On 6/4/24 10:48, Adrián Moreno wrote: > On Tue, Jun 04, 2024 at 12:05:20AM GMT, Dumitru Ceara wrote: >> It improves the debugging experience if we can easily get a list of >> OpenFlow rules and groups that contribute to the creation of a datapath >> flow. >> >> The suggested workflow is: >> a. dump datapath flows (along with UUIDs), this also prints the core IDs >> (PMD IDs) when applicable. >> >> $ ovs-appctl dpctl/dump-flows -m >> flow-dump from pmd on cpu core: 7 >> ufid:7460db8f..., recirc_id(0), .... >> >> b. dump related OpenFlow rules and groups: >> $ ovs-appctl upcall/dump-ufid-rules ufid:7460db8f... pmd=7 >> cookie=0x12345678, table=0 >> priority=100,ip,in_port=2,nw_dst=10.0.0.2,actions=resubmit(,1) >> cookie=0x0, table=1 priority=200,actions=group:1 >> group_id=1,bucket=bucket_id:0,actions=ct(commit,table=2,nat(dst=20.0.0.2)) >> cookie=0x0, table=2 actions=output:1 >> >> The new command only shows rules and groups attached to ukeys that are >> in states UKEY_VISIBLE or UKEY_OPERATIONAL. That should be fine as all >> other ukeys should not be relevant for the use case presented above. >> >> For ukeys that don't have an xcache populated yet, the command goes >> ahead and populates one. In theory this is creates a slight overhead as >> those ukeys might not need an xcache until they see traffic (and get >> revalidated) but in practice the overhead should be minimal. >> >> This commit tries to mimic the output format of the ovs-ofctl >> dump-flows/dump-groups commands. For groups it actually uses >> ofputil_group/_bucket functions for formatting. For rules it's not that >> straightforward or clean as 'struct rule' uses its own versions of >> fields to describe the match, actions, and other relevant fields and >> 'ovs-ofctl dump-flows' operates on the equivalent 'struct >> ofputil_flow_stats'. >> > > Thanks Dumitru, this is very much needed! > See some comments inline. >
Thanks, Ales and Adrian for the review! >> Signed-off-by: Dumitru Ceara <[email protected]> >> --- >> NEWS | 3 + >> include/openvswitch/ofp-group.h | 7 ++ >> lib/ofp-group.c | 110 +++++++++++++++++++------------- >> ofproto/ofproto-dpif-upcall.c | 87 +++++++++++++++++++++++++ >> ofproto/ofproto-dpif.c | 33 ++++++++++ >> ofproto/ofproto-dpif.h | 4 ++ >> tests/ofproto-dpif.at | 39 +++++++++++ >> tests/ofproto-macros.at | 4 ++ >> 8 files changed, 243 insertions(+), 44 deletions(-) >> >> diff --git a/NEWS b/NEWS >> index 5ae0108d552b..1bc085f97045 100644 >> --- a/NEWS >> +++ b/NEWS >> @@ -9,6 +9,9 @@ Post-v3.3.0 >> https://github.com/openvswitch/ovs.git >> - DPDK: >> * OVS validated with DPDK 23.11.1. >> + - ovs-appctl: >> + * Added 'upcall/dump-ufid-rules' to output the set of OpenFlow rules >> and >> + groups that contributed to the creation of a specific datapath flow. >> >> >> v3.3.0 - 16 Feb 2024 >> diff --git a/include/openvswitch/ofp-group.h >> b/include/openvswitch/ofp-group.h >> index cd7af0ebff9c..79fcb3a4c0d1 100644 >> --- a/include/openvswitch/ofp-group.h >> +++ b/include/openvswitch/ofp-group.h >> @@ -70,6 +70,11 @@ struct ofputil_bucket *ofputil_bucket_find(const struct >> ovs_list *, >> bool ofputil_bucket_check_duplicate_id(const struct ovs_list *); >> struct ofputil_bucket *ofputil_bucket_list_front(const struct ovs_list *); >> struct ofputil_bucket *ofputil_bucket_list_back(const struct ovs_list *); >> +void ofputil_bucket_format(const struct ofputil_bucket *, >> + enum