From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
To: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: intel-xe@lists.freedesktop.org, Lukasz Laguna <lukasz.laguna@intel.com>
Subject: Re: [PATCH 07/12] drm/xe: Add device flag to indicate standalone MERT
Date: Tue, 28 Oct 2025 14:53:01 -0700 [thread overview]
Message-ID: <87zf9an16q.wl-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <20251021-cri-v1-7-bf11e61d9f49@intel.com>
On Tue, 21 Oct 2025 22:17:39 -0700, Lucas De Marchi wrote:
>
> From: Lukasz Laguna <lukasz.laguna@intel.com>
>
> The MERT subsystem manages memory accesses between host and device. On
> the Crescent Island platform, it requires direct management by the
> driver.
>
> Introduce a device flag and corresponding helpers to identify platforms
> with standalone MERT, enabling proper initialization and handling.
Looks like all future dGfx products may or may not have this, so we need a
has_mert flag, such as introduced here. So this LGTM:
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
>
> Signed-off-by: Lukasz Laguna <lukasz.laguna@intel.com>
> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
> ---
> drivers/gpu/drm/xe/xe_device.h | 5 +++++
> drivers/gpu/drm/xe/xe_device_types.h | 2 ++
> drivers/gpu/drm/xe/xe_pci.c | 2 ++
> drivers/gpu/drm/xe/xe_pci_types.h | 1 +
> 4 files changed, 10 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h
> index 32cc6323b7f64..6604b89330d51 100644
> --- a/drivers/gpu/drm/xe/xe_device.h
> +++ b/drivers/gpu/drm/xe/xe_device.h
> @@ -172,6 +172,11 @@ static inline bool xe_device_has_lmtt(struct xe_device *xe)
> return IS_DGFX(xe);
> }
>
> +static inline bool xe_device_has_mert(struct xe_device *xe)
> +{
> + return xe->info.has_mert;
> +}
> +
> u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size);
>
> void xe_device_snapshot_print(struct xe_device *xe, struct drm_printer *p);
> diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
> index f1e3b16897d1b..fb401809fae5a 100644
> --- a/drivers/gpu/drm/xe/xe_device_types.h
> +++ b/drivers/gpu/drm/xe/xe_device_types.h
> @@ -305,6 +305,8 @@ struct xe_device {
> * pcode mailbox commands.
> */
> u8 has_mbx_power_limits:1;
> + /** @info.has_mert: Device has standalone MERT */
> + u8 has_mert:1;
> /** @info.has_pxp: Device has PXP support */
> u8 has_pxp:1;
> /** @info.has_range_tlb_inval: Has range based TLB invalidations */
> diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
> index 000bb30b97712..ef27d17d904b8 100644
> --- a/drivers/gpu/drm/xe/xe_pci.c
> +++ b/drivers/gpu/drm/xe/xe_pci.c
> @@ -403,6 +403,7 @@ static const struct xe_device_desc cri_desc = {
> .has_display = false,
> .has_flat_ccs = false,
> .has_mbx_power_limits = true,
> + .has_mert = true,
> .has_sriov = true,
> .max_gt_per_tile = 2,
> .require_force_probe = true,
> @@ -669,6 +670,7 @@ static int xe_info_init_early(struct xe_device *xe,
> xe->info.has_heci_cscfi = desc->has_heci_cscfi;
> xe->info.has_late_bind = desc->has_late_bind;
> xe->info.has_llc = desc->has_llc;
