git: 31896115b557 - main - hn: Forward capability changes to transparent VFs
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Thu, 17 Sep 2026 00:13:34 UTC
The branch main has been updated by kbowling:
URL: https://cgit.FreeBSD.org/src/commit/?id=31896115b5574944718cc4f5260efe47b622b55b
commit 31896115b5574944718cc4f5260efe47b622b55b
Author: Kevin Bowling <kbowling@FreeBSD.org>
AuthorDate: 2026-09-16 17:13:40 +0000
Commit: Kevin Bowling <kbowling@FreeBSD.org>
CommitDate: 2026-09-17 00:12:07 +0000
hn: Forward capability changes to transparent VFs
The transparent VF capability handler ignored the requested change and
only copied the VF's enabled capabilities. Forward SIOCSIFCAP to the VF
and return its result. Preserve VF capabilities which hn does not expose.
Limit advertised capabilities to those supported by the transparent
packet path and VLAN relay. Do not inherit VF services such as send tags
or the extended capability ioctl when hn has no corresponding methods.
At handoff, adopt the VF's enabled offloads without reconfiguring it.
Mark the datapath as switching while the VF applies a capability change,
since its ioctl may reinitialize the device. If the association is still
ready and unchanged afterwards, synchronize hn with the actual VF state
even on error and restrict checksum assistance to the forwarded offloads.
Republish link state suppressed during the transition when the VF is
still ready. Refresh VLAN child capabilities after adoption and when
restoring the synthetic path.
Run the VF capability ioctl without hn_lock to avoid reversing the VLAN
configuration lock order. Hold an ifnet reference, serialize capability
changes and VF initialization, and make detach wait with hn_lock released.
Revalidate the VF association before publishing the result. Defer
VLAN-child capability notifications to the existing VF taskqueue so they
also run without hn_lock, and drain that work before ifnet teardown.
MFC after: 2 weeks
Sponsored by: BBOX.io
---
sys/dev/hyperv/netvsc/if_hn.c | 132 +++++++++++++++++++++++++++++++++------
sys/dev/hyperv/netvsc/if_hnvar.h | 2 +
2 files changed, 114 insertions(+), 20 deletions(-)
diff --git a/sys/dev/hyperv/netvsc/if_hn.c b/sys/dev/hyperv/netvsc/if_hn.c
index 73ad8b978c7e..d8ac23a4af6e 100644
--- a/sys/dev/hyperv/netvsc/if_hn.c
+++ b/sys/dev/hyperv/netvsc/if_hn.c
@@ -184,6 +184,12 @@ do { \
#define HN_CSUM_IP6_HWASSIST(sc) \
((sc)->hn_tx_ring[0].hn_csum_assist & HN_CSUM_IP6_MASK)
+/* Packet offloads can follow the VF; services requiring hn methods cannot. */
+#define HN_XPNT_VF_CAPS (IFCAP_HWCSUM | IFCAP_HWCSUM_IPV6 | \
+ IFCAP_TSO | IFCAP_LRO | IFCAP_VLAN_MTU | IFCAP_VLAN_HWTAGGING | \
+ IFCAP_VLAN_HWCSUM | IFCAP_VLAN_HWTSO | IFCAP_VLAN_HWFILTER | \
+ IFCAP_JUMBO_MTU | IFCAP_LINKSTATE | IFCAP_HWSTATS | IFCAP_MEXTPG)
+
#define HN_PKTSIZE_MIN(align) \
roundup2(ETHER_MIN_LEN + ETHER_VLAN_ENCAP_LEN - ETHER_CRC_LEN + \
HN_RNDIS_PKT_LEN, (align))
@@ -299,6 +305,8 @@ static void hn_xpnt_vf_input(if_t, struct mbuf *);
static int hn_xpnt_vf_iocsetflags(struct hn_softc *);
static int hn_xpnt_vf_iocsetcaps(struct hn_softc *,
struct ifreq *);
+static void hn_xpnt_vf_synccaps(struct hn_softc *);
+static void hn_xpnt_vf_vlancap_taskfunc(void *, int);
static void hn_xpnt_vf_saveifflags(struct hn_softc *);
