mvneta: Fix MTU update sequence

After MTU is updated we might start using allocating RX buffers from different pool. (MJUM9BYTES vs MCLBYTES)
Because of that we need to update the RX buffer size in hardware.
Previously it was done only when the interface was up, which is incorrect since MTU can be changed at any time.

Differential revision:	https://reviews.freebsd.org/D31724
Sponsored by:		Stormshield
Obtained from:		Semihalf
MFC after:		2 weeks
Reviewed by:		wma
This commit is contained in:
Kornel Duleba 2021-08-31 08:22:29 +02:00 committed by Wojciech Macek
parent bb536de6c0
commit 5438ef47e3

View file

@ -2168,29 +2168,28 @@ mvneta_ioctl(struct ifnet *ifp, u_long cmd, caddr_t data)
ifp->if_hwassist = CSUM_IP | CSUM_TCP |
CSUM_UDP;
}
if (ifp->if_drv_flags & IFF_DRV_RUNNING) {
/* Stop hardware */
/*
* Reinitialize RX queues.
* We need to update RX descriptor size.
*/
if (ifp->if_drv_flags & IFF_DRV_RUNNING)
mvneta_stop_locked(sc);
/*
* Reinitialize RX queues.
* We need to update RX descriptor size.
*/
for (q = 0; q < MVNETA_RX_QNUM_MAX; q++) {
mvneta_rx_lockq(sc, q);
if (mvneta_rx_queue_init(ifp, q) != 0) {
device_printf(sc->dev,
"initialization failed:"
" cannot initialize queue\n");
mvneta_rx_unlockq(sc, q);
error = ENOBUFS;
break;
}
for (q = 0; q < MVNETA_RX_QNUM_MAX; q++) {
mvneta_rx_lockq(sc, q);
if (mvneta_rx_queue_init(ifp, q) != 0) {
device_printf(sc->dev,
"initialization failed:"
" cannot initialize queue\n");
mvneta_rx_unlockq(sc, q);
error = ENOBUFS;
break;
}
/* Trigger reinitialization */
mvneta_init_locked(sc);
mvneta_rx_unlockq(sc, q);
}
if (ifp->if_drv_flags & IFF_DRV_RUNNING)
mvneta_init_locked(sc);
mvneta_sc_unlock(sc);
}
break;