Date: Mon, 21 Apr 2014 16:15:34 -0700 From: John-Mark Gurney <jmg@funkthat.com> To: Chris Torek <torek@torek.net> Cc: freebsd-hackers@freebsd.org Subject: Re: MAXPHYS in md(4) Message-ID: <20140421231534.GY43976@funkthat.com> In-Reply-To: <201404212230.s3LMUWif045734@elf.torek.net> References: <20140421222210.GW43976@funkthat.com> <201404212230.s3LMUWif045734@elf.torek.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Chris Torek wrote this message on Mon, Apr 21, 2014 at 16:30 -0600:
> >This really should be moved up into the generic GEOM layer so geom
> >module ever sees IO that is larger than MAXPHYS...
>
> Well, maybe, but if you're going to do something like that it
> might be nice to have the device equivalent of MTU discovery....
Why? GEOM modules are written on the assumption that no IO larger
than MAXPHYS will ever be seen... They allocate arrays of structures
or other items based upon MAXPHYS, and will smash the stack/crash//do
bad things if it receives an IO larger than MAXPHYS...
The change would not break anything that isn't already broken (or
working by luck)...
Now if you're talking about wanting to increase MAXPHYS, there are many
threads talking about what needs to be done about it, but that is
completely different than this issue...
--
John-Mark Gurney Voice: +1 415 225 5579
"All that I will do, has been done, All that I have, has not."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20140421231534.GY43976>