diff options
| author | riastradh <riastradh@NetBSD.org> | 2021-06-16 00:21:17 +0000 |
|---|---|---|
| committer | riastradh <riastradh@NetBSD.org> | 2021-06-16 00:21:17 +0000 |
| commit | bbdda93be2f0bfefe91466b398fe42700a6ea284 (patch) | |
| tree | e75e77046575209dbb21f73d0bc44c61ec0ec6c0 /sys/dev/scsipi | |
| parent | 5a96957fa793651c8e7093a5a4c2fcae87232af2 (diff) | |
if_attach and if_initialize cannot fail, don't test return value
These were originally made failable back in 2017 when if_initialize
allocated a softint in every interface for link state changes, so
that it could fail gracefully instead of panicking:
https://mail-index.NetBSD.org/source-changes/2017/10/23/msg089053.html
However, this spawned many seldom- or never-tested error branches,
which are risky to have around. And that softint in every interface
has since been replaced by a single global workqueue, because link
state changes require thread context but not low latency or high
throughput:
https://mail-index.NetBSD.org/source-changes/2020/02/06/msg113759.html
So there is no longer any reason for if_initialize to fail. (The
subroutine if_stats_init can't fail because percpu_alloc can't fail
either.)
There is a snag: the softint_establish in if_percpuq_create could
fail, potentially leading to bad consequences later on trying to use
the softint. This change doesn't introduce any new bugs because of
the snag -- if_percpuq_attach was already broken. However, the snag
can be better addressed without spawning error branches, either by
using a single softint or making softints less scarce.
(Separate commit will change the signatures of if_attach and
if_initialize to return void, scheduled to ride whatever is the next
convenient kernel bump.)
Patch and testing on amd64 and evbmips64-eb by maya@; commit message
soliloquy, and compile-testing on evbppc/i386/earmv7hf, by me.
Diffstat (limited to 'sys/dev/scsipi')
| -rw-r--r-- | sys/dev/scsipi/if_se.c | 10 |
1 files changed, 3 insertions, 7 deletions
diff --git a/sys/dev/scsipi/if_se.c b/sys/dev/scsipi/if_se.c index ac6940a7cfd..beb21a83fad 100644 --- a/sys/dev/scsipi/if_se.c +++ b/sys/dev/scsipi/if_se.c @@ -1,4 +1,4 @@ -/* $NetBSD: if_se.c,v 1.112 2020/09/29 02:58:52 msaitoh Exp $ */ +/* $NetBSD: if_se.c,v 1.113 2021/06/16 00:21:19 riastradh Exp $ */ /* * Copyright (c) 1997 Ian W. Dall <ian.dall@dsto.defence.gov.au> @@ -59,7 +59,7 @@ */ #include <sys/cdefs.h> -__KERNEL_RCSID(0, "$NetBSD: if_se.c,v 1.112 2020/09/29 02:58:52 msaitoh Exp $"); +__KERNEL_RCSID(0, "$NetBSD: if_se.c,v 1.113 2021/06/16 00:21:19 riastradh Exp $"); #ifdef _KERNEL_OPT #include "opt_inet.h" @@ -368,11 +368,7 @@ seattach(device_t parent, device_t self, void *aux) sc->sc_attach_state = 1; /* Attach the interface. */ - rv = if_initialize(ifp); - if (rv != 0) { - sedetach(sc->sc_dev, 0); - return; /* Error */ - } + if_initialize(ifp); snprintf(wqname, sizeof(wqname), "%sRx", device_xname(sc->sc_dev)); rv = workqueue_create(&sc->sc_recv_wq, wqname, se_recv_worker, sc, |
