Re: [ovs-dev] [PATCH] netdev: Remove modulo operation in dpdk/vhost/afxdp tx path

2023-10-07 Thread Li,Rongqing via dev
> -Original Message- > From: Simon Horman > Sent: Friday, October 6, 2023 6:52 PM > To: Li,Rongqing > Cc: d...@openvswitch.org > Subject: Re: [ovs-dev] [PATCH] netdev: Remove modulo operation in > dpdk/vhost/afxdp tx path > > On Tue, Mar 23, 2021 at 11

[ovs-dev] [PATCH] netdev: Remove modulo operation in dpdk/vhost/afxdp tx path

2021-03-22 Thread Li RongQing
After 324c8374852a ("dpif-netdev: XPS (Transmit Packet Steering) implementation."), it is impossible that qid is greater than or equal to n_txq of netdev. so remove this modulo operation which is in hot path Signed-off-by: Li RongQing --- lib/netdev-afxdp.c | 1 - lib/netdev-dpdk.c

Re: [ovs-dev] [PATCH][v2] conntrack: Fix conntrack tw expiration

2021-01-14 Thread Li,Rongqing
> -Original Message- > From: Aaron Conole [mailto:acon...@redhat.com] > Sent: Thursday, January 14, 2021 11:23 PM > To: Li,Rongqing > Cc: ovs-dev@openvswitch.org; William Tu > Subject: Re: [ovs-dev] [PATCH][v2] conntrack: Fix conntrack tw expiration > &g

Re: [ovs-dev] [PATCH][v2] conntrack: Fix conntrack tw expiration

2021-01-08 Thread Li,Rongqing
struct conntrack *ct, struct conn > > *conn_, > > > > if (tcp_flags & TCP_RST) { > > src->state = dst->state = CT_DPIF_TCPS_TIME_WAIT; > > +conn_update_expiration(ct, >up, > CT_TM_TCP_CLOSED, > > + now); > > } > > } else { > >

[ovs-dev] [PATCH][v2] conntrack: Fix conntrack tw expiration

2021-01-06 Thread Li RongQing
space connection tracker.") Signed-off-by: Li RongQing Co-authored-by: Mao YingMing Signed-off-by: Mao YingMing --- resend with maoyingming signature lib/conntrack-tcp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/lib/conntrack-tcp.c b/lib/conntrack-tcp.c index 18a2aa7c7..f1595a

[ovs-dev] [PATCH] conntrack: Fix conntrack tw expiration

2021-01-06 Thread Li RongQing
In connection tracking system, a connection will enter timewait status when one side receive a fin, then a reset packet reached But the expiration time is not updated, still is the previous expiration time Signed-off-by: Li RongQing --- lib/conntrack-tcp.c | 1 + 1 file changed, 1 insertion

[ovs-dev] 答复: ipf question

2019-12-03 Thread Li,Rongqing
Thank you I will continue as your suggestion when have time -RongQing 发件人: Darrell Ball [mailto:dlu...@gmail.com] 发送时间: 2019年11月30日 3:19 收件人: Li,Rongqing ; ovs dev 主题: Re: ipf question -- Forwarded message - From: Li,Rongqing mailto:lirongq...@baidu.com>> Date: Thu,

[ovs-dev] 答复: [RFC PATCH] netdev-dpdk: Narrow down txq critical section.

2019-12-03 Thread Li,Rongqing
> -邮件原件- > 发件人: David Marchand [mailto:david.march...@redhat.com] > 发送时间: 2019年12月3日 20:40 > 收件人: Ilya Maximets > 抄送: ovs dev ; Ian Stokes ; > Li,Rongqing > 主题: Re: [RFC PATCH] netdev-dpdk: Narrow down txq critical section. > > On Tue, Dec 3, 2019 at 12

[ovs-dev] 答复: ipf question

2019-11-18 Thread Li,Rongqing
are received, the remaining 8 fragment will in complete list, and can not send out until expired to clean them, is it true? -RongQing 发件人: Darrell Ball [mailto:dlu...@gmail.com] 发送时间: 2019年11月19日 10:22 收件人: Li,Rongqing 抄送: ovs dev 主题: Re: ipf question Thanks; I had a look and I noticed ipf

[ovs-dev] 答复: ipf question

2019-11-15 Thread Li,Rongqing
发件人: Li,Rongqing 发送时间: 2019年11月16日 10:03 收件人: Darrell Ball 抄送: ovs dev 主题: 答复: ipf question 发件人: Darrell Ball 发送时间: 2019年11月15日 22:58 收件人: Li,Rongqing 抄送: ovs dev 主题: Re: ipf question >Let me paraphrase, just to confirm we are on the same page. &g