> + xe->info.has_mert = desc->has_mert;
> xe->info.has_pxp = desc->has_pxp;
> xe->info.has_sriov = xe_configfs_primary_gt_allowed(to_pci_dev(xe->drm.dev)) &&
> desc->has_sriov;
> diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h
> index a4451bdc79fb3..3e3bcdd1d1cf1 100644
> --- a/drivers/gpu/drm/xe/xe_pci_types.h
> +++ b/drivers/gpu/drm/xe/xe_pci_types.h
> @@ -46,6 +46,7 @@ struct xe_device_desc {
> u8 has_late_bind:1;
> u8 has_llc:1;
> u8 has_mbx_power_limits:1;
> + u8 has_mert:1;
> u8 has_pxp:1;
> u8 has_sriov:1;
> u8 needs_scratch:1;
>
> --
> 2.51.0
>
next prev parent reply other threads:[~2025-10-28 21:53 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-22 5:17 [PATCH 00/12] drm/xe: Add support for Crescent Island Lucas De Marchi
2025-10-22 5:17 ` [PATCH 01/12] drm/xe/cri: Add CRI platform definition Lucas De Marchi
2025-10-22 14:58 ` Shekhar Chauhan
2025-10-22 19:45 ` Lucas De Marchi
2025-10-23 10:57 ` Lucas De Marchi
2025-10-23 11:13 ` Lucas De Marchi
2025-10-22 5:17 ` [PATCH 02/12] topic/for-xe-CI: drm/xe/cri: Define GuC firmware for CRI Lucas De Marchi
2025-10-23 10:33 ` Lucas De Marchi
2025-10-22 5:17 ` [PATCH 03/12] drm/xe/cri: Setup MOCS table Lucas De Marchi
2025-10-22 8:06 ` Vivekanandan, Balasubramani
2025-10-22 5:17 ` [PATCH 04/12] drm/xe/cri: Add new performance limit reasons bits Lucas De Marchi
2025-10-22 6:31 ` [PATCH 4/12] " Raag Jadav
2025-10-22 21:22 ` Lucas De Marchi
2025-10-22 5:17 ` [PATCH 05/12] drm/xe/cri: Add check to verify if CSC is a PCIe endpoint Lucas De Marchi
2025-10-22 5:17 ` [PATCH 06/12] drm/xe/pm: Enable D3cold WAKE# support Lucas De Marchi
2025-10-22 6:35 ` [PATCH 6/12] " Raag Jadav
2025-10-22 19:53 ` Lucas De Marchi
2025-10-22 5:17 ` [PATCH 07/12] drm/xe: Add device flag to indicate standalone MERT Lucas De Marchi
2025-10-28 21:53 ` Dixit, Ashutosh [this message]
2025-10-22 5:17 ` [PATCH 08/12] drm/xe/oa/uapi: Expose MERT OA unit Lucas De Marchi
2025-10-22 23:09 ` Umesh Nerlige Ramappa
2025-11-24 21:34 ` Dixit, Ashutosh
2025-10-22 5:17 ` [PATCH 09/12] drm/xe/pf: Configure LMTT in MERT Lucas De Marchi
2025-10-22 5:17 ` [PATCH 10/12] drm/xe: Handle MERT interrupts Lucas De Marchi
2025-10-22 23:19 ` Matt Roper
2025-10-23 14:42 ` Lucas De Marchi
2025-10-28 9:30 ` Laguna, Lukasz
2025-10-22 5:17 ` [PATCH 11/12] drm/xe/pf: Add TLB invalidation support for MERT Lucas De Marchi
2025-10-22 18:28 ` Matthew Brost
2025-10-23 15:11 ` Lucas De Marchi
2025-10-28 9:33 ` Laguna, Lukasz
2025-10-22 5:17 ` [PATCH 12/12] drm/xe/pf: Handle MERT catastrophic errors Lucas De Marchi
2025-10-22 5:40 ` ✗ CI.checkpatch: warning for drm/xe: Add support for Crescent Island Patchwork
2025-10-22 5:41 ` ✓ CI.KUnit: success " Patchwork
2025-10-22 6:29 ` ✓ Xe.CI.BAT: " Patchwork
2025-10-22 7:56 ` ✗ 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=87zf9an16q.wl-ashutosh.dixit@intel.com \
--to=ashutosh.dixit@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=lucas.demarchi@intel.com \
--cc=lukasz.laguna@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