, 2019 14:49 > To: dev@dpdk.org > Cc: Pei, Andy ; Xu, Rosen ; > Kovacevic, Marko ; liq...@163.com; > sta...@dpdk.org > Subject: [PATCH] raw/ifpga: fix fd leak in rte fpga do pr Pls use raw/ifpga_rawdev prefix. I suggest

Re: [dpdk-dev] [PATCH v2] crypto/virtio: fix iv offset

On 4/5/19 6:04 PM, Fan Zhang wrote: This patch fixes the incorrect iv offset calculation. Fixes: b063e843fa03 ("crypto/virtio: fix IV physical address") Cc: sta...@dpdk.org Signed-off-by: Fan Zhang --- v2: fixed a typo drivers/crypto/virtio/virtio_rxtx.c | 4 ++-- 1 file changed, 2 i

Re: [dpdk-dev] [PATCH v2] virtio: fix buffer leak on vlan insert

On 4/5/19 6:31 PM, Stephen Hemminger wrote: The function rte_vlan_insert may allocate a new buffer for the vlan header and return a different mbuf than originally passed. In this case, the stored mbuf in txm[] array could point to wrong buffer. Fixes: dd856dfcb9e7 ("virtio: use any layout on

Re: [dpdk-dev] [PATCH] raw/ifpga: fix fd leak in rte fpga do pr

Okay -Original Message- From: Xu, Rosen Sent: Wednesday, April 17, 2019 3:14 PM To: Pei, Andy ; dev@dpdk.org Cc: Kovacevic, Marko ; liq...@163.com; sta...@dpdk.org Subject: RE: [PATCH] raw/ifpga: fix fd leak in rte fpga do pr Hi, > -Original Message- > From: Pei, Andy > Sent: W

[dpdk-dev] [PATCH v2] raw/ifpga_rawdev: fix fd leak in rte fpga do pr

In rte_fpga_do_pr() function, if 'stat' returns error, rte_fpga_do_pr() returns -EINVAL without closing the 'file_fd' that has been opened. After this patch, 'file_fd' is closed before rte_fpga_do_pr() returns -EINVAL when 'stat' returns error Coverity issue: 279441 Fixes: ef1e8ede3da5 ("raw/ifpga

Re: [dpdk-dev] [PATCH] vhost: fix silent queue enabling with legacy guests

On 4/12/19 3:09 PM, Ilya Maximets wrote: vhost should notify the application in case of all vring state changes. In general, application should not care about negotiation of VHOST_USER_F_PROTOCOL_FEATURES. Protocol details like this should be hidden by the vhost library. With this patch appl

[dpdk-dev] [PATCH] cryptodev: add an option to support both iv and J0 for GCM

This patch adds an option to support both IV (of all supported sizes) and J0 when using Galois Counter Mode of crypto operation. Signed-off-by: Arek Kusztal --- lib/librte_cryptodev/rte_crypto_sym.h | 37 ++- 1 file changed, 19 insertions(+), 18 deletions(-) diff

[dpdk-dev] [PATCH v2] ethdev: fix QinQ strip offload support

From: Vivek Sharma Enable missing support for QinQ strip rx offload in vlan offload set/get methods. Fixes: cc9d0456b870 ("i40e: support double vlan stripping and insertion") Cc: sta...@dpdk.org Signed-off-by: Vivek Sharma --- v2: * Use pointer to dereference dev->data->dev_conf.rxmode.offload

[dpdk-dev] [PATCH] doc: add cryptodev gcm iv deprecation notice

This patch adds deprecation notice of changing iv behaviour when using Galois Counter Mode of operation. Right now IV of all supported sizes can be used. Signed-off-by: Arek Kusztal --- doc/guides/rel_notes/deprecation.rst | 6 ++ 1 file changed, 6 insertions(+) diff --git a/doc/guides/rel_

Re: [dpdk-dev] [PATCH v2 2/2] net/af_xdp: make reserve/submit peek/release consistent

On Tue, Apr 16, 2019 at 5:10 PM Xiaolong Ye wrote: > As David pointed out, if we reserve N slots for Tx, but only submit n > slots, we would end up with an incorrect opinion of the number of available > slots later, we also would get wrong idx when we call > xsk_ring_prod__reserve next time. It a

Re: [dpdk-dev] [PATCH v2 1/2] net/af_xdp: enqueue buf ring when allocate Tx queue fails

On Tue, Apr 16, 2019 at 5:09 PM Xiaolong Ye wrote: > When it fails to allocate enough slots in Tx queue for transmitting > packets, we need to return the dequeued addrs to buf ring. > > Fixes: f1debd77efaf ("net/af_xdp: introduce AF_XDP PMD") > > Signed-off-by: Xiaolong Ye > --- > drivers/net/a

Re: [dpdk-dev] [PATCH] vhost: fix silent queue enabling with legacy guests

On 4/12/19 3:09 PM, Ilya Maximets wrote: vhost should notify the application in case of all vring state changes. In general, application should not care about negotiation of VHOST_USER_F_PROTOCOL_FEATURES. Protocol details like this should be hidden by the vhost library. With this patch appl

