Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
To: Tomasz Lis <tomasz.lis@intel.com>, <intel-xe@lists.freedesktop.org>
Cc: "Michał Winiarski" <michal.winiarski@intel.com>,
	"Piotr Piórkowski" <piotr.piorkowski@intel.com>,
	"Matthew Brost" <matthew.brost@intel.com>,
	"Satyanarayana K V P" <satyanarayana.k.v.p@intel.com>
Subject: Re: [PATCH v3 5/5] drm/xe/vf: Do not disable VF migration on ATS-M
Date: Fri, 17 Oct 2025 00:25:04 +0200	[thread overview]
Message-ID: <0ed38cf2-4b3e-41f2-a1b7-a16533e3896a@intel.com> (raw)
In-Reply-To: <20251016120511.856792-6-tomasz.lis@intel.com>



On 10/16/2025 2:05 PM, Tomasz Lis wrote:
> The ATS-M does support VF migration, and it has graphics ver below
> the currently allowed range. Experimental support of this platform
> should therefore be available.
> 
> This change allows experimental VF migration support on ATS-M. It
> is also explicitly not allowing ADL, by adding condition on MEMIRQ.
> Supporting VF migration through MMIO interrupts would require
> some special handling in order to achieve reliability.
> 
> v2: Add MEMIRQ condition
> 
> Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_sriov_vf.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> index 5fb042c05112..fb2b3d41c610 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> @@ -159,10 +159,13 @@ static void vf_migration_init_early(struct xe_device *xe)
>  		return vf_disable_migration(xe,
>  					    "experimental feature not available on production builds");
>  
> -	if (GRAPHICS_VER(xe) < 20)
> -		return vf_disable_migration(xe, "requires gfx version >= 20, but only %u found",
> +	if (GRAPHICS_VER(xe) < 12)
> +		return vf_disable_migration(xe, "requires gfx version >= 12, but only %u found",
>  					    GRAPHICS_VER(xe));

maybe just drop it completely and only rely on memirq support ?

it's available from 12.50, see

static inline bool xe_device_has_memirq(struct xe_device *xe)
{
	return GRAPHICS_VERx100(xe) >= 1250;
}

>  > +	if (!xe_device_has_memirq(xe))
> +		return vf_disable_migration(xe, "requires memory-based IRQ support");
> +
>  	xe->sriov.vf.migration.enabled = true;
>  	xe_sriov_dbg(xe, "migration support enabled\n");
>  }


  reply	other threads:[~2025-10-16 22:25 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-16 12:05 [PATCH v3 0/5] drm/xe/vf: Minor fixes to post-migration recovery Tomasz Lis
2025-10-16 12:05 ` [PATCH v3 1/5] drm/xe/vf: Helper for telling whether CCS migration BBs are needed Tomasz Lis
2025-10-16 19:57   ` Michal Wajdeczko
2025-10-16 21:51     ` Lis, Tomasz
2025-10-16 12:05 ` [PATCH v3 2/5] drm/xe/vf: Fix GuC FW check for VF migration support Tomasz Lis
2025-10-16 21:55   ` Michal Wajdeczko
2025-10-17 15:31     ` Lis, Tomasz
2025-10-17 15:40       ` Michal Wajdeczko
2025-10-16 12:05 ` [PATCH v3 3/5] drm/xe/vf: Skip fixups on VF migration before getting GGTT info Tomasz Lis
2025-10-16 22:20   ` Michal Wajdeczko
2025-10-20 19:30     ` Lis, Tomasz
2025-10-16 12:05 ` [PATCH v3 4/5] drm/xe: Assert that VF will never use fixed placement of BOs Tomasz Lis
2025-10-16 12:05 ` [PATCH v3 5/5] drm/xe/vf: Do not disable VF migration on ATS-M Tomasz Lis
2025-10-16 22:25   ` Michal Wajdeczko [this message]
2025-10-20 19:22     ` Lis, Tomasz
2025-10-16 12:11 ` ✓ CI.KUnit: success for drm/xe/vf: Minor fixes to post-migration recovery (rev3) Patchwork
2025-10-16 12:57 ` ✓ Xe.CI.BAT: " Patchwork
2025-10-17  8:23 ` ✗ Xe.CI.Full: failure " Patchwork

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=0ed38cf2-4b3e-41f2-a1b7-a16533e3896a@intel.com \
    --to=michal.wajdeczko@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.brost@intel.com \
    --cc=michal.winiarski@intel.com \
    --cc=piotr.piorkowski@intel.com \
    --cc=satyanarayana.k.v.p@intel.com \
    --cc=tomasz.lis@intel.com \
    /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