Re: [PATCH] net: sched: sch_teql: fix null-pointer dereference

On 4/8/21 6:26 PM, Eric Dumazet wrote: On 4/8/21 5:14 PM, Pavel Tikhomirov wrote: Reproduce: modprobe sch_teql tc qdisc add dev teql0 root teql0 This leads to (for instance in Centos 7 VM) OOPS: Null pointer dereference happens on master->slaves dereference in teql_destroy()

[PATCH] net: sched: sch_teql: fix null-pointer dereference

get OOPS. Signed-off-by: Pavel Tikhomirov --- net/sched/sch_teql.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index 2f1f0a378408..6af6b95bdb67 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -134,6 +134,9 @@ teql_

Re: [PATCH] vhost: return EINVAL if iovecs size does not match the message size

On 12/13/2018 10:55 PM, Michael S. Tsirkin wrote: > On Thu, Dec 13, 2018 at 05:53:50PM +0300, Pavel Tikhomirov wrote: >> We've failed to copy and process vhost_iotlb_msg so let userspace at >> least know about it. For instance before these patch the code below runs >> wi

[PATCH] vhost: return EINVAL if iovecs size does not match the message size

1) { perror("open"); return 1; } iov.iov_base = &msg; iov.iov_len = sizeof(msg)-4; if (writev(fd, &iov,1) == -1) { perror("writev"); return 1; } return 0; } Signed-off-by: Pavel Tikhomirov --- drivers/vhost/vhost.c | 8 ++-- 1 file cha

[PATCH v2] ipv4: make tcp_notsent_lowat sysctl knob behave as true unsigned int

umentation we have "tcp_notsent_lowat - UNSIGNED INTEGER" v2: simplify to just proc_douintvec Signed-off-by: Pavel Tikhomirov --- net/ipv4/sysctl_net_ipv4.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 80bc36b..5

[PATCH] ipv4: make tcp_notsent_lowat sysctl knob behave as true unsigned int

umentation we have "tcp_notsent_lowat - UNSIGNED INTEGER" Signed-off-by: Pavel Tikhomirov --- net/ipv4/sysctl_net_ipv4.c | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 80bc36b..5361373 100644 --- a/net/ipv4

[PATCH 4/6] ip xfrm policy: add save/restore

eckpoint/restore for container migration. Signed-off-by: Pavel Tikhomirov --- ip/xfrm_policy.c | 73 ++ man/man8/ip-xfrm.8 | 8 ++ 2 files changed, 76 insertions(+), 5 deletions(-) diff --git a/ip/xfrm_policy.c b/ip/xfrm_policy.c index cc

[PATCH 1/6] ip: merge together all save_xxx_preps and xxx_dump_check_magics

want to reuse them in xfrm state/policy save Signed-off-by: Pavel Tikhomirov --- ip/ip_common.h | 3 +++ ip/ipaddress.c | 43 +++ ip/iproute.c | 18 +- ip/iprule.c| 42 ++ 4 files changed, 17

[PATCH 0/6] ip: add save/restore to xfrm tables

Pavel Tikhomirov (6): ip: merge together all save_xxx_preps and xxx_dump_check_magics ip: merge together all save_xxx to save_nlmsg ip xfrm state: add save/restore ip xfrm policy: add save/restore ip xfrm state: fixup hard/soft timeouts/limits while save ip xfrm policy: fixup hard/soft

[PATCH 3/6] ip xfrm state: add save/restore

This patch adds save and restore commands to "ip xfrm state" similar to commit 2f4e171f7df2 and commit f4ff11e3e298 for "ip rule" and "ip route" respectively. The feature is useful in checkpoint/restore for container migration. Signed-off-by: Pavel Tikhomirov

[PATCH 6/6] ip xfrm policy: fixup hard/soft timeouts/limits while save

same as for states, though I beleive that plicies do not actually support byte-soft | byte-hard | packet-soft | packet-hard, but they still can be set. Signed-off-by: Pavel Tikhomirov --- ip/ip_common.h | 5 + ip/xfrm_policy.c | 35 +-- ip/xfrm_state.c

[PATCH 5/6] ip xfrm state: fixup hard/soft timeouts/limits while save

timeout of 30sec and now we do C/R for the whole container. As the process in question knows nothing about C/R, after restore it might expect that xfrm state dissappears in 30-10=20sec, so we need to set timeout to 20. If difference is <=0 set 1 at least to do not lose the limit completely

[PATCH 2/6] ip: merge together all save_xxx to save_nlmsg

want to reuse them in xfrm state/policy save Signed-off-by: Pavel Tikhomirov --- ip/ip_common.h | 2 ++ ip/ipaddress.c | 14 -- ip/iproute.c | 23 +++ ip/iprule.c| 16 +--- 4 files changed, 18 insertions(+), 37 deletions(-) diff --git a/ip

[PATCH v2] netfilter: nf_log: fix error on write NONE to logger choice sysctl

cat /proc/sys/net/netfilter/nf_log/0 nfnetlink_log echo NONE > /proc/sys/net/netfilter/nf_log/0 cat /proc/sys/net/netfilter/nf_log/0 NONE v2: add missed error check for proc_dostring Signed-off-by: Pavel Tikhomirov --- net/netfilter/nf_log.c | 11 ++- 1 file changed, 6 inser

[PATCH] netfilter: nf_log: fix error on write NONE to logger choice sysctl

cat /proc/sys/net/netfilter/nf_log/0 nfnetlink_log echo NONE > /proc/sys/net/netfilter/nf_log/0 cat /proc/sys/net/netfilter/nf_log/0 NONE Signed-off-by: Pavel Tikhomirov --- net/netfilter/nf_log.c | 9 - 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/net/netfilt

Re: [Intel-wired-lan] [PATCH v2] ixgbe: on recv increment rx.ring->stats.yields

Hi, please, let me know if I need to fix something else here. On 12/11/2015 05:05 PM, Pavel Tikhomirov wrote: It seem to be non intentionaly changed to tx in commit adc810900a70 ("ixgbe: Refactor busy poll socket code to address multiple issues") Lock is taken from ixgbe_low_latency

[PATCH v2] ixgbe: on recv increment rx.ring->stats.yields

ield stats can be shown through ethtool: ethtool -S enp129s0 | grep yield v2: follow commit citing style Signed-off-by: Pavel Tikhomirov --- drivers/net/ethernet/intel/ixgbe/ixgbe.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/d

[PATCH] ixgbe: on recv increment rx.ring->stats.yields instead of tx

tx counter. Yield stats can be shown through ethtool: ethtool -S enp129s0 | grep yield Signed-off-by: Pavel Tikhomirov --- drivers/net/ethernet/intel/ixgbe/ixgbe.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net