From: sashiko-bot@kernel.org
To: "Stuart Summers" <stuart.summers@intel.com>
Cc: intel-xe@lists.freedesktop.org
Subject: Re: [PATCH 11/16] drm/xe: Run page table updates on the media GT when there is no primary
Date: Fri, 20 Nov 2026 06:10:03 +0000 [thread overview]
Message-ID: <sashiko-outbox-224538@kernel.org> (raw)
In-Reply-To: <20261120054712.377573-29-stuart.summers@intel.com>
Thank you for your contribution! Sashiko AI review found 3 potential
issues to consider:
- [Medium] drm/xe: xe_gt_flr_prepare() and xe_gt_reinit() skip
media-only tile migration
- [Medium] drm/xe: NULL dereference of usm.bb_pool in
xe_migrate_prepare_vm()
- [Medium] drm/xe: silent no-op blit copy and clear on media-only tiles
in xe_migrate
--
commit e4d0026f61d092e34250b892e47f82036445a7f5
Author: Stuart Summers <stuart.summers@intel.com>
drm/xe: Run page table updates on the media GT when there is no primary
This commit introduces xe_tile_pt_gt() and xe_tile_pt_class() so that tile
migration and bind exec queues fall back to the media GT and a video
decode engine when a tile has no primary GT.
> diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
> index fe71cb011d3a0..327e4985d8586 100644
> --- a/drivers/gpu/drm/xe/xe_gt.c
> +++ b/drivers/gpu/drm/xe/xe_gt.c
[ ... ]
> @@ -657,10 +658,12 @@ static int gt_init_with_all_forcewake(struct xe_gt *gt)
> }
> }
>
> - if (xe_gt_is_main_type(gt)) {
> - struct xe_tile *tile = gt_to_tile(gt);
> -
> - err = xe_migrate_init(tile->migrate);
> + /*
> + * Normally the migration context runs on the primary GT, but a tile
> + * with no graphics IP has only its media GT to run it on.
> + */
> + if (gt == xe_tile_pt_gt(gt_to_tile(gt))) {
> + err = xe_migrate_init(gt_to_tile(gt)->migrate);
> if (err)
> return err;
> }
[Severity: Medium]
Should xe_gt_flr_prepare() and xe_gt_reinit() also check
gt == xe_tile_pt_gt(gt_to_tile(gt)) instead of xe_gt_is_main_type(gt)?
On a tile with no primary GT, xe_migrate_init() now sets up tile->migrate
on the media GT, where xe_gt_is_main_type(gt) is false. Across a PCIe
FLR, xe_uc_flr_prepare() tears down the media GT's GuC queues (including
tile->migrate->q), while xe_gt_flr_prepare() and xe_gt_reinit() skip
xe_migrate_flr_prepare() and xe_migrate_reinit(), leaving the migration VM
page tables and queue uninitialized after resume.
> diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
> index d7a8d8ecffe63..8fc3e5d500a2b 100644
> --- a/drivers/gpu/drm/xe/xe_migrate.c
> +++ b/drivers/gpu/drm/xe/xe_migrate.c
[ ... ]
> @@ -485,7 +486,7 @@ static void xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
> if (xe->info.has_usm) {
> xe_tile_assert(tile, xe_bo_size(batch) == SZ_1M);
>
> - batch = tile->primary_gt->usm.bb_pool->bo;
> + batch = xe_tile_pt_gt(tile)->usm.bb_pool->bo;
> m->usm_batch_base_ofs = m->batch_base_ofs + SZ_1M;
> xe_tile_assert(tile, xe_bo_size(batch) == SZ_512K);
>
> @@ -506,7 +507,7 @@ static void xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
> m->batch_base_ofs = xe_migrate_vram_ofs(xe, batch_addr, false);
>
> if (xe->info.has_usm) {
> - batch = tile->primary_gt->usm.bb_pool->bo;
> + batch = xe_tile_pt_gt(tile)->usm.bb_pool->bo;
> batch_addr = xe_bo_addr(batch, 0, XE_PAGE_SIZE);
> m->usm_batch_base_ofs = xe_migrate_vram_ofs(xe, batch_addr, false);
> }
[Severity: Medium]
Can xe_tile_pt_gt(tile)->usm.bb_pool be NULL here when tile->primary_gt is
NULL and xe->info.has_usm is set (for example, with gt_types_allowed=media
in configfs on an Xe2+ platform)?
gt_init_with_all_forcewake() only allocates gt->usm.bb_pool when
xe_gt_is_main_type(gt) is true, so tile->media_gt->usm.bb_pool is NULL and
xe_migrate_prepare_vm() dereferences a NULL pointer. Even if bb_pool were
allocated on xe_tile_pt_gt(tile), media GTs have no copy engines, so
gt->usm.paging_hwe0 is also NULL and the xe->info.has_usm paths in
xe_migrate_init() and xe_exec_queue_create_bind() will fail with -EINVAL
instead of falling back to xe_tile_pt_class(tile).
