Re: [ovs-dev] [PATCH] netdev-afxdp: Fix umem creation failure due to uninitialized config.

2019-10-11 Thread Ilya Maximets
On 10.10.2019 20:21, William Tu wrote: On Wed, Oct 09, 2019 at 04:17:58PM +0200, Ilya Maximets wrote: Later version of 'struct xsk_umem_config' contains additional field 'flags'. OVS doesn't use that field passing uninitialized stack memory to the 'xsk_umem__create()' call that could fail

Re: [ovs-dev] [PATCH] netdev-afxdp: Update memory locking limits unconditionally.

2019-10-11 Thread Ilya Maximets
On 10.10.2019 20:29, William Tu wrote: On Wed, Oct 09, 2019 at 04:23:31PM +0200, Ilya Maximets wrote: Any type of AF_XDP socket in all modes implies creation of BPF map of type BPF_MAP_TYPE_XSKMAP. This leads to BPF_MAP_CREATE syscall and subsequently 'xsk_map_alloc()' function

Re: [ovs-dev] [PATCH] tests: Allow valgrind check for afxdp testsuite.

2019-10-11 Thread Ilya Maximets
On 10.10.2019 22:51, William Tu wrote: On Wed, Oct 09, 2019 at 04:18:38PM -0700, William Tu wrote: On Wed, Oct 09, 2019 at 11:20:00AM -0700, Ben Pfaff wrote: On Wed, Oct 09, 2019 at 04:36:49PM +0200, Ilya Maximets wrote: New 'make' target 'check-afxdp-valgrind'. Signed-off-by: Ilya Maximets

Re: [ovs-dev] [PATCH] cirrus: Use latest stable FreeBSD images.

2019-10-10 Thread Ilya Maximets
On 10.10.2019 16:31, Aaron Conole wrote: Ilya Maximets writes: CirrusCI recently introduced [1] new feature to use image families instead of bare image names for gCloud based instances. This allows us to use most recent stable builds. All the stable builds are in the same image family

[ovs-dev] [PATCH] cirrus: Use latest stable FreeBSD images.

2019-10-10 Thread Ilya Maximets
/cirrus-ci-docs/issues/422 [2] https://github.com/cirruslabs/cirrus-ci-docs/issues/359 Signed-off-by: Ilya Maximets --- .cirrus.yml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/.cirrus.yml b/.cirrus.yml index 251f6734d..281bd75de 100644 --- a/.cirrus.yml +++ b

[ovs-dev] [PATCH ovn] pinctrl: Update icmp6 header structures.

2019-10-10 Thread Ilya Maximets
commit 361a47d66 ("flow: Fix using pointer to member of packed struct icmp6_hdr.") in OVS updated 'struct icmp6_error_header' with renaming it to 'struct icmp6_data_header'. OVN should update names of a structure and constants accordingly. Signed-off-by: Ilya Maximets --- This is a

Re: [ovs-dev] [PATCH] flow: Fix using pointer to member of packed struct icmp6_hdr.

2019-10-10 Thread Ilya Maximets
On 09.10.2019 20:18, Ben Pfaff wrote: On Wed, Oct 09, 2019 at 06:54:29PM +0200, Ilya Maximets wrote: On 08.10.2019 18:55, William Tu wrote: On Tue, Oct 01, 2019 at 08:04:00PM +0300, Ilya Maximets wrote: OVS has no structure definition for ICMPv6 header with additional data. More precisely

Re: [ovs-dev] [PATCH] flow: Fix using pointer to member of packed struct icmp6_hdr.

2019-10-09 Thread Ilya Maximets
On 08.10.2019 18:55, William Tu wrote: On Tue, Oct 01, 2019 at 08:04:00PM +0300, Ilya Maximets wrote: OVS has no structure definition for ICMPv6 header with additional data. More precisely, it has, but this structure named as 'icmp6_error_header' and only suitable to store error related

[ovs-dev] [PATCH] tests: Allow valgrind check for afxdp testsuite.

2019-10-09 Thread Ilya Maximets
New 'make' target 'check-afxdp-valgrind'. Signed-off-by: Ilya Maximets --- tests/automake.mk | 6 ++ 1 file changed, 6 insertions(+) diff --git a/tests/automake.mk b/tests/automake.mk index 9dd07a2d9..4bf8f00d5 100644 --- a/tests/automake.mk +++ b/tests/automake.mk @@ -282,6 +282,12

[ovs-dev] [PATCH] netdev-afxdp: Update memory locking limits unconditionally.

2019-10-09 Thread Ilya Maximets
type for AF_XDP.") Signed-off-by: Ilya Maximets --- lib/netdev-afxdp.c | 19 ++- 1 file changed, 6 insertions(+), 13 deletions(-) diff --git a/lib/netdev-afxdp.c b/lib/netdev-afxdp.c index 6e0180327..50ac1c5a5 100644 --- a/lib/netdev-afxdp.c +++ b/lib/netdev-afxdp.c @@ -524,

[ovs-dev] [PATCH] netdev-afxdp: Fix umem creation failure due to uninitialized config.

2019-10-09 Thread Ilya Maximets
need to clear the whole structure explicitly to avoid this kind of issues. Fixes: 0de1b425962d ("netdev-afxdp: add new netdev type for AF_XDP.") Signed-off-by: Ilya Maximets --- lib/netdev-afxdp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/lib/netdev-afxdp.c b/lib/netdev-afx

Re: [ovs-dev] [PATCH 2/2] backtrace: Fix 32-bit libunwind build.

