* [PATCH net] netfilter: flowtable: Fix expired flow not being deleted from software @ 2020-05-06 11:27 Paul Blakey 2020-05-10 22:26 ` Pablo Neira Ayuso 0 siblings, 1 reply; 5+ messages in thread From: Paul Blakey @ 2020-05-06 11:27 UTC (permalink / raw) To: Paul Blakey, Oz Shlomo, Pablo Neira Ayuso, Roi Dayan, netdev, Saeed Mahameed Cc: netfilter-devel Once a flow is considered expired, it is marked as DYING, and scheduled a delete from hardware. The flow will be deleted from software, in the next gc_step after hardware deletes the flow (and flow is marked DEAD). Till that happens, the flow's timeout might be updated from a previous scheduled stats, or software packets (refresh). This will cause the gc_step to no longer consider the flow expired, and it will not be deleted from software. Fix that by looking at the DYING flag as in deciding a flow should be deleted from software. Fixes: c29f74e0df7a ("netfilter: nf_flow_table: hardware offload support") Signed-off-by: Paul Blakey <paulb@mellanox.com> Reviewed-by: Oz Shlomo <ozsh@mellanox.com> Reviewed-by: Roi Dayan <roid@mellanox.com> --- net/netfilter/nf_flow_table_core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index c0cb7949..b0e9f7a 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -362,7 +362,8 @@ static void nf_flow_offload_gc_step(struct flow_offload *flow, void *data) struct nf_flowtable *flow_table = data; if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct) || - test_bit(NF_FLOW_TEARDOWN, &flow->flags)) { + test_bit(NF_FLOW_TEARDOWN, &flow->flags) || + test_bit(NF_FLOW_HW_DYING, &flow->flags)) { if (test_bit(NF_FLOW_HW, &flow->flags)) { if (!test_bit(NF_FLOW_HW_DYING, &flow->flags)) nf_flow_offload_del(flow_table, flow); -- 1.8.3.1 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH net] netfilter: flowtable: Fix expired flow not being deleted from software 2020-05-06 11:27 [PATCH net] netfilter: flowtable: Fix expired flow not being deleted from software Paul Blakey @ 2020-05-10 22:26 ` Pablo Neira Ayuso 2020-05-11 7:24 ` Paul Blakey 0 siblings, 1 reply; 5+ messages in thread From: Pablo Neira Ayuso @ 2020-05-10 22:26 UTC (permalink / raw) To: Paul Blakey; +Cc: Oz Shlomo, Roi Dayan, netdev, Saeed Mahameed, netfilter-devel [-- Attachment #1: Type: text/plain, Size: 716 bytes --] On Wed, May 06, 2020 at 02:27:29PM +0300, Paul Blakey wrote: > Once a flow is considered expired, it is marked as DYING, and > scheduled a delete from hardware. The flow will be deleted from > software, in the next gc_step after hardware deletes the flow > (and flow is marked DEAD). Till that happens, the flow's timeout > might be updated from a previous scheduled stats, or software packets > (refresh). This will cause the gc_step to no longer consider the flow > expired, and it will not be deleted from software. > > Fix that by looking at the DYING flag as in deciding > a flow should be deleted from software. Would this work for you? The idea is to skip the refresh if this has already expired. Thanks. [-- Attachment #2: expired.patch --] [-- Type: text/x-diff, Size: 1235 bytes --] diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index 4344e572b7f9..862efa7c606d 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -252,10 +252,18 @@ int flow_offload_add(struct nf_flowtable *flow_table, struct flow_offload *flow) } EXPORT_SYMBOL_GPL(flow_offload_add); +static inline bool nf_flow_has_expired(const struct flow_offload *flow) +{ + return nf_flow_timeout_delta(flow->timeout) <= 0; +} + void flow_offload_refresh(struct nf_flowtable *flow_table, struct flow_offload *flow) { - flow->timeout = nf_flowtable_time_stamp + NF_FLOW_TIMEOUT; + if (!nf_flow_has_expired(flow)) { + flow->timeout = nf_flowtable_time_stamp + NF_FLOW_TIMEOUT; + return; + } if (likely(!nf_flowtable_hw_offload(flow_table) || !test_and_clear_bit(NF_FLOW_HW_REFRESH, &flow->flags))) @@ -265,11 +273,6 @@ void flow_offload_refresh(struct nf_flowtable *flow_table, } EXPORT_SYMBOL_GPL(flow_offload_refresh); -static inline bool nf_flow_has_expired(const struct flow_offload *flow) -{ - return nf_flow_timeout_delta(flow->timeout) <= 0; -} - static void flow_offload_del(struct nf_flowtable *flow_table, struct flow_offload *flow) { ^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH net] netfilter: flowtable: Fix expired flow not being deleted from software 2020-05-10 22:26 ` Pablo Neira Ayuso @ 2020-05-11 7:24 ` Paul Blakey 2020-05-11 8:42 ` Pablo Neira Ayuso 0 siblings, 1 reply; 5+ messages in thread From: Paul Blakey @ 2020-05-11 7:24 UTC (permalink / raw) To: Pablo Neira Ayuso Cc: Oz Shlomo, Roi Dayan, netdev, Saeed Mahameed, netfilter-devel On 5/11/2020 1:26 AM, Pablo Neira Ayuso wrote: > On Wed, May 06, 2020 at 02:27:29PM +0300, Paul Blakey wrote: >> Once a flow is considered expired, it is marked as DYING, and >> scheduled a delete from hardware. The flow will be deleted from >> software, in the next gc_step after hardware deletes the flow >> (and flow is marked DEAD). Till that happens, the flow's timeout >> might be updated from a previous scheduled stats, or software packets >> (refresh). This will cause the gc_step to no longer consider the flow >> expired, and it will not be deleted from software. >> >> Fix that by looking at the DYING flag as in deciding >> a flow should be deleted from software. > Would this work for you? > > The idea is to skip the refresh if this has already expired. > > Thanks. The idea is ok, but timeout check + update isn't atomic (need atomic_inc_unlesss or something like that), and there is also the hardware stats which if comes too late (after gc finds it expired) might bring a flow back to life. ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH net] netfilter: flowtable: Fix expired flow not being deleted from software 2020-05-11 7:24 ` Paul Blakey @ 2020-05-11 8:42 ` Pablo Neira Ayuso 2020-05-11 9:50 ` Paul Blakey 0 siblings, 1 reply; 5+ messages in thread From: Pablo Neira Ayuso @ 2020-05-11 8:42 UTC (permalink / raw) To: Paul Blakey; +Cc: Oz Shlomo, Roi Dayan, netdev, Saeed Mahameed, netfilter-devel [-- Attachment #1: Type: text/plain, Size: 1346 bytes --] On Mon, May 11, 2020 at 10:24:44AM +0300, Paul Blakey wrote: > > > On 5/11/2020 1:26 AM, Pablo Neira Ayuso wrote: > > On Wed, May 06, 2020 at 02:27:29PM +0300, Paul Blakey wrote: > >> Once a flow is considered expired, it is marked as DYING, and > >> scheduled a delete from hardware. The flow will be deleted from > >> software, in the next gc_step after hardware deletes the flow > >> (and flow is marked DEAD). Till that happens, the flow's timeout > >> might be updated from a previous scheduled stats, or software packets > >> (refresh). This will cause the gc_step to no longer consider the flow > >> expired, and it will not be deleted from software. > >> > >> Fix that by looking at the DYING flag as in deciding > >> a flow should be deleted from software. > > Would this work for you? > > > > The idea is to skip the refresh if this has already expired. > > > > Thanks. > > The idea is ok, but timeout check + update isn't atomic (need atomic_inc_unlesss > or something like that), and there is also > the hardware stats which if comes too late (after gc finds it expired) might > bring a flow back to life. Right. Once the entry has expired, there should not be a way turning back. I'm attaching a new sketch, it's basically using the teardown state to specify that the gc already made the decision to remove this entry. Thanks. [-- Attachment #2: x.patch --] [-- Type: text/x-diff, Size: 1046 bytes --] diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index 4344e572b7f9..42da6e337276 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -284,7 +284,7 @@ static void flow_offload_del(struct nf_flowtable *flow_table, if (nf_flow_has_expired(flow)) flow_offload_fixup_ct(flow->ct); - else if (test_bit(NF_FLOW_TEARDOWN, &flow->flags)) + else flow_offload_fixup_ct_timeout(flow->ct); flow_offload_free(flow); @@ -361,8 +361,10 @@ static void nf_flow_offload_gc_step(struct flow_offload *flow, void *data) { struct nf_flowtable *flow_table = data; - if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct) || - test_bit(NF_FLOW_TEARDOWN, &flow->flags)) { + if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct)) + set_bit(NF_FLOW_TEARDOWN, &flow->flags); + + if (test_bit(NF_FLOW_TEARDOWN, &flow->flags)) { if (test_bit(NF_FLOW_HW, &flow->flags)) { if (!test_bit(NF_FLOW_HW_DYING, &flow->flags)) nf_flow_offload_del(flow_table, flow); ^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH net] netfilter: flowtable: Fix expired flow not being deleted from software 2020-05-11 8:42 ` Pablo Neira Ayuso @ 2020-05-11 9:50 ` Paul Blakey 0 siblings, 0 replies; 5+ messages in thread From: Paul Blakey @ 2020-05-11 9:50 UTC (permalink / raw) To: Pablo Neira Ayuso Cc: Oz Shlomo, Roi Dayan, netdev, Saeed Mahameed, netfilter-devel On 5/11/2020 11:42 AM, Pablo Neira Ayuso wrote: > On Mon, May 11, 2020 at 10:24:44AM +0300, Paul Blakey wrote: >> >> On 5/11/2020 1:26 AM, Pablo Neira Ayuso wrote: >>> On Wed, May 06, 2020 at 02:27:29PM +0300, Paul Blakey wrote: >>>> Once a flow is considered expired, it is marked as DYING, and >>>> scheduled a delete from hardware. The flow will be deleted from >>>> software, in the next gc_step after hardware deletes the flow >>>> (and flow is marked DEAD). Till that happens, the flow's timeout >>>> might be updated from a previous scheduled stats, or software packets >>>> (refresh). This will cause the gc_step to no longer consider the flow >>>> expired, and it will not be deleted from software. >>>> >>>> Fix that by looking at the DYING flag as in deciding >>>> a flow should be deleted from software. >>> Would this work for you? >>> >>> The idea is to skip the refresh if this has already expired. >>> >>> Thanks. >> The idea is ok, but timeout check + update isn't atomic (need atomic_inc_unlesss >> or something like that), and there is also >> the hardware stats which if comes too late (after gc finds it expired) might >> bring a flow back to life. > Right. Once the entry has expired, there should not be a way turning > back. > > I'm attaching a new sketch, it's basically using the teardown state to > specify that the gc already made the decision to remove this entry. > > Thanks. Looks fine to me, are you submitting that instead? ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-05-11 9:50 UTC | newest] Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2020-05-06 11:27 [PATCH net] netfilter: flowtable: Fix expired flow not being deleted from software Paul Blakey 2020-05-10 22:26 ` Pablo Neira Ayuso 2020-05-11 7:24 ` Paul Blakey 2020-05-11 8:42 ` Pablo Neira Ayuso 2020-05-11 9:50 ` Paul Blakey
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).