Re: [dpdk-dev] [PATCH] net/virtio: fix dangling pointer on failure

On 4/15/19 4:48 PM, Aaron Conole wrote: When eth_virtio_dev_init() is cleaning up, it does not correctly set the mac_addrs variable to NULL, which will lead to a double free. Found during unit-test fixes. Fixes: 43d18765c027 ("net/virtio: fix memory leak on failure") Cc: sta...@dpdk.org Repo

Re: [dpdk-dev] [PATCH v3] vhost: fix null pointer checking

On 4/11/19 4:48 PM, Mohammad Abdul Awal wrote: Null value for parameters will cause segfault. Fixes: d7280c9fff ("vhost: support selective datapath") Fixes: 72e8543093 ("vhost: add API to get MTU value") Fixes: a277c71598 ("vhost: refactor code structure") Fixes: ca33faf9ef ("vhost: introduce

Re: [dpdk-dev] [PATCH v2] vhost: fix passing destroyed device to destroy callback

On 4/12/19 10:32 AM, Ilya Maximets wrote: Application should be able to obtain information like 'ifname' from the 'vid' passed to 'destroy_connection' callback. Currently, all the API calls with passed 'vid' fails with 'device not found'. Fixes: efba12a78ddf ("vhost: add user callbacks for so

Re: [dpdk-dev] [PATCH] vhost: fix device leak on connection add failure

On 4/11/19 12:23 PM, Ilya Maximets wrote: Need to destroy allocated device if application fails to add new connection or we have fdset failure. Fixes: acbff5c67ea7 ("vhost: fix crash when exceeding file descriptors") Fixes: efba12a78ddf ("vhost: add user callbacks for socket open/close") Cc:s

Re: [dpdk-dev] [PATCH v1] net/virtio-user: fix return value check

On 4/10/19 4:44 AM, Chenbo Xia wrote: Fix unchecked return value for fcntl. Coverity issue: 277210 Fixes: bd8f50a45d0f ("net/virtio-user: support server mode") Cc:sta...@dpdk.org Signed-off-by: Chenbo Xia --- drivers/net/virtio/virtio_user/vhost_user.c | 5 - 1 file changed, 4 inserti

Re: [dpdk-dev] [PATCH v2] virtio: fix buffer leak on vlan insert

On 4/5/19 6:31 PM, Stephen Hemminger wrote: The function rte_vlan_insert may allocate a new buffer for the vlan header and return a different mbuf than originally passed. In this case, the stored mbuf in txm[] array could point to wrong buffer. Fixes: dd856dfcb9e7 ("virtio: use any layout on

Re: [dpdk-dev] [PATCH v2] crypto/virtio: fix iv offset

On 4/5/19 6:04 PM, Fan Zhang wrote: This patch fixes the incorrect iv offset calculation. Fixes: b063e843fa03 ("crypto/virtio: fix IV physical address") Cc: sta...@dpdk.org Signed-off-by: Fan Zhang --- v2: fixed a typo drivers/crypto/virtio/virtio_rxtx.c | 4 ++-- 1 file changed, 2 i

Re: [dpdk-dev] [PATCH v2 2/2] net/af_xdp: make reserve/submit peek/release consistent

On 04/17, David Marchand wrote: >On Tue, Apr 16, 2019 at 5:10 PM Xiaolong Ye wrote: > >> As David pointed out, if we reserve N slots for Tx, but only submit n >> slots, we would end up with an incorrect opinion of the number of available >> slots later, we also would get wrong idx when we call >>

Re: [dpdk-dev] [PATCH v2] raw/ifpga_rawdev: fix fd leak in rte fpga do pr

> -Original Message- > From: Pei, Andy > Sent: Wednesday, April 17, 2019 15:16 > To: dev@dpdk.org > Cc: Pei, Andy ; Xu, Rosen ; > Kovacevic, Marko ; liq...@163.com; > sta...@dpdk.org > Subject: [PATCH v2] raw/ifpga_rawdev: fix fd leak in rte fpga do pr > > In rte_fpga_do_pr() function,

Re: [dpdk-dev] [PATCH] net: do not insert VLAN tag to shared mbufs

On Tue, Apr 16, 2019 at 02:32:18PM -0400, Chas Williams wrote: > > > On 4/16/19 12:28 PM, Bruce Richardson wrote: > > On Tue, Apr 16, 2019 at 04:51:26PM +0100, Ferruh Yigit wrote: > > > The vlan_insert() is buggy when it tires to handle the shared mbufs, > > > instead don't support inserting VLAN

[dpdk-dev] Coverity defects in DPDK 19.05 RC1

Hi, The Coverity database is reporting the following open items against your name. If you have already submitted a patch then please update the "Action" in the database as "Fix submitted". If you haven't submitted a fix please do so soon. John CIDType Impact Owne

Re: [dpdk-dev] [RFC 3/3] ci: enable tests on non-arm platforms

> -Original Message- > From: dev On Behalf Of Aaron Conole > Sent: Friday, March 29, 2019 10:53 PM > To: dev@dpdk.org > Subject: [dpdk-dev] [RFC 3/3] ci: enable tests on non-arm platforms > > When building under Travis (or another linux CI service), enable running the > tests > for each

