From: Marcelo Tosatti <mtosatti@redhat.com> To: Paolo Bonzini <pbonzini@redhat.com> Cc: kvm@vger.kernel.org, qemu-devel@nongnu.org, "Dr. David Alan Gilbert" <dgilbert@redhat.com>, Juan Quintela <quintela@redhat.com>, Radim Krcmar <rkrcmar@redhat.com>, Eduardo Habkost <ehabkost@redhat.com> Subject: Re: [qemu patch 2/2] kvmclock: reduce kvmclock difference on migration Date: Mon, 14 Nov 2016 15:13:21 -0200 [thread overview] Message-ID: <20161114171318.GA6336@amt.cnet> (raw) In-Reply-To: <cbcac3d2-88aa-3cb4-7257-e70061a22a2c@redhat.com> On Mon, Nov 14, 2016 at 05:43:33PM +0100, Paolo Bonzini wrote: > > > On 14/11/2016 16:40, Marcelo Tosatti wrote: > > static bool kvmclock_src_use_reliable_get_clock(void *opaque) > > { > > KVMClockState *s = opaque; > > > > /* > > * On machine types that support reliable KVM_GET_CLOCK, > > * if host kernel does provide reliable KVM_GET_CLOCK, > > * set src_use_reliable_get_clock=true so that destination > > * avoids reading kvmclock from memory. > > */ > > if (s->mach_use_reliable_get_clock && kvm_has_adjust_clock_stable()) > > { > > s->src_use_reliable_get_clock = true; > > } > > > > return s->mach_use_reliable_get_clock; > > } > > > > > > Ah, OK, done. > > s->src_use_reliable_get_clock should not be set with > KVM_CHECK_EXTENSION, but rather from the flags returned by KVM_GET_CLOCK. Well, thats not right: What matters is the presence of get_kvmclock_ns which returns a value that the guest sees. get_kernel_monotonic_clock() + kvmclock_offset + (rdtsc() - tsc_timestamp) IOW what the guest sees. And you changed that in commit 108b249c453dd7132599ab6dc7e435a7036c193f Author: Paolo Bonzini <pbonzini@redhat.com> Date: Thu Sep 1 14:21:03 2016 +0200 KVM: x86: introduce get_kvmclock_ns And the correct behaviour (once KVM_GET_CLOCK is fixed per previous message to return rdtsc - tsc_timestamp for the non masterclock case) depends on this commit above, not on masterclock. > > So s->src_use_reliable_get_clock is only used to indicate > > to the destination that: "you can use KVM_GET_CLOCK value, > > its safe". > > Yes, we agree. I was listing all the points, not just those where we > disagree. Actually I'm not sure where we disagree, except on using > flags from KVM_CHECK_EXTENSION vs. flags from KVM_GET_CLOCK... > > Paolo
WARNING: multiple messages have this Message-ID (diff)
From: Marcelo Tosatti <mtosatti@redhat.com> To: Paolo Bonzini <pbonzini@redhat.com> Cc: kvm@vger.kernel.org, qemu-devel@nongnu.org, "Dr. David Alan Gilbert" <dgilbert@redhat.com>, Juan Quintela <quintela@redhat.com>, Radim Krcmar <rkrcmar@redhat.com>, Eduardo Habkost <ehabkost@redhat.com> Subject: Re: [Qemu-devel] [qemu patch 2/2] kvmclock: reduce kvmclock difference on migration Date: Mon, 14 Nov 2016 15:13:21 -0200 [thread overview] Message-ID: <20161114171318.GA6336@amt.cnet> (raw) In-Reply-To: <cbcac3d2-88aa-3cb4-7257-e70061a22a2c@redhat.com> On Mon, Nov 14, 2016 at 05:43:33PM +0100, Paolo Bonzini wrote: > > > On 14/11/2016 16:40, Marcelo Tosatti wrote: > > static bool kvmclock_src_use_reliable_get_clock(void *opaque) > > { > > KVMClockState *s = opaque; > > > > /* > > * On machine types that support reliable KVM_GET_CLOCK, > > * if host kernel does provide reliable KVM_GET_CLOCK, > > * set src_use_reliable_get_clock=true so that destination > > * avoids reading kvmclock from memory. > > */ > > if (s->mach_use_reliable_get_clock && kvm_has_adjust_clock_stable()) > > { > > s->src_use_reliable_get_clock = true; > > } > > > > return s->mach_use_reliable_get_clock; > > } > > > > > > Ah, OK, done. > > s->src_use_reliable_get_clock