2019-10-08 Thread Ilya Maximets
On 08.10.2019 18:13, William Tu wrote: On Mon, Oct 07, 2019 at 07:10:26PM +0200, Ilya Maximets wrote: On 07.10.2019 18:11, Ilya Maximets wrote: On 07.10.2019 18:07, William Tu wrote: On Mon, Oct 7, 2019 at 5:06 AM Ilya Maximets wrote: On 04.10.2019 23:21, William Tu wrote: The libunwind

Re: [ovs-dev] [PATCH 2/2] backtrace: Fix 32-bit libunwind build.

2019-10-07 Thread Ilya Maximets
On 07.10.2019 18:11, Ilya Maximets wrote: On 07.10.2019 18:07, William Tu wrote: On Mon, Oct 7, 2019 at 5:06 AM Ilya Maximets wrote: On 04.10.2019 23:21, William Tu wrote: The libunwind unw_word_t type is defined as uint32_t for 32-bit system and uint64_t for 64-bit system.  The patch fixes

Re: [ovs-dev] [PATCH 2/2] backtrace: Fix 32-bit libunwind build.

2019-10-07 Thread Ilya Maximets
On 07.10.2019 18:07, William Tu wrote: On Mon, Oct 7, 2019 at 5:06 AM Ilya Maximets wrote: On 04.10.2019 23:21, William Tu wrote: The libunwind unw_word_t type is defined as uint32_t for 32-bit system and uint64_t for 64-bit system. The patch fixes the compile error using PRIxPTR to print

Re: [ovs-dev] [PATCH 2/2] backtrace: Fix 32-bit libunwind build.

2019-10-07 Thread Ilya Maximets
Signed-off-by: William Tu Acked-by: Ilya Maximets --- Thanks! I applied both patches to master, but in the reverse order. Best regards, Ilya Maximets. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] travis: Fix 32-bit libunwind system build.

2019-10-04 Thread Ilya Maximets
On 04.10.2019 2:34, William Tu wrote: On Thu, Oct 3, 2019 at 10:15 AM Ilya Maximets wrote: On 03.10.2019 18:13, Ilya Maximets wrote: On 02.10.2019 20:15, William Tu wrote: 32-bit and 64-bit libunwind can not be installed at the same time. For 32-bit build, this patch removes the 64-bit

Re: [ovs-dev] [PATCH v2] ci: Get rid of OVS_CFLAGS in CI scripts.

2019-10-04 Thread Ilya Maximets
On 04.10.2019 19:31, Ben Pfaff wrote: On Fri, Oct 04, 2019 at 04:37:38PM +0200, Ilya Maximets wrote: Our CI scripts uses OVS_CFLAGS variable that is intended for internal usage by 'configure' script only. Usual CFLAGS should be used instead to avoid giving bad example to users. Additionally

Re: [ovs-dev] [PATCH] doc: Fix incorrect reference for dpdk-testpmd.

2019-10-04 Thread Ilya Maximets
Signed-off-by: David Marchand Acked-by: Kevin Traynor Thanks David and Kevin! Applied to master and branch-2.12. Best regards, Ilya Maximets. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

[ovs-dev] [PATCH v2] ci: Get rid of OVS_CFLAGS in CI scripts.

2019-10-04 Thread Ilya Maximets
. Suggested-by: Ben Pfaff Signed-off-by: Ilya Maximets --- Version 2: * CFLAGS passed to './configure' where possible (except 'distcheck' case). .cirrus.yml| 3 ++- .travis.yml| 2 +- .travis/linux-build.sh | 22 +- 3 files changed, 16 insertions

Re: [ovs-dev] [PATCH] ci: Get rid of OVS_CFLAGS in CI scripts.

2019-10-03 Thread Ilya Maximets
On 03.10.2019 15:36, Ben Pfaff wrote: On Thu, Oct 03, 2019 at 05:47:29PM +0200, Ilya Maximets wrote: Our CI scripts uses OVS_CFLAGS variable that is intended for internal usage by 'configure' script only. Usual CFLAGS should be used instead to avoid giving bad example to users. Additionally

Re: [ovs-dev] [PATCH] travis: Fix 32-bit libunwind system build.

2019-10-03 Thread Ilya Maximets
On 03.10.2019 18:13, Ilya Maximets wrote: On 02.10.2019 20:15, William Tu wrote: 32-bit and 64-bit libunwind can not be installed at the same time. For 32-bit build, this patch removes the 64-bit libunwind and install 32-bit version. Tested-at: https://travis-ci.org/williamtu/ovs-travis/builds

Re: [ovs-dev] [PATCH] travis: Fix 32-bit libunwind system build.

2019-10-03 Thread Ilya Maximets
he order of applying. i.e. following diff should be squashed to one of them: -if [[ $BUILD_ENV =~ "-m32" ]]; then +if [ "$M32" ]; then Other than that, Acked-by: Ilya Maximets ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [dpdk-latest PATCH 2/2] travis: Switch to dpdk master branch.

2019-10-03 Thread Ilya Maximets
ut it might be better to just export DPDK_VER=refs/heads/master in the 'before_script:' section like we're exporting PATH. This will be less invasive and easier to maintain. What do you think? Best regards, Ilya Maximets. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

[ovs-dev] [PATCH] ci: Get rid of OVS_CFLAGS in CI scripts.

