From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vitaly Kuznetsov Subject: Re: [PATCH 2/2] x86/vdso: Add VCLOCK_HVCLOCK vDSO clock read method Date: Fri, 10 Feb 2017 13:15:00 +0100 Message-ID: <87h942qm2j.fsf@vitty.brq.redhat.com> References: <20170209141052.18694-1-vkuznets@redhat.com> <20170209141052.18694-3-vkuznets@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: (Andy Lutomirski's message of "Thu, 9 Feb 2017 14:55:50 -0800") List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org To: Andy Lutomirski Cc: Stephen Hemminger , Haiyang Zhang , "x86@kernel.org" , "linux-kernel@vger.kernel.org" , "virtualization@lists.linux-foundation.org" , Ingo Molnar , "H. Peter Anvin" , "devel@linuxdriverproject.org" , Thomas Gleixner List-Id: virtualization@lists.linuxfoundation.org Andy Lutomirski writes: > On Thu, Feb 9, 2017 at 12:45 PM, KY Srinivasan wrote: >> >> >>> -----Original Message----- >>> From: Thomas Gleixner [mailto:tglx@linutronix.de] >>> Sent: Thursday, February 9, 2017 9:08 AM >>> To: Vitaly Kuznetsov >>> Cc: x86@kernel.org; Andy Lutomirski ; Ingo Molnar >>> ; H. Peter Anvin ; KY Srinivasan >>> ; Haiyang Zhang ; Stephen >>> Hemminger ; Dexuan Cui >>> ; linux-kernel@vger.kernel.org; >>> devel@linuxdriverproject.org; virtualization@lists.linux-foundation.org >>> Subject: Re: [PATCH 2/2] x86/vdso: Add VCLOCK_HVCLOCK vDSO clock read >>> method >>> >>> On Thu, 9 Feb 2017, Vitaly Kuznetsov wrote: >>> > +#ifdef CONFIG_HYPERV_TSCPAGE >>> > +static notrace u64 vread_hvclock(int *mode) >>> > +{ >>> > + const struct ms_hyperv_tsc_page *tsc_pg = >>> > + (const struct ms_hyperv_tsc_page *)&hvclock_page; >>> > + u64 sequence, scale, offset, current_tick, cur_tsc; >>> > + >>> > + while (1) { >>> > + sequence = READ_ONCE(tsc_pg->tsc_sequence); >>> > + if (!sequence) >>> > + break; >>> > + >>> > + scale = READ_ONCE(tsc_pg->tsc_scale); >>> > + offset = READ_ONCE(tsc_pg->tsc_offset); >>> > + rdtscll(cur_tsc); >>> > + >>> > + current_tick = mul_u64_u64_shr(cur_tsc, scale, 64) + offset; >>> > + >>> > + if (READ_ONCE(tsc_pg->tsc_sequence) == sequence) >>> > + return current_tick; >>> >>> That sequence stuff lacks still a sensible explanation. It's fundamentally >>> different from the sequence counting we do in the kernel, so documentation >>> for it is really required. >> >> The host is updating multiple fields in this shared TSC page and the sequence number is >> used to ensure that the guest sees a consistent set values published. If I remember >> correctly, Xen has a similar mechanism. > > So what's the actual protocol? When the hypervisor updates the page, > does it freeze all guest cpus? If not, how does it maintain > atomicity? I don't really know how it is implemented server-side but I *think* that freezing all CPUs is only required when we want to update *both* ReferenceTscScale and ReferenceTscOffset at the same time (as Hyper-V is 64-bit only so it can always atomically update 64-bit values)... -- Vitaly