Re: [dpdk-dev] [PATCH v2] ethdev: fix QinQ strip offload support

17/04/2019 09:40, viveksha...@marvell.com: > From: Vivek Sharma > > Enable missing support for QinQ strip rx offload > in vlan offload set/get methods. > > Fixes: cc9d0456b870 ("i40e: support double vlan stripping and insertion") > Cc: sta...@dpdk.org Not sure it is a fix. The commit mentioned

Re: [dpdk-dev] ABI and inline functions

On Wed, Apr 17, 2019 at 05:12:43AM +, Honnappa Nagarahalli wrote: > Hello, > There was a conversation [1] in the context of RCU library. I thought > it needs participation from broader audience. Summary for the context (please > look at [1] for full details) > Thanks for kicking off th

[dpdk-dev] [PATCH] net/iavf: fix info get

iavf driver reset the dev_info structure. This action will clear the setting in core library. This patch fix that issue. Fixes: 69dd4c3d0898 ("net/avf: enable queue and device") Signed-off-by: Qiming Yang --- drivers/net/iavf/iavf_ethdev.c | 1 - 1 file changed, 1 deletion(-) diff --git a/driv

Re: [dpdk-dev] [PATCH] net/iavf: fix info get

On Wed, Apr 17, 2019 at 10:41 AM Qiming Yang wrote: > iavf driver reset the dev_info structure. This action will clear > the setting in core library. This patch fix that issue. > > Fixes: 69dd4c3d0898 ("net/avf: enable queue and device") > "Cc: sta...@dpdk.org" is supposed to be part of the commi

Re: [dpdk-dev] [PATCH v2] app/test/ipsec: fix logic around dequeue burst

> > > Added crypto_dequeue_burst() function to call > rte_crypto_dequeue_burst() in a loop with a > delay to ensure that all the packets are > dequeued from the crtpto device. > > Fixes: 59d7353b0df0 ("test/ipsec: fix test suite setup") > Cc: sta...@dpdk.org > Signed-off-by: Bernard Iremonger

Re: [dpdk-dev] [PATCH v2] app/test/ipsec: fix logic around dequeue burst

> > > > Added crypto_dequeue_burst() function to call > > rte_crypto_dequeue_burst() in a loop with a > > delay to ensure that all the packets are > > dequeued from the crtpto device. > > > > Fixes: 59d7353b0df0 ("test/ipsec: fix test suite setup") > > Cc: sta...@dpdk.org > > Signed-off-by: Bernar

[dpdk-dev] [PATCH v3 1/4] net/af_xdp: enqueue buf ring when allocate Tx queue fails

When it fails to allocate enough slots in Tx queue for transmitting packets, we need to return the dequeued addrs to buf ring. Fixes: f1debd77efaf ("net/af_xdp: introduce AF_XDP PMD") Signed-off-by: Xiaolong Ye --- drivers/net/af_xdp/rte_eth_af_xdp.c | 1 + 1 file changed, 1 insertion(+) diff

[dpdk-dev] [PATCH v3 0/4] some fixes for AF_XDP pmd

This patchset provides some fixes to af_xdp pmd, at first, I just added a simple error handling when Tx queue allocation fails, then David suggested a better way to do it and pointed out the inconsistent issue of reserve/submit ops (for Tx queue) and peek/release ops (for Rx queue), the third pat

[dpdk-dev] [PATCH v3 2/4] net/af_xdp: specify minimal and maximal MTU

Since AF_XDP pmd doesn't support multi segment, we need to add a valid check in eth_dev_mtu_set. Signed-off-by: Xiaolong Ye --- drivers/net/af_xdp/rte_eth_af_xdp.c | 12 1 file changed, 12 insertions(+) diff --git a/drivers/net/af_xdp/rte_eth_af_xdp.c b/drivers/net/af_xdp/rte_eth_

[dpdk-dev] [PATCH v3 3/4] net/af_xdp: make reserve/submit peek/release consistent

As David pointed out, if we reserve N slots for Tx, but only submit n slots, we would end up with an incorrect opinion of the number of available slots later, we also would get wrong idx when we call xsk_ring_prod__reserve next time. It also applies to xsk_ring_cons__peek()/xsk_ring_cons__release()

[dpdk-dev] [PATCH v3 4/4] net/af_xdp: fix typos in Rx function

We should use ETH_AF_XDP_RX_BATCH_SIZE in Rx function rather than ETH_AF_XDP_TX_BATCH_SIZE. Fixes: f1debd77efaf ("net/af_xdp: introduce AF_XDP PMD") Signed-off-by: Xiaolong Ye --- drivers/net/af_xdp/rte_eth_af_xdp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers

Re: [dpdk-dev] [PATCH v3 1/4] net/af_xdp: enqueue buf ring when allocate Tx queue fails

On Wed, Apr 17, 2019 at 11:02 AM Xiaolong Ye wrote: > When it fails to allocate enough slots in Tx queue for transmitting > packets, we need to return the dequeued addrs to buf ring. > > Fixes: f1debd77efaf ("net/af_xdp: introduce AF_XDP PMD") > > Signed-off-by: Xiaolong Ye > --- > drivers/net/

