Re: kern/55115 (siisata(4) broken after enabling MSI support)

[Jaromír Doleček <jaromir.dolecek%gmail.com@localhost>]

The following reply was made to PR kern/55115; it has been noted by GNATS.

From: =?UTF-8?B?SmFyb23DrXIgRG9sZcSNZWs=?= <jaromir.dolecek%gmail.com@localhost>
To: "Jonathan A. Kollasch" <jakllsch%kollasch.net@localhost>
Cc: "gnats-bugs%NetBSD.org@localhost" <gnats-bugs%netbsd.org@localhost>, Jaromir Dolecek <jdolecek%netbsd.org@localhost>, 
	"John D. Baker" <jdbaker%consolidated.net@localhost>
Subject: Re: kern/55115 (siisata(4) broken after enabling MSI support)
Date: Sat, 28 Mar 2020 17:13:30 +0100

 This particular card is not even multi vector.
 
 I have just the PCI Express one (Adaptec 1220A with SiI3132 chip), and
 it works fine with MSI.
 
 I think it's unlikely it's board problem (wm0 is using MSI and works),
 but perhaps worth checking if by change there would be any BIOS
 settings enabling MSI just for particular slots.
 
 Another option would be to simply disable MSI for Sii3124. It's an old
 chip used for old PCI/PCI-X cards, so not much of those left in wild.
 
 Jaromir
 
 Le sam. 28 mars 2020 =C3=A0 13:53, Jonathan A. Kollasch
 <jakllsch%kollasch.net@localhost> a =C3=A9crit :
 >
 > Some if not all siisata(4) parts support multi-vector MSI.  But I don't
 > think I enabled it, and I should think the chip/channel resets we do
 > would put it in a single interrupt mode.