From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH] iommu/vt-d: Remove unnecassary qi clflushes Date: Tue, 5 Jul 2016 18:57:09 +0200 Message-ID: References: <1466773994-242872-1-git-send-email-namit@vmware.com> <20160627113014.GL11432@8bytes.org> <926A8CFB-D22D-4DC6-9EE0-D912451365BB@gmail.com> <5F48A3E6-99EE-4DC5-BF38-4F0338E9E8D0@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <5F48A3E6-99EE-4DC5-BF38-4F0338E9E8D0-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: iommu-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: iommu-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Nadav Amit Cc: Nadav Amit , iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org List-Id: iommu@lists.linux-foundation.org On 05/07/2016 18:50, Nadav Amit wrote: > Paolo Bonzini wrote: > >> >> >> On 05/07/2016 18:27, Nadav Amit wrote: >>>> Although such hardware is old, there are some hypervisors that do not set >>>> the ecap.coherency of emulated IOMMUs. Yes, it is unwise, but there is no >>>> reason to further punish these hypervisors. >> >> QEMU will need the kernel to respect ecap.coherency in order to support >> nested VFIO, for example. > > To clarify - the kernel respects the coherency, but performs more clflushes > than necessary. It has no functional impact, but induces performance > degradation (which I did not measure, but is likely to be several hundreds > of cycles per flush). Oh, then QEMU indeed doesn't care about flushing the invalidation queue. Paolo