sctp: make the PLPMTUD probe more effective and efficient
Subsystems (4)
Xin Long
maintainerv2 cover sctp: make the PLPMTUD probe more effective and efficient
As David Laight noticed, it currently takes quite some time to find the optimal pmtu in the Search state, and also lacks the black hole detection in the Search Complete state. This patchset is to address them to mke the PLPMTUD probe more effective and efficient.
v1->v2:
- see Patch 1/2.
Xin Long (2):
sctp: do black hole detection in search complete state
sctp: send the next probe immediately once the last one is acked
Documentation/networking/ip-sysctl.rst | 12 ++++++++---- include/net/sctp/structs.h | 3 ++- net/sctp/sm_statefuns.c | 5 ++++- net/sctp/transport.c | 16 ++++++++-------- 4 files changed, 22 insertions(+), 14 deletions(-)
Signature · 1 line
2.27.0
Xin Long
maintainerPATCH v2 1/2 sctp: do black hole detection in search complete state
Currently the PLPMUTD probe will stop for a long period (interval * 30) after it enters search complete state. If there's a pmtu change on the route path, it takes a long time to be aware if the ICMP TooBig packet is lost or filtered.
As it says in rfc8899#section-4.3:
"A DPLPMTUD method MUST NOT rely solely on this method." (ICMP PTB message).
This patch is to enable the other method for search complete state:
"A PL can use the DPLPMTUD probing mechanism to periodically
generate probe packets of the size of the current PLPMTU."
With this patch, the probe will continue with the current pmtu every 'interval' until the PMTU_RAISE_TIMER 'timeout', which we implement by adding raise_count to raise the probe size when it counts to 30 and removing the SCTP_PL_COMPLETE check for PMTU_RAISE_TIMER.
v1->v2:
- do t->pl.raise_count++ and t->pl.raise_count check separately, as Marcelo suggested.
--- include/net/sctp/structs.h | 3 ++- net/sctp/transport.c | 16 ++++++++-------- 2 files changed, 10 insertions(+), 9 deletions(-)
include/net/sctp/structs.h +2 -1
@@ -987,7 +987,8 @@ struct sctp_transport {__u16 pmtu;__u16 probe_size;__u16 probe_high; -__u8 probe_count; + __u8 probe_count:3; + __u8 raise_count:5;__u8 state; } pl;/* plpmtud related */
net/sctp/transport.c +8 -8
@@ -213,15 +213,10 @@ void sctp_transport_reset_reconf_timer(struct sctp_transport *transport)void sctp_transport_reset_probe_timer (struct sctp_transport *transport) { -int scale =1 ; -if (timer_pending (&transport->probe_timer))return ; -if (transport->pl.state ==SCTP_PL_COMPLETE && - transport->pl.probe_count ==1 ) - scale =30 ;/* works as PMTU_RAISE_TIMER */ if (!mod_timer (&transport->probe_timer, - jiffies + transport->probe_interval * scale)) + jiffies + transport->probe_interval))sctp_transport_hold (transport); }
@@ -333,6 +328,7 @@ void sctp_transport_pl_recv(struct sctp_transport *t) t->pl.probe_size +=SCTP_PL_MIN_STEP ;if (t->pl.probe_size >= t->pl.probe_high) { t->pl.probe_high =0 ; + t->pl.raise_count =0 ; t->pl.state =SCTP_PL_COMPLETE ;/* Search -> Search Complete */ t->pl.probe_size = t->pl.pmtu;
@@ -340,8 +336,12 @@ void sctp_transport_pl_recv(struct sctp_transport *t)sctp_assoc_sync_pmtu (t->asoc); } }else if (t->pl.state ==SCTP_PL_COMPLETE ) { - t->pl.state =SCTP_PL_SEARCH ;/* Search Complete -> Search */ - t->pl.probe_size +=SCTP_PL_MIN_STEP ; + t->pl.raise_count++; +if (t->pl.raise_count ==30 ) { +/* Raise probe_size again after 30 * interval in Search Complete */ + t->pl.state =SCTP_PL_SEARCH ;/* Search Complete -> Search */ + t->pl.probe_size +=SCTP_PL_MIN_STEP ; + } } }
Signature · 1 line
2.27.0
Xin Long
maintainerPATCH v2 2/2 sctp: send the next probe immediately once the last one is acked
In reply to Xin LongThese is no need to wait for 'interval' period for the next probe if the last probe is already acked in search state. The 'interval' period waiting should be only for probe failure timeout and the current pmtu check when it's in search complete state.
This change will shorten the probe time a lot in search state, and also fix the document accordingly.
--- Documentation/networking/ip-sysctl.rst | 12 ++++++++---- net/sctp/sm_statefuns.c | 5 ++++- 2 files changed, 12 insertions(+), 5 deletions(-)
Documentation/networking/ip-sysctl.rst +8 -4
@@ -2835,10 +2835,14 @@ encap_port - INTEGER Default: 0 plpmtud_probe_interval - INTEGER - The time interval (in milliseconds) for sending PLPMTUD probe chunks. - These chunks are sent at the specified interval with a variable size - to probe the mtu of a given path between 2 endpoints. PLPMTUD will - be disabled when 0 is set, and other values for it must be >= 5000. + The time interval (in milliseconds) for the PLPMTUD probe timer, + which is configured to expire after this period to receive an + acknowledgment to a probe packet. This is also the time interval + between the probes for the current pmtu when the probe search + is done. + + PLPMTUD will be disabled when 0 is set, and other values for it + must be >= 5000. Default: 0
net/sctp/sm_statefuns.c +4 -1
@@ -1275,7 +1275,10 @@ enum sctp_disposition sctp_sf_backbeat_8_3(struct net *net,return SCTP_DISPOSITION_DISCARD ;sctp_transport_pl_recv (link); -return SCTP_DISPOSITION_CONSUME ; +if (link->pl.state ==SCTP_PL_COMPLETE ) +return SCTP_DISPOSITION_CONSUME ; + +return sctp_sf_send_probe (net, ep, asoc, type, link, commands); } max_interval = link->hbinterval + link->rto;
Signature · 1 line
2.27.0
Quoted 11 lines · Xin Long
As David Laight noticed, it currently takes quite some time to find the optimal pmtu in the Search state, and also lacks the black hole detection in the Search Complete state. This patchset is to address them to mke the PLPMTUD probe more effective and efficient.
v1->v2:
- see Patch 1/2.Xin Long (2):
sctp: do black hole detection in search complete state
sctp: send the next probe immediately once the last one is acked
David Miller
maintainerI applied v1 of this so I'll need relative fixups.
Thank you.