should not be set with > KVM_CHECK_EXTENSION, but rather from the flags returned by KVM_GET_CLOCK. Well, thats not right: What matters is the presence of get_kvmclock_ns which returns a value that the guest sees. get_kernel_monotonic_clock() + kvmclock_offset + (rdtsc() - tsc_timestamp) IOW what the guest sees. And you changed that in commit 108b249c453dd7132599ab6dc7e435a7036c193f Author: Paolo Bonzini <pbonzini@redhat.com> Date: Thu Sep 1 14:21:03 2016 +0200 KVM: x86: introduce get_kvmclock_ns And the correct behaviour (once KVM_GET_CLOCK is fixed per previous message to return rdtsc - tsc_timestamp for the non masterclock case) depends on this commit above, not on masterclock. > > So s->src_use_reliable_get_clock is only used to indicate > > to the destination that: "you can use KVM_GET_CLOCK value, > > its safe". > > Yes, we agree. I was listing all the points, not just those where we > disagree. Actually I'm not sure where we disagree, except on using > flags from KVM_CHECK_EXTENSION vs. flags from KVM_GET_CLOCK... > > Paolo
next prev parent reply other threads:[~2016-11-14 17:13 UTC|newest] Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-11-14 12:36 [qemu patch 0/2] improve kvmclock difference on migration Marcelo Tosatti 2016-11-14 12:36 ` [Qemu-devel] " Marcelo Tosatti 2016-11-14 12:36 ` [qemu patch 1/2] kvm: sync linux headers Marcelo Tosatti 2016-11-14 12:36 ` [Qemu-devel] " Marcelo Tosatti 2016-11-14 12:36 ` [qemu patch 2/2] kvmclock: reduce kvmclock difference on migration Marcelo Tosatti 2016-11-14 12:36 ` [Qemu-devel] " Marcelo Tosatti 2016-11-14 13:54 ` Paolo Bonzini 2016-11-14 13:54 ` [Qemu-devel] " Paolo Bonzini 2016-11-14 14:00 ` Marcelo Tosatti 2016-11-14 14:00 ` [Qemu-devel] " Marcelo Tosatti 2016-11-14 14:22 ` Paolo Bonzini 2016-11-14 14:22 ` [Qemu-devel] " Paolo Bonzini 2016-11-14 14:50 ` Marcelo Tosatti 2016-11-14 14:50 ` [Qemu-devel] " Marcelo Tosatti 2016-11-14 15:00 ` Paolo Bonzini 2016-11-14 15:00 ` [Qemu-devel] " Paolo Bonzini 2016-11-14 15:40 ` Marcelo Tosatti 2016-11-14 15:40 ` [Qemu-devel] " Marcelo Tosatti 2016-11-14 16:43 ` Paolo Bonzini 2016-11-14 16:43 ` [Qemu-devel] " Paolo Bonzini 2016-11-14 17:13 ` Marcelo Tosatti [this message] 2016-11-14 17:13 ` Marcelo Tosatti 2016-11-14 17:20 ` Paolo Bonzini 2016-11-14 17:20 ` [Qemu-devel] " Paolo Bonzini 2016-11-14 18:15 ` Marcelo Tosatti 2016-11-14 18:15 ` [Qemu-devel] " Marcelo Tosatti 2016-11-17 12:16 ` Marcelo Tosatti 2016-11-17 12:16 ` [Qemu-devel] " Marcelo Tosatti 2016-11-17 13:03 ` Paolo Bonzini 2016-11-17 13:03 ` [Qemu-devel] " Paolo Bonzini 2016-11-28 13:47 ` Paolo Bonzini 2016-11-28 13:47 ` [Qemu-devel] " Paolo Bonzini 2016-11-28 14:28 ` Eduardo Habkost 2016-11-28 14:28 ` [Qemu-devel] " Eduardo Habkost 2016-11-28 15:12 ` Paolo Bonzini 2016-11-28 15:12 ` [Qemu-devel] " Paolo Bonzini 2016-11-28 16:36 ` Marcelo Tosatti 2016-11-28 16:36 ` [Qemu-devel] " Marcelo Tosatti 2016-11-28 17:30 ` Paolo Bonzini 2016-11-28 17:30 ` [Qemu-devel] " Paolo Bonzini 2016-11-14 14:11 ` Juan Quintela 2016-11-14 14:11 ` [Qemu-devel] " Juan Quintela 2016-11-14 14:09 ` Juan Quintela 2016-11-14 14:09 ` [Qemu-devel] " Juan Quintela 2016-11-14 15:37 ` Marcelo Tosatti 2016-11-14 15:37 ` [Qemu-devel] " Marcelo Tosatti
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20161114171318.GA6336@amt.cnet \ --to=mtosatti@redhat.com \ --cc=dgilbert@redhat.com \ --cc=ehabkost@redhat.com \ --cc=kvm@vger.kernel.org \ --cc=pbonzini@redhat.com \ --cc=qemu-devel@nongnu.org \ --cc=quintela@redhat.com \ --cc=rkrcmar@redhat.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.