dpif-netdev: Set PMD load based sleep start/inc to 1 us.

Now that the timer slack for the PMD threads is reduced we can also
reduce the start/increment for PMD load based sleeping to match it.

This will further reduce initial sleep times making it more resilient
to interfaces that might be sensitive to large sleep times.

Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>

4 files changed, 10 insertions, 18 deletions

diff --git a/Documentation/topics/dpdk/pmd.rst b/Documentation/topics/dpdk/pmd.rst
index 0c3bb717f..e70986d16 100644
--- a/Documentation/topics/dpdk/pmd.rst
+++ b/Documentation/topics/dpdk/pmd.rst
@@ -334,10 +334,7 @@ when there is no load or very-low load on all the Rx queues they poll.
 This can be enabled by setting the max requested sleep time (in microseconds)
 for a PMD thread::
 
-    $ ovs-vsctl set open_vswitch . other_config:pmd-maxsleep=500
-
-Non-zero values will be rounded up to the nearest 10 microseconds to avoid
-requesting very small sleep times.
+    $ ovs-vsctl set open_vswitch . other_config:pmd-maxsleep=50
 
 With a non-zero max value a PMD may request to sleep by an incrementing amount
 of time up to the maximum time. If at any point the threshold of at least half
@@ -356,12 +353,12 @@ Sleep time statistics for 10 secs can be seen with::
     $ ovs-appctl dpif-netdev/pmd-stats-clear \
         && sleep 10 && ovs-appctl dpif-netdev/pmd-perf-show
 
-Example output, showing that during the last 10 seconds, 76.8% of iterations
-had a sleep of some length. The total amount of sleep time was 9.15 seconds and
-the average sleep time per iteration was 46 microseconds::
+Example output, showing that during the last 10 seconds, 74.5% of iterations
+had a sleep of some length. The total amount of sleep time was 9.06 seconds
+and the average sleep time where a sleep was requested was 9 microseconds::
 
-   - sleep iterations:       153994  ( 76.8 % of iterations)
-   Sleep time (us):         9159399  ( 59 us/iteration avg.)
+   - sleep iterations:       977037  ( 74.5 % of iterations)
+   Sleep time (us):         9068841  (  9 us/iteration avg.)
 
 Any potential power saving from PMD load based sleeping is dependent on the
 system configuration (e.g. enabling processor C-states) and workloads.
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 4f06e3f4e..c9f7179c3 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -177,7 +177,7 @@ static struct odp_support dp_netdev_support = {
 /* Number of pkts Rx on an interface that will stop pmd thread sleeping. */
 #define PMD_SLEEP_THRESH (NETDEV_MAX_BURST / 2)
 /* Time in uS to increment a pmd thread sleep time. */
-#define PMD_SLEEP_INC_US 10
+#define PMD_SLEEP_INC_US 1
 
 struct dpcls {
     struct cmap_node node;      /* Within dp_netdev_pmd_thread.classifiers */
@@ -4983,7 +4983,6 @@ dpif_netdev_set_config(struct dpif *dpif, const struct smap *other_config)
     set_pmd_auto_lb(dp, autolb_state, log_autolb);
 
     pmd_max_sleep = smap_get_ullong(other_config, "pmd-maxsleep", 0);
-    pmd_max_sleep = ROUND_UP(pmd_max_sleep, 10);
     pmd_max_sleep = MIN(PMD_RCU_QUIESCE_INTERVAL, pmd_max_sleep);
     atomic_read_relaxed(&dp->pmd_max_sleep, &cur_pmd_max_sleep);
     if (first_set_config || pmd_max_sleep != cur_pmd_max_sleep) {
diff --git a/tests/pmd.at b/tests/pmd.at
index e0f58f7a6..c707f762c 100644
--- a/tests/pmd.at
+++ b/tests/pmd.at
@@ -1266,7 +1266,7 @@ OVS_WAIT_UNTIL([tail ovs-vswitchd.log | grep "PMD load based sleeps are disabled
 dnl Check low value max sleep
 get_log_next_line_num
 AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-maxsleep="1"])
-OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD max sleep request is 10 usecs."])
+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD max sleep request is 1 usecs."])
 OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD load based sleeps are enabled."])
 
 dnl Check high value max sleep
@@ -1294,8 +1294,8 @@ OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD max sleep request
 OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD load based sleeps are enabled."])
 dnl Check rounding
 get_log_next_line_num
-AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-maxsleep="491"])
-OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD max sleep request is 500 usecs."])
+AT_CHECK([ovs-vsctl set open_vswitch . other_config:pmd-maxsleep="499"])
+OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD max sleep request is 499 usecs."])
 OVS_WAIT_UNTIL([tail -n +$LINENUM ovs-vswitchd.log | grep "PMD load based sleeps are enabled."])
 
 OVS_VSWITCHD_STOP
diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
index 8c4acfb18..2b57fc0e3 100644
--- a/vswitchd/vswitch.xml
+++ b/vswitchd/vswitch.xml
@@ -807,10 +807,6 @@
           Rx queues that it polls.
         </p>
         <p>
-          To avoid requesting very small sleeps (e.g. less than 10 us) the
-          value will be rounded up to the nearest 10 us.
-        </p>
-        <p>
           The maximum value is <code>10000 microseconds</code>.
         </p>
       </column>