static bool hn_xpnt_vf_isready(struct hn_softc *);
static bool hn_xpnt_vf_caninit(struct hn_softc *);
@@ -1295,22 +1303,93 @@ hn_ifaddr_event(void *arg, if_t ifp)
hn_rxvf_change(arg, ifp);
}
-static int
-hn_xpnt_vf_iocsetcaps(struct hn_softc *sc, struct ifreq *ifr __unused)
+static void
+hn_xpnt_vf_synccaps(struct hn_softc *sc)
{
if_t ifp, vf_ifp;
+ int caps;
HN_LOCK_ASSERT(sc);
ifp = sc->hn_ifp;
vf_ifp = sc->hn_vf_ifp;
+ caps = if_getcapabilities(ifp);
+
+ /* Reflect the actual VF state even if an ioctl failed partway through. */
+ if_setcapenable(ifp, if_getcapenable(vf_ifp) & caps);
+ if_sethwassist(ifp, if_gethwassist(vf_ifp) &
+ (HN_CSUM_IP_MASK | HN_CSUM_IP6_MASK | CSUM_TSO));
+ /* The worker refreshes VLAN children without holding hn_lock. */
+ if (!sc->hn_detaching)
+ taskqueue_enqueue(sc->hn_vf_taskq, &sc->hn_vf_vlancap);
+}
+
+static void
+hn_xpnt_vf_vlancap_taskfunc(void *xsc, int pending __unused)
+{
+ struct hn_softc *sc = xsc;
+
+ /* VLAN configuration takes vlan_sx before entering hn ioctls. */
+ sx_assert(&sc->hn_lock, SA_UNLOCKED);
+ if_vlancap(sc->hn_ifp);
+}
+
+static int
+hn_xpnt_vf_iocsetcaps(struct hn_softc *sc, struct ifreq *ifr)
+{
+ if_t vf_ifp;
+ u_int assoc;
+ int caps, error;
+
+ HN_LOCK_ASSERT(sc);
+ if (sc->hn_vf_caps_busy)
+ return (EBUSY);
+ if (sc->hn_detaching || sc->hn_vf_detaching ||
+ !hn_xpnt_vf_isready(sc))
+ return (ENXIO);
+
+ vf_ifp = sc->hn_vf_ifp;
+ if_ref(vf_ifp);
+ assoc = sc->hn_vf_active_assoc;
+ caps = if_getcapabilities(sc->hn_ifp);
+ /* Leave capabilities which hn does not expose unchanged on the VF. */
+ ifr->ifr_reqcap = (ifr->ifr_reqcap & caps) |
+ (if_getcapenable(vf_ifp) & ~caps);
+ sc->hn_vf_caps_busy = true;
+ rm_wlock(&sc->hn_vf_lock);
+ sc->hn_xvf_flags |= HN_XVFFLAG_SWITCHING;
+ rm_wunlock(&sc->hn_vf_lock);
/*
- * Just sync up with VF's enabled capabilities.
+ * The VF ioctl may acquire vlan_sx, whose callers enter hn ioctls.
+ * Detach waits for hn_vf_caps_busy with hn_lock released; new VF
+ * initialization and capability changes cannot overlap this ioctl.
*/
- if_setcapenable(ifp, if_getcapenable(vf_ifp));
- if_sethwassist(ifp, if_gethwassist(vf_ifp));
+ HN_UNLOCK(sc);
+ error = ifhwioctl(SIOCSIFCAP, vf_ifp, (caddr_t)ifr, curthread);
+ HN_LOCK(sc);
- return (0);
+ if (!sc->hn_detaching && !sc->hn_vf_detaching &&
+ sc->hn_vf_ifp == vf_ifp && hn_xpnt_vf_isready(sc) &&
+ sc->hn_vf_active_assoc == assoc) {
+ hn_xpnt_vf_synccaps(sc);
+ rm_wlock(&sc->hn_vf_lock);
+ sc->hn_xvf_flags &= ~HN_XVFFLAG_SWITCHING;
+ /* A link event during the ioctl was suppressed by SWITCHING. */
+ if ((sc->hn_xvf_flags & HN_XVFFLAG_ENABLED) &&
+ hn_xpnt_vf_isready(sc))
+ if_link_state_change(sc->hn_ifp,
+ if_getlinkstate(vf_ifp));
+ rm_wunlock(&sc->hn_vf_lock);
+ } else if (error == 0) {
+ error = sc->hn_detaching || sc->hn_vf_detaching ? ENXIO : EAGAIN;
+ }
+ if_rele(vf_ifp);
+ sc->hn_vf_caps_busy = false;
+ wakeup(&sc->hn_vf_caps_busy);
+ /* Retry association work deferred while the ioctl was in flight. */
+ if (!sc->hn_detaching && !sc->hn_vf_detaching)
+ taskqueue_enqueue_timeout(sc->hn_vf_taskq, &sc->hn_vf_init, 0);
+ return (error);
}
static int
@@ -1698,13 +1777,11 @@ hn_xpnt_vf_setready(struct hn_softc *sc)
sc->hn_saved_hwassist = if_gethwassist(ifp);
/*
- * Intersect supported/enabled capabilities.