2019-10-03 Thread Ilya Maximets
. Suggested-by: Ben Pfaff Signed-off-by: Ilya Maximets --- .cirrus.yml| 4 ++-- .travis.yml| 2 +- .travis/linux-build.sh | 23 +-- 3 files changed, 16 insertions(+), 13 deletions(-) diff --git a/.cirrus.yml b/.cirrus.yml index 654e63ee0..1080729fe

Re: [ovs-dev] [PATCH] m4: Fix 32bit build with 64bit version of libunwind-dev installed.

2019-10-02 Thread Ilya Maximets
On 02.10.2019 17:46, Ben Pfaff wrote: On Tue, Oct 01, 2019 at 06:40:10PM -0700, William Tu wrote: On Tue, Oct 1, 2019 at 4:19 PM William Tu wrote: On Tue, Oct 01, 2019 at 08:29:41PM +0300, Ilya Maximets wrote: On 01.10.2019 19:56, Ben Pfaff wrote: On Tue, Oct 01, 2019 at 07:46:10PM +0300

Re: [ovs-dev] [PATCH] m4: Fix 32bit build with 64bit version of libunwind-dev installed.

2019-10-01 Thread Ilya Maximets
On 01.10.2019 19:56, Ben Pfaff wrote: On Tue, Oct 01, 2019 at 07:46:10PM +0300, Ilya Maximets wrote: On 01.10.2019 19:17, Ben Pfaff wrote: On Tue, Oct 01, 2019 at 02:22:20PM +0300, Ilya Maximets wrote: 32 and 64 bit versions of libunwind-dev package can not be installed at the same time

[ovs-dev] [PATCH] flow: Fix using pointer to member of packed struct icmp6_hdr.

2019-10-01 Thread Ilya Maximets
uot;Support for match & set ICMPv6 reserved and options type fields") Signed-off-by: Ilya Maximets --- lib/conntrack.c | 2 +- lib/flow.c | 77 + lib/packets.h | 12 +--- 3 files changed, 48 insertions(+), 43 deletions(-) diff --g

Re: [ovs-dev] [PATCH] m4: Fix 32bit build with 64bit version of libunwind-dev installed.

2019-10-01 Thread Ilya Maximets
On 01.10.2019 19:17, Ben Pfaff wrote: On Tue, Oct 01, 2019 at 02:22:20PM +0300, Ilya Maximets wrote: 32 and 64 bit versions of libunwind-dev package can not be installed at the same time at least in Debian and Ubuntu. AC_CHECK_LIB macro doesn't know if we have '-m32' in OVS_CFLAGS, so we need

Re: [ovs-dev] [PATCHv4] fatal-signal: Catch SIGSEGV and print backtrace.

2019-10-01 Thread Ilya Maximets
On 01.10.2019 12:01, Ilya Maximets wrote: On Fri, Sep 27, 2019 at 10:22:55AM -0700, William Tu wrote: The patch catches the SIGSEGV signal and prints the backtrace using libunwind at the monitor daemon. This makes debugging easier when there is no debug symbol package or gdb installed

[ovs-dev] [PATCH] m4: Fix 32bit build with 64bit version of libunwind-dev installed.

2019-10-01 Thread Ilya Maximets
print backtrace.") Signed-off-by: Ilya Maximets --- m4/openvswitch.m4 | 16 +++- 1 file changed, 15 insertions(+), 1 development) diff --git a/m4/openvswitch.m4 b/m4/openvswitch.m4 index 79e0be5a3..fa822eab7 100644 --- a/m4/openvswitch.m4 +++ b/m4/openvswitch.m4 @@ -640

Re: [ovs-dev] [PATCH] cirrus: Fix typo in the name of sphinx package.

2019-10-01 Thread Ilya Maximets
On 01.10.2019 13:02, Numan Siddique wrote: On Tue, Oct 1, 2019, 2:58 PM Ilya Maximets wrote: This fixes broken build on CirrusCI: https://cirrus-ci.com/build/4677435792031744 CC: Ben Pfaff mailto:b...@ovn.org>> Fixes: 1ca0323e7c29 ("Require Python 3 and rem

[ovs-dev] [PATCH] cirrus: Fix typo in the name of sphinx package.

2019-10-01 Thread Ilya Maximets
This fixes broken build on CirrusCI: https://cirrus-ci.com/build/4677435792031744 CC: Ben Pfaff Fixes: 1ca0323e7c29 ("Require Python 3 and remove support for Python 2.") Signed-off-by: Ilya Maximets --- .cirrus.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.

Re: [ovs-dev] [PATCHv4] fatal-signal: Catch SIGSEGV and print backtrace.

2019-10-01 Thread Ilya Maximets
applied this to master. Hi. This patch broke the 32bit gcc build on travis: https://travis-ci.org/openvswitch/ovs/jobs/590580374 /usr/include/libunwind.h:23:28: fatal error: libunwind-x86.h: No such file or directory Could you, please, take a look? Best regards, Ilya Maximets

Re: [ovs-dev] [PATCHv2] netdev-afxdp: Detect numa node id.

2019-09-30 Thread Ilya Maximets
from the filesystem? 3. More comments inline. Best regards, Ilya Maximets. On 27.09.2019 20:26, William Tu wrote: The patch detects the numa node id from the name of the netdev, by reading the '/sys/class/net//device/numa_node'. If not available, ex: virtual device, or any error happens

Re: [ovs-dev] dpdk-latest rebase

2019-09-27 Thread Ilya Maximets
On 27.09.2019 10:43, David Marchand wrote: On Thu, Sep 26, 2019 at 9:45 PM Stokes, Ian wrote: On 9/26/2019 1:09 PM, David Marchand wrote: On Thu, Sep 26, 2019 at 11:41 AM Ilya Maximets wrote: On 26.09.2019 12:08, Stokes, Ian wrote: Hi all, the dpdk-latest branch has been lagging behind

Re: [ovs-dev] [PATCH] travis: Drop -MD related workaround for sparse.

