Hi,
In hvn_alloc_cmd() we spin within a mutex while the freelist is empty.
Because we're using a mutex there is no way to miss the wakeup(9) from
hvn_free_cmd(), so we don't even need a timeout here. Instead of
doing msleep(9) for 1 tick repeatedly we can msleep_nsec(9) without a
timeout (INFSLP) repeatedly and poll less in this spot.
Once again I can't test this. But the principle of protecting the
sleep with a mutex is sound.
ok?
Index: if_hvn.c
===================================================================
RCS file: /cvs/src/sys/dev/pv/if_hvn.c,v
retrieving revision 1.43
diff -u -p -r1.43 if_hvn.c
--- if_hvn.c 4 Dec 2020 03:22:46 -0000 1.43
+++ if_hvn.c 4 Dec 2020 03:35:25 -0000
@@ -1127,8 +1127,8 @@ hvn_alloc_cmd(struct hvn_softc *sc)
mtx_enter(&sc->sc_cntl_fqlck);
while ((rc = TAILQ_FIRST(&sc->sc_cntl_fq)) == NULL)
- msleep(&sc->sc_cntl_fq, &sc->sc_cntl_fqlck,
- PRIBIO, "nvsalloc", 1);
+ msleep_nsec(&sc->sc_cntl_fq, &sc->sc_cntl_fqlck,
+ PRIBIO, "nvsalloc", INFSLP);
TAILQ_REMOVE(&sc->sc_cntl_fq, rc, rc_entry);
mtx_leave(&sc->sc_cntl_fqlck);
return (rc);