[ ... ]
> @@ -1276,7 +1277,7 @@ static struct dma_fence *__xe_migrate_copy(struct xe_migrate *m,
> bool copy_only_ccs,
> bool is_vram_resolve)
> {
> - struct xe_gt *gt = m->tile->primary_gt;
> + struct xe_gt *gt = xe_tile_pt_gt(m->tile);
> struct xe_device *xe = gt_to_xe(gt);
> struct dma_fence *fence = NULL;
> u64 size = xe_bo_size(src_bo);
[ ... ]
> @@ -1992,7 +1993,7 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m,
> bool clear_vram = is_devmem(bo, dst);
> bool clear_bo_data = XE_MIGRATE_CLEAR_FLAG_BO_DATA & clear_flags;
> bool clear_ccs = XE_MIGRATE_CLEAR_FLAG_CCS_DATA & clear_flags;
> - struct xe_gt *gt = m->tile->primary_gt;
> + struct xe_gt *gt = xe_tile_pt_gt(m->tile);
> struct xe_device *xe = gt_to_xe(gt);
> bool clear_only_system_ccs = false;
> struct dma_fence *fence = NULL;
> @@ -2357,7 +2358,7 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m,
> struct dma_fence *deps,
> const enum xe_migrate_copy_dir dir)
> {
> - struct xe_gt *gt = m->tile->primary_gt;
> + struct xe_gt *gt = xe_tile_pt_gt(m->tile);
> struct xe_device *xe = gt_to_xe(gt);
> bool use_usm_batch = xe->info.has_usm;
> struct dma_fence *fence = NULL;
[Severity: Medium]
When m->q is created on XE_ENGINE_CLASS_VIDEO_DECODE, do these blit copy
and clear paths silently succeed without copying or clearing memory?
A migration job puts the PTE updates in job->ptrs[0].batch_addr and the
blitter commands in job->ptrs[1].batch_addr, but the video decode ring
ops only emit job->ptrs[0]:
drivers/gpu/drm/xe/xe_ring_ops.c:emit_job_gen12_video() {
for (i = 0; i < job->q->width; ++i)
__emit_job_gen12_video(job, job->q->lrc[i],
job->ptrs[i].batch_addr,
&job->ptrs[i].head,
xe_sched_job_lrc_seqno(job));
}
With q->width == 1, the engine executes only the PTE batch and signals the
job fence without running the copy or clear batch. Later in the series,
commit "drm/xe: Fall back to CPU copies when the tile cannot blit" adds
xe_migrate_can_blit() checks to TTM moves, but other callers remain
unguarded: xe_ttm_access_memory() still calls xe_migrate_access_memory()
-> xe_migrate_vram() when len >= SZ_16K (silently dropping ptrace or
/proc/<pid>/mem reads and writes on VRAM BOs), and xe_vram_memtest() calls
xe_migrate_clear() and fails probe with -EIO when CONFIG_DRM_XE_DEBUG_MEM
is enabled.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261120054712.377573-18-stuart.summers@intel.com?part=11
next prev parent reply other threads:[~2026-11-20 6:10 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-20 5:47 [PATCH 00/16] Work around media 52-bit limitation on CRI Stuart Summers
2026-11-20 5:47 ` [PATCH 01/16] drm/xe/sysctrl: Use preallocated mailbox staging buffers Stuart Summers
2026-11-20 5:52 ` sashiko-bot
2026-11-20 5:47 ` [PATCH 02/16] drm/xe: Process device OOB workarounds during early info init Stuart Summers
2026-11-20 5:47 ` [PATCH 03/16] drm/xe/sysctrl: Restore the SoC remapper on every runtime resume Stuart Summers
2026-11-20 5:47 ` [PATCH 04/16] drm/xe: Add enable_media module parameter Stuart Summers
2026-11-20 5:47 ` [PATCH 05/16] drm/xe: Detect each GMD_ID IP in a loop Stuart Summers
2026-11-20 5:47 ` [PATCH 06/16] drm/xe: Tolerate a missing graphics IP in xe_info_init() Stuart Summers
2026-11-20 5:47 ` [PATCH 07/16] drm/xe: Verify IP presence against GU_CNTL Stuart Summers
2026-11-20 5:47 ` [PATCH 08/16] drm/xe: Guard per-tile primary GT accesses Stuart Summers
2026-11-20 5:47 ` [PATCH 09/16] drm/xe: Stop assuming the root tile has a primary GT Stuart Summers
2026-11-20 5:47 ` [PATCH 10/16] drm/xe: Do not claim flat CCS without a graphics IP Stuart Summers
2026-11-20 5:47 ` [PATCH 11/16] drm/xe: Run page table updates on the media GT when there is no primary Stuart Summers
2026-11-20 6:10 ` sashiko-bot [this message]
2026-11-20 5:47 ` [PATCH 12/16] drm/xe: Document that gt_types_allowed=media now probes Stuart Summers
2026-11-20 5:47 ` [PATCH 13/16] drm/xe: Fall back to CPU copies when the tile cannot blit Stuart Summers
2026-11-20 6:09 ` sashiko-bot
2026-11-20 5:47 ` [PATCH 14/16] drm/xe: Require a fully visible VRAM BAR without a primary GT Stuart Summers
2026-11-20 5:47 ` [PATCH 15/16] drm/xe: Allow probe " Stuart Summers
2026-11-20 5:47