2019-09-26 Thread Ilya Maximets
On 26.09.2019 13:33, Stokes, Ian wrote: On 9/26/2019 10:34 AM, Ilya Maximets wrote: The issue was fixed in upstream sparse by the following commit: d90c0838c101 ("cgcc: fix wrong processing of -MD & -MMD") This patch is required to fix our travis build. Signed-off-by:

Re: [ovs-dev] [PATCH v3] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

2019-09-26 Thread Ilya Maximets
On 24.09.2019 16:15, Eelco Chaudron wrote: On 12 Sep 2019, at 12:24, Ilya Maximets wrote: On 12.09.2019 13:19, Ilya Maximets wrote: On 12.09.2019 13:07, Eelco Chaudron wrote: On 12 Sep 2019, at 10:39, Ilya Maximets wrote: On 11.09.2019 16:20, Eelco Chaudron wrote: Currently, OVS does

Re: [ovs-dev] [PATCH branch-2.10] dpdk: Use DPDK 17.11.6 release.

2019-09-26 Thread Ilya Maximets
On 26.09.2019 12:46, Kevin Traynor wrote: On 26/09/2019 09:44, Stokes, Ian wrote: On 9/25/2019 10:50 PM, Ben Pfaff wrote: It looks like these DPDK version update patches have been overlooked. Ilya, Ian, do you want to look at them? Hi Ben, sure I can look at them, I had worked with

Re: [ovs-dev] dpdk-latest rebase

2019-09-26 Thread Ilya Maximets
bility that some additional fixes required. Best regards, Ilya Maximets. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

[ovs-dev] [PATCH] travis: Drop -MD related workaround for sparse.

2019-09-26 Thread Ilya Maximets
The issue was fixed in upstream sparse by the following commit: d90c0838c101 ("cgcc: fix wrong processing of -MD & -MMD") This patch is required to fix our travis build. Signed-off-by: Ilya Maximets --- This should be applied to master and backported to 2.12. .travis/linux-

Re: [ovs-dev] [PATCH 1/2] cgcc: gendeps for -MM, -MD & -MMD too

2019-09-26 Thread Ilya Maximets
On Wed, Apr 24, 2019 at 04:12:32PM +0300, Ilya Maximets wrote: On 20.02.2019 16:34, Luc Van Oostenryck wrote: > These flags must set '$gendeps', just like a plain '-M' do, > since they implies '-M'. > > Signed-off-by: Luc Van Oostenryck > --- Hi. Unlike simple '-M', '-MD' and

Re: [ovs-dev] [PATCHv4] netdev-afxdp: Add need_wakeup supprt.

2019-09-24 Thread Ilya Maximets
Hi. Thanks for a new version. Comments inline. Best regards, Ilya Maximets. On 17.09.2019 21:36, William Tu wrote: The patch adds support for using need_wakeup flag in AF_XDP rings. A new option, use_need_wakeup, is added. When this option is used, it means that OVS has to explicitly wake up

Re: [ovs-dev] [PATCH v7 1/1] Avoid dp_hash recirculation for balance-tcp bond selection mode

2019-09-24 Thread Ilya Maximets
n one reloading point makes code more complex and less readable. One more thought is that we could introduce some version sequence number for a rx/tx config for PMD to not reload if it's not changed. [1] https://mail.openvswitch.org/pipermail/ovs-dev/2019-August/362001.html Best regards, Ilya Maximets. ___

Re: [ovs-dev] [PATCH] faq: Update OVS/DPDK version table for OVS 2.12.

2019-09-24 Thread Ilya Maximets
regards, Ilya Maximets. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] Virtio Crypto with OVS-DPDK

2019-09-12 Thread Ilya Maximets
t; >   > > Regards, > > Harish > > > -Original Message- > From: ovs-dev-boun...@openvswitch.org On > Behalf Of Ilya Maximets > Sent: Tuesday, September 10, 2019 2:02 PM > To: ovs-dev@openvswitch.org > Subject: Re: [ovs-dev] Virtio Crypto with OVS

Re: [ovs-dev] [PATCH v8] netdev-dpdk:Detailed packet drop statistics

2019-09-12 Thread Ilya Maximets
On 10.09.2019 20:31, Sriram Vatala wrote: > -Original Message- > From: Ilya Maximets > Sent: 10 September 2019 19:29 > To: Sriram Vatala ; ovs-dev@openvswitch.org > Cc: ktray...@redhat.com; ian.sto...@intel.com > Subject: Re: [PATCH v8] netdev-dpdk:Detailed pack

[ovs-dev] [PATCH] netdev-dpdk: Reuse vhost function for dpdk ETH custom stats.

