All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jan Beulich <jbeulich@suse.com>
To: "Penny, Zheng" <penny.zheng@amd.com>
Cc: "Huang, Ray" <Ray.Huang@amd.com>,
	"oleksii.kurochko@gmail.com" <oleksii.kurochko@gmail.com>,
	"Andrew Cooper" <andrew.cooper3@citrix.com>,
	"Roger Pau Monné" <roger.pau@citrix.com>,
	"xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>
Subject: Re: [PATCH v3 27/28] xen/domctl: make HVM_PARAM_IDENT_PT conditional upon CONFIG_MGMT_HYPERCALLS
Date: Tue, 18 Nov 2025 08:12:08 +0100	[thread overview]
Message-ID: <0ff1d79b-a7c2-4fc0-a335-17cbdb977dfc@suse.com> (raw)
In-Reply-To: <DM4PR12MB8451BC18FAF3D9B97F84B604E1D6A@DM4PR12MB8451.namprd12.prod.outlook.com>

On 18.11.2025 07:45, Penny, Zheng wrote:
> [Public]
> 
>> -----Original Message-----
>> From: Jan Beulich <jbeulich@suse.com>
>> Sent: Thursday, October 30, 2025 9:35 PM
>> To: Penny, Zheng <penny.zheng@amd.com>
>> Cc: Huang, Ray <Ray.Huang@amd.com>; oleksii.kurochko@gmail.com; Andrew
>> Cooper <andrew.cooper3@citrix.com>; Roger Pau Monné <roger.pau@citrix.com>;
>> xen-devel@lists.xenproject.org
>> Subject: Re: [PATCH v3 27/28] xen/domctl: make HVM_PARAM_IDENT_PT
>> conditional upon CONFIG_MGMT_HYPERCALLS
>>
>> On 13.10.2025 12:15, Penny Zheng wrote:
>>> Helper domctl_lock_{acquire,release} is domctl_lock, which
>>> HVM_PARAM_IDENT_PT uses to ensure synchronization and hence being a
>> toolstack-only operation.
>>> So we shall make HVM_PARAM_IDENT_PT conditional upon
>>> CONFIG_MGMT_HYPERCALLS, returning -EOPNOTSUPP when
>> MGMT_HYPERCALLS=n.
>>>
>>> Suggested-by: Jan Beulich <jbeulich@suse.com>
>>
>> I fear this isn't quite what I suggested. The param get/set are XSM_TARGET, i.e.
>> can be used by DM as well. The particular one here shouldn't be used by a DM, but
>> that's a different question. Similarly in principle the PVH Dom0 building code should
>> be able to use this path; it doesn't right now in favor of some open- coding.
>>
>> What iirc I did suggest was that the serialization isn't needed when no domctl can
>> be used to otherwise alter (relevant) guest state.
> 
> Ah, true, serialization isn't needed when MGMT_HYPERCALLS=n, as no domctl-op could alter the guest state at the same time.
> Then maybe adding IS_ENABLED() checking is enough:

Yes, that or indeed introducing stubs. Which one is the lesser evil I'm having
a hard time determining. Hence I'd suggest that you go with the below, unless
someone else chimes in.

Jan

> ```
> diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c
> index 5a50721bd0..4e1b3ee5f4 100644
> --- a/xen/arch/x86/hvm/hvm.c
> +++ b/xen/arch/x86/hvm/hvm.c
> @@ -4324,7 +4324,7 @@ static int hvm_set_param(struct domain *d, uint32_t index, uint64_t value)
>           * the domctl_lock.
>           */
>          rc = -ERESTART;
> -        if ( !domctl_lock_acquire() )
> +        if ( IS_ENABLED(CONFIG_MGMT_HYPERCALLS) && !domctl_lock_acquire() )
>              break;
> 
>          rc = 0;
> @@ -4334,7 +4334,8 @@ static int hvm_set_param(struct domain *d, uint32_t index, uint64_t value)
>              paging_update_cr3(v, false);
>          domain_unpause(d);
> 
> -        domctl_lock_release();
> +        if ( IS_ENABLED(CONFIG_MGMT_HYPERCALLS) )
> +            domctl_lock_release();
>          break;
>      case HVM_PARAM_DM_DOMAIN:
>          /* The only value this should ever be set to is DOMID_SELF */
> ```
> 
>>
>> Jan



  reply	other threads:[~2025-11-18  7:12 UTC|newest]