Re: [dpdk-dev] [PATCH v3 4/4] net/af_xdp: fix typos in Rx function

On Wed, Apr 17, 2019 at 11:02 AM Xiaolong Ye wrote: > We should use ETH_AF_XDP_RX_BATCH_SIZE in Rx function rather than > ETH_AF_XDP_TX_BATCH_SIZE. > > Fixes: f1debd77efaf ("net/af_xdp: introduce AF_XDP PMD") > > Reported-by: David Marchand Signed-off-by: Xiaolong Ye > --- > drivers/net/af_xd

Re: [dpdk-dev] [PATCH] doc: fix interactive commands in testpmd guide

> -Original Message- > From: Yigit, Ferruh > Sent: Tuesday, April 16, 2019 4:06 PM > To: Babu Radhakrishnan, AgalyaX ; > dev@dpdk.org > Cc: Pattan, Reshma ; olivier.m...@6wind.com; > Iremonger, Bernard > Subject: Re: [dpdk-dev] [PATCH] doc: fix interactive commands in testpmd > guide > >

Re: [dpdk-dev] [PATCH v3 3/4] net/af_xdp: make reserve/submit peek/release consistent

On Wed, Apr 17, 2019 at 11:02 AM Xiaolong Ye wrote: > As David pointed out, if we reserve N slots for Tx, but only submit n > slots, we would end up with an incorrect opinion of the number of available > slots later, we also would get wrong idx when we call > xsk_ring_prod__reserve next time. It

Re: [dpdk-dev] [PATCH v3 2/4] net/af_xdp: specify minimal and maximal MTU

On Wed, Apr 17, 2019 at 11:02 AM Xiaolong Ye wrote: > Since AF_XDP pmd doesn't support multi segment, we need to add a valid > check in eth_dev_mtu_set. > How about: Properly report mtu capability in port device info. Reported-by: David Marchand Signed-off-by: Xiaolong Ye > --- > drivers/n

[dpdk-dev] [PATCH v2] net/bonding: fix potential out of bounds read

Add validation to pointer constructed from the IPv4 header length in order to prevent malformed packets from generating a potential out of bounds memory read. Fixes: 09150784a776 ("net/bonding: burst mode hash calculation") Cc: sta...@dpdk.org Signed-off-by: Radu Nicolau --- v2: add fixes lines

Re: [dpdk-dev] [PATCH v1] fix alphabetical ordering of headers

Hi Ayuj, > -Original Message- > From: Ayuj Verma [mailto:ayve...@marvell.com] > Sent: Friday, April 12, 2019 8:03 AM > To: akhil.go...@nxp.com; Kusztal, ArkadiuszX > mailto:arkadiuszx.kusz...@intel.com>>; Trahe, > Fiona > mailto:fiona.tr...@intel.com>> > Cc:

[dpdk-dev] [PATCH] drivers/crypto: fix session destroy for dpaa and dpaa2

Fixes: 1f14d500bce1 ("crypto/dpaa_sec: support IPsec protocol offload") Fixes: 0a23d4b6f4c2 ("crypto/dpaa2_sec: support protocol offload IPsec") Coverity issue: 336798 Cc: sta...@dpdk.org Signed-off-by: Akhil Goyal --- drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 2 +- drivers/crypto/dpaa_sec/d

[dpdk-dev] [PATCH v2] drivers/crypto: fix session destroy for dpaa and dpaa2

Fixes: 1f14d500bce1 ("crypto/dpaa_sec: support IPsec protocol offload") Fixes: 0a23d4b6f4c2 ("crypto/dpaa2_sec: support protocol offload IPsec") Coverity issue: 336798 Cc: sta...@dpdk.org Signed-off-by: Akhil Goyal --- v2: fix checkpatch drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 2 +- drive

Re: [dpdk-dev] [PATCH] examples/l3fwd: support separate buffer pool per port

Hi > > > As you have stated below, it's just the same thing with two different > > views. > > > > > > > I think it would be plausible for both cases: > > > > - one port per core (your case). > > > > - multiple ports per core. > > > > > > Indeed. For this particular patch, I just chose the firs

Re: [dpdk-dev] [PATCH v1] fix alphabetical ordering of headers

Hi Ayuj, Akhil, From: Akhil Goyal [mailto:akhil.go...@nxp.com] Sent: Wednesday, April 17, 2019 10:58 AM To: Ayuj Verma ; Trahe, Fiona ; Kusztal, ArkadiuszX Cc: Shally Verma ; Sunila Sahu ; Kanaka Durga Kotamarthy ; Arvind Desai ; dev@dpdk.org Subject: Re: [PATCH v1] fix alphabetical ordering o

Re: [dpdk-dev] [PATCH] doc: add cryptodev gcm iv deprecation notice

> -Original Message- > From: Kusztal, ArkadiuszX > Sent: Wednesday, April 17, 2019 8:41 AM > To: dev@dpdk.org > Cc: akhil.go...@nxp.com; Trahe, Fiona ; De Lara > Guarch, Pablo > ; Kusztal, ArkadiuszX > > Subject: [PATCH] doc: add cryptodev gcm iv deprecation notice > > This patch add

