diff options
| author | Reyk Floeter <reyk@cvs.openbsd.org> | 2016-01-25 10:39:21 +0000 |
|---|---|---|
| committer | Reyk Floeter <reyk@cvs.openbsd.org> | 2016-01-25 10:39:21 +0000 |
| commit | bc35c9c0a6dd58a74d4e2a1a2d86d1f62bbc49c2 (patch) | |
| tree | d64a16da0f340e2c6d8bd3fd66cc3ab16233b34d /sys | |
| parent | b16598fcec4662c5c8ade2339946556313ab4451 (diff) | |
In vmxnet3_start(), do not send the mbuf to bpf after passing it to
the hardware. This could have resulted in a page fault when the mbuf
has already been freed by the TX interrupt handler on another CPU.
This has the slight drawback that bpf can be called before the packet
is eventually dropped by vmxnet3_load_mbuf() - but I'm getting this
simple and verified fix in before doing further optimizations on the
start handler.
As discussed with mikeb@ jsg@ dlg@
Diffstat (limited to 'sys')
| -rw-r--r-- | sys/dev/pci/if_vmx.c | 12 |
1 files changed, 6 insertions, 6 deletions
diff --git a/sys/dev/pci/if_vmx.c b/sys/dev/pci/if_vmx.c index 2a3367c35c2..9fd77cc7a6b 100644 --- a/sys/dev/pci/if_vmx.c +++ b/sys/dev/pci/if_vmx.c @@ -1,4 +1,4 @@ -/* $OpenBSD: if_vmx.c,v 1.41 2016/01/04 16:16:56 mikeb Exp $ */ +/* $OpenBSD: if_vmx.c,v 1.42 2016/01/25 10:39:20 reyk Exp $ */ /* * Copyright (c) 2013 Tsubai Masanari @@ -1065,17 +1065,17 @@ vmxnet3_start(struct ifnet *ifp) if (m == NULL) break; +#if NBPFILTER > 0 + if (ifp->if_bpf) + bpf_mtap_ether(ifp->if_bpf, m, BPF_DIRECTION_OUT); +#endif + n = vmxnet3_load_mbuf(sc, ring, &m); if (n == -1) { ifp->if_oerrors++; continue; } -#if NBPFILTER > 0 - if (ifp->if_bpf) - bpf_mtap_ether(ifp->if_bpf, m, BPF_DIRECTION_OUT); -#endif - ifp->if_opackets++; used += n; } |