From: Joao Martins <firstname.lastname@example.org> To: Jan Beulich <JBeulich@suse.com> Cc: xen-devel <email@example.com>, Wei Liu <firstname.lastname@example.org>, Dario Faggioli <email@example.com>, Andrew Cooper <firstname.lastname@example.org> Subject: Re: [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path Date: Thu, 9 Jun 2016 13:11:51 +0100 [thread overview] Message-ID: <57595D07.email@example.com> (raw) In-Reply-To: <575976AE02000078000F3695@prv-mh.provo.novell.com> On 06/09/2016 01:01 PM, Jan Beulich wrote: > This looks like a copy and paste mistake in commit 1b6a99892d ("x86: > Simpler time handling when TSC is constant across all power saving > states"), responsible for occasional many-microsecond cross-CPU skew of > what NOW() returns. > > Also improve the correlation between local TSC and stime stamps > obtained at the end of the two calibration handlers: Compute the stime > one from the TSC one, instead of doing another rdtsc() for that > compuation. > > Signed-off-by: Jan Beulich <firstname.lastname@example.org> > --- > As to 4.7 inclusion: This of course looks like a pretty blatant mistake > that has been there for many years (and hence many releases). There's > certainly non-zero risk that I'm overlooking something here (despite > Joao apparently having come to the same conclusion), so I can't really > make up my mind on whether to request this patch to be put there right > away, or rather having linger in -unstable for a while. > Initially I thought of this as a fix too, but then wouldn't having t->stime_local_stamp be c->stime_local_stamp, render no use to the platform timer reads done on calibration? Unless we would change update_vcpu_system to use stime_master_stamp instead? stime_master_stamp field isn't used anywhere other than the dom0 injected cpu_frequency_change or when at boot seeding the cpu_time struct on init_percpu_time (and the already mentioned use on local_time_calibration) ? init_percpu_time also takes a different read of the platform timer per cpu and could probably be inherited by a read done on the boot processor and written on remaining CPUs, so that all would start from the same stamp. IOW - this sounds like time we are turning stime to be totally TSC except when initially seeding each cpu_time? > --- a/xen/arch/x86/time.c > +++ b/xen/arch/x86/time.c > @@ -998,7 +998,7 @@ static void local_time_calibration(void) > /* Atomically read cpu_calibration struct and write cpu_time struct. */ > local_irq_disable(); > t->local_tsc_stamp = c->local_tsc_stamp; > - t->stime_local_stamp = c->stime_master_stamp; > + t->stime_local_stamp = c->stime_local_stamp; > t->stime_master_stamp = c->stime_master_stamp; > local_irq_enable(); > update_vcpu_system_time(current); > @@ -1275,7 +1275,7 @@ static void time_calibration_tsc_rendezv > } > > c->local_tsc_stamp = rdtsc(); > - c->stime_local_stamp = get_s_time(); > + c->stime_local_stamp = get_s_time_fixed(c->local_tsc_stamp); > c->stime_master_stamp = r->master_stime; > > raise_softirq(TIME_CALIBRATE_SOFTIRQ); > @@ -1305,7 +1305,7 @@ static void time_calibration_std_rendezv > } > > c->local_tsc_stamp = rdtsc(); > - c->stime_local_stamp = get_s_time(); > + c->stime_local_stamp = get_s_time_fixed(c->local_tsc_stamp); > c->stime_master_stamp = r->master_stime; > > raise_softirq(TIME_CALIBRATE_SOFTIRQ); > > > _______________________________________________ Xen-devel mailing list Xenemail@example.com http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2016-06-09 12:11 UTC|newest] Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-06-09 12:01 Jan Beulich 2016-06-09 12:10 ` Andrew Cooper 2016-06-09 12:11 ` Joao Martins [this message] 2016-06-09 12:57 ` Jan Beulich 2016-06-09 15:00 ` Joao Martins 2016-06-09 15:52 ` Jan Beulich 2016-06-09 18:19 ` Joao Martins 2016-06-10 6:59 ` Jan Beulich 2016-06-10 9:29 ` Jan Beulich 2016-06-10 17:07 ` Joao Martins 2016-06-09 12:12 ` Wei Liu
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=57595D07.firstname.lastname@example.org \ --email@example.com \ --cc=JBeulich@suse.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --subject='Re: [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path' \ /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: link
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).