Thread overview: 68+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-13 10:15 [PATCH v3 00/28] Disable domctl-op via CONFIG_MGMT_HYPERCALLS Penny Zheng
2025-10-13 10:15 ` [PATCH v3 01/28] xen/xsm: remove redundant xsm_iomem_mapping() Penny Zheng
2025-10-13 11:18   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 02/28] xen/mem_sharing: wrap hvm_copy_context_and_params() with CONFIG_MEM_SHARING Penny Zheng
2025-10-14 14:43   ` Grygorii Strashko
2025-10-28 17:01   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 03/28] xen/altp2m: move p2m_set_suppress_ve_multi() forward Penny Zheng
2025-10-28 17:03   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 04/28] xen/sched: remove vcpu_set_soft_affinity() Penny Zheng
2025-10-13 10:24   ` Jürgen Groß
2025-10-13 10:15 ` [PATCH v3 05/28] xen/sysctl: replace CONFIG_SYSCTL with CONFIG_MGMT_DOMCTL Penny Zheng
2025-10-13 10:41   ` Jürgen Groß
2025-10-29 14:30   ` Jan Beulich
2025-10-29 21:26   ` Stewart Hildebrand
2025-11-19  6:33     ` Penny, Zheng
2025-10-13 10:15 ` [PATCH v3 06/28] xen/x86: move domctl.o out of PV_SHIM_EXCLUSIVE Penny Zheng
2025-10-29 14:33   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 07/28] xen/domctl: make MGMT_HYPERCALLS transiently def_bool Penny Zheng
2025-10-29 14:37   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 09/28] xen/vm_event: consolidate CONFIG_VM_EVENT Penny Zheng
2025-10-21 13:24   ` Grygorii Strashko
2025-10-29 15:57   ` Jan Beulich
2025-11-11  7:08     ` Penny, Zheng
2025-11-11  8:13       ` Jan Beulich
2025-11-11  9:46         ` Penny, Zheng
2025-10-13 10:15 ` [PATCH v3 10/28] xen/vm_event: make VM_EVENT depend on CONFIG_MGMT_HYPERCALLS Penny Zheng
2025-10-13 10:15 ` [PATCH v3 11/28] xen/xsm: wrap xsm_vm_event_control() with CONFIG_VM_EVENT Penny Zheng
2025-10-13 10:15 ` [PATCH v3 12/28] xen/domctl: wrap domain_pause_by_systemcontroller() with MGMT_HYPERCALLS Penny Zheng
2025-10-30 11:28   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 13/28] xen/domctl: wrap domain_soft_reset() with CONFIG_MGMT_HYPERCALLS Penny Zheng
2025-10-30 12:14   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 14/28] xen/domctl: wrap domain_resume() " Penny Zheng
2025-10-13 10:15 ` [PATCH v3 15/28] xen/domctl: wrap domain_kill() " Penny Zheng
2025-10-30 12:43   ` Jan Beulich
2025-11-12  8:58     ` Penny, Zheng
2025-11-12 10:02       ` Jan Beulich
2025-11-13  4:11         ` Penny, Zheng
2025-11-13  4:40     ` Penny, Zheng
2025-10-13 10:15 ` [PATCH v3 16/28] xen/domctl: wrap domain_set_node_affinity() " Penny Zheng
2025-10-13 10:15 ` [PATCH v3 17/28] xen/domctl: wrap vcpu_affinity_domctl() " Penny Zheng
2025-10-13 10:44   ` Jürgen Groß
2025-10-13 10:15 ` [PATCH v3 18/28] xen/domctl: wrap sched_adjust() " Penny Zheng
2025-10-13 11:03   ` Jürgen Groß
2025-10-13 11:13     ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 19/28] xen/domctl: wrap xsm_irq_permission " Penny Zheng
2025-10-13 10:15 ` [PATCH v3 20/28] xen/domctl: wrap arch-specific domain_set_time_offset() " Penny Zheng
2025-10-13 10:15 ` [PATCH v3 21/28] xen/domctl: wrap xsm_set_target() " Penny Zheng
2025-10-13 10:15 ` [PATCH v3 22/28] xen/domctl: wrap iommu-related domctl op " Penny Zheng
2025-10-30 13:09   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 23/28] xen/domctl: wrap arch_{get,set}_paging_mempool_size() " Penny Zheng
2025-10-13 10:15 ` [PATCH v3 24/28] xen/domctl: make CONFIG_X86_PSR depend on CONFIG_MGMT_HYPERCALLS Penny Zheng
2025-10-13 10:15 ` [PATCH v3 25/28] xen/domctl: avoid unreachable codes when both MGMT_HYPERCALLS and MEM_SHARING unset Penny Zheng
2025-10-30 13:13   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 26/28] xen/domctl: wrap arch-specific domctl-op with CONFIG_MGMT_HYPERCALLS Penny Zheng
2025-10-30 13:24   ` Jan Beulich
2025-10-13 10:15 ` [PATCH v3 27/28] xen/domctl: make HVM_PARAM_IDENT_PT conditional upon CONFIG_MGMT_HYPERCALLS Penny Zheng
2025-10-30 13:34   ` Jan Beulich
2025-11-18  6:45     ` Penny, Zheng
2025-11-18  7:12       ` Jan Beulich [this message]
2025-10-13 10:15 ` [PATCH v3 28/28] xen/domctl: wrap common/domctl.c with CONFIG_MGMT_HYPERCALLS Penny Zheng
2025-10-30 13:40   ` Jan Beulich
2025-11-18  6:43     ` Penny, Zheng
2025-11-18  7:14       ` Jan Beulich
2025-11-18  7:51         ` Penny, Zheng
2025-11-18 19:29           ` Jason Andryuk
2025-11-20  4:09             ` Penny, Zheng
     [not found] ` <20251013101540.3502842-9-Penny.Zheng@amd.com>
2025-10-29 15:02   ` [PATCH v3 08/28] xen/vm_event: introduce vm_event_is_enabled() Jan Beulich
2025-10-30 11:10     ` Grygorii Strashko

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=0ff1d79b-a7c2-4fc0-a335-17cbdb977dfc@suse.com \
    --to=jbeulich@suse.com \
    --cc=Ray.Huang@amd.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=oleksii.kurochko@gmail.com \
    --cc=penny.zheng@amd.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.