From mboxrd@z Thu Jan 1 00:00:00 1970 From: trooper ryan Subject: Re: PCI: Cannot allocate resource region 0 of device 0000:00:0d.0 Date: Sun, 25 Apr 2004 19:13:56 +0000 Sender: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Message-ID: <408C0DF4.4070502@dmtsystems.net> References: <408A85DE.8090700@dmtsystems.net> <20040424125753.GJ22558@parcelfarce.linux.theplanet.co.uk> <408AF79C.1050804@dmtsystems.net> <20040424150533.GL22558@parcelfarce.linux.theplanet.co.uk> <408BE531.8080503@dmtsystems.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <408BE531.8080503-QPf0Bz723pkjHdulDSg1yw@public.gmane.org> Errors-To: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , List-Archive: To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Cc: Matthew Wilcox List-Id: linux-acpi@vger.kernel.org This turned out to be a hardware problem - I gave the riser contacts a good clean and now the device is recognised correctly. Many thanks for your help Matthew! trooper ryan wrote: > The card does work on an XP box I have, so I suspect it might be the PCI > riser card I'm using to slot it in to a 1RU box. I'll try and locate > another PCI riser to use and let you know how I go. > > Thanks for all your help! > > > > Matthew Wilcox wrote: > >> On Sat, Apr 24, 2004 at 11:26:20PM +0000, trooper ryan wrote: >> >>> Anyhow, the info you requested: >>> # setpci -s 00:0d.0 BASE_ADDRESS_0=ffffffff >>> setpci -s 00:0d.0 BASE_ADDRESS_0 returns 00d7f008 sometimes and >>> ffffffff others. >> >> >> >> This is quite suspicious. ffffffff would indicate this device is not >> connected to the bus [1] . The other value returned indicates it's a >> 32-bit BAR (entirely plausible) which is prefetchable (seems kosher for >> a video capture card). The base address would then be 0x00d7f000 which >> is slightly under the 16MB boundary. Very curious as this is normally >> assigned to system RAM. >> >> All in all, this looks like a pretty broken piece of hardware to me. >> Possibly something else is going on to cause this, like the system being >> over its power-budget. I advide you to treat this as a hardware problem >> and try doing some things to narrow it down, like taking all other cards >> out of the machine and seeing if the problem persists; try the card in >> a different slot; check the card connectors for damage, etc. >> >> It'd also be worth trying a different OS with it, just to be sure it's >> a hardware problem ;-) >> >> [1] Due to how PCI BARs work, there must be at least one bit clear. >> If it's a memory BAR, bit 0 must be clear, and if it's a port BAR, bit 1 >> must be 0. If there's no reponse from a device, the PCI bus floats high, >> so you get back 0xffffffff. When you write 0xffffffff to a BAR, the >> device >> ignores the bits that are less than the size of the aperture so when you >> read it back, you should get only the implemented bits set. >> > > > ------------------------------------------------------- > This SF.net email is sponsored by: The Robotic Monkeys at ThinkGeek > For a limited time only, get FREE Ground shipping on all orders of $35 > or more. Hurry up and shop folks, this offer expires April 30th! > http://www.thinkgeek.com/freeshipping/?cpg=12297 > _______________________________________________ > Acpi-devel mailing list > Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org > https://lists.sourceforge.net/lists/listinfo/acpi-devel > > ------------------------------------------------------- This SF.net email is sponsored by: The Robotic Monkeys at ThinkGeek For a limited time only, get FREE Ground shipping on all orders of $35 or more. Hurry up and shop folks, this offer expires April 30th! http://www.thinkgeek.com/freeshipping/?cpg=12297