All of lore.kernel.org
 help / color / mirror / Atom feed
From: Haozhong Zhang <haozhong.zhang@intel.com>
To: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Kevin Tian <kevin.tian@intel.com>, Keir Fraser <keir@xen.org>,
	Jan Beulich <jbeulich@suse.com>,
	Jun Nakajima <jun.nakajima@intel.com>,
	Andrew Cooper <andrew.cooper3@citrix.com>,
	xen-devel@lists.xen.org,
	Aravind Gopalakrishnan <Aravind.Gopalakrishnan@amd.com>,
	Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Subject: Re: [PATCH v2 04/14] x86/time.c: Use correct guest TSC frequency in tsc_get_info()
Date: Thu, 10 Dec 2015 08:14:04 +0800	[thread overview]
Message-ID: <20151210001404.GB4704@hz-desktop.sh.intel.com> (raw)
In-Reply-To: <566720F3.4090107@oracle.com>

On 12/08/15 13:26, Boris Ostrovsky wrote:
> On 12/07/2015 08:08 PM, Haozhong Zhang wrote:
> >On 12/07/15 12:53, Boris Ostrovsky wrote:
> >>On 12/06/2015 03:58 PM, Haozhong Zhang wrote:
> >>>When the TSC mode of a HVM container is TSC_MODE_DEFAULT or
> >>>TSC_MODE_PVRDTSCP and no TSC emulation is used, the existing
> >>>tsc_get_info() uses the host TSC frequency (cpu_khz) as the guest TSC
> >>>frequency. However, tsc_set_info() may set the guest TSC frequency to a
> >>>value different than the host. In order to keep consistent to
> >>>tsc_set_info(), this patch makes tsc_get_info() use the value set by
> >>>tsc_set_info() as the guest TSC frequency.
> >>>
> >>>Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
> >>>---
> >>>  xen/arch/x86/time.c | 12 +++++++++---
> >>>  1 file changed, 9 insertions(+), 3 deletions(-)
> >>>
> >>>diff --git a/xen/arch/x86/time.c b/xen/arch/x86/time.c
> >>>index 1091e69..95df4f1 100644
> >>>--- a/xen/arch/x86/time.c
> >>>+++ b/xen/arch/x86/time.c
> >>>@@ -1749,6 +1749,9 @@ void tsc_get_info(struct domain *d, uint32_t *tsc_mode,
> >>>                    uint64_t *elapsed_nsec, uint32_t *gtsc_khz,
> >>>                    uint32_t *incarnation)
> >>>  {
> >>>+    bool_t enable_tsc_scaling = has_hvm_container_domain(d) &&
> >>>+                                cpu_has_tsc_ratio;
> >>&& !d->arch.vtsc ?
> >>
> >>(assuming my comment to the previous patch is correct).
> >>
> >enable_tsc_scaling in tsc_get_info() is always used in the condition
> >!d->arch.vtsc, so it's not necessary to include it again in
> >enable_tsc_scaling.
> >
> >>>+
> >>>      *incarnation = d->arch.incarnation;
> >>>      *tsc_mode = d->arch.tsc_mode;
> >>>@@ -1769,7 +1772,7 @@ void tsc_get_info(struct domain *d, uint32_t *tsc_mode,
> >>>          }
> >>>          tsc = rdtsc();
> >>>          *elapsed_nsec = scale_delta(tsc, &d->arch.vtsc_to_ns);
> >>>-        *gtsc_khz = cpu_khz;
> >>>+        *gtsc_khz = enable_tsc_scaling ? d->arch.tsc_khz : cpu_khz;
> >>>          break;
> >>>      case TSC_MODE_PVRDTSCP:
> >>>          if ( d->arch.vtsc )
> >>>@@ -1779,10 +1782,13 @@ void tsc_get_info(struct domain *d, uint32_t *tsc_mode,
> >>>          }
> >>>          else
> >>>          {
> >>>+            struct time_scale *scale = enable_tsc_scaling ?
> >>>+                                       &this_cpu(cpu_time).tsc_scale :
> >>>+                                       &d->arch.vtsc_to_ns;
> >>IIUIC tsc_scale is host property and so why would it be used if TSC scaling
> >>is available to guests?
> >>
> >scale is used below to convert a host TSC to nanosec. When TSC scaling is used,
> >d->arch.vtsc_to_ns may not base on the host TSC frequency, so I turn to the host
> >tsc_scale instead.
> 
> OK. Can we then use tsc_scale for both with and without TSC scaling? (on the
> set side too).
>
Yes, we can, because early_time_init() calculates tsc_scale based on
cpu_khz. I'll change it in the next version.

> (And as a side note, I think that the comment in include/asm-x86/domain.h
> describing vtsc_to_ns as being used for emulated cases is rather misleading.
> We use it for both emulated and native)
>
agree, I'll update those comments in the next version.

