xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Boris Ostrovsky <boris.ostrovsky@oracle.com>
To: Haozhong Zhang <haozhong.zhang@intel.com>, xen-devel@lists.xen.org
Cc: Kevin Tian <kevin.tian@intel.com>, Wei Liu <wei.liu2@citrix.com>,
	Ian Campbell <ian.campbell@citrix.com>,
	Stefano Stabellini <stefano.stabellini@eu.citrix.com>,
	Jun Nakajima <jun.nakajima@intel.com>,
	Andrew Cooper <andrew.cooper3@citrix.com>,
	Ian Jackson <ian.jackson@eu.citrix.com>,
	Aravind Gopalakrishnan <Aravind.Gopalakrishnan@amd.com>,
	Jan Beulich <jbeulich@suse.com>, Keir Fraser <keir@xen.org>,
	Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Subject: Re: [PATCH 11/13] vmx: Use scaled host TSC to calculate TSC offset
Date: Thu, 22 Oct 2015 11:55:00 -0400	[thread overview]
Message-ID: <562906D4.3090700@oracle.com> (raw)
In-Reply-To: <1443424438-13404-12-git-send-email-haozhong.zhang@intel.com>

On 09/28/2015 03:13 AM, Haozhong Zhang wrote:
> If VMX TSC scaling is enabled and no TSC emulation is used,
> vmx_set_tsc_offset() will calculate the TSC offset by substracting the
> scaled host TSC from the current guest TSC.
>
> Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
> ---
>   xen/arch/x86/hvm/vmx/vmx.c | 15 +++++++++++++++
>   1 file changed, 15 insertions(+)
>
> diff --git a/xen/arch/x86/hvm/vmx/vmx.c b/xen/arch/x86/hvm/vmx/vmx.c
> index 454440e..163974d 100644
> --- a/xen/arch/x86/hvm/vmx/vmx.c
> +++ b/xen/arch/x86/hvm/vmx/vmx.c
> @@ -1102,11 +1102,26 @@ static void vmx_handle_cd(struct vcpu *v, unsigned long value)
>   
>   static void vmx_set_tsc_offset(struct vcpu *v, u64 offset, u64 at_tsc)
>   {
> +    uint64_t host_tsc, guest_tsc;
> +    struct domain *d = v->domain;
> +
> +    guest_tsc = hvm_get_guest_tsc_fixed(v, at_tsc);
> +
> +    if ( cpu_has_vmx_tsc_scaling && !d->arch.vtsc )
> +    {
> +        host_tsc = at_tsc ? at_tsc : rdtsc();
> +        offset = guest_tsc - hvm_scale_tsc(v, host_tsc);
> +    }
> +
>       vmx_vmcs_enter(v);
>   
> +    if ( !nestedhvm_enabled(d) )
> +        goto out;
> +
>       if ( nestedhvm_vcpu_in_guestmode(v) )
>           offset += nvmx_get_tsc_offset(v);
>   
> +out:
>       __vmwrite(TSC_OFFSET, offset);
>       vmx_vmcs_exit(v);
>   }


This (and corresponding SVM code) looks somewhat suspect to me: if the 
processor supports scaling we are ignoring caller-provided offset.

Besides, at least when called from hvm_set_guest_tsc_fixed() --- we've 
already taken scaling into account, that's what patch 6 does, doesn't it?


-boris

  reply	other threads:[~2015-10-22 15:55 UTC|newest]

