Date: Mon, 29 Jun 2009 17:53:05 +0300 From: Alexander Motin <mav@FreeBSD.org> To: Kamigishi Rei <spambox@haruhiism.net> Cc: freebsd-current@freebsd.org Subject: Re: RFC: ATA to CAM integration patch Message-ID: <4A48D551.5090509@FreeBSD.org> In-Reply-To: <4A48D136.3050309@haruhiism.net> References: <1246206181.00132972.1246192801@10.7.7.3> <1246206186.00132982.1246194002@10.7.7.3> <1246209783.00133001.1246197001@10.7.7.3> <1246260183.00133237.1246247402@10.7.7.3> <4A48798A.5070604@FreeBSD.org> <4A487BF7.8060103@haruhiism.net> <4A48922D.5090507@FreeBSD.org> <4A48C0BC.3030104@haruhiism.net> <4A48CD19.8080606@FreeBSD.org> <4A48D136.3050309@haruhiism.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Kamigishi Rei wrote: > Alexander Motin wrote: >>> ahcich1: ALL SLOTS BUSY! >>> Is it purely informational or does it indicate a problem? (I'm using >>> Intel Q35+ICH9 in AHCI mode.) >> It is bad. It must never happen. It means that driver got more >> commands then it has empty slots available. Results can be unpredictable. >> How have you manage it? Were there any other messages around? > No other messages. Going to try your patch from the followup message. > Basically, I did nothing out of the ordinary. I'm running mysqld, > postgres, and two lighttpds in jails, one of the lighttpds is serving > huge media files. > The messages appeared mostly during "make buildworld" for one of the > GEOM_MIRROR drives (/usr/obj operations probably) and for one of raidz1 > drives (where /usr/src resides). It could be not a real bug, but incorrect diagnostic, in situation, when the last of 32 submitted command finishes first. NCQ in practice. :) -- Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4A48D551.5090509>