Re: [dpdk-dev] [PATCH] cryptodev: add an option to support both iv and J0 for GCM

> > This patch adds an option to support both IV (of all supported sizes) > and J0 when using Galois Counter Mode of crypto operation. > > Signed-off-by: Arek Kusztal > --- > lib/librte_cryptodev/rte_crypto_sym.h | 37 > ++- > 1 file changed, 19 insertions(+), 1

Re: [dpdk-dev] [PATCH] cryptodev: add an option to support both iv and J0 for GCM

Hi Arek, > -Original Message- > From: Kusztal, ArkadiuszX > Sent: Wednesday, April 17, 2019 8:37 AM > To: dev@dpdk.org > Cc: akhil.go...@nxp.com; Trahe, Fiona ; De Lara > Guarch, Pablo > ; Kusztal, ArkadiuszX > > Subject: [PATCH] cryptodev: add an option to support both iv and J0 for GC

Re: [dpdk-dev] [PATCH v3 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto

> > Inline crypto installs a flow rule in the NIC. This flow > rule must be installed before the first inbound packet is > received. > > The create_session() function installs the flow rule. > > Refactor ipsec-secgw.c, sa.c, ipsec.h and ipsec.c to create > sessions at startup to fix the issue of

Re: [dpdk-dev] [PATCH v2 0/3] travis: enhancements for build (plus a meson fix)

12/04/2019 18:21, Aaron Conole: > This series improves the travis integration a bit, and adds the ability for > the meson system to auto-detect the number of cores. > > v1->v2 (all patch 3/3 changes): > * Fix a spelling mistake > * Add support for FreeBSD > * Include a default fallback > * Use a m

Re: [dpdk-dev] [PATCH v4] power: update for handling fifo path string

On 16/4/2019 11:20 AM, Lukasz Gosiewski wrote: From: Lukasz Krakowiak Removed doubled created fifo path string for channel info. Signed-off-by: Lukasz Krakowiak Signed-off-by: Lukasz Gosiewski --- v4: *improvement to coding style v3: *improvement to coding style v2: * rebase to master c

[dpdk-dev] [BUG] net/af_xdp: Current code can only create one af_xdp device

I tested the new af_xdp based device on the current master branch and noticed, that the usage of static mempool names allows only for the creation of a single af_xdp vdev. If a second vdev of the same type gets created, the mempool allocation fails. Best regards, Markus Theil

Re: [dpdk-dev] [PATCH] net/iavf: fix info get

On 4/17/2019 9:46 AM, David Marchand wrote: > On Wed, Apr 17, 2019 at 10:41 AM Qiming Yang wrote: > >> iavf driver reset the dev_info structure. This action will clear >> the setting in core library. This patch fix that issue. >> >> Fixes: 69dd4c3d0898 ("net/avf: enable queue and device") >> > "C

Re: [dpdk-dev] [RFC 3/3] ci: enable tests on non-arm platforms

Jerin Jacob Kollanukkaran writes: >> -Original Message- >> From: dev On Behalf Of Aaron Conole >> Sent: Friday, March 29, 2019 10:53 PM >> To: dev@dpdk.org >> Subject: [dpdk-dev] [RFC 3/3] ci: enable tests on non-arm platforms >> >> When building under Travis (or another linux CI servic

Re: [dpdk-dev] [PATCH v3 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto

Hi Akhil, > -Original Message- > From: Akhil Goyal [mailto:akhil.go...@nxp.com] > Sent: Wednesday, April 17, 2019 12:52 PM > To: Iremonger, Bernard ; dev@dpdk.org; > Ananyev, Konstantin > Cc: sta...@dpdk.org > Subject: RE: [PATCH v3 1/2] examples/ipsec-secgw: fix 1st packet dropped > for

Re: [dpdk-dev] [EXT] RE: [PATCH v3 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto

> > > + rte_eth_dev_rss_hash_conf_get(sa->portid, > > > + &rss_conf); > > > + for (i = 0, j = 0; i < eth_dev->data->nb_rx_queues; > > > + ++i) > > > + if (eth_dev->dat

Re: [dpdk-dev] [PATCH] cryptodev: add an option to support both iv and J0 for GCM

Hi Akhil, Yes, it's intended for 19.08 > -Original Message- > From: Akhil Goyal [mailto:akhil.go...@nxp.com] > Sent: Wednesday, April 17, 2019 12:34 PM > To: Kusztal, ArkadiuszX ; dev@dpdk.org > Cc: Trahe, Fiona ; De Lara Guarch, Pablo > > Subject: RE: [PATCH] cryptodev: add an option to

Re: [dpdk-dev] [PATCH v3 2/4] net/af_xdp: specify minimal and maximal MTU

On 04/17, David Marchand wrote: >On Wed, Apr 17, 2019 at 11:02 AM Xiaolong Ye wrote: > >> Since AF_XDP pmd doesn't support multi segment, we need to add a valid >> check in eth_dev_mtu_set. >> > >How about: >Properly report mtu capability in port device info. Sounds better. > > >Reported-by: Dav