Thread overview: 117+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-28  7:13 [PATCH 00/13] Add VMX TSC scaling support Haozhong Zhang
2015-09-28  7:13 ` [PATCH 01/13] x86/time.c: Use system time to calculate elapsed_nsec in tsc_get_info() Haozhong Zhang
2015-10-09  6:51   ` Jan Beulich
2015-10-09 13:41     ` Boris Ostrovsky
2015-10-09 14:00       ` Haozhong Zhang
2015-10-09 15:11         ` Jan Beulich
2015-10-09 16:09           ` Boris Ostrovsky
2015-10-09 16:19             ` Jan Beulich
2015-10-09 16:31               ` Boris Ostrovsky
2015-10-09 16:51                 ` Haozhong Zhang
2015-10-09 18:59                   ` Boris Ostrovsky
2015-10-09 14:39       ` Jan Beulich
2015-10-09 15:37         ` Boris Ostrovsky
2015-10-09 16:39           ` Haozhong Zhang
2015-10-09 16:44             ` Boris Ostrovsky
2015-10-09 14:35     ` Haozhong Zhang
2015-10-09 14:43       ` Jan Beulich
2015-10-09 15:56         ` Boris Ostrovsky
2015-10-14  2:45     ` Haozhong Zhang
2015-10-14  9:40       ` Jan Beulich
2015-10-14 10:00         ` Haozhong Zhang
2015-10-14 10:20           ` Jan Beulich
2015-10-14 10:24             ` Haozhong Zhang
2015-09-28  7:13 ` [PATCH 02/13] x86/time.c: Get the correct guest TSC rate " Haozhong Zhang
2015-09-28  7:13 ` [PATCH 03/13] x86/hvm: Collect information of TSC scaling ratio Haozhong Zhang
2015-10-22 12:53   ` Jan Beulich
2015-10-22 14:40     ` Haozhong Zhang
2015-10-22 14:51       ` Jan Beulich
2015-10-22 15:57         ` Haozhong Zhang
2015-09-28  7:13 ` [PATCH 04/13] x86/hvm: Setup " Haozhong Zhang
2015-10-22 13:13   ` Jan Beulich
2015-10-22 15:55     ` Haozhong Zhang
2015-10-22 16:05       ` Jan Beulich
2015-10-22 16:39         ` Haozhong Zhang
2015-10-23  7:44     ` Haozhong Zhang
2015-10-23  7:59       ` Jan Beulich
2015-10-23  8:18         ` Haozhong Zhang
2015-10-23  8:31           ` Jan Beulich
2015-10-23  8:40             ` Haozhong Zhang
2015-10-23  9:18               ` Jan Beulich
2015-09-28  7:13 ` [PATCH 05/13] x86/hvm: Replace architecture TSC scaling by a common function Haozhong Zhang
2015-10-22 13:52   ` Jan Beulich
2015-10-23  0:49     ` Haozhong Zhang
2015-09-28  7:13 ` [PATCH 06/13] x86/hvm: Scale host TSC when setting/getting guest TSC Haozhong Zhang
2015-10-22 14:17   ` Jan Beulich
2015-10-22 15:44     ` Boris Ostrovsky
2015-10-22 16:23       ` Haozhong Zhang
2015-10-27 20:16       ` Aravind Gopalakrishnan
2015-10-28  1:51         ` Haozhong Zhang
2015-11-09  7:43         ` Haozhong Zhang
2015-11-12 13:50           ` George Dunlap
2015-10-22 16:03     ` Haozhong Zhang
2015-10-27  1:54     ` Haozhong Zhang
2015-10-27  8:15       ` Jan Beulich
2015-10-27  8:25         ` Haozhong Zhang
2015-10-27  8:44     ` Haozhong Zhang
2015-10-27 13:10       ` Boris Ostrovsky
2015-10-27 13:55         ` Boris Ostrovsky
2015-10-27 16:13           ` haozhong.zhang
2015-10-27 16:13         ` haozhong.zhang
2015-09-28  7:13 ` [PATCH 07/13] x86/hvm: Move saving/loading vcpu's TSC to common code Haozhong Zhang
2015-10-22 14:54   ` Jan Beulich
2015-09-28  7:13 ` [PATCH 08/13] x86/hvm: Detect TSC scaling through hvm_funcs in tsc_set_info() Haozhong Zhang