Haozhong

> -boris
> 
> 
> >
> >Haozhong
> >
> >>-boris
> >>
> >>
> >>>              tsc = rdtsc();
> >>>-            *elapsed_nsec = scale_delta(tsc, &d->arch.vtsc_to_ns) -
> >>>+            *elapsed_nsec = scale_delta(tsc, scale) -
> >>>                              d->arch.vtsc_offset;
> >>>-            *gtsc_khz = 0; /* ignored by tsc_set_info */
> >>>+            *gtsc_khz = enable_tsc_scaling ? d->arch.tsc_khz : 0;
> >>>          }
> >>>          break;
> >>>      }
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel

  reply	other threads:[~2015-12-10  0:14 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-06 20:58 [PATCH v2 00/14] Add VMX TSC scaling support Haozhong Zhang
2015-12-06 20:58 ` [PATCH v2 01/14] svm: Fix incorrect TSC scaling Haozhong Zhang
2015-12-07 16:49   ` Boris Ostrovsky
2015-12-06 20:58 ` [PATCH v2 02/14] x86/time.c: Fix domain type check in tsc_set_info() Haozhong Zhang
2015-12-07 16:49   ` Boris Ostrovsky
2015-12-06 20:58 ` [PATCH v2 03/14] x86/time.c: Use correct guest TSC frequency " Haozhong Zhang
2015-12-07 16:57   ` Boris Ostrovsky
2015-12-08  0:30     ` Haozhong Zhang
2015-12-06 20:58 ` [PATCH v2 04/14] x86/time.c: Use correct guest TSC frequency in tsc_get_info() Haozhong Zhang
2015-12-07 17:53   ` Boris Ostrovsky
2015-12-08  1:08     ` Haozhong Zhang
2015-12-08 18:26       ` Boris Ostrovsky
2015-12-10  0:14         ` Haozhong Zhang [this message]
2015-12-06 20:58 ` [PATCH v2 05/14] x86/hvm: Scale host TSC when setting/getting guest TSC Haozhong Zhang
2015-12-07 17:54   ` Boris Ostrovsky
2015-12-06 20:58 ` [PATCH v2 06/14] x86/time.c: Scale host TSC in pvclock properly Haozhong Zhang
2015-12-07 18:14   ` Boris Ostrovsky
2015-12-08  1:52     ` Haozhong Zhang
2015-12-08 19:21       ` Boris Ostrovsky
2015-12-10  0:23         ` Haozhong Zhang
2015-12-10 11:56           ` Joao Martins
2015-12-29 18:22             ` Joao Martins
2015-12-30  1:07               ` Haozhong Zhang
2015-12-06 20:58 ` [PATCH v2 07/14] svm: Remove redundant TSC scaling in svm_set_tsc_offset() Haozhong Zhang
2015-12-07 18:25   ` Boris Ostrovsky
2015-12-08  1:54     ` Haozhong Zhang
2015-12-06 20:58 ` [PATCH v2 08/14] x86/hvm: Collect information of TSC scaling ratio Haozhong Zhang
2015-12-07 18:31   ` Boris Ostrovsky
2015-12-10 10:19   ` Tian, Kevin
2015-12-10 10:27     ` Zhang, Haozhong
2015-12-06 20:58 ` [PATCH v2 09/14] x86/hvm: Setup " Haozhong Zhang
2015-12-07 18:42   ` Boris Ostrovsky
2015-12-08  1:58     ` Haozhong Zhang
2015-12-10 10:27   ` Tian, Kevin
2015-12-10 10:37     ` Zhang, Haozhong
2015-12-06 20:58 ` [PATCH v2 10/14] x86/hvm: Replace architecture TSC scaling by a common function Haozhong Zhang
2015-12-07 18:55   ` Boris Ostrovsky
2015-12-08  2:06     ` Haozhong Zhang
2015-12-10 10:29   ` Tian, Kevin
2015-12-06 20:58 ` [PATCH v2 11/14] x86/hvm: Move saving/loading vcpu's TSC to common code Haozhong Zhang
2015-12-07 18:56   ` Boris Ostrovsky
2015-12-10 10:30   ` Tian, Kevin
2015-12-06 20:58 ` [PATCH v2 12/14] x86/hvm: Detect TSC scaling through hvm_funcs Haozhong Zhang
2015-12-07 18:58   ` Boris Ostrovsky
2015-12-10 10:31   ` Tian, Kevin
2015-12-06 20:58 ` [PATCH v2 13/14] vmx: Add VMX RDTSC(P) scaling support Haozhong Zhang
2015-12-10 10:33   ` Tian, Kevin
2015-12-06 20:58 ` [PATCH v2 14/14] docs: Add descriptions of TSC scaling in xl.cfg and tscmode.txt Haozhong Zhang
2015-12-10 10:40   ` Tian, Kevin
2015-12-10 10:56     ` Zhang, Haozhong
2015-12-06 21:14 ` [PATCH v2 00/14] Add VMX TSC scaling support Haozhong Zhang
2015-12-07  9:03   ` Egger, Christoph
2015-12-07 10:16     ` Haozhong Zhang
2015-12-07 17:04       ` Andrew Cooper
2015-12-08  2:13         ` Haozhong Zhang
2015-12-10 10:43         ` Tian, Kevin
2015-12-10 11:13           ` Haozhong Zhang
2015-12-11  7:35             ` Tian, Kevin
2015-12-11 10:09               ` Andrew Cooper
2015-12-07 10:37   ` Jan Beulich
2015-12-07 10:56     ` Haozhong Zhang

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=20151210001404.GB4704@hz-desktop.sh.intel.com \
    --to=haozhong.zhang@intel.com \
    --cc=Aravind.Gopalakrishnan@amd.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=boris.ostrovsky@oracle.com \
    --cc=jbeulich@suse.com \
    --cc=jun.nakajima@intel.com \
    --cc=keir@xen.org \
    --cc=kevin.tian@intel.com \
    --cc=suravee.suthikulpanit@amd.com \
    --cc=xen-devel@lists.xen.org \
    /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
Be 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.