Re: [dpdk-dev] [PATCH v3 1/4] net/af_xdp: enqueue buf ring when allocate Tx queue fails

On 04/17, David Marchand wrote: >On Wed, Apr 17, 2019 at 11:02 AM Xiaolong Ye wrote: > >> When it fails to allocate enough slots in Tx queue for transmitting >> packets, we need to return the dequeued addrs to buf ring. >> >> Fixes: f1debd77efaf ("net/af_xdp: introduce AF_XDP PMD") >> >> Signed-of

Re: [dpdk-dev] [PATCH v5 1/3] rcu: add RCU library supporting QSBR mechanism

> > > > > > > > > > > > > > > > > > > > > > > > After evaluating long term API/ABI issues, I think > > > > > > > > > > > > you need to get rid of almost all use of inline and > > > > > > > > > > > > visible structures. Yes it might be marginally > > > > > > > > > > > > slower, but you thank me >

[dpdk-dev] [PATCH v4 0/2] examples/ipsec-secgw: fix 1st pkt dropped

This patchset fixes the issue of the first inbound packet being dropped for inline crypto. Changes in v4: -- Fix indentation at line 173 in ipsec.c to solve compile error on ARM. Changes in v3: -- The previous refactoring of the create_session() function has been dropped.

[dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto

Inline crypto installs a flow rule in the NIC. This flow rule must be installed before the first inbound packet is received. The create_session() function installs the flow rule. Refactor ipsec-secgw.c, sa.c, ipsec.h and ipsec.c to create sessions at startup to fix the issue of the first packet b

[dpdk-dev] [PATCH v4 2/2] examples/ipsec-secgw/test: fix inline test scripts

Remove workaround in tun_aesgcm_defs.sh and trs_aesgcm_defs.sh to get around the bug where the first inbound packet is dropped for inline crypto. Fixes: 929784452094 ("examples/ipsec-secgw: add scripts for functional test") Cc: sta...@dpdk.org Signed-off-by: Bernard Iremonger Acked-by: Konstanti

[dpdk-dev] [PATCH v4 3/4] net/af_xdp: make reserve/submit peek/release consistent

As David pointed out, if we reserve N slots for Tx, but only submit n slots, we would end up with an incorrect opinion of the number of available slots later, we also would get wrong idx when we call xsk_ring_prod__reserve next time. It also applies to xsk_ring_cons__peek()/xsk_ring_cons__release()

[dpdk-dev] [PATCH v4 4/4] net/af_xdp: fix typos in Rx function

We should use ETH_AF_XDP_RX_BATCH_SIZE in Rx function rather than ETH_AF_XDP_TX_BATCH_SIZE. Fixes: f1debd77efaf ("net/af_xdp: introduce AF_XDP PMD") Reported-by: David Marchand Reviewed-by: David Marchand Signed-off-by: Xiaolong Ye --- drivers/net/af_xdp/rte_eth_af_xdp.c | 4 ++-- 1 file chan

[dpdk-dev] [PATCH v4 1/4] net/af_xdp: enqueue buf ring when allocate Tx queue fails

When it fails to allocate enough slots in Tx queue for transmitting packets, we need to return the dequeued addrs to buf ring. Fixes: f1debd77efaf ("net/af_xdp: introduce AF_XDP PMD") Reviewed-by: David Marchand Signed-off-by: Xiaolong Ye --- drivers/net/af_xdp/rte_eth_af_xdp.c | 1 + 1 file c

[dpdk-dev] [PATCH v4 0/4] some fixes for AF_XDP pmd

This patchset provides some fixes to af_xdp pmd, at first, I just added a simple error handling when Tx queue allocation fails, then David suggested a better way to do it and pointed out the inconsistent issue of reserve/submit ops (for Tx queue) and peek/release ops (for Rx queue), the third patch

[dpdk-dev] [PATCH v4 2/4] net/af_xdp: specify minimal and maximal MTU

Properly report mtu capability in port device info. Reported-by: David Marchand Reviewed-by: David Marchand Signed-off-by: Xiaolong Ye --- drivers/net/af_xdp/rte_eth_af_xdp.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/af_xdp/rte_eth_af_xdp.c b/drivers/net/af_xdp/rte_et

Re: [dpdk-dev] [PATCH v5 1/3] rcu: add RCU library supporting QSBR mechanism

> > > > > > > > > > > > > > > > > > > > > > > > > > After evaluating long term API/ABI issues, I > > > > > > > > > > > > > think you need to get rid of almost all use of > > > > > > > > > > > > > inline and visible structures. Yes it might be > > > > > > > > > > > > > marginally slower, but you tha

Re: [dpdk-dev] [PATCH v4 0/4] some fixes for AF_XDP pmd

On Wed, Apr 17, 2019 at 3:55 PM Xiaolong Ye wrote: > This patchset provides some fixes to af_xdp pmd, at first, I just added > a simple error handling when Tx queue allocation fails, then David > suggested a better way to do it and pointed out the inconsistent issue > of reserve/submit ops (for T

