From: Matthew Auld <matthew.auld@intel.com>
To: Nitin Gote <nitin.r.gote@intel.com>, intel-xe@lists.freedesktop.org
Cc: matthew.d.roper@intel.com
Subject: Re: [PATCH] drm/xe: derive mem copy capability from graphics version
Date: Mon, 19 Jan 2026 13:17:27 +0000 [thread overview]
Message-ID: <ccc20c84-9c83-4068-82be-dc693b397cc5@intel.com> (raw)
In-Reply-To: <20260119121819.1676700-2-nitin.r.gote@intel.com>
On 19/01/2026 12:18, Nitin Gote wrote:
> Drop .has_mem_copy_instr from the platform descriptors and compute
> it in xe_info_init() once handle_gmdid() populates graphics_verx100.
> Early init was running with graphics_verx100==0, so the old placement
> silently disabled MEM_COPY on every GMDID platform. Centralizing the
> GRAPHICS_VER(xe) >= 20 || platform == XE_PVC check keeps Xe2+/PVC enabled
> and removes redundant per-platform plumbing for future Xe platforms.
PVC had a slightly different instruction layout for this IIRC, which is
why I kept it to xe2+. I doubt it is worth the effort, but if you want
to add support for PVC I think best keep as a separate patch (need to
also change the instruction emit part).
>
> Bspec: 57561
>
> Fixes: 1e12dbae9d72 ("drm/xe/migrate: support MEM_COPY instruction")
> Cc: Matt Roper <matthew.d.roper@intel.com>
> Suggested-by: Matthew Auld <matthew.auld@intel.com>
> Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> ---
> drivers/gpu/drm/xe/xe_pci.c | 7 ++-----
> drivers/gpu/drm/xe/xe_pci_types.h | 1 -
> 2 files changed, 2 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
> index 34df063024fe..b41e3afc0431 100644
> --- a/drivers/gpu/drm/xe/xe_pci.c
> +++ b/drivers/gpu/drm/xe/xe_pci.c
> @@ -350,7 +350,6 @@ static const struct xe_device_desc lnl_desc = {
> .has_display = true,
> .has_flat_ccs = 1,
> .has_pxp = true,
> - .has_mem_copy_instr = true,
> .max_gt_per_tile = 2,
> .needs_scratch = true,
> .va_bits = 48,
> @@ -375,7 +374,6 @@ static const struct xe_device_desc bmg_desc = {
> .has_pre_prod_wa = 1,
> .has_soc_remapper_telem = true,
> .has_sriov = true,
> - .has_mem_copy_instr = true,
> .max_gt_per_tile = 2,
> .needs_scratch = true,
> .subplatforms = (const struct xe_subplatform_desc[]) {
> @@ -392,7 +390,6 @@ static const struct xe_device_desc ptl_desc = {
> .has_display = true,
> .has_flat_ccs = 1,
> .has_sriov = true,
> - .has_mem_copy_instr = true,
> .has_pre_prod_wa = 1,
> .has_pxp = true,
> .max_gt_per_tile = 2,
> @@ -407,7 +404,6 @@ static const struct xe_device_desc nvls_desc = {
> .dma_mask_size = 46,
> .has_display = true,
> .has_flat_ccs = 1,
> - .has_mem_copy_instr = true,
> .has_pre_prod_wa = 1,
> .max_gt_per_tile = 2,
> .require_force_probe = true,
> @@ -706,7 +702,6 @@ static int xe_info_init_early(struct xe_device *xe,
> xe->info.has_soc_remapper_telem = desc->has_soc_remapper_telem;
> xe->info.has_sriov = xe_configfs_primary_gt_allowed(to_pci_dev(xe->drm.dev)) &&
> desc->has_sriov;
> - xe->info.has_mem_copy_instr = desc->has_mem_copy_instr;
> xe->info.skip_guc_pc = desc->skip_guc_pc;
> xe->info.skip_mtcfg = desc->skip_mtcfg;
> xe->info.skip_pcode = desc->skip_pcode;
> @@ -897,6 +892,8 @@ static int xe_info_init(struct xe_device *xe,
> xe->info.has_ctx_tlb_inval = graphics_desc->has_ctx_tlb_inval;
> xe->info.has_usm = graphics_desc->has_usm;
> xe->info.has_64bit_timestamp = graphics_desc->has_64bit_timestamp;
> + xe->info.has_mem_copy_instr = GRAPHICS_VER(xe) >= 20 ||
> + xe->info.platform == XE_PVC;
>
> xe_info_probe_tile_count(xe);
>
> diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h
> index 7ccb0ab7a53b..8b2ff3f25607 100644
> --- a/drivers/gpu/drm/xe/xe_pci_types.h
> +++ b/drivers/gpu/drm/xe/xe_pci_types.h
> @@ -49,7 +49,6 @@ struct xe_device_desc {
> u8 has_llc:1;
> u8 has_mbx_power_limits:1;
> u8 has_mbx_thermal_info:1;
> - u8 has_mem_copy_instr:1;
> u8 has_mert:1;
> u8 has_pre_prod_wa:1;
> u8 has_page_reclaim_hw_assist:1;
next prev parent reply other threads:[~2026-01-19 13:17 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-19 12:18 [PATCH] drm/xe: derive mem copy capability from graphics version Nitin Gote
2026-01-19 13:17 ` Matthew Auld [this message]
2026-01-19 14:24 ` Gote, Nitin R
2026-01-19 13:37 ` ✗ CI.KUnit: failure for " Patchwork
-- strict thread matches above, loose matches on Subject: below --
2026-01-19 15:03 [PATCH] " Nitin Gote
2026-01-19 15:19 ` Matthew Auld
2026-01-20 5:47 Nitin Gote
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=ccc20c84-9c83-4068-82be-dc693b397cc5@intel.com \
--to=matthew.auld@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.d.roper@intel.com \
--cc=nitin.r.gote@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