2019-09-12 Thread Ilya Maximets
netdev stats. Cc: Sriram Vatala Signed-off-by: Ilya Maximets --- Sriram, you could rebase your patch on top of this and send both patches as a patch-set. So the changes will be logically splitted. If you're using 'git format-patch' + 'git send-email' it should preserve authorship correctly

Re: [ovs-dev] [PATCH v3] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

2019-09-12 Thread Ilya Maximets
On 12.09.2019 13:19, Ilya Maximets wrote: > On 12.09.2019 13:07, Eelco Chaudron wrote: >> >> >> On 12 Sep 2019, at 10:39, Ilya Maximets wrote: >> >>> On 11.09.2019 16:20, Eelco Chaudron wrote: >>>> Currently, OVS does not register and therefor

Re: [ovs-dev] [PATCH v3] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

2019-09-12 Thread Ilya Maximets
On 12.09.2019 13:07, Eelco Chaudron wrote: > > > On 12 Sep 2019, at 10:39, Ilya Maximets wrote: > >> On 11.09.2019 16:20, Eelco Chaudron wrote: >>> Currently, OVS does not register and therefore not handle the >>> interface reset event from the DPDK frame

Re: [ovs-dev] [PATCH] MAINTAINERS: Update email for Ilya Maximets.

2019-09-12 Thread Ilya Maximets
On 11.09.2019 20:34, Ben Pfaff wrote: > On Wed, Sep 11, 2019 at 06:51:44PM +0300, Ilya Maximets wrote: >> CC: Ilya Maximets >> Signed-off-by: Ilya Maximets >> --- >> >> As someone could already know, my samsung email will be deactivated >> soon. So, the

Re: [ovs-dev] [PATCH v3] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

2019-09-12 Thread Ilya Maximets
On 11.09.2019 16:20, Eelco Chaudron wrote: > Currently, OVS does not register and therefore not handle the > interface reset event from the DPDK framework. This would cause a > problem in cases where a VF is used as an interface, and its > configuration changes. > > As an example in the following

[ovs-dev] [PATCH] MAINTAINERS: Update email for Ilya Maximets.

2019-09-11 Thread Ilya Maximets
CC: Ilya Maximets Signed-off-by: Ilya Maximets --- As someone could already know, my samsung email will be deactivated soon. So, the ovn.org address should be used instead. BTW, I'll continue working with upstream OVS, i.e. patch review, preparing fixes. Stay tuned! However, I might

Re: [ovs-dev] [PATCH] ovn: Fix broken dist* targets.

2019-09-11 Thread Ilya Maximets
On 11.09.2019 14:31, Numan Siddique wrote: > > > On Tue, Sep 10, 2019 at 1:31 PM Ilya Maximets <mailto:i.maxim...@samsung.com>> wrote: > > Redundant line continuation was missed while removing OVN > that broke distdir target and subsequently all the d

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

2019-09-11 Thread Ilya Maximets
On 11.09.2019 15:21, Konieczny, TomaszX wrote: >> -Original Message- >> From: Ilya Maximets >> Sent: 11 September 2019 13:04 >> To: Konieczny, TomaszX ; d...@openvswitch.org >> Cc: Stokes, Ian >> Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk

Re: [ovs-dev] [PATCH] ovn: Fix broken dist* targets.

2019-09-11 Thread Ilya Maximets
Kind reminder. This fix is important, because the issue breaks all the testsuite targets in TravisCI. It'll be good to make it work. Best regards, Ilya Maximets. On 10.09.2019 11:01, Ilya Maximets wrote: > Redundant line continuation was missed while removing OVN > that broke distdir

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

2019-09-11 Thread Ilya Maximets
On 11.09.2019 13:50, Konieczny, TomaszX wrote: >> -Original Message- >> From: Ilya Maximets >> Sent: 11 September 2019 12:26 >> To: Konieczny, TomaszX ; d...@openvswitch.org >> Cc: Stokes, Ian >> Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

2019-09-11 Thread Ilya Maximets
On 11.09.2019 13:16, Konieczny, TomaszX wrote: > > >> -Original Message----- >> From: Ilya Maximets >> Sent: 11 September 2019 12:08 >> To: Konieczny, TomaszX ; d...@openvswitch.org >> Cc: Stokes, Ian >> Subject: Re: [ovs-dev] [PATCH v1] netde

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

2019-09-11 Thread Ilya Maximets
On 11.09.2019 12:53, Konieczny, TomaszX wrote: >> -Original Message- >> From: Ilya Maximets >> Sent: 10 September 2019 14:29 >> To: Konieczny, TomaszX ; d...@openvswitch.org >> Cc: Stokes, Ian >> Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk

Re: [ovs-dev] [PATCH v2] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

2019-09-10 Thread Ilya Maximets
On 10.09.2019 15:12, Eelco Chaudron wrote: > Currently, OVS does not register and therefore not handle the > interface reset event from the DPDK framework. This would cause a > problem in cases where a VF is used as an interface, and its > configuration changes. > > As an example in the following

Re: [ovs-dev] [PATCH v8] netdev-dpdk:Detailed packet drop statistics

2019-09-10 Thread Ilya Maximets
On 08.09.2019 19:01, Sriram Vatala wrote: > OVS may be unable to transmit packets for multiple reasons and > today there is a single counter to track packets dropped due to > any of those reasons. The most common reason is that a VM is > unable to read packets fast enough causing the vhostuser

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

2019-09-10 Thread Ilya Maximets
On 10.09.2019 13:43, Konieczny, TomaszX wrote: > -Original Message- > From: Ilya Maximets > Sent: 10 September 2019 11:44 > To: Konieczny, TomaszX ; d...@openvswitch.org > Cc: Stokes, Ian > Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

