[3346] in linux-scsi channel archive
Re: SCSI host numbering (patch)
daemon@ATHENA.MIT.EDU (Leonard N. Zubkoff)
Mon Feb 16 12:31:13 1998
Date: Mon, 16 Feb 1998 09:11:48 -0800
From: "Leonard N. Zubkoff" <lnz@dandelion.com>
To: eric@andante.jic.com
CC: Dario_Ballabio@milano.europe.dg.com, linux-scsi@vger.rutgers.edu,
ankry@green.mif.pg.gda.pl
In-reply-to: <Pine.LNX.3.95.980214181414.3033C-100000@andante.jic.com>
(message from Eric Youngdale on Sat, 14 Feb 1998 19:13:13 -0500 (EST))
Date: Sat, 14 Feb 1998 19:13:13 -0500 (EST)
From: Eric Youngdale <eric@andante.jic.com>
Is the goal to control probe order, or is the goal to control
device naming? In other words, if the problem isn't strictly probe
order, then it isn't stricly a requirement that the actual probe order be
controlled - itn't it simply the host *numbering* that we want to control.
I have to wonder at this point how much work we should put into making the host
adapter probe order configurable from the command line. My impression, perhaps
incorrect, is that unlike me many people prefer to use modules for most
everything, and the loading order is explicit with modules so that this problem
does not occur. For those not using modules, it's trivial to select any order
you choose with a trivial edit to the source.
I think we should separate out the question of preserving the SCSI host
numbering on module unload/reload from the question of probe order.
Leonard
