Author: davidcs
Date: Thu Apr 2 21:55:03 2015
New Revision: 281006
URL: https://svnweb.freebsd.org/changeset/base/281006
Log:
When an mbuf allocation fails in the receive path, the mbuf containing the
received packet is not sent to the host net
work stack and is reused again on the receive ring. Remaining received
packets in the ring are not processed in that
invocation of bxe_rxeof() and defered to the task thread.
MFC after: 5 days
Modified:
head/sys/dev/bxe/bxe.c
Modified: head/sys/dev/bxe/bxe.c
==============================================================================
--- head/sys/dev/bxe/bxe.c Thu Apr 2 21:35:36 2015 (r281005)
+++ head/sys/dev/bxe/bxe.c Thu Apr 2 21:55:03 2015 (r281006)
@@ -3246,7 +3246,7 @@ bxe_rxeof(struct bxe_softc *sc,
uint16_t bd_cons, bd_prod, bd_prod_fw, comp_ring_cons;
uint16_t hw_cq_cons, sw_cq_cons, sw_cq_prod;
int rx_pkts = 0;
- int rc;
+ int rc = 0;
BXE_FP_RX_LOCK(fp);
@@ -3388,6 +3388,10 @@ bxe_rxeof(struct bxe_softc *sc,
(sc->max_rx_bufs != RX_BD_USABLE) ?
bd_prod : bd_cons);
if (rc != 0) {
+
+ /* we simply reuse the received mbuf and don't post it to the
stack */
+ m = NULL;
+
BLOGE(sc, "mbuf alloc fail for fp[%02d] rx chain (%d)\n",
fp->index, rc);
fp->eth_q_stats.rx_soft_errors++;
@@ -3476,6 +3480,9 @@ next_cqe:
sw_cq_cons = RCQ_NEXT(sw_cq_cons);
/* limit spinning on the queue */
+ if (rc != 0)
+ break;
+
if (rx_pkts == sc->rx_budget) {
fp->eth_q_stats.rx_budget_reached++;
break;
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "[email protected]"
