Author: sephe
Date: Mon Jun 13 06:24:41 2016
New Revision: 301858
URL: https://svnweb.freebsd.org/changeset/base/301858
Log:
MFC 295304,295305,295306
295304
hyperv/hn: Increase LRO entry count to 128 by default
hn(4) only has one RX ring currently, so default 8 LRO entries
are too small.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5166
295305
hyperv/hn: Move LRO flush to the channel processing rollup
This significantly increases LRO aggregation ratio when there are
large amount of connections (improves reception performance a lot).
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5167
295306
hyperv/hn: Add an option to always do transmission scheduling
It is off by default. This eases more experiment on hn(4).
Reviewed by: adrian, Hongjiang Zhang <honzhan microsoft com>
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5175
Modified:
stable/10/sys/dev/hyperv/netvsc/hv_net_vsc.h
stable/10/sys/dev/hyperv/netvsc/hv_netvsc_drv_freebsd.c
Directory Properties:
stable/10/ (props changed)
Modified: stable/10/sys/dev/hyperv/netvsc/hv_net_vsc.h
==============================================================================
--- stable/10/sys/dev/hyperv/netvsc/hv_net_vsc.h Mon Jun 13 06:09:02
2016 (r301857)
+++ stable/10/sys/dev/hyperv/netvsc/hv_net_vsc.h Mon Jun 13 06:24:41
2016 (r301858)
@@ -1024,6 +1024,7 @@ typedef struct hn_softc {
int hn_txdesc_avail;
int hn_txeof;
+ int hn_sched_tx;
int hn_direct_tx_size;
struct taskqueue *hn_tx_taskq;
struct task hn_start_task;
Modified: stable/10/sys/dev/hyperv/netvsc/hv_netvsc_drv_freebsd.c
==============================================================================
--- stable/10/sys/dev/hyperv/netvsc/hv_netvsc_drv_freebsd.c Mon Jun 13
06:09:02 2016 (r301857)
+++ stable/10/sys/dev/hyperv/netvsc/hv_netvsc_drv_freebsd.c Mon Jun 13
06:24:41 2016 (r301858)
@@ -132,6 +132,8 @@ __FBSDID("$FreeBSD$");
/* YYY should get it from the underlying channel */
#define HN_TX_DESC_CNT 512
+#define HN_LROENT_CNT_DEF 128
+
#define HN_RNDIS_MSG_LEN \
(sizeof(rndis_msg) + \
RNDIS_VLAN_PPI_SIZE + \
@@ -232,6 +234,13 @@ TUNABLE_INT("dev.hn.tx_chimney_size", &h
static int hn_direct_tx_size = HN_DIRECT_TX_SIZE_DEF;
TUNABLE_INT("dev.hn.direct_tx_size", &hn_direct_tx_size);
+#if defined(INET) || defined(INET6)
+#if __FreeBSD_version >= 1100095
+static int hn_lro_entry_count = HN_LROENT_CNT_DEF;
+TUNABLE_INT("dev.hn.lro_entry_count", &hn_lro_entry_count);
+#endif
+#endif
+
/*
* Forward declarations
*/
@@ -335,6 +344,11 @@ netvsc_attach(device_t dev)
#if __FreeBSD_version >= 1100045
int tso_maxlen;
#endif
+#if defined(INET) || defined(INET6)
+#if __FreeBSD_version >= 1100095
+ int lroent_cnt;
+#endif
+#endif
sc = device_get_softc(dev);
if (sc == NULL) {
@@ -417,9 +431,17 @@ netvsc_attach(device_t dev)
}
#if defined(INET) || defined(INET6)
+#if __FreeBSD_version >= 1100095
+ lroent_cnt = hn_lro_entry_count;
+ if (lroent_cnt < TCP_LRO_ENTRIES)
+ lroent_cnt = TCP_LRO_ENTRIES;
