From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Hildenbrand Subject: Re: Balloon pressuring page cache Date: Wed, 5 Feb 2020 10:58:14 +0100 Message-ID: References: <91270a68-ff48-88b0-219c-69801f0c252f@redhat.com> <539B606A-A0CA-4AA4-B99A-759C874A1EF8@vmware.com> <20200204033657-mutt-send-email-mst@kernel.org> <286AC319A985734F985F78AFA26841F73E41F0F0@shsmsx102.ccr.corp.intel.com> <1eff30a0-a38a-cd45-2fc1-80cfd0bf5f04@redhat.com> <286AC319A985734F985F78AFA26841F73E41F306@shsmsx102.ccr.corp.intel.com> <2b388a78-79cd-f99a-6f87-6446dcb4b819@redhat.com> <286AC319A985734F985F78AFA26841F73E41F367@shsmsx102.ccr.corp.intel.com> <605bef3e-373f-f39a-4849-930326564b5b@redhat.com> <286AC319A985734F985F78AFA26841F73E41F3DE@shsmsx102.ccr.corp.intel.com> <286AC319A985734F985F78AFA26841F73E41F490@shsmsx102.ccr.corp.intel.com> <5b184893-014c-35a1-928b-37b8f4647116@redhat.com> <286AC319A985734F985F78AFA26841F73E41F562@shsmsx102.ccr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <286AC319A985734F985F78AFA26841F73E41F562@shsmsx102.ccr.corp.intel.com> Content-Language: en-US List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: virtualization-bounces@lists.linux-foundation.org Sender: "Virtualization" To: "Wang, Wei W" , "Michael S. Tsirkin" Cc: "virtualization@lists.linux-foundation.org" , Tyler Sanderson , "linux-mm@kvack.org" , Nadav Amit , David Rientjes , Alexander Duyck , Michal Hocko List-Id: virtualization@lists.linuxfoundation.org On 05.02.20 10:49, Wang, Wei W wrote: > On Wednesday, February 5, 2020 5:37 PM, David Hildenbrand wrote: >>> >>> Not sure how TCG tracks the dirty bits. But In whatever >>> implementation, the hypervisor should have >> >> There is only a single bitmap for that purpose. (well, the one where KVM >> syncs to) >> >>> already dealt with the race between he current round and the previous >> round dirty recording. >>> (the race isn't brought by this feature essentially) >> >> It is guaranteed to work reliably without this feature as you only clear what >> *has been migrated*, > > Not "clear what has been migrated" (that skips nothing..) > Anyway, it's a hint used for optimization. Yes, an optimization that might easily lead to data corruption when the two bitmaps are either not in place or don't play along in that specific way (and I suspect this is the case under TCG). -- Thanks, David / dhildenb