When user_frac in the polling subsystem is low it is going to busy the

CPU for too long period than necessary.  Additively, interfaces are kept
polled (in the tick) even if no more packets are available.
In order to avoid such situations a new generic mechanism can be
implemented in proactive way, keeping track of the time spent on any
packet and fragmenting the time for any tick, stopping the processing
as soon as possible.

In order to implement such mechanism, the polling handler needs to
change, returning the number of packets processed.
While the intended logic is not part of this patch, the polling KPI is
broken by this commit, adding an int return value and the new flag
IFCAP_POLLING_NOCOUNT (which will signal that the return value is
meaningless for the installed handler and checking should be skipped).

Bump __FreeBSD_version in order to signal such situation.

Reviewed by:	emaste
Sponsored by:	Sandvine Incorporated

1 files changed, 2 insertions, 1 deletions

diff --git a/sys/dev/lmc/if_lmc.c b/sys/dev/lmc/if_lmc.c
index c97bdfa..1df51ad 100644
--- a/sys/dev/lmc/if_lmc.c
+++ b/sys/dev/lmc/if_lmc.c
@@ -3960,7 +3960,7 @@ user_interrupt(softc_t *sc, int check_status)
 # if (defined(__FreeBSD__) && defined(DEVICE_POLLING))
 
 /* Service the card from the kernel idle loop without interrupts. */
-static void
+static int
 fbsd_poll(struct ifnet *ifp, enum poll_cmd cmd, int count)
   {
   softc_t *sc = IFP2SC(ifp);
@@ -4828,6 +4828,7 @@ setup_ifnet(struct ifnet *ifp)
 
 # if (defined(__FreeBSD__) && defined(DEVICE_POLLING))
   ifp->if_capabilities |= IFCAP_POLLING;
+  ifp->if_capenable    |= IFCAP_POLLING_NOCOUNT;
 # if (__FreeBSD_version < 500000)
   ifp->if_capenable    |= IFCAP_POLLING;
 # endif