- *
- * NOTE:
- * if_hwassist is not changed here.
+ * Expose only capabilities supported by the transparent packet path.
+ * VF services such as send tags and the extended capability ioctl
+ * require methods which hn does not implement.
*/
- if_setcapabilitiesbit(ifp, 0, if_getcapabilities(vf_ifp));
- if_setcapenablebit(ifp, 0, if_getcapabilities(ifp));
+ if_setcapabilities(ifp, if_getcapabilities(vf_ifp) & HN_XPNT_VF_CAPS);
/*
* Fix TSO settings.
@@ -1717,12 +1794,9 @@ hn_xpnt_vf_setready(struct hn_softc *sc)
if_sethwtsomaxsegsize(ifp, if_gethwtsomaxsegsize(vf_ifp));
/*
- * Change VF's enabled capabilities.
+ * Adopt the VF's enabled capabilities without changing its settings.
*/
- memset(&ifr, 0, sizeof(ifr));
- strlcpy(ifr.ifr_name, if_name(vf_ifp), sizeof(ifr.ifr_name));
- ifr.ifr_reqcap = if_getcapenable(ifp);
- hn_xpnt_vf_iocsetcaps(sc, &ifr);
+ hn_xpnt_vf_synccaps(sc);
if (if_getmtu(ifp) != ETHERMTU) {
int error;
@@ -1773,6 +1847,7 @@ hn_xpnt_vf_caninit(struct hn_softc *sc)
HN_LOCK_ASSERT(sc);
return (hn_xpnt_vf && sc->hn_vf_ifp != NULL &&
!sc->hn_detaching && !sc->hn_vf_detaching &&
+ !sc->hn_vf_caps_busy &&
(atomic_load_acq_int(&sc->hn_vf_assoc) & HN_VF_ASSOC_ALLOCATED) &&
(int)(ticks - sc->hn_vf_rdytick) >= 0);
}
@@ -1791,6 +1866,8 @@ hn_xpnt_vf_restore(struct hn_softc *sc)
if_sethwtsomaxsegsize(ifp, sc->hn_saved_tsosegsz);
if_setcapenable(ifp, sc->hn_saved_capenable);
if_sethwassist(ifp, sc->hn_saved_hwassist);
+ if (!sc->hn_detaching)
+ taskqueue_enqueue(sc->hn_vf_taskq, &sc->hn_vf_vlancap);
sc->hn_vf_ready = false;
}
@@ -1849,8 +1926,10 @@ hn_xpnt_vf_setdisable(struct hn_softc *sc, bool clear_vf)
/* NOTE: hn_vf_lock for hn_transmit()/hn_qflush() */
rm_wlock(&sc->hn_vf_lock);
sc->hn_xvf_flags &= ~HN_XVFFLAG_ENABLED;
- if (clear_vf)
+ if (clear_vf) {
+ sc->hn_xvf_flags &= ~HN_XVFFLAG_SWITCHING;
sc->hn_vf_ifp = NULL;
+ }
rm_wunlock(&sc->hn_vf_lock);
for (i = 0; i < sc->hn_rx_ring_cnt; ++i)
@@ -1952,7 +2031,7 @@ hn_xpnt_vf_init(struct hn_softc *sc)
* Some VF drivers initialize hwassist only when brought up. Refresh
* the offload state before allowing transmit through the VF.