Re: [dpdk-dev] [PATCH v5 1/3] rcu: add RCU library supporting QSBR mechanism

17/04/2019 15:39, Ananyev, Konstantin: > In general, I still think that sync primitives better to stay inlined - there > is no much point to create ones > and then figure out that no-one using them because they are too slow. > Though if there is no real perf difference between inlined and normal -

Re: [dpdk-dev] [PATCH v2] net/bonding: fix potential out of bounds read

On 4/17/2019 10:33 AM, Radu Nicolau wrote: > Add validation to pointer constructed from the IPv4 header length > in order to prevent malformed packets from generating a potential > out of bounds memory read. > > Fixes: 09150784a776 ("net/bonding: burst mode hash calculation") > Cc: sta...@dpdk.org

[dpdk-dev] [PATCH 1/8] ipc: fix rte_mp_request_sync memleak

When sending multiple requests, rte_mp_request_sync can succeed sending a few of those requests, but then fail on a later one and in the end return with rc=-1. The upper layers - e.g. device hotplug - currently handles this case as if no messages were sent and no memory for response buffers was all

[dpdk-dev] [PATCH 2/8] ipc: fix hotplug memleak

When sending multiple requests, rte_mp_request_sync can succeed sending a few of those requests, but then fail on a later one and in the end return with rc=-1. The upper layers - e.g. device hotplug - currently handles this case as if no messages were sent and no memory for response buffers was all

[dpdk-dev] [PATCH 3/8] ipc: fix vdev memleak

When sending multiple requests, rte_mp_request_sync can succeed sending a few of those requests, but then fail on a later one and in the end return with rc=-1. The upper layers - e.g. device hotplug - currently handles this case as if no messages were sent and no memory for response buffers was all

[dpdk-dev] [PATCH 4/8] ipc: fix vfio memleak

When sending multiple requests, rte_mp_request_sync can succeed sending a few of those requests, but then fail on a later one and in the end return with rc=-1. The upper layers - e.g. device hotplug - currently handles this case as if no messages were sent and no memory for response buffers was all

[dpdk-dev] [PATCH 5/8] ipc: fix pdump memleak

When sending multiple requests, rte_mp_request_sync can succeed sending a few of those requests, but then fail on a later one and in the end return with rc=-1. The upper layers - e.g. device hotplug - currently handles this case as if no messages were sent and no memory for response buffers was all

[dpdk-dev] [PATCH 6/8] ipc: fix tap pmd memleak

When sending multiple requests, rte_mp_request_sync can succeed sending a few of those requests, but then fail on a later one and in the end return with rc=-1. The upper layers - e.g. device hotplug - currently handles this case as if no messages were sent and no memory for response buffers was all

[dpdk-dev] [PATCH v3] net/bonding: fix potential out of bounds read

Add validation to pointer constructed from the IPv4 header length in order to prevent malformed packets from generating a potential out of bounds memory read. Fixes: 09150784a776 ("net/bonding: burst mode hash calculation") Cc: sta...@dpdk.org Signed-off-by: Radu Nicolau --- v2: add fixes lines

[dpdk-dev] [PATCH 7/8] ipc: fix net/mlx4 memleak

When sending multiple requests, rte_mp_request_sync can succeed sending a few of those requests, but then fail on a later one and in the end return with rc=-1. The upper layers - e.g. device hotplug - currently handles this case as if no messages were sent and no memory for response buffers was all

[dpdk-dev] [PATCH 8/8] ipc: fix net/mlx5 memleak

When sending multiple requests, rte_mp_request_sync can succeed sending a few of those requests, but then fail on a later one and in the end return with rc=-1. The upper layers - e.g. device hotplug - currently handles this case as if no messages were sent and no memory for response buffers was all

Re: [dpdk-dev] [dpdk-stable] [PATCH 1/3] devtools/test-meson-builds: skip missing compilers

10/04/2019 22:52, Bruce Richardson: > if either gcc or clang are missing, skip doing those builds. > This allows a setup to only do, e.g. gcc tests. > > CC: sta...@dpdk.org > Signed-off-by: Bruce Richardson > --- > --- a/devtools/test-meson-builds.sh > +++ b/devtools/test-meson-builds.sh > # sha

Re: [dpdk-dev] [dpdk-stable] [PATCH 1/3] devtools/test-meson-builds: skip missing compilers

On Wed, Apr 17, 2019 at 04:59:53PM +0200, Thomas Monjalon wrote: > 10/04/2019 22:52, Bruce Richardson: > > if either gcc or clang are missing, skip doing those builds. > > This allows a setup to only do, e.g. gcc tests. > > > > CC: sta...@dpdk.org > > Signed-off-by: Bruce Richardson > > --- > > -

Re: [dpdk-dev] [PATCH 0/3] fix test-meson-builds for older distros

11/04/2019 10:48, Luca Boccassi: > On Wed, 2019-04-10 at 21:52 +0100, Bruce Richardson wrote: > > In order to ensure good test coverage, we sometimes need to run > > builds > > on distros with older compilers, such as gcc 4.8 on RHEL/Centos 7. > > This > > patchset does a small amount of clean-up t

