xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: "Roger Pau Monné" <roger.pau@citrix.com>
To: Jan Beulich <jbeulich@suse.com>
Cc: George Dunlap <George.Dunlap@eu.citrix.com>,
	"xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>,
	Wei Liu <wl@xen.org>, Andrew Cooper <andrew.cooper3@citrix.com>
Subject: Re: [Xen-devel] [PATCH 3/9] x86/mm: honor opt_pcid also for 32-bit PV domains
Date: Thu, 12 Sep 2019 12:34:41 +0200	[thread overview]
Message-ID: <20190912103441.le4zets5ey73uxwl@Air-de-Roger> (raw)
In-Reply-To: <f58f9215-4005-9c1d-0701-1e7fe705b786@suse.com>

On Wed, Sep 11, 2019 at 05:22:51PM +0200, Jan Beulich wrote:
> I can't see any technical or performance reason why we should treat
> 32-bit PV different from 64-bit PV in this regard.
> 
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

The original commit mentions that PCID doesn't improve performance for
non-XPTI domains, but it doesn't mention whether it makes performance
worse. The change LGTM, if you are fine performance wise:

Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>

> 
> --- a/xen/arch/x86/pv/domain.c
> +++ b/xen/arch/x86/pv/domain.c
> @@ -180,7 +180,24 @@ int switch_compat(struct domain *d)
>      d->arch.x87_fip_width = 4;
>  
>      d->arch.pv.xpti = false;
> -    d->arch.pv.pcid = false;
> +
> +    if ( use_invpcid && cpu_has_pcid )
> +        switch ( ACCESS_ONCE(opt_pcid) )
> +        {
> +        case PCID_OFF:
> +        case PCID_XPTI:
> +            d->arch.pv.pcid = false;
> +            break;
> +
> +        case PCID_ALL:
> +        case PCID_NOXPTI:
> +            d->arch.pv.pcid = true;
> +            break;
> +
> +        default:
> +            ASSERT_UNREACHABLE();
> +            break;
> +        }

This chunk is (functionality wise) exactly the same as the one in
pv_domain_initialise, it might be good to put this in a separate
helper?

>  
>      return 0;
>  
> @@ -312,7 +329,7 @@ int pv_domain_initialise(struct domain *
>  
>      d->arch.pv.xpti = is_hardware_domain(d) ? opt_xpti_hwdom : opt_xpti_domu;
>  
> -    if ( !is_pv_32bit_domain(d) && use_invpcid && cpu_has_pcid )

This is_pv_32bit_domain is already pointless, is_32bit_pv gets
unconditionally set to 0 just two lines above.

Thanks, Roger.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

  reply	other threads:[~2019-09-12 10:35 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-11 15:15 [Xen-devel] [PATCH RESEND/PING 0/9] XSA-292 follow-up Jan Beulich
2019-09-11 15:21 ` [Xen-devel] [PATCH 1/9] x86: adjust cr3_pcid() return type Jan Beulich
2019-09-12  9:19   ` Roger Pau Monné
2019-09-11 15:22 ` [Xen-devel] [PATCH 2/9] x86: limit the amount of TLB flushing in switch_cr3_cr4() Jan Beulich
2019-09-12  9:54   ` Roger Pau Monné
2019-09-12 10:11     ` Jan Beulich
2019-09-12 10:38       ` Roger Pau Monné
2019-09-11 15:22 ` [Xen-devel] [PATCH 3/9] x86/mm: honor opt_pcid also for 32-bit PV domains Jan Beulich
2019-09-12 10:34   ` Roger Pau Monné [this message]
2019-09-12 10:45     ` Jan Beulich
2019-09-11 15:23 ` [Xen-devel] [PATCH 4/9] x86/HVM: move NOFLUSH handling out of hvm_set_cr3() Jan Beulich
2019-09-12 11:35   ` Roger Pau Monné
2019-09-12 11:52     ` Jan Beulich
2019-09-12 14:44       ` Roger Pau Monné
2019-09-12 14:47         ` Jan Beulich
2019-09-12 15:42           ` Roger Pau Monné
2019-09-12 15:52             ` Jan Beulich
2019-09-11 15:24 ` [Xen-devel] [PATCH 5/9] x86/HVM: refuse CR3 loads with reserved (upper) bits set Jan Beulich
2019-09-12 11:45   ` Roger Pau Monné
2019-09-12 12:01     ` Jan Beulich
2019-09-11 15:25 ` [Xen-devel] [PATCH 6/9] x86/HVM: relax shadow mode check in hvm_set_cr3() Jan Beulich
2019-09-12 14:50   ` Roger Pau Monné
2019-09-11 15:25 ` [Xen-devel] [PATCH 7/9] x86/HVM: cosmetics to hvm_set_cr3() Jan Beulich
2019-09-12 15:04   ` Roger Pau Monné
2019-09-11 15:26 ` [Xen-devel] [PATCH 8/9] x86/CPUID: drop INVPCID dependency on PCID Jan Beulich
2019-09-12 15:11   ` Roger Pau Monné
2019-09-11 15:26 ` [Xen-devel] [PATCH 9/9] x86: PCID is unused when !PV Jan Beulich
2019-09-12 15:31   ` Roger Pau Monné
2019-09-12 15:46     ` Jan Beulich
2019-09-12 15:48     ` Jan Beulich
2019-09-12 15:57       ` Roger Pau Monné
  -- strict thread matches above, loose matches on Subject: below --
2019-05-02 11:35 [PATCH 0/9] XSA-292 follow-up Jan Beulich
2019-05-02 12:19 ` [PATCH 3/9] x86/mm: honor opt_pcid also for 32-bit PV domains Jan Beulich
2019-05-02 12:19   ` [Xen-devel] " Jan Beulich

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=20190912103441.le4zets5ey73uxwl@Air-de-Roger \
    --to=roger.pau@citrix.com \
    --cc=George.Dunlap@eu.citrix.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=wl@xen.org \
    --cc=xen-devel@lists.xenproject.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).