All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Jan Beulich <jbeulich@suse.com>
Cc: "Roger Pau Monné" <roger.pau@citrix.com>,
	"Alejandro Vallejo" <alejandro.vallejo@cloud.com>,
	Xen-devel <xen-devel@lists.xenproject.org>
Subject: Re: [PATCH] x86/cpu-policy: Extend the guest max policy max leaf/subleaves
Date: Wed, 30 Oct 2024 10:30:40 +0000	[thread overview]
Message-ID: <eae9a653-eb8f-45fd-bce2-495c50f0abde@citrix.com> (raw)
In-Reply-To: <14173e8a-ec20-455a-87bf-ae6d3d9a9051@suse.com>

On 30/10/2024 6:51 am, Jan Beulich wrote:
> On 29.10.2024 18:55, Andrew Cooper wrote:
>> We already have one migration case opencoded (feat.max_subleaf).  A more
>> recent discovery is that we advertise x2APIC to guests without ensuring that
>> we provide max_leaf >= 0xb.
>>
>> In general, any leaf known to Xen can be safely configured by the toolstack if
>> it doesn't violate other constraints.
>>
>> Therefore, introduce guest_common_{max,default}_leaves() to generalise the
>> special case we currently have for feat.max_subleaf, in preparation to be able
>> to provide x2APIC topology in leaf 0xb even on older hardware.
>>
>> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
> Reviewed-by: Jan Beulich <jbeulich@suse.com>
>
> I'll have to update the AMX logic accordingly (maybe also the AVX10 one).

Yeah - I need to get back to your shrinking series too.

> I'd like to point out that this highlights a naming anomaly in
> x86_cpu_policies_are_compatible(): update_domain_cpu_policy() passes in
> the respective max policy as first argument. Imo the first parameter of
> the function would better be named "max" there.

That's covered in the documentation.  It made sense when I first planned
things, but that was many many iterations ago.

>
>> --- a/xen/arch/x86/cpu-policy.c
>> +++ b/xen/arch/x86/cpu-policy.c
>> @@ -391,6 +391,27 @@ static void __init calculate_host_policy(void)
>>      p->platform_info.cpuid_faulting = cpu_has_cpuid_faulting;
>>  }
>>  
>> +/*
>> + * Guest max policies can have any max leaf/subleaf within bounds.
>> + *
>> + * - Some incoming VMs have a larger-than-necessary feat max_subleaf.
>> + * - Some VMs we'd like to synthesise leaves not present on the host.
>> + */
>> +static void __init guest_common_max_leaves(struct cpu_policy *p)
>> +{
>> +    p->basic.max_leaf       = ARRAY_SIZE(p->basic.raw) - 1;
>> +    p->feat.max_subleaf     = ARRAY_SIZE(p->feat.raw) - 1;
>> +    p->extd.max_leaf        = 0x80000000U + ARRAY_SIZE(p->extd.raw) - 1;
>> +}
>> +
>> +/* Guest default policies inherit the host max leaf/subleaf settings. */
>> +static void __init guest_common_default_leaves(struct cpu_policy *p)
>> +{
>> +    p->basic.max_leaf       = host_cpu_policy.basic.max_leaf;
>> +    p->feat.max_subleaf     = host_cpu_policy.feat.max_subleaf;
>> +    p->extd.max_leaf        = host_cpu_policy.extd.max_leaf;
>> +}
> Which sadly still leaves open how to suitably shrink the max values,
> when they're larger than necessary (for the guest).

Only the toolstack can do the shrinking, and only as the about the final
step after optional features have been activated.

~Andrew


  reply	other threads:[~2024-10-30 10:31 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-29 17:55 [PATCH] x86/cpu-policy: Extend the guest max policy max leaf/subleaves Andrew Cooper
2024-10-29 18:26 ` Alejandro Vallejo
2024-10-30  6:51 ` Jan Beulich
2024-10-30 10:30   ` Andrew Cooper [this message]
2024-10-30  8:59 ` Roger Pau Monné
2024-10-30 10:39   ` Andrew Cooper
2024-10-30 11:03     ` Roger Pau Monné
2024-10-30 14:45       ` Andrew Cooper
2024-10-30 15:13         ` Roger Pau Monné
2024-10-30 15:59           ` Alejandro Vallejo
2024-10-30 16:30             ` Roger Pau Monné
2024-10-30 16:51           ` Andrew Cooper
2024-10-30 17:10             ` Roger Pau Monné
2024-10-30 17:24               ` Andrew Cooper
2024-10-31  9:20               ` Jan Beulich
2024-10-31  9:16             ` 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=eae9a653-eb8f-45fd-bce2-495c50f0abde@citrix.com \
    --to=andrew.cooper3@citrix.com \
    --cc=alejandro.vallejo@cloud.com \
    --cc=jbeulich@suse.com \
    --cc=roger.pau@citrix.com \
    --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 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.