tc: Add support for ce_threshold_value/mask in fq_codel

Commit dfcb63ce1de6 ("fq_codel: generalise ce_threshold marking for subset
of traffic") added support in fq_codel for setting a value and mask that
will be applied to the diffserv/ECN byte to turn on the ce_threshold
feature for a subset of traffic.

This adds support to iproute for setting these values. The parameter is
called ce_threshold_selector and takes a value followed by a
slash-separated mask. Some examples:

 # apply ce_threshold to ECT(1) traffic
 tc qdisc replace dev eth0 root fq_codel ce_threshold 1ms ce_threshold_selector 0x1/0x3

 # apply ce_threshold to ECN-capable traffic marked as diffserv AF22
 tc qdisc replace dev eth0 root fq_codel ce_threshold 1ms ce_threshold_selector 0x50/0xfc

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: David Ahern <dsahern@kernel.org>

diff --git a/man/man8/tc-fq_codel.8 b/man/man8/tc-fq_codel.8
index 7ee6c269ed424ddb0273740d42cedddda6cef02c..84340fe57498a58532033c634c0e24098603ad57 100644 (file)
--- a/man/man8/tc-fq_codel.8
+++ b/man/man8/tc-fq_codel.8
@@ -20,6 +20,8 @@ BYTES ] [
 ] [
 .B ce_threshold
 TIME ] [
+.B ce_threshold_selector
+VALUE/MASK ] [
 .B memory_limit
 BYTES ]
 
@@ -89,6 +91,15 @@ sets a threshold above which all packets are marked with ECN Congestion
 Experienced. This is useful for DCTCP-style congestion control algorithms that
 require marking at very shallow queueing thresholds.
 
+.SS ce_threshold_selector
+sets a filter so that the
+.B ce_threshold
+feature is applied to only a subset of the traffic seen by the qdisc. If set, the MASK value
+will be applied as a bitwise AND to the diffserv/ECN byte of the IP header, and only if the
+result of this masking equals VALUE, will the
+.B ce_threshold
+logic be applied to the packet.
+
 .SH EXAMPLES
 #tc qdisc add   dev eth0 root fq_codel
 .br

diff --git a/tc/q_fq_codel.c b/tc/q_fq_codel.c
index 30098065224330e3e05dc576d9f3febdc99cd912..b7552e294fd08e180608b096537f717b2a4f8629 100644 (file)
--- a/tc/q_fq_codel.c
+++ b/tc/q_fq_codel.c
@@ -55,6 +55,7 @@ static void explain(void)
                                        "[ target TIME ] [ interval TIME ]\n"
                                        "[ quantum BYTES ] [ [no]ecn ]\n"
                                        "[ ce_threshold TIME ]\n"
+                                       "[ ce_threshold_selector VALUE/MASK ]\n"
                                        "[ drop_batch SIZE ]\n");
 }
 
@@ -69,6 +70,8 @@ static int fq_codel_parse_opt(struct qdisc_util *qu, int argc, char **argv,
        unsigned int quantum = 0;
        unsigned int ce_threshold = ~0U;
        unsigned int memory = ~0U;
+       __u8 ce_threshold_mask = 0;
+       __u8 ce_threshold_selector = 0xFF;
        int ecn = -1;
        struct rtattr *tail;
 
@@ -109,6 +112,24 @@ static int fq_codel_parse_opt(struct qdisc_util *qu, int argc, char **argv,
                                fprintf(stderr, "Illegal \"ce_threshold\"\n");
                                return -1;
                        }
+               } else if (strcmp(*argv, "ce_threshold_selector") == 0) {
+                       char *sep;
+
+                       NEXT_ARG();
+                       sep = strchr(*argv, '/');
+                       if (!sep) {
+                               fprintf(stderr, "Missing mask for \"ce_threshold_selector\"\n");
+                               return -1;
+                       }
+                       *sep++ = '\0';
+                       if (get_u8(&ce_threshold_mask, sep, 0)) {
+                               fprintf(stderr, "Illegal mask for \"ce_threshold_selector\"\n");
+                               return -1;
+                       }
+                       if (get_u8(&ce_threshold_selector, *argv, 0)) {
+                               fprintf(stderr, "Illegal \"ce_threshold_selector\"\n");
+                               return -1;
+                       }
                } else if (strcmp(*argv, "memory_limit") == 0) {
                        NEXT_ARG();
                        if (get_size(&memory, *argv)) {
@@ -152,6 +173,10 @@ static int fq_codel_parse_opt(struct qdisc_util *qu, int argc, char **argv,
        if (ce_threshold != ~0U)
                addattr_l(n, 1024, TCA_FQ_CODEL_CE_THRESHOLD,
                          &ce_threshold, sizeof(ce_threshold));
+       if (ce_threshold_selector != 0xFF) {
+               addattr8(n, 1024, TCA_FQ_CODEL_CE_THRESHOLD_MASK, ce_threshold_mask);
+               addattr8(n, 1024, TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR, ce_threshold_selector);
+       }
        if (memory != ~0U)
                addattr_l(n, 1024, TCA_FQ_CODEL_MEMORY_LIMIT,
                          &memory, sizeof(memory));
@@ -172,6 +197,8 @@ static int fq_codel_print_opt(struct qdisc_util *qu, FILE *f, struct rtattr *opt
        unsigned int ecn;
        unsigned int quantum;
        unsigned int ce_threshold;
+       __u8 ce_threshold_selector = 0;
+       __u8 ce_threshold_mask = 0;
        unsigned int memory_limit;
        unsigned int drop_batch;
 
@@ -211,6 +238,19 @@ static int fq_codel_print_opt(struct qdisc_util *qu, FILE *f, struct rtattr *opt
                print_string(PRINT_FP, NULL, "ce_threshold %s ",
                             sprint_time(ce_threshold, b1));
        }
+       if (tb[TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR] &&
+           RTA_PAYLOAD(tb[TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR]) >= sizeof(__u8))
+               ce_threshold_selector = rta_getattr_u8(tb[TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR]);
+       if (tb[TCA_FQ_CODEL_CE_THRESHOLD_MASK] &&
+           RTA_PAYLOAD(tb[TCA_FQ_CODEL_CE_THRESHOLD_MASK]) >= sizeof(__u8))
+               ce_threshold_mask = rta_getattr_u8(tb[TCA_FQ_CODEL_CE_THRESHOLD_MASK]);
+       if (ce_threshold_mask || ce_threshold_selector) {
+               print_hhu(PRINT_ANY, "ce_threshold_selector", "ce_threshold_selector %#x",
+                         ce_threshold_selector);
+               print_hhu(PRINT_ANY, "ce_threshold_mask", "/%#x ",
+                         ce_threshold_mask);
+       }
+
        if (tb[TCA_FQ_CODEL_INTERVAL] &&
            RTA_PAYLOAD(tb[TCA_FQ_CODEL_INTERVAL]) >= sizeof(__u32)) {
                interval = rta_getattr_u32(tb[TCA_FQ_CODEL_INTERVAL]);