From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthew Wilcox Subject: Re: [PATCH v3 0/6] ATS capability support for Intel IOMMU Date: Thu, 12 Feb 2009 20:44:57 -0700 Message-ID: <20090213034457.GM3624@parisc-linux.org> References: <1234443038-15437-1-git-send-email-yu.zhao@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: jbarnes@virtuousgeek.org, dwmw2@infradead.org, linux-pci@vger.kernel.org, iommu@lists.linux-foundation.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org To: Yu Zhao Return-path: Content-Disposition: inline In-Reply-To: <1234443038-15437-1-git-send-email-yu.zhao@intel.com> Sender: linux-pci-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On Thu, Feb 12, 2009 at 08:50:32PM +0800, Yu Zhao wrote: > 2, avoid using pci_find_ext_capability every time when reading ATS > Invalidate Queue Depth (Matthew Wilcox) er ... I didn't say it was a problem. I said I couldn't tell if it was a problem. There's no point in taking up an extra 4 bytes per pci_dev if it's not a performance problem. How often do we query the queue depth? Is this something a device driver will call once per device and then remember for itself, or only use at setup? Or is it something we call every millisecond? -- Matthew Wilcox Intel Open Source Technology Centre "Bill, look, we understand that you're interested in selling us this operating system, but compare it to ours. We can't possibly take such a retrograde step."