ofp11_group_type, enum ofp_version, >> + const struct ofputil_port_map *, >> + const struct ofputil_table_map *, >> + struct ds *); >> >> static inline bool >> ofputil_bucket_has_liveness(const struct ofputil_bucket *bucket) >> @@ -88,6 +93,8 @@ struct ofputil_group_props { >> void ofputil_group_properties_destroy(struct ofputil_group_props *); >> void ofputil_group_properties_copy(struct ofputil_group_props *to, >> const struct ofputil_group_props *from); >> +void ofputil_group_properties_format(const struct ofputil_group_props *, >> + struct ds *); >> /* Protocol-independent group_mod. */ >> struct ofputil_group_mod { >> uint16_t command; /* One of OFPGC15_*. */ >> diff --git a/lib/ofp-group.c b/lib/ofp-group.c >> index 737f48047b10..8f7614c22f0e 100644 >> --- a/lib/ofp-group.c >> +++ b/lib/ofp-group.c >> @@ -58,14 +58,16 @@ ofputil_group_from_string(const char *s, uint32_t >> *group_idp) >> return true; >> } >> >> -/* Appends to 's' a string representation of the OpenFlow group ID >> 'group_id'. >> - * Most groups' string representation is just the number, but for special >> - * groups, e.g. OFPG_ALL, it is the name, e.g. "ALL". */ >> +/* Appends to 's' a string representation of the OpenFlow group. 'group_id'. > > nit: s/group. 'group_id'/group 'group_id'/ > Ack. >> + * Most groups' string representation is just 'group_id=' followed by the >> ID, >> + * but for special groups, e.g. OFPG_ALL, the ID is replaced by the name, >> + * e.g. "ALL". */ >> void >> ofputil_format_group(uint32_t group_id, struct ds *s) >> { >> char name[MAX_GROUP_NAME_LEN + 1]; >> >> + ds_put_cstr(s, "group_id="); >> ofputil_group_to_string(group_id, name, sizeof name); >> ds_put_cstr(s, name); >> } >> @@ -297,7 +299,7 @@ ofputil_group_desc_request_format(struct ds *string, >> const struct ofp_header *oh) >> { >> uint32_t group_id = ofputil_decode_group_desc_request(oh); >> - ds_put_cstr(string, " group_id="); >> + ds_put_cstr(string, " "); >> ofputil_format_group(group_id, string); >> >> return 0; >> @@ -585,7 +587,7 @@ ofputil_group_stats_request_format(struct ds *string, >> return error; >> } >> >> - ds_put_cstr(string, " group_id="); >> + ds_put_cstr(string, " "); >> ofputil_format_group(group_id, string); >> return 0; >> } >> @@ -1526,6 +1528,31 @@ ofputil_group_properties_destroy(struct >> ofputil_group_props *gp) >> free(gp->fields.values); >> } >> >> +void >> +ofputil_group_properties_format(const struct ofputil_group_props *gp, >> + struct ds *ds) >> +{ >> + if (!gp->selection_method[0]) { >> + return; >> + } >> + >> + ds_put_format(ds, ",selection_method=%s", gp->selection_method); >> + if (gp->selection_method_param) { >> + ds_put_format(ds, ",selection_method_param=%"PRIu64, >> + gp->selection_method_param); >> + } >> + >> + size_t n = bitmap_count1(gp->fields.used.bm, MFF_N_IDS); >> + if (n == 1) { >> + ds_put_cstr(ds, ",fields="); >> + oxm_format_field_array(ds, &gp->fields); >> + } else if (n > 1) { >> + ds_put_cstr(ds, ",fields("); >> + oxm_format_field_array(ds, &gp->fields); >> + ds_put_char(ds, ')'); >> + } >> +} >> + >> static enum ofperr >> parse_group_prop_ntr_selection_method(struct ofpbuf *payload, >> enum ofp11_group_type group_type, >> @@ -1813,6 +1840,37 @@ ofp_print_bucket_id(struct ds *s, const char *label, >> uint32_t bucket_id, >> ds_put_char(s, ','); >> } >> >> +void >> +ofputil_bucket_format(const struct ofputil_bucket *bucket, >> + enum ofp11_group_type type, enum ofp_version >> ofp_version, >> + const struct ofputil_port_map *port_map, >> + const struct ofputil_table_map *table_map, >> + struct ds *s) >> +{ >> + ds_put_cstr(s, "bucket="); >> + >> + ofp_print_bucket_id(s, "bucket_id:", bucket->bucket_id, ofp_version); >> + if (bucket->weight != (type == OFPGT11_SELECT ? 