+ tcp_lro_init_args(&sc->hn_lro, ifp, lroent_cnt, 0);
+ device_printf(dev, "LRO: entry count %d\n", lroent_cnt);
+#else
tcp_lro_init(&sc->hn_lro);
/* Driver private LRO settings */
sc->hn_lro.ifp = ifp;
+#endif
#ifdef HN_LRO_HIWAT
sc->hn_lro.lro_hiwat = sc->hn_lro_hiwat;
#endif
@@ -512,6 +534,10 @@ netvsc_attach(device_t dev)
SYSCTL_ADD_INT(ctx, child, OID_AUTO, "direct_tx_size",
CTLFLAG_RW, &sc->hn_direct_tx_size, 0,
"Size of the packet for direct transmission");
+ SYSCTL_ADD_INT(ctx, child, OID_AUTO, "sched_tx",
+ CTLFLAG_RW, &sc->hn_sched_tx, 0,
+ "Always schedule transmission "
+ "instead of doing direct transmission");
if (unit == 0) {
struct sysctl_ctx_list *dc_ctx;
@@ -547,6 +573,12 @@ netvsc_attach(device_t dev)
SYSCTL_ADD_INT(dc_ctx, dc_child, OID_AUTO, "direct_tx_size",
CTLFLAG_RD, &hn_direct_tx_size, 0,
"Size of the packet for direct transmission");
+#if defined(INET) || defined(INET6)
+#if __FreeBSD_version >= 1100095
+ SYSCTL_ADD_INT(dc_ctx, dc_child, OID_AUTO, "lro_entry_count",
+ CTLFLAG_RD, &hn_lro_entry_count, 0, "LRO entry count");
+#endif
+#endif
}
return (0);
@@ -736,6 +768,15 @@ void
netvsc_channel_rollup(struct hv_device *device_ctx)
{
struct hn_softc *sc = device_get_softc(device_ctx->device);
+#if defined(INET) || defined(INET6)
+ struct lro_ctrl *lro = &sc->hn_lro;
+ struct lro_entry *queued;
+
+ while ((queued = SLIST_FIRST(&lro->lro_active)) != NULL) {
+ SLIST_REMOVE_HEAD(&lro->lro_active, next);
+ tcp_lro_flush(lro, queued);
+ }
+#endif
if (!sc->hn_txeof)
return;
@@ -1312,18 +1353,8 @@ skip:
}
void
-netvsc_recv_rollup(struct hv_device *device_ctx)
+netvsc_recv_rollup(struct hv_device *device_ctx __unused)
{
-#if defined(INET) || defined(INET6)
- hn_softc_t *sc = device_get_softc(device_ctx->device);
- struct lro_ctrl *lro = &sc->hn_lro;
- struct lro_entry *queued;
-
- while ((queued = SLIST_FIRST(&lro->lro_active)) != NULL) {
- SLIST_REMOVE_HEAD(&lro->lro_active, next);
- tcp_lro_flush(lro, queued);
- }
-#endif
}
/*
@@ -1577,9 +1608,11 @@ hn_stop(hn_softc_t *sc)
static void
hn_start(struct ifnet *ifp)
{
- hn_softc_t *sc;
+ struct hn_softc *sc = ifp->if_softc;
+
+ if (sc->hn_sched_tx)
+ goto do_sched;
- sc = ifp->if_softc;
if (NV_TRYLOCK(sc)) {
int sched;
@@ -1588,15 +1621,18 @@ hn_start(struct ifnet *ifp)
if (!sched)
return;
}
+do_sched:
taskqueue_enqueue_fast(sc->hn_tx_taskq, &sc->hn_start_task);
}
static void
hn_start_txeof(struct ifnet *ifp)
{
- hn_softc_t *sc;
+ struct hn_softc *sc = ifp->if_softc;
+
+ if (sc->hn_sched_tx)
+ goto do_sched;
- sc = ifp->if_softc;
if (NV_TRYLOCK(sc)) {
int sched;
@@ -1608,6 +1644,7 @@ hn_start_txeof(struct ifnet *ifp)
&sc->hn_start_task);
}
} else {
+do_sched:
/*
* Release the OACTIVE earlier, with the hope, that
* others could catch up. The task will clear the
_______________________________________________
[email protected] mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-stable-10
To unsubscribe, send any mail to "[email protected]"