*/
- hn_xpnt_vf_iocsetcaps(sc, NULL);
+ hn_xpnt_vf_synccaps(sc);
/*
* NOTE:
@@ -1996,7 +2075,8 @@ hn_xpnt_vf_init_taskfunc(void *xsc, int pending __unused)
if ((sc->hn_flags & HN_FLAG_SYNTH_ATTACHED) == 0)
goto done;
- if (sc->hn_vf_ifp == NULL || sc->hn_vf_detaching)
+ if (sc->hn_vf_ifp == NULL || sc->hn_vf_detaching ||
+ sc->hn_vf_caps_busy)
goto done;
if (!hn_xpnt_vf) {
if ((sc->hn_flags & HN_FLAG_RXVF) && sc->hn_vf_active_assoc !=
@@ -2133,6 +2213,9 @@ hn_ifnet_detevent(void *xsc, if_t ifp)
if (hn_xpnt_vf) {
sc->hn_vf_detaching = true;
+ while (sc->hn_vf_caps_busy)
+ sx_sleep(&sc->hn_vf_caps_busy, &sc->hn_lock, 0,
+ "hnvfcap", 0);
/*
* Make sure that the delayed initialization is not running.
*
@@ -2316,6 +2399,7 @@ hn_attach(device_t dev)
device_get_nameunit(dev));
TIMEOUT_TASK_INIT(sc->hn_vf_taskq, &sc->hn_vf_init, 0,
hn_xpnt_vf_init_taskfunc, sc);
+ TASK_INIT(&sc->hn_vf_vlancap, 0, hn_xpnt_vf_vlancap_taskfunc, sc);
/*
* Allocate ifnet and setup its name earlier, so that if_printf
@@ -2674,12 +2758,15 @@ hn_detach(device_t dev)
HN_LOCK(sc);
sc->hn_detaching = true;
+ while (sc->hn_vf_caps_busy)
+ sx_sleep(&sc->hn_vf_caps_busy, &sc->hn_lock, 0, "hnvfcap", 0);
HN_UNLOCK(sc);
if (sc->hn_vlan_atthand != NULL)
EVENTHANDLER_DEREGISTER(vlan_config, sc->hn_vlan_atthand);
if (sc->hn_vlan_dethand != NULL)
EVENTHANDLER_DEREGISTER(vlan_unconfig, sc->hn_vlan_dethand);
taskqueue_drain_timeout(sc->hn_vf_taskq, &sc->hn_vf_init);
+ taskqueue_drain(sc->hn_vf_taskq, &sc->hn_vf_vlancap);
if (sc->hn_ifaddr_evthand != NULL)
EVENTHANDLER_DEREGISTER(ifaddr_event, sc->hn_ifaddr_evthand);
@@ -4076,6 +4163,11 @@ hn_ioctl(if_t ifp, u_long cmd, caddr_t data)
case SIOCSIFCAP:
HN_LOCK(sc);
+ if (sc->hn_detaching || sc->hn_vf_caps_busy) {
+ error = sc->hn_vf_caps_busy ? EBUSY : ENXIO;
+ HN_UNLOCK(sc);
+ break;
+ }
if (hn_xpnt_vf_isready(sc)) {
ifr_vf = *ifr;
diff --git a/sys/dev/hyperv/netvsc/if_hnvar.h b/sys/dev/hyperv/netvsc/if_hnvar.h
index 4be884d5522c..7fde5c8fe30e 100644
--- a/sys/dev/hyperv/netvsc/if_hnvar.h
+++ b/sys/dev/hyperv/netvsc/if_hnvar.h
@@ -268,6 +268,7 @@ struct hn_softc {
/* Applied VLANs and VF lifetime use hn_lock. */
u_int hn_vf_vlans[HN_VLAN_WORDS];
bool hn_vf_detaching;
+ bool hn_vf_caps_busy; /* unlocked capability ioctl */
/*
* Transparent VF delayed initialization.
@@ -279,6 +280,7 @@ struct hn_softc {
u_int hn_vf_active_assoc; /* confirmed VF generation */
struct taskqueue *hn_vf_taskq;
struct timeout_task hn_vf_init;
+ struct task hn_vf_vlancap;
/*
* Saved information for VF under transparent mode.