Re: [PATCH v2] sched: report if filter is too large to dump

2018-02-21 Thread Roman Kapl
On 02/21/2018 09:42 AM, Phil Sutter wrote: Hi Roman, On Wed, Feb 21, 2018 at 09:38:52AM +0100, Roman Kapl wrote: On 02/21/2018 08:45 AM, Phil Sutter wrote: On Mon, Feb 19, 2018 at 09:32:51PM +0100, Roman Kapl wrote: Note: The error case can happen pretty easily if you create a filter with 32

Re: [PATCH v2] sched: report if filter is too large to dump

2018-02-21 Thread Roman Kapl
On 02/21/2018 08:45 AM, Phil Sutter wrote: Hi Roman, On Mon, Feb 19, 2018 at 09:32:51PM +0100, Roman Kapl wrote: So far, if the filter was too large to fit in the allocated skb, the kernel did not return any error and stopped dumping. Modify the dumper so that it returns -EMSGSIZE when

[PATCH v2] sched: report if filter is too large to dump

2018-02-19 Thread Roman Kapl
allocation size, which in turn leads to Signed-off-by: Roman Kapl <c...@rkapl.cz> --- v1 -> v2: add the "progress" comment, fixed error name in commit message. I've looked at other dumpers in rtnetnlink, there are various ways to handle that. For example rtnl_stats_dump ha

[PATCH] net: sched: report if filter is too large to dump

2018-02-18 Thread Roman Kapl
allocation size, which in turn leads to smaller SKBs in kernel. Signed-off-by: Roman Kapl <c...@rkapl.cz> --- net/sched/cls_api.c | 6 +- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index f21610c5da1a..b5771a586c2d 100644 ---

Re: [PATCH] net: make sure skb_dst is valid before using

2018-01-24 Thread Roman Kapl
On 01/24/2018 10:16 AM, Xin Long wrote: On Wed, Jan 24, 2018 at 6:42 AM, Roman Kapl <c...@rkapl.cz> wrote: Tunnel devices often use skb_dst(skb)->ops, but ops are not implemented for metadata tunnel destinations. Use skb_valid_dst to check if skb_dst is a real (non-metadata) de

[PATCH] net: make sure skb_dst is valid before using

2018-01-23 Thread Roman Kapl
tch is not exhaustive. Fixes: 52a589d51f10 ("geneve: update skb dst pmtu on tx path") Fixes: a93bf0ff4490 ("vxlan: update skb dst pmtu on tx path") Signed-off-by: Roman Kapl <c...@rkapl.cz> --- drivers/infiniband/ulp/ipoib/ipoib_cm.c | 3 ++- drivers/net/geneve.c

Re: [PATCH net] vxlan: update skb dst pmtu on tx path

2018-01-22 Thread Roman Kapl
ars some thought was given to this. Or maybe md_dst_ops should have update pmtu? Thanks, Roman Kapl

[PATCH v3] net: sched: crash on blocks with goto chain action

2017-11-29 Thread Roman Kapl
ns: ip link add dtest type dummy tc qdisc add dev dtest ingress tc filter add dev dtest chain 1 parent : handle 1 prio 1 flower action goto chain 2 Fixes: 623859ae06b8 ("Merge branch 'net-sched-race-fix'") Signed-off-by: Roman Kapl <c...@rkapl.cz> --- v1 -&

[PATCH v2] net: sched: crash on blocks with goto chain action

2017-11-24 Thread Roman Kapl
: 822e86d997 ("net_sched: remove tcf_block_put_deferred()") Signed-off-by: Roman Kapl <c...@rkapl.cz> --- v1 -> v2: Hold all chains instead of just the currently iterated one, the code should be more clear this way. --- net/sched/cls_api.c | 17 -

Re: [PATCH] net: sched: crash on blocks with goto chain action

2017-11-21 Thread Roman Kapl
On 11/21/2017 08:31 PM, Cong Wang wrote: On Mon, Nov 20, 2017 at 1:41 PM, Roman Kapl <c...@rkapl.cz> wrote: On 11/20/2017 06:54 PM, Cong Wang wrote: On Sun, Nov 19, 2017 at 8:17 AM, Roman Kapl <c...@rkapl.cz> wrote: tcf_block_put_ext has assumed that all filters (and thus their

Re: [PATCH] net: sched: crash on blocks with goto chain action

2017-11-20 Thread Roman Kapl
On 11/20/2017 06:54 PM, Cong Wang wrote: On Sun, Nov 19, 2017 at 8:17 AM, Roman Kapl <c...@rkapl.cz> wrote: tcf_block_put_ext has assumed that all filters (and thus their goto actions) are destroyed in RCU callback and thus can not race with our list iteration. However, that is not true

[PATCH] net: sched: crash on blocks with goto chain action

2017-11-20 Thread Roman Kapl
-by: Roman Kapl <c...@rkapl.cz> --- The mail was original rejected by vger, this is a re-send to netdev@vger only (with the same message ID). Sorry for any confusion. --- net/sched/cls_api.c | 18 ++ 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/net/sched/cls_api.c

[PATCH v2] net: sched: fix crash when deleting secondary chains

2017-11-20 Thread Roman Kapl
et_net() before call_rcu()") Fixes: f93e1cdcf42c ("net/sched: fix filter flushing") Acked-by: Jiri Pirko <j...@mellanox.com> Signed-off-by: Roman Kapl <c...@rkapl.cz> --- v1 -> v2: Added Fixes and Acked-by tags The mail was original rejected by vger, this is a re-send

[PATCH v2] net: move somaxconn init from sysctl code

2017-05-24 Thread Roman Kapl
Signed-off-by: Roman Kapl <roman.k...@sysgo.com> --- net/core/net_namespace.c | 19 +++ net/core/sysctl_net_core.c | 2 -- 2 files changed, 19 insertions(+), 2 deletions(-) diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 1934efd..26bbfab 100644 --- a/net/core

Re: [PATCH] net: set default value for somaxconn

2017-05-23 Thread Roman Kapl
On 05/22/2017 08:18 PM, David Miller wrote: From: Roman Kapl <roman.k...@sysgo.com> Date: Mon, 22 May 2017 14:22:41 +0200 The default value for somaxconn is set in sysctl_core_net_init(), but this function is not called when kernel is configured without CONFIG_SYSCTL. This r

[PATCH] net: set default value for somaxconn

2017-05-22 Thread Roman Kapl
t;TCP: request_sock_TCP: Possible SYN flooding on port 7. Dropping request. Check SNMP counters." Before ef547f2ac16 (tcp: remove max_qlen_log), the effects were less severe, because the backlog was always at least eight slots long. Signed-off-by: Roman Kapl <roman.k...@sysgo.com> --- net/core