[ovs-dev] 答复: ipf question

2019-11-15 Thread Li,Rongqing
发件人: Darrell Ball 发送时间: 2019年11月15日 22:58 收件人: Li,Rongqing 抄送: ovs dev 主题: Re: ipf question   >Let me paraphrase, just to confirm we are on the same page. >IIUC, for example, in the case of a 33 fragment packet, in the first pass all >33 fragments enter ipf, then are >reassemble

[ovs-dev] [PATCH] ipf: bail out when ipf state is COMPLETED

2019-11-14 Thread Li RongQing
in ipf_list_state_transition at lib/ipf.c:545 Co-authored-by: Wang Li Signed-off-by: Wang Li Signed-off-by: Li RongQing --- lib/ipf.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/lib/ipf.c b/lib/ipf.c index 4cc0f2df6..45c489122 100644 --- a/lib/ipf.c +++ b/lib/ipf.c

[ovs-dev] 答复: [PATCH] conntrack: check the result of extract_l3_ipv4/6

2019-08-21 Thread Li,Rongqing
发件人: Darrell Ball [mailto:dlu...@gmail.com] 发送时间: 2019年8月22日 9:00 收件人: Ben Pfaff 抄送: Li,Rongqing ; ovs-dev@openvswitch.org 主题: Re: [ovs-dev] [PATCH] conntrack: check the result of extract_l3_ipv4/6 On Wed, Aug 21, 2019 at 3:13 PM Ben Pfaff mailto:b...@ovn.org>> wrote: On Mon, Aug 19

[ovs-dev] [PATCH] conntrack: check the result of extract_l3_ipv4/6

2019-08-19 Thread Li RongQing
the result of extract_l3_ipv4/6 should be checked in reverse_nat_packet when it is false, meaning this packet is wrong, should not do handle it continually Signed-off-by: Li RongQing --- lib/conntrack.c | 17 +++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/lib

[ovs-dev] [PATCH] conntrack: check the result of extract_l3_ipv4/6

2019-08-18 Thread Li RongQing
the result of extract_l3_ipv4/6 should be checked in reverse_nat_packet when it is false, meaning this packet is wrong, should not do handle it continually Signed-off-by: Li RongQing --- lib/conntrack.c | 17 +++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/lib

[ovs-dev] [PATCH] netdev-dpdk: ensure that qid is valid before using it

2019-05-30 Thread Li RongQing
otherwise it maybe lead to out of bound access in tx_q Signed-off-by: Li RongQing --- lib/netdev-dpdk.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c index c06f46931..e6cdcd454 100644 --- a/lib/netdev-dpdk.c +++ b/lib/netdev-dpdk.c

[ovs-dev] [PATCH] ofctl: break the loop if ovs_pcap_read returns error

2019-02-17 Thread Li RongQing
otherwise packet is NULL, and dereference it to cause segfault Signed-off-by: Li RongQing --- utilities/ovs-ofctl.c | 1 + 1 file changed, 1 insertion(+) diff --git a/utilities/ovs-ofctl.c b/utilities/ovs-ofctl.c index d8a31e394..63620e4fb 100644 --- a/utilities/ovs-ofctl.c +++ b/utilities/ovs

[ovs-dev] [PATCH] netdev-dpdk: remove constant arguments for ingress_policer_run

2019-02-15 Thread Li RongQing
should_steal is always true, not need to be passed Signed-off-by: Li RongQing --- lib/netdev-dpdk.c | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c index f07b10c88..7ae29f3c2 100644 --- a/lib/netdev-dpdk.c +++ b/lib/netdev-dpdk.c

[ovs-dev] [PATCH] conntrack: force packet length check in extract_l3_ipv4

2019-02-13 Thread Li RongQing
Assure the packet size is larger than ip total length Signed-off-by: Li RongQing --- lib/conntrack.c | 4 1 file changed, 4 insertions(+) diff --git a/lib/conntrack.c b/lib/conntrack.c index 5c43410ec..4a39327b3 100644 --- a/lib/conntrack.c +++ b/lib/conntrack.c @@ -1525,6 +1525,10

[ovs-dev] 答复: [PATCH] netdev-dpdk: shrink critical region under tx_q[qid].tx_lock

2019-02-11 Thread Li,Rongqing
> -邮件原件- > 发件人: Ian Stokes [mailto:ian.sto...@intel.com] > 发送时间: 2019年2月6日 23:34 > 收件人: Li,Rongqing ; d...@openvswitch.org > 主题: Re: [ovs-dev] [PATCH] netdev-dpdk: shrink critical region under > tx_q[qid].tx_lock > > On 1/31/2019 2:

[ovs-dev] [PATCH][v2] conntrack: Remove unnecessary check in process_ftp_ctl_v4

2019-02-10 Thread Li RongQing
It has been assured that both first and second int from ftp command are not bigger than 255, so their combination(first int << 8 +second int) must not bigger than 65535 Co-authored-by: Wang Li Signed-off-by: Wang Li Signed-off-by: Li RongQing Cc: Darrell Ball --- lib/conntrack

[ovs-dev] 答复: [PATCH] conntrack: Remove unnecessary check in process_ftp_ctl_v4

2019-02-10 Thread Li,Rongqing
发件人: Darrell Ball [mailto:dlu...@gmail.com] 发送时间: 2019年2月1日 23:25 收件人: Li,Rongqing 抄送: ovs dev 主题: Re: [ovs-dev] [PATCH] conntrack: Remove unnecessary check in process_ftp_ctl_v4 On Fri, Feb 1, 2019 at 1:07 AM Li,Rongqing mailto:lirongq...@baidu.com>> wrote: 发件人: Darrel

[ovs-dev] 答复: [PATCH] conntrack: Remove unnecessary check in process_ftp_ctl_v4

2019-02-01 Thread Li,Rongqing
发件人: Darrell Ball [mailto:dlu...@gmail.com] 发送时间: 2019年2月1日 16:15 收件人: Li,Rongqing 抄送: ovs dev 主题: Re: [ovs-dev] [PATCH] conntrack: Remove unnecessary check in process_ftp_ctl_v4 >This was intentionally done to be documentative and also make it hard to break; >this code path sees

[ovs-dev] 答复: [PATCH] conntrack: compute hash value once when calling conn_key_lookup

2019-02-01 Thread Li,Rongqing
发件人: Darrell Ball [mailto:dlu...@gmail.com] 发送时间: 2019年2月1日 16:25 收件人: Li,Rongqing 抄送: ovs dev 主题: Re: [ovs-dev] [PATCH] conntrack: compute hash value once when calling conn_key_lookup > The function comment says it all: > /* Typical usage of this helper is in non per-packe

[ovs-dev] [PATCH] netdev-dpdk: shrink critical region under tx_q[qid].tx_lock

2019-01-30 Thread Li RongQing
netdev_dpdk_filter_packet_len() does not need to be protected by tx_q[].tx_lock, and tx_q[].tx_lock can not protect it too, same to netdev_dpdk_qos_run so move them out of this lock to improve the scalability Signed-off-by: Li RongQing --- lib/netdev-dpdk.c | 33

[ovs-dev] [PATCH] flow: fix a possible memory leak in parse_ct_state

2019-01-27 Thread Li RongQing
state_s should be freed always before exit parse_ct_state Signed-off-by: Li RongQing --- lib/flow.c | 1 + 1 file changed, 1 insertion(+) diff --git a/lib/flow.c b/lib/flow.c index c6e47781b..479b7fe46 100644 --- a/lib/flow.c +++ b/lib/flow.c @@ -1261,6 +1261,7 @@ parse_ct_state(const char

[ovs-dev] [PATCH] flow: fix udp checksum

2019-01-25 Thread Li RongQing
As per RFC 768, if the calculated UDP checksum is 0, it should be instead set as 0x in the frame. A value of 0 in the checksum field indicates to the receiver that no checksum was calculated and hence it should not verify the checksum. Signed-off-by: Li RongQing --- lib/flow.c | 3 +++ 1

[ovs-dev] [PATCH] conntrack: Remove unnecessary check in process_ftp_ctl_v4

2019-01-18 Thread Li RongQing
It has been assured that both first and second int from ftp command are not bigger than 255, so their combination(first int << 8 +second int) must not bigger than 65535 Signed-off-by: Wang Li Signed-off-by: Li RongQing --- lib/conntrack.c | 7 +-- 1 file changed, 1 insertion

[ovs-dev] [PATCH] conntrack: compute hash value once when calling conn_key_lookup

2019-01-18 Thread Li RongQing
it is expensive to compute hash value, and When call conn_lookup, hash value is computed twice, in fact, we can cache it and pass it to conn_lookup Signed-off-by: Zhang Yu Signed-off-by: Li RongQing --- lib/conntrack.c | 19 +++ 1 file changed, 11 insertions(+), 8 deletions

[ovs-dev] [PATCH][net-next] openvswitch: remove BUG_ON from get_dpdev

2018-11-08 Thread Li RongQing
if local is NULL pointer, and the following access of local's dev will trigger panic, which is same as BUG_ON Signed-off-by: Li RongQing --- net/openvswitch/vport-netdev.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index