From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752832AbaBLRAv (ORCPT ); Wed, 12 Feb 2014 12:00:51 -0500 Received: from mail-ob0-f173.google.com ([209.85.214.173]:42366 "EHLO mail-ob0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751638AbaBLRAu (ORCPT ); Wed, 12 Feb 2014 12:00:50 -0500 MIME-Version: 1.0 In-Reply-To: <20140212132640.GA3147@dhcp-26-207.brq.redhat.com> References: <434a34e01c353441131204edef36126a95e8928a.1391172839.git.agordeev@redhat.com> <20140131.133051.483295479425700960.davem@davemloft.net> <20140211003814.GC12851@google.com> <20140211010803.GD12851@google.com> <20140212132640.GA3147@dhcp-26-207.brq.redhat.com> From: Bjorn Helgaas Date: Wed, 12 Feb 2014 10:00:28 -0700 Message-ID: Subject: Re: [PATCH 01/34] bnx2: Use pci_enable_msix_range() To: Alexander Gordeev Cc: "linux-kernel@vger.kernel.org" , "linux-pci@vger.kernel.org" Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Feb 12, 2014 at 6:26 AM, Alexander Gordeev wrote: > On Mon, Feb 10, 2014 at 06:08:03PM -0700, Bjorn Helgaas wrote: >> I skimmed these and the scsi patches, and I think you were right in >> proposing an MSI-X enable function that takes a single "number of vectors" >> argument, in addition to pci_enable_msix_range(), which takes a minimum and >> a maximum. Obviously the pci_enable_msix_fixed() or whatever could be a >> simple #define wrapper or something. >> >> Of the fifty-some net and scsi patches, I counted 23 that use the min == >> max pattern, and it seems a shame to have to repeat that expression. > > (un-CCing netdev@vger.kernel.org) > > Hi Bjorn, > > I propose pci_enable_msix_exact(pdev, entries, nvec) rather than > pci_enable_msix_fixed(). OK, that sounds fine. > Do you prefer this one to return 0/-errno or nvec/-errno? 0/-errno seems right to me. We are asking for a very specific thing, and returning nvec doesn't give the caller any additional information (since the caller supplied nvec in the first place), so simple success/failure is what I would expect. > Do you want pci_enable_msi_exact() in addition to pci_enable_msix_exact()? If there are cases where pci_enable_msi_exact() would be used, I guess the same arguments would apply, so yes. I haven't looked at the archives, but I have this nagging feeling that this is exactly what you proposed initially, and I had some objection. If so, I'm sorry about changing my mind :) Bjorn