Re: [ovs-dev] [PATCH] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

2019-09-10 Thread Ilya Maximets
On 10.09.2019 14:11, Eelco Chaudron wrote: > > > On 5 Sep 2019, at 14:40, Ilya Maximets wrote: > >> Hi Eelco, > > >>> , 2 deletions(-) >>> >>> diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c >>> index bc20d6843..a23150387 10

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

2019-09-10 Thread Ilya Maximets
On 10.09.2019 12:29, Ilya Maximets wrote: > On 09.09.2019 14:39, Tomasz Konieczny wrote: >> Currently OVS is unable to change flow control configuration in DPDK >> because new settings are being overwritten by current settings >> with rte_eth_dev_flow_ctrl_get(). The fix r

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

2019-09-10 Thread Ilya Maximets
t;) Regarding the fix itself: Can we just move following two lines: dev->fc_conf.mode = fc_mode; dev->fc_conf.autoneg = autoneg; below the rte_eth_dev_flow_ctrl_get() ? Current version of the patch will re-setup flow control on each call if it is not in initial state. Be

Re: [ovs-dev] Virtio Crypto with OVS-DPDK

2019-09-10 Thread Ilya Maximets
rish, Why do you want to integrate virtio-crypto into OVS? OVS is an OpenFlow Network Switch, but crypto devices are not network devices. They will not fit in OVS purposes nor OVS architecture. Best regards, Ilya Maximets. > > Below is the DPDK sample application we tried. > > https://

[ovs-dev] [PATCH] ovn: Fix broken dist* targets.

2019-09-10 Thread Ilya Maximets
("Remove OVN.") Signed-off-by: Ilya Maximets --- ovn/lib/automake.mk | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ovn/lib/automake.mk b/ovn/lib/automake.mk index c74430005..53a10e211 100644 --- a/ovn/lib/automake.mk +++ b/ovn/lib/automake.mk @@

Re: [ovs-dev] [PATCH] dpdk: Remove unneeded log message copy.

2019-09-06 Thread Ilya Maximets
at a first glance. One small nit is that you need a space between "(int)" and "size", but this could be, probably, fixed while applying. Best regards, Ilya Maximets. > lib/dpdk.c | 12 +--- > 1 file changed, 5 insertions(+), 7 deletions(-) > > diff --git

Re: [ovs-dev] [branch-2.12 PATCH] netdev-dpdk: Fix padding info comment.

2019-09-06 Thread Ilya Maximets
/* 352 */ > uint8_tpad52[384]; /* 384 */ > }; /* 320 384 */ > > Fixes: c161357d5d96 ("netdev-dpdk: Add custom stat for vhost tx retries.") > Reported-by: Ilya Maximets > Signed-of

Re: [ovs-dev] [PATCH v7] Detailed packet drop statistics per dpdk and vhostuser ports

2019-09-06 Thread Ilya Maximets
ill not vary with the > introduction of new counters in future. > > @Kevin Traynor : I will change the comment line for the number of padding > bytes in my next patch instead of you sending a new patch just for changing > the comment line. > > Thanks & Regards, > Sriram. > > --

Re: [ovs-dev] [PATCH] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

2019-09-06 Thread Ilya Maximets
On 05.09.2019 17:44, Eelco Chaudron wrote: > See inlines below, and will sent a v2 early next week. > > On 5 Sep 2019, at 14:40, Ilya Maximets wrote: > >> Hi Eelco, >> Thanks for the patch! Looks reasonable. >> >> One comment is that it's better

Re: [ovs-dev] [PATCH] tests: Add track-origins flag to valgrind.

2019-09-06 Thread Ilya Maximets
On 05.09.2019 18:18, Ben Pfaff wrote: > On Tue, Sep 03, 2019 at 06:07:26PM +0300, Ilya Maximets wrote: >> Useful for tracking where the uninitialized memory came from. >> Report example: >> >> Thread 13 revalidator11: >> Conditional jump or move

Re: [ovs-dev] [PATCH v2 0/3] Fix TSC frequency if DPDK is not available.

2019-09-06 Thread Ilya Maximets
On 02.09.2019 14:27, Ilya Maximets wrote: > > Version 2: > * Removed redundant OVS_UNUSED. > * 'ovs_numa_thread_setaffinity_core' reimplemented via > 'ovs_numa_thread_setaffinity_dump' to reduce code duplication. > * Added Ack from William to patches #2 and #3 as t

Re: [ovs-dev] [PATCH] dpif-netdev: Add core id in the PMD thread name.

2019-09-06 Thread Ilya Maximets
On 13.08.2019 19:37, Ilya Maximets wrote: > This is highly useful to see on which core PMD is running by > only looking at the thread name. Thread Id still allows to > distinguish different threads running on the same core over the time: > >|dpif_netdev(pmd-c10/id:53)|DBG|Creat

Re: [ovs-dev] [PATCH] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

2019-09-05 Thread Ilya Maximets
Hi Eelco, Thanks for the patch! Looks reasonable. One comment is that it's better to explicitly initialize the flag in common_construct. I see that we doesn't initialize 'started' flag, but this might be fixed too. More comments inline. Best regards, Ilya Maximets. On 05.09.2019 14:48, Eelco

Re: [ovs-dev] [PATCH v12] Improved Packet Drop Statistics in OVS

2019-09-05 Thread Ilya Maximets
Not a code review. Just a few notes about patch formatting. Best regards, Ilya Maximets. On 25.07.2019 15:16, Anju Thomas wrote: > Currently OVS maintains explicit packet drop/error counters only on > port level. Packets that are dropped as part of normal OpenFlow > processing ar

Re: [ovs-dev] SIGILL ovs branch-2.12/arm64/DPDK

2019-09-05 Thread Ilya Maximets
of DPDK code are in header inlines and built within the application build process. > > Best Regards, > Wei Yanqin > > -Original Message- > From: Ilya Maximets > Sent: Wednesday, September 4, 2019 8:46 PM > To: Yanqin Wei (Arm Technology China) ; James Page > >

Re: [ovs-dev] [branch-2.12 1/2] Set release date for 2.12.0.