[dpdk-dev] [PATCH] lib/bitratestats: fix unchecked return value

Checking the return value of rte_metrics_update_values, if failed returning that value. Coverity had picked up that that the return value wasn't being checked. Coverity issue: 336863 Fixes: 2ad7ba9a6567 ("bitrate: add bitrate statistics library") Cc: remy.hor...@intel.com Signed-off-by: Andrius S

Re: [dpdk-dev] [PATCH v4 4/4] net/af_xdp: fix typos in Rx function

Acked-by: Rami Rosen

Re: [dpdk-dev] [EXT] Re: [RFC 3/3] ci: enable tests on non-arm platforms

> -Original Message- > From: Aaron Conole > Sent: Wednesday, April 17, 2019 6:10 PM > To: Jerin Jacob Kollanukkaran > Cc: dev@dpdk.org > Subject: [EXT] Re: [dpdk-dev] [RFC 3/3] ci: enable tests on non-arm platforms > > -

Re: [dpdk-dev] [PATCH v4 0/4] some fixes for AF_XDP pmd

On 04/17, David Marchand wrote: >On Wed, Apr 17, 2019 at 3:55 PM Xiaolong Ye wrote: > >> This patchset provides some fixes to af_xdp pmd, at first, I just added >> a simple error handling when Tx queue allocation fails, then David >> suggested a better way to do it and pointed out the inconsistent

Re: [dpdk-dev] [PATCH] cryptodev: add an option to support both iv and J0 for GCM

Hi Fiona, > [Fiona] IVs greater than 16 bytes can be used, right? If so change to "a > minimum of 16 bytes must be >allocated". Same below. [AK] - yes, this comment may be misleading, of course it means 16 bytes minimum. This is because of legacy reasons, i.e. aesni-gcm once had formation of p

[dpdk-dev] [PATCH v2 1/7] net/atlantic: macsec hardware structures declaration

From: Pavel Belous Here we define hardware and software configuration structures for macsec interface. MACSEC itself is implemented in Phy module, but its configuration is done via firmware interface Signed-off-by: Pavel Belous Signed-off-by: Igor Russkikh --- drivers/net/atlantic/atl_types.h

[dpdk-dev] [PATCH v2 0/7] add MACSEC hw offload to atlantic PMD

This patchset implements MACSEC hardware offload configuration in Atlantic PMD via experimental PMD API. There is an agreement to work on 19.08 timeframe for the generic MACSEC API inside of rte_security. v2 changes: - Removed public API changes for macsec, leave only PMD experimental API for M

[dpdk-dev] [PATCH v2 2/7] net/atlantic: enable macsec configuration

From: Pavel Belous This is a driver side of macsec configuration routines. It fills in config structures and sends requests to FW for configuration activities. We declare macsec offload bits in DPDK offload capabilities. Also update documentation with feature matrix for the enabled feature. Si

[dpdk-dev] [PATCH v2 3/7] net/atlantic: macsec firmware interface

From: Pavel Belous Implementation of firmware interface for macsec configuration. Structure with config data is written into FW memory, then we trigger FW to execute the request and wait for result. Signed-off-by: Pavel Belous Signed-off-by: Igor Russkikh --- drivers/net/atlantic/atl_types.h

[dpdk-dev] [PATCH v2 4/7] net/atlantic: interrupt handling of macsec events

From: Pavel Belous MACSEC should be configured only after link up event, thus we use link interrupt to file an alarm for configuration. FW also uses link interrupt line to indicate incoming events from MACSEC. These may include key expiration, packet counter wrap, etc. We pass these events to th

[dpdk-dev] [PATCH v2 5/7] net/atlantic: implement macsec statistics

From: Pavel Belous We add extra xstat fields to include macsec counters and stats Signed-off-by: Pavel Belous Signed-off-by: Igor Russkikh --- drivers/net/atlantic/atl_ethdev.c | 82 +-- 1 file changed, 77 insertions(+), 5 deletions(-) diff --git a/drivers/net/atl

[dpdk-dev] [PATCH v2 7/7] net/atlantic: public experimental macsec API

Add public visible macsec API functions. These are wrappers over internal configuration routines, validating and converting port no into dev pointer Signed-off-by: Igor Russkikh --- drivers/net/atlantic/Makefile | 1 + drivers/net/atlantic/meson.build | 1 + driv

[dpdk-dev] [PATCH v2 6/7] net/atlantic: bump internal driver version

Version is synced with internal Aquantia's driver versioning Signed-off-by: Pavel Belous Signed-off-by: Igor Russkikh --- drivers/net/atlantic/atl_common.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/atlantic/atl_common.h b/drivers/net/atlantic/atl_common.h

[dpdk-dev] [PATCH v2] cryptodev: add an option to support both iv and J0 for GCM

This patch adds an option to support both IV (of all supported sizes) and J0 when using Galois Counter Mode of crypto operation. Signed-off-by: Arek Kusztal --- Note: This patch is intended for 19.08 release lib/librte_cryptodev/rte_crypto_sym.h | 37 ++- 1 file

  1   2   >