From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sergei Shtylyov Subject: Re: [PATCH v2] 3c59x: fix PCI resource management Date: Thu, 02 May 2013 21:40:53 +0400 Message-ID: <5182A525.8020601@cogentembedded.com> References: <201305020240.34487.sergei.shtylyov@cogentembedded.com> <51819C12.2060103@cogentembedded.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit To: netdev@vger.kernel.org, klassert@mathematik.tu-chemnitz.de, tedheadster@gmail.com, David Miller Return-path: Received: from mail-la0-f44.google.com ([209.85.215.44]:35353 "EHLO mail-la0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759009Ab3EBRlW (ORCPT ); Thu, 2 May 2013 13:41:22 -0400 Received: by mail-la0-f44.google.com with SMTP id ed20so786526lab.31 for ; Thu, 02 May 2013 10:41:20 -0700 (PDT) In-Reply-To: <51819C12.2060103@cogentembedded.com> Sender: netdev-owner@vger.kernel.org List-ID: Hello. On 02-05-2013 2:49, Sergei Shtylyov wrote: >> From: Sergei Shtylyov >> The driver wrongly claimed I/O ports at an address returned by pci_iomap() -- >> even if it was passed an MMIO address. Fix this by claiming/releasing all PCI >> resources in the PCI driver probe()/remove() methods instead and get rid of the >> 'must_free_region' flag weirdness (why would Cardbus claim anything for us?). >> Also, the remove() method was trying to talk to the chip after having disabled >> its address decoders (at least on x86) -- fix this and while doing it get rid >> of the useless VORTEX_PCI() invocations. >> While at it, fix some cases of the overly indented code and the code crossing >> a 80-column boundary... >> [akpm@linux-foundation.org: coding-style fixes] >> Signed-off-by: Sergei Shtylyov >> Cc: Steffen Klassert >> Signed-off-by: Andrew Morton >> --- >> The patch is against David Miller's 'net-next.git' repo. >> Matthew, please test it on one of your 3c59x EISA boards. >> Steffen, here's my second try to get this patch past you, the last one was back >> in 2009... > This is mostly a request for testing at this point, I forgot to add [RFT] > to the subject... David and others, your comments are welcome however. The patch probably needs some more trimming before it can be considered a pure fix. It has been somewhat tested now on EISA (and once more on PCI) board, and as a result of testing another one, EISA specific error was reported, fix for which I'm going to publish for testing RSN. WBR, Sergei