2019-09-04 Thread Ilya Maximets
+1,5 @@ > -v2.12.0 - xx xxx > -- > +v2.12.0 - 03 Sep 2019 > + Not a merge blocker, but the size of the line is bothering me. :) It's one symbol shorter than the date. Best regards, Ilya Maximets. __

Re: [ovs-dev] [PATCH 2/2] netdev-afxdp: Add need_wakeup supprt.

2019-09-04 Thread Ilya Maximets
On 04.09.2019 17:04, William Tu wrote: > Hi Ilya, > > Thanks for the feedback. > > On Wed, Sep 4, 2019 at 5:19 AM Ilya Maximets wrote: >> >> Hi William, >> >> Thanks for the patch! >> One general comment is that we, probably, should make this

Re: [ovs-dev] [PATCH 2/2] netdev-afxdp: Add need_wakeup supprt.

2019-09-04 Thread Ilya Maximets
-+-+-++ >> >> Did not research why PVP is slower, maybe related to the TAP interface >> with AF_XDP? >> > I haven't tried PVP with this feature. > Maybe for virtual device, we don't need "need_wakeup" feature. > Let me invest

Re: [ovs-dev] [PATCH v7] Detailed packet drop statistics per dpdk and vhostuser ports

2019-09-04 Thread Ilya Maximets
ll useful to move these stats to a separate structure to avoid big padding in case we'll want to add one more. And I'm still thinking that we need to drop paddings at all for most of the structure fields, but this should be a separate change. Best regards, Ilya Maximets. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] SIGILL ovs branch-2.12/arm64/DPDK

2019-09-04 Thread Ilya Maximets
BTW, I submitted a bug to DPDK: https://bugs.dpdk.org/show_bug.cgi?id=344 Best regards, Ilya Maximets. On 04.09.2019 12:53, Yanqin Wei (Arm Technology China) wrote: > Understood. Thanks for the information. > >   > > Best Regards, > > Wei Yanqin > >   >

Re: [ovs-dev] [PATCH 2/2] netdev-afxdp: Add need_wakeup supprt.

2019-09-04 Thread Ilya Maximets
/ disabling the feature. However, I think that it's better to keep it enabled by default. Some comments inline. Best regards, Ilya Maximets. On 27.08.2019 2:02, William Tu wrote: > The patch adds support for using need_wakeup flag in AF_XDP rings. > When this flag is used, it means that O

Re: [ovs-dev] SIGILL ovs branch-2.12/arm64/DPDK

2019-09-03 Thread Ilya Maximets
in the list of cpu features in /proc/cpuinfo ? Best regards, Ilya Maximets. > > Program received signal SIGILL, Illegal instruction. > hash_bytes (p_=p_ at entry=0xaac36e50, n=7, basis=basis at entry=0) at > ../lib/hash.c:38 > 38 ../lib/hash.c: No such file or directory. > &

[ovs-dev] [PATCH] tests: Add track-origins flag to valgrind.

2019-09-03 Thread Ilya Maximets
: revalidate_ukey__ (ofproto-dpif-upcall.c:2219) <...> by 0x6AF488E: clone (clone.S:95) Uninitialised value was created by a stack allocation at 0x9D4450: compose_slow_path (ofproto-dpif-upcall.c:1062) Signed-off-by: Ilya Maximets --- tests/automake.mk | 3 ++- 1 file chan

Re: [ovs-dev] [PATCH v2 3/3] dpdk: Use ovs-numa provided functions to manage thread affinity.

2019-09-02 Thread Ilya Maximets
On 02.09.2019 16:17, David Marchand wrote: > On Mon, Sep 2, 2019 at 3:15 PM Ilya Maximets wrote: >> >> On 02.09.2019 16:09, David Marchand wrote: >>> On Mon, Sep 2, 2019 at 1:27 PM Ilya Maximets wrote: >>>> >>>> This allows to decrease

Re: [ovs-dev] [PATCH v2 3/3] dpdk: Use ovs-numa provided functions to manage thread affinity.

2019-09-02 Thread Ilya Maximets
On 02.09.2019 16:09, David Marchand wrote: > On Mon, Sep 2, 2019 at 1:27 PM Ilya Maximets wrote: >> >> This allows to decrease code duplication and avoid using Linux-specific >> functions (this might be useful in the future if we'll try to allow >> running OvS+DPDK on

[ovs-dev] [PATCH v2 3/3] dpdk: Use ovs-numa provided functions to manage thread affinity.

2019-09-02 Thread Ilya Maximets
This allows to decrease code duplication and avoid using Linux-specific functions (this might be useful in the future if we'll try to allow running OvS+DPDK on FreeBSD). Signed-off-by: Ilya Maximets Acked-by: William Tu --- lib/dpdk.c | 27 --- 1 file changed, 12

[ovs-dev] [PATCH v2 2/3] dpif-netdev-perf: Fix TSC frequency for non-DPDK case.