2015-10-22 15:01   ` Jan Beulich
2015-09-28  7:13 ` [PATCH 09/13] x86/time.c: Scale host TSC in pvclock properly Haozhong Zhang
2015-09-28 16:36   ` Boris Ostrovsky
2015-09-29  0:19     ` Haozhong Zhang
2015-10-22 15:50   ` Boris Ostrovsky
2015-10-22 16:44     ` Haozhong Zhang
2015-10-22 19:15       ` Boris Ostrovsky
2015-09-28  7:13 ` [PATCH 10/13] vmx: Detect and initialize VMX RDTSC(P) scaling Haozhong Zhang
2015-10-27 13:19   ` Jan Beulich
2015-10-27 16:17     ` Haozhong Zhang
2015-09-28  7:13 ` [PATCH 11/13] vmx: Use scaled host TSC to calculate TSC offset Haozhong Zhang
2015-10-22 15:55   ` Boris Ostrovsky [this message]
2015-10-22 17:12     ` Haozhong Zhang
2015-10-22 19:19       ` Boris Ostrovsky
2015-10-23  0:52         ` Haozhong Zhang
2015-10-27 13:29   ` Jan Beulich
2015-10-27 16:21     ` Haozhong Zhang
2015-09-28  7:13 ` [PATCH 12/13] vmx: Add a call-back to apply TSC scaling ratio to hardware Haozhong Zhang
2015-09-28 16:02   ` Boris Ostrovsky
2015-09-29  1:07     ` Haozhong Zhang
2015-09-29  9:33       ` Andrew Cooper
2015-09-29 10:02         ` Haozhong Zhang
2015-09-29 10:25           ` Andrew Cooper
2015-09-29 13:59             ` Haozhong Zhang
2015-10-27 13:33   ` Jan Beulich
2015-10-28  2:41     ` Haozhong Zhang
2015-09-28  7:13 ` [PATCH 13/13] tools/libxl: Add 'vtsc_khz' option to set guest TSC rate Haozhong Zhang
2015-09-28 11:47   ` Julien Grall
2015-09-28 12:11     ` Haozhong Zhang
2015-09-28 14:19   ` Wei Liu
2015-09-29  0:40     ` Haozhong Zhang
2015-09-29  9:20       ` Wei Liu
2015-09-29  9:50         ` Haozhong Zhang
2015-09-29 10:24           ` Julien Grall
2015-09-29 10:07       ` Ian Campbell
2015-09-29 10:33         ` Wei Liu
2015-09-29 12:57         ` Haozhong Zhang
2015-09-29 10:04   ` Ian Campbell
2015-09-29 10:13     ` Haozhong Zhang
2015-09-29 10:24       ` Andrew Cooper
2015-09-29 10:28         ` Ian Campbell
2015-09-29 10:31           ` Andrew Cooper
2015-09-29 13:53           ` Haozhong Zhang
2015-09-29 13:56             ` Andrew Cooper
2015-09-29 14:01               ` Haozhong Zhang
2015-09-29 14:37                 ` Ian Campbell
2015-09-29 15:16                   ` Haozhong Zhang
2015-09-28 10:51 ` [PATCH 00/13] Add VMX TSC scaling support Andrew Cooper
2015-09-28 13:48   ` Boris Ostrovsky
2015-11-22 17:54 ` Haozhong Zhang
2015-11-23 15:37   ` Boris Ostrovsky
2015-11-24 13:05     ` Haozhong Zhang
2015-11-24 14:19       ` Boris Ostrovsky
2015-11-24 14:25       ` 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=562906D4.3090700@oracle.com \
    --to=boris.ostrovsky@oracle.com \
    --cc=Aravind.Gopalakrishnan@amd.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=haozhong.zhang@intel.com \
    --cc=ian.campbell@citrix.com \
    --cc=ian.jackson@eu.citrix.com \
    --cc=jbeulich@suse.com \
    --cc=jun.nakajima@intel.com \
    --cc=keir@xen.org \
    --cc=kevin.tian@intel.com \
    --cc=stefano.stabellini@eu.citrix.com \
    --cc=suravee.suthikulpanit@amd.com \
    --cc=wei.liu2@citrix.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 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).