Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 30 Sep 1998 13:29:33 -0500
From:      Karl Denninger <karl@Denninger.Net>
To:        "Justin T. Gibbs" <gibbs@plutotech.com>
Cc:        scsi@FreeBSD.ORG
Subject:   Re: Long IDE probes?
Message-ID:  <19980930132933.B4481@Denninger.Net>
In-Reply-To: <199809301819.MAA14837@pluto.plutotech.com>; from Justin T. Gibbs on Wed, Sep 30, 1998 at 12:13:04PM -0600
References:  <19980930125858.B4304@Denninger.Net> <199809301819.MAA14837@pluto.plutotech.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Sep 30, 1998 at 12:13:04PM -0600, Justin T. Gibbs wrote:
> >The default was recently increased (read: made longer) - like doubled.
> 
> The delay in the generic kernel has always been 15 seconds.  The default
> with no delay specified has always been 2 seconds.  CAM effectively reduces
> this delay over the old SCSI code as all bus delays during the initial
> prove phase occur in parallel.
> 
> >I *KNOW* that things like Exabytes "go away" for short periods of time 
> >after a bus reset.  But we're talking about *defaults* here, and modern
> >disks and tape drives, while they may kvetch internally for a while before
> >they'll generally be useful, but virtually all will respond to an inquiry
> >within 3-5 seconds - maximum.
> 
> Not everyone who uses FreeBSD has "modern devices".  Not everyone
> who uses FreeBSD knows that their devices will only work with a longish
> delay.  Are you saying that we should lose the ability to install on
> these machines simply because you, a user that knows how to modify this
> behavior, finds the default behavior annoying?
> 
> --
> Justin

Justin, quit being a pompous ass and lose the attitude.  Its uncalled for.

You and I both know damn well that to install you need ONLY, WORST CASE:
	1.	A floppy that works (to boot from)
	2.	A CDROM
	3.	A disk

Some folks need only a floppy, disk, and network card - they load off the 
network.

Kvetching about how someone's 1980's scanner or ancient tape drive won't 
come up under GENERIC on initial boot is both pointless and inappropriate,
given that you *know* these facts to be true, and that for every one of
those people there are a dozen others who can't boot without manual 
override of GENERIC's defaults due to ISA bus conflicts.

Don't give me crap about how "the defaults must accomodate everyone".  
They don't RIGHT NOW and FreeBSD has NEVER had that as a criteria for 
driver probe behavior.  I have loaded FreeBSD on a dozen machines over 
the last two years in which the default network card settings in GENERIC - 
over which I wanted to load the system - were in conflict with other found
devices and as such I had to MANUALLY go into the config screen and reset
them.

There are a half-dozen machines in MCSNet's computer room that have ISA 
cards in them that simply WILL NOT run GENERIC without manual intervention - 
not because the devices aren't there, but because the ONLY possible 
combination of IRQs and shared RAM addresses aren't the pre-ordained set 
of ones which are in GENERIC's idea of where devices have to be in order 
to show up.

If you want to make a point about ancient hardware, then explain why in the 
heck was CAM integrated without ALL the legacy SCSI devices being supported?

Don't get me wrong - I'm not bitching about the CAM integration - that was
the right choice.  It was just as much right as the current settings for bus
settling time and probe delays in the IDE code are wrong.


--
-- 
Karl Denninger (karl@denninger.net) http://www.mcs.net/~karl
I ain't even *authorized* to speak for anyone other than myself, so give
up now on trying to associate my words with any particular organization.

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19980930132933.B4481>