2019-09-02 Thread Ilya Maximets
. other_config:dpdk-init=false This change is mostly useful for AF_XDP netdev support, i.e. allows to use dpif-netdev/pmd-perf-show command and various PMD perf metrics. Signed-off-by: Ilya Maximets Acked-by: William Tu --- lib/dpdk-stub.c| 6 lib/dpdk.c | 6 lib/dpdk.h

[ovs-dev] [PATCH v2 1/3] ovs-numa: Add dump based thread affinity functions.

2019-09-02 Thread Ilya Maximets
New functions to get and set CPU affinity using CPU dumps. This will abstract OS specific implementation details from the cross-platform code. Signed-off-by: Ilya Maximets --- lib/ovs-numa.c | 72 ++ lib/ovs-numa.h | 2 ++ 2 files changed, 69

[ovs-dev] [PATCH v2 0/3] Fix TSC frequency if DPDK is not available.

2019-09-02 Thread Ilya Maximets
Version 2: * Removed redundant OVS_UNUSED. * 'ovs_numa_thread_setaffinity_core' reimplemented via 'ovs_numa_thread_setaffinity_dump' to reduce code duplication. * Added Ack from William to patches #2 and #3 as they wasn't changed. Ilya Maximets (3): ovs-numa: Add dump based thread

Re: [ovs-dev] [PATCH v7] Detailed packet drop statistics per dpdk and vhostuser ports

2019-08-30 Thread Ilya Maximets
tistics.". This patch adds statistics for all the dpdk based interfaces, so there is no need to list them in the commit name. Prefix clearly describes the patch area. (Please, do not drop the version number because of patch re-naming. Just add a note about renaming in a version hist

Re: [ovs-dev] [PATCH v2] Make pid_exists() more robust against empty pid argument

2019-08-30 Thread Ilya Maximets
On 29.08.2019 17:46, Ben Pfaff wrote: > On Tue, Aug 27, 2019 at 02:43:21PM +0300, Ilya Maximets wrote: >> On 14.08.2019 18:47, Michele Baldessari wrote: >>> In some of our destructive testing of ovn-dbs inside containers managed >>> by pacemaker we reached a situation

Re: [ovs-dev] [PATCH] dpif-netdev: Fail port addition if reconfiguration failed.

2019-08-29 Thread Ilya Maximets
On 26.08.2019 19:11, Stokes, Ian wrote: > > On 8/26/2019 3:12 PM, Ilya Maximets wrote: >> On 23.07.2019 14:20, Ilya Maximets wrote: >>> If the port was destroyed during the initial reconfiguration, we should >>> report an error to the upper layers. Otherwise, succe

Re: [ovs-dev] [PATCH V3] packets: Fix using outdated RSS hash after MPLS decapsulation.

2019-08-29 Thread Ilya Maximets
rom the same MPLS tunnel would hit the same EMC entry. > > This patch invalidates RSS hash (by resetting offload flags) after MPLS > header is popped. > > Signed-off-by: Nitin Katiyar Thanks! Applied to master and branch-2.12. Best regards, Ilya Maximets. _

Re: [ovs-dev] [PATCH v2] flow: miniflow_extract metadata branchless optimization

2019-08-29 Thread Ilya Maximets
ode duplication, so I'm trying to figure out what are the possibilities here. Best regards, Ilya Maximets. > > Best Regards, > Wei Yanqin > > -Original Message- > From: Ben Pfaff > Sent: Thursday, August 29, 2019 6:11 AM > To: Yanqin Wei (Arm Technology China) ; Ilya

Re: [ovs-dev] Open vSwitch 2.12 release ready?

2019-08-29 Thread Ilya Maximets
ed if you'll apply it as is. Best regards, Ilya Maximets. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH v1] ofproto: Fix OVS crash when packets hit Openflow rules with certain combinations of nested actions

2019-08-29 Thread Ilya Maximets
K under the ofproto_mutex and revert is still possible under the mutex. The only thing we need to take out of the mutex is real action execution by the datapath triggered by ofproto_packet_out_finish(). Callers never check the status of this function so it should be not so hard. So, possible solution: *

Re: [ovs-dev] [PATCH v1] ofproto: Fix OVS crash when packets hit Openflow rules with certain combinations of nested actions

2019-08-28 Thread Ilya Maximets
Hi. Thanks for the patch. But I cannot reproduce the issue on current master using commands you provided. Could you, please, provide also a stack trace of the asserted thread? This should make it easier to understand the issue. Best regards, Ilya Maximets. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] [PATCH] packets: Fix using outdated RSS hash after MPLS decapsulation.

2019-08-28 Thread Ilya Maximets
On 28.08.2019 11:27, Nitin Katiyar wrote: > > >> -Original Message----- >> From: Ilya Maximets [mailto:i.maxim...@samsung.com] >> Sent: Wednesday, August 28, 2019 12:44 PM >> To: Nitin Katiyar ; ovs-dev@openvswitch.org >> Cc: Stokes, Ian >> Subject

Re: [ovs-dev] [PATCH] packets: Fix using outdated RSS hash after MPLS decapsulation.

2019-08-28 Thread Ilya Maximets
On 28.08.2019 10:03, Nitin Katiyar wrote: > > >> -Original Message----- >> From: Ilya Maximets [mailto:i.maxim...@samsung.com] >> Sent: Tuesday, August 27, 2019 5:37 PM >> To: Nitin Katiyar ; ovs-dev@openvswitch.org >> Cc: Stokes, Ian >> Subject

  1   2   3   4   5   6   7   8   9   10   >