git: a42a0b77f0de - stable/11 - Fix array out of bound panic introduced in r306219.
Ryan Moeller
freqlabs at FreeBSD.org
Tue Mar 30 15:04:17 UTC 2021
The branch stable/11 has been updated by freqlabs:
URL: https://cgit.FreeBSD.org/src/commit/?id=a42a0b77f0de636a91f79fa2fde8a507d88b79b7
commit a42a0b77f0de636a91f79fa2fde8a507d88b79b7
Author: Alexander Motin <mav at FreeBSD.org>
AuthorDate: 2019-05-28 18:32:04 +0000
Commit: Ryan Moeller <freqlabs at FreeBSD.org>
CommitDate: 2021-03-30 15:02:24 +0000
Fix array out of bound panic introduced in r306219.
As I see, different NICs in different configurations may have different
numbers of TX and RX queues. The code was assuming 1:1 mapping between
event queues (interrupts) and TX/RX queues. Since number of interrupts
is set to maximum of TX and RX queues, when those two are different, the
system is doomed.
I have no documentation or deep knowledge about this hardware, so this
change is based on general observations and code reading. If some of my
guesses are wrong, please do better. I just confirmed HP NC550SFP NICs
are working now.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
(cherry picked from commit 3582828053556ca0e05ed9aab3e78008a0595e09)
---
sys/dev/oce/oce_if.c | 18 ++++++++++++++----
1 file changed, 14 insertions(+), 4 deletions(-)
diff --git a/sys/dev/oce/oce_if.c b/sys/dev/oce/oce_if.c
index 1de38d4528fc..362bc1776890 100644
--- a/sys/dev/oce/oce_if.c
+++ b/sys/dev/oce/oce_if.c
@@ -2395,10 +2395,20 @@ oce_eqd_set_periodic(POCE_SOFTC sc)
goto modify_eqd;
}
- rq = sc->rq[i];
- rxpkts = rq->rx_stats.rx_pkts;
- wq = sc->wq[i];
- tx_reqs = wq->tx_stats.tx_reqs;
+ if (i == 0) {
+ rq = sc->rq[0];
+ rxpkts = rq->rx_stats.rx_pkts;
+ } else
+ rxpkts = 0;
+ if (i + 1 < sc->nrqs) {
+ rq = sc->rq[i + 1];
+ rxpkts += rq->rx_stats.rx_pkts;
+ }
+ if (i < sc->nwqs) {
+ wq = sc->wq[i];
+ tx_reqs = wq->tx_stats.tx_reqs;
+ } else
+ tx_reqs = 0;
now = ticks;
if (!aic->ticks || now < aic->ticks ||
More information about the dev-commits-src-branches
mailing list