On Wed, Mar 23, 2011 at 10:51:34AM +1300, [email protected] wrote:
>
> # cvs -d [email protected]:/cvs diff -u bge.4
> Index: bge.4
> ===================================================================
> RCS file: /cvs/src/share/man/man4/bge.4,v
> retrieving revision 1.50
> diff -u -r1.50 bge.4
> --- bge.4 9 Jul 2010 07:10:55 -0000 1.50
> +++ bge.4 22 Mar 2011 18:57:36 -0000
> @@ -193,10 +193,7 @@
> A fatal initialization error has occurred.
> .It "bge%d: couldn't map interrupt"
> A fatal initialization error has occurred.
> -.It "bge%d: no memory for jumbo buffer queue!"
> -The driver failed to allocate memory for jumbo frames during
> -initialization.
> -.It "bge%d: watchdog timeout"
> +.It "bge%d: watchdog timeout -- resetting"
> The device has stopped responding to the network, or there is a problem with
> the network connection (cable).
> .El
>
stuart has already removed the jumbo buffer diagnostic. that leaves the
watchdog timeout...
digging further in src/sys:
/usr/src/sys/dev/ic/ti.c: printf("%s: watchdog timeout -- resetting\n",
sc ->sc_dv.dv_xname);
/usr/src/sys/dev/pci/if_ix.c: printf("%s: Watchdog timeout -- resetting\n",
if p->if_xname);
/usr/src/sys/dev/pci/if_bge.c: printf("%s: watchdog timeout -- resetting\n",
sc ->bge_dev.dv_xname);
/usr/src/sys/dev/pci/if_em.c: printf("%s: watchdog timeout -- resetting\n",
sc ->sc_dv.dv_xname);
/usr/src/sys/dev/pci/if_ixgb.c: printf("%s: watchdog timeout -- resetting\n",
sc ->sc_dv.dv_xname);
of these devices, only ti(4) and bge(4) have DIAGNOSTICS sections. both
omit "-- resetting".
so, would any developer like to comment on whether there is a reason for
it being documented this way, or whether we need to change the docs or
not?
jmc
