linux-kernel - Re: [Regression] PCI resources allocation problem on HP nx6325

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <alpine.LFD.2.01.0908041555120.3270@localhost.localdomain>
Date:	Tue, 4 Aug 2009 16:04:16 -0700 (PDT)
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Manuel Lauss <mano@...rinelk.homelinux.net>
cc:	"Rafael J. Wysocki" <rjw@...k.pl>,
	Matthew Wilcox <willy@...ux.intel.com>,
	LKML <linux-kernel@...r.kernel.org>,
	Linux PCI <linux-pci@...r.kernel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Andrew Patterson <andrew.patterson@...com>
Subject: Re: [Regression] PCI resources allocation problem on HP nx6325

On Mon, 3 Aug 2009, Manuel Lauss wrote:
> 
> This patch introduces a warning on my system:
> 
> pci 0000:02:03.0: BAR 6: address space collision on of device [0xff680000-0xff69ffff]

Heh, that's funny. I suspect you have always had the error, it's just that 
back before the commit that started using "pci_claim_resource()", we did 
that "find_parent_resource()+request_resource()" by hand.

So before commit a76117dfd687ec4be0a9a05214f3009cc5f73a42, the ROM 
resources were done with:

	pr = pci_find_parent_resource(dev, r);
	if (!pr || request_resource(pr, r) < 0) {
		r->end -= r->start;
		r->start = 0;
	}

inside pcibios_assign_resources(), and it would never warn about it, it 
would just silently mark the resource unregistered (and then we'd 
re-allocate it later).

So I _think_ that you actually are getting the same layout as with 2.6.30, 
but with a warning that didn't exist in 2.6.30. Can you verify?

With the change to use "pci_claim_resource()", it initially changed things 
to use "insert_resource()", and that shouldn't have even succeeded, but 
apparently did! That's a bit scary. The patch to make it use 
pci_request_resource() should have made it have the same behavior as we 
had in 2.6.30 - albeit with the warning (that we didn't use to have).

I do wonder why the insert_resource() seems to have worked, though, so 
maybe I'm misdiagnosing this. Can you post your /proc/iomem from 2.6.30?

		Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/