1 : 0)) { >> + ds_put_format(s, "weight:%"PRIu16",", bucket->weight); >> + } >> + if (bucket->watch_port != OFPP_NONE) { >> + ds_put_cstr(s, "watch_port:"); >> + ofputil_format_port(bucket->watch_port, port_map, s); >> + ds_put_char(s, ','); >> + } >> + if (bucket->watch_group != OFPG_ANY) { >> + ds_put_format(s, "watch_group:%"PRIu32",", bucket->watch_group); >> + } >> + >> + ds_put_cstr(s, "actions="); >> + struct ofpact_format_params fp = { >> + .port_map = port_map, >> + .table_map = table_map, >> + .s = s, >> + }; >> + ofpacts_format(bucket->ofpacts, bucket->ofpacts_len, &fp); >> +} >> + >> static void >> ofp_print_group(struct ds *s, uint32_t group_id, uint8_t type, >> const struct ovs_list *p_buckets, >> @@ -1831,23 +1889,7 @@ ofp_print_group(struct ds *s, uint32_t group_id, >> uint8_t type, >> ds_put_format(s, ",type=%s", type_str[type > 4 ? 4 : type]); >> } >> >> - if (props->selection_method[0]) { >> - ds_put_format(s, ",selection_method=%s", props->selection_method); >> - if (props->selection_method_param) { >> - ds_put_format(s, ",selection_method_param=%"PRIu64, >> - props->selection_method_param); >> - } >> - >> - size_t n = bitmap_count1(props->fields.used.bm, MFF_N_IDS); >> - if (n == 1) { >> - ds_put_cstr(s, ",fields="); >> - oxm_format_field_array(s, &props->fields); >> - } else if (n > 1) { >> - ds_put_cstr(s, ",fields("); >> - oxm_format_field_array(s, &props->fields); >> - ds_put_char(s, ')'); >> - } >> - } >> + ofputil_group_properties_format(props, s); >> >> if (!p_buckets) { >> return; >> @@ -1856,28 +1898,8 @@ ofp_print_group(struct ds *s, uint32_t group_id, >> uint8_t type, >> ds_put_char(s, ','); >> >> LIST_FOR_EACH (bucket, list_node, p_buckets) { >> - ds_put_cstr(s, "bucket="); >> - >> - ofp_print_bucket_id(s, "bucket_id:", bucket->bucket_id, >> ofp_version); >> - if (bucket->weight != (type == OFPGT11_SELECT ? 1 : 0)) { >> - ds_put_format(s, "weight:%"PRIu16",", bucket->weight); >> - } >> - if (bucket->watch_port != OFPP_NONE) { >> - ds_put_cstr(s, "watch_port:"); >> - ofputil_format_port(bucket->watch_port, port_map, s); >> - ds_put_char(s, ','); >> - } >> - if (bucket->watch_group != OFPG_ANY) { >> - ds_put_format(s, "watch_group:%"PRIu32",", bucket->watch_group); >> - } >> - >> - ds_put_cstr(s, "actions="); >> - struct ofpact_format_params fp = { >> - .port_map = port_map, >> - .table_map = table_map, >> - .s = s, >> - }; >> - ofpacts_format(bucket->ofpacts, bucket->ofpacts_len, &fp); >> + ofputil_bucket_format(bucket, type, ofp_version, >> + port_map, table_map, s); >> ds_put_char(s, ','); >> } >> >> diff --git a/ofproto/ofproto-dpif-upcall.c b/ofproto/ofproto-dpif-upcall.c >> index 83609ec62b63..810fad159d2c 100644 >> --- a/ofproto/ofproto-dpif-upcall.c >> +++ b/ofproto/ofproto-dpif-upcall.c >> @@ -383,6 +383,9 @@ static void upcall_unixctl_disable_ufid(struct >> unixctl_conn *, int argc, >> const char *argv[], void >> *aux); >> static void upcall_unixctl_enable_ufid(struct unixctl_conn *, int argc, >> const char *argv[], void *aux); >> +static void upcall_unixctl_dump_ufid_rules(struct unixctl_conn *, int argc, >> + const char *argv[], void *aux); >> + >> static void upcall_unixctl_set_flow_limit(struct unixctl_conn *conn, int >> argc, >> const char *argv[], void *aux); >> static void upcall_unixctl_dump_wait(struct unixctl_conn *conn, int argc, >> @@ -460,6 +463,8 @@ udpif_init(void) >> upcall_unixctl_disable_ufid, NULL); >> unixctl_command_register("upcall/enable-ufid", "", 0, 0, >> upcall_unixctl_enable_ufid, NULL); >> + unixctl_command_register("upcall/dump-ufid-rules", "UFID PMD-ID", >> 1, 2, >> + upcall_unixctl_dump_ufid_rules, NULL); >> unixctl_command_register("upcall/set-flow-limit", >> "flow-limit-number", >> 1, 1, upcall_unixctl_set_flow_limit, NULL); >> unixctl_command_register("revalidator/wait", "", 0, 0, >> @@ -2479,6 +2484,48 @@ revalidate_ukey(struct udpif *udpif, struct udpif_key >> *ukey, >> return result; >> } >> >> +static void >> +ukey_xcache_format(struct udpif *udpif, struct udpif_key *ukey, struct ds >> *ds) >> + OVS_REQUIRES(ukey->mutex) >> +{ >> + /* It only makes sense to format rules for ukeys that are (still) >> + * in use. */ >> + if (ukey->state != UKEY_VISIBLE && ukey->state != UKEY_OPERATIONAL) { >> + return; >> + } > > I agree it's probably OK not to print ukeys that are about to disappear. > However, I don't know if that should be decided by "ukey_xcache_format" > or by its caller. > Makes sense, I moved it to the caller. >> + >> + if (!ukey->xcache) { >> + populate_xcache(udpif, ukey, ukey->stats.tcp_flags); > > Not that it'll be very common but if populate_xcache returns non-zero, > this ukey cannot could not be xlated and the xcache might have partially > initialized or invalid data. We should probably not try to print its > content as the revalidator will remove it soon enough. > Oops I missed that. I added a check now. >> + } >> + >> + struct ofpbuf entries = ukey->xcache->entries; >> + struct xc_entry *entry; >> + >> + XC_ENTRY_FOR_EACH (entry, &entries) { >> + switch (entry->type) { >> + case XC_RULE: >> + rule_dpif_format(entry->rule, ds); >> + ds_put_char(ds, '\n'); >> + break; >> + case XC_GROUP: >> + group_dpif_format(entry->group.group, entry->group.bucket, ds); >> + ds_put_char(ds, '\n'); >> + break; >> + case XC_TABLE: >> + case XC_BOND: >> + case XC_NETDEV: >> + case XC_NETFLOW: >> + case XC_MIRROR: >> + case XC_LEARN: >> + case XC_NORMAL: >> + case XC_FIN_TIMEOUT: >> + case XC_TNL_NEIGH: >> + case XC_TUNNEL_HEADER: >> + break; >> + } >> + } >> +} >> + >> static void >> delete_op_init__(struct udpif *udpif, struct ukey_op *op, >> const struct dpif_flow *flow) >> @@ -3220,6 +3267,46 @@ upcall_unixctl_enable_ufid(struct unixctl_conn *conn, >> int argc OVS_UNUSED, >> "for supported datapaths"); >> } >> >> +static void >> +upcall_unixctl_dump_ufid_rules(struct unixctl_conn *conn, int argc, >> + const char *argv[], void *aux OVS_UNUSED) >> +{ >> + unsigned int pmd_id = PMD_ID_NULL; >> + const char *key_s = argv[1]; >> + ovs_u128 ufid; >> + >> + if (odp_ufid_from_string(key_s, &ufid) <= 0) { >> + unixctl_command_reply_error(conn, "failed to parse ufid"); >> + return; >> + } >> + >> + if (argc == 3) { >> + const char *pmd_str = argv[2]; >> + if (!ovs_scan(pmd_str, "pmd=%d", &pmd_id)) { >> + unixctl_command_reply_error(conn, >> + "Invalid pmd argument format. " >> + "Expecting 'pmd=PMD-ID'"); >> + return; >> + } >> + } >> + >> + struct ds ds = DS_EMPTY_INITIALIZER; >> + struct udpif *udpif; >> + >> + LIST_FOR_EACH (udpif, list_node, &all_udpifs) { >> + struct udpif_key *ukey = ukey_lookup(udpif, &ufid, pmd_id); >> + if (!ukey) { >> + continue; >> + } >> + >> + ovs_mutex_lock(&ukey->mutex); >> + ukey_xcache_format(udpif, ukey, &ds); >> + ovs_mutex_unlock(&ukey->mutex); >> + } >> + unixctl_command_reply(conn, ds_cstr(&ds)); >> + ds_destroy(&ds); >> +} >> + >> /* Set the flow limit. >> * >> * This command is only needed for advanced debugging, so it's not >> diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c >> index fcd7cd753ca4..c554c2607515 100644 >> --- a/ofproto/ofproto-dpif.c >> +++ b/ofproto/ofproto-dpif.c >> @@ -4457,6 +4457,27 @@ rule_set_recirc_id(struct rule *rule_, uint32_t id) >> ovs_mutex_unlock(&rule->up.mutex); >> } >> >> +void >> +rule_dpif_format(const struct rule_dpif *rule, struct ds *ds) >> +{ > > Have you considered using "flow_stats_ds"? Defined in ofproto layer, > it pretty much does the same as this function (plus prints the stats). > It used by "ovs-appctl bridge/dump-flows" which is not as popular as its > OFP equivalent but using it would still save us a function. > I haven't because I didn't see it, thanks for pointing it out! I was only looking in the ofp layer. :) Now that you pointed it out I used it. I did have to export it and rename it to ofproto_rule_stats_ds() to match the rest of the public functions in ofproto-provider.h. >> + const struct tun_table *tun_table = >> ofproto_get_tun_tab(rule->up.ofproto); >> + ds_put_format(ds, "cookie=0x%"PRIx64", table=%"PRIu8" ", >> + ntohll(rule->up.flow_cookie), >> + rule->up.table_id); >> + >> + size_t rule_header_len = ds->length; >> + cls_rule_format(&rule->up.cr, tun_table, NULL, ds); >> + if (ds->length != rule_header_len) { >> + /* Only add comma if the match is non default. */ >> + ds_put_char(ds, ','); >> + } >> + >> + ds_put_cstr(ds, "actions="); >> + struct ofpact_format_params fp = { .s = ds }; >> + ofpacts_format(rule->up.actions->ofpacts, rule->up.actions->ofpacts_len, >> + &fp); >> +} >> + >> ovs_version_t >> ofproto_dpif_get_tables_version(struct ofproto_dpif *ofproto) >> { >> @@ -5407,6 +5428,18 @@ group_dpif_lookup(struct ofproto_dpif *ofproto, >> uint32_t group_id, >> version, take_ref); >> return ofgroup ? group_dpif_cast(ofgroup) : NULL; >> } >> + >> +void >> +group_dpif_format(struct group_dpif *group, struct ofputil_bucket *bucket, >> + struct ds *ds) >> +{ >> + struct ofgroup *ofg = &group->up; >> + >> + ofputil_format_group(ofg->group_id, ds); >> + ofputil_group_properties_format(&ofg->props, ds); >> + ds_put_char(ds, ','); >> + ofputil_bucket_format(bucket, ofg->type, OFP15_VERSION, NULL, NULL, ds); > > Coming from the entry in the xcache, "bucket" could be NULL for buckets > of type "all" or "indirect". If so, all buckets should be printed. > > I think in this function we should check for that condition or else > ofputil_bucket_format could segfault. > Nice catch! I changed it to print all buckets in that case. > Also, a test for this case might be useful. > Done. >> +} >> >> /* Sends 'packet' out 'ofport'. If 'port' is a tunnel and that tunnel type >> * supports a notion of an OAM flag, sets it if 'oam' is true. >> diff --git a/ofproto/ofproto-dpif.h b/ofproto/ofproto-dpif.h >> index d33f73df8aed..76255b089e76 100644 >> --- a/ofproto/ofproto-dpif.h >> +++ b/ofproto/ofproto-dpif.h >> @@ -118,6 +118,8 @@ rule_dpif_is_internal(const struct rule_dpif *rule) >> { >> return rule->up.table_id == TBL_INTERNAL; >> } >> + >> +void rule_dpif_format(const struct rule_dpif *, struct ds *); >> >> /* Groups. */ >> >> @@ -151,6 +153,8 @@ void group_dpif_credit_stats(struct group_dpif *, >> struct group_dpif *group_dpif_lookup(struct ofproto_dpif *, >> uint32_t group_id, ovs_version_t >> version, >> bool take_ref); >> +void group_dpif_format(struct group_dpif *, struct ofputil_bucket *, >> + struct ds *); >> >> >> /* Backers. >> diff --git a/tests/ofproto-dpif.at b/tests/ofproto-dpif.at >> index 0b23fd6c5ea6..47bbde35e5d3 100644 >> --- a/tests/ofproto-dpif.at >> +++ b/tests/ofproto-dpif.at >> @@ -12136,3 +12136,42 @@ AT_CHECK([test 1 = `ovs-ofctl parse-pcap p2-tx.pcap >> | wc -l`]) >> >> OVS_VSWITCHD_STOP >> AT_CLEANUP >> + >> +AT_SETUP([ofproto-dpif - dump-ufid-rules]) >> +OVS_VSWITCHD_START( >> +[add-port br0 p1 \ >> + -- set bridge br0 datapath-type=dummy \ >> + -- set interface p1 type=dummy-pmd \ >> + -- add-port br0 p2 \ >> + -- set interface p2 type=dummy-pmd >> +], [], [], [DUMMY_NUMA]) >> + >> +dnl Add some OpenFlow rules and groups. >> +AT_DATA([groups.txt], [dnl >> + >> group_id=1,type=select,selection_method=dp_hash,bucket=bucket_id:0,weight:100,actions=ct(commit,table=2,nat(dst=20.0.0.2)) >> +]) >> +AT_DATA([flows.txt], [dnl >> +table=0,priority=100,cookie=0x12345678,in_port=p1,ip,nw_dst=10.0.0.2,actions=resubmit(,1) >> +table=1,priority=200,actions=group:1 >> +table=2,actions=p2 >> +]) >> +AT_CHECK([ovs-ofctl add-groups br0 groups.txt]) >> +AT_CHECK([ovs-ofctl add-flows br0 flows.txt]) >> + >> +AT_CHECK([ovs-appctl netdev-dummy/receive p1 >> 'ipv4(src=10.0.0.1,dst=10.0.0.2,proto=6),tcp(src=1,dst=2)']) >> + >> +OVS_WAIT_UNTIL_EQUAL([ovs-appctl dpctl/dump-flows | sed 's/.*core: >> [[0-9]]*//' | sort], [ >> +recirc_id(0),in_port(1),packet_type(ns=0,id=0),eth_type(0x0800),ipv4(dst=10.0.0.2,frag=no), >> packets:0, bytes:0, used:never, actions:hash(l4(0)),recirc(0x1) >> +recirc_id(0x1),dp_hash(0xfbe73382/0xf),in_port(1),packet_type(ns=0,id=0),eth_type(0x0800),ipv4(frag=no), >> packets:0, bytes:0, used:never, >> actions:ct(commit,nat(dst=20.0.0.2)),recirc(0x2) >> +recirc_id(0x2),in_port(1),packet_type(ns=0,id=0),eth_type(0x0800),ipv4(frag=no), >> packets:0, bytes:0, used:never, actions:2]) >> + >> +ufids=$(ovs-appctl dpctl/dump-flows -m | match_ufid) >> +AT_CHECK([for ufid in $ufids; do ovs-appctl upcall/dump-ufid-rules $ufid >> pmd=0; done | sort], [0], [dnl >> +cookie=0x0, table=1 priority=200,actions=group:1 >> +cookie=0x0, table=2 actions=output:1 >> +cookie=0x12345678, table=0 >> priority=100,ip,in_port=2,nw_dst=10.0.0.2,actions=resubmit(,1) >> +group_id=1,selection_method=dp_hash,bucket=bucket_id:0,weight:100,actions=ct(commit,table=2,nat(dst=20.0.0.2)) >> +]) >> + >> +OVS_VSWITCHD_STOP >> +AT_CLEANUP >> diff --git a/tests/ofproto-macros.at b/tests/ofproto-macros.at >> index c22fb3c79c3f..a5997902c09c 100644 >> --- a/tests/ofproto-macros.at >> +++ b/tests/ofproto-macros.at >> @@ -140,6 +140,10 @@ strip_ufid () { >> s/ufid:[[-0-9a-f]]* //' >> } >> >> +match_ufid () { >> + grep -oE 'ufid:[[-0-9a-f]]+' | sort -u >> +} >> + >> # Strips packets: and bytes: from output >> strip_stats () { >> sed 's/packets:[[0-9]]*/packets:0/ >> -- >> 2.44.0 >> > V2 posted: https://patchwork.ozlabs.org/project/openvswitch/patch/[email protected]/ Thanks again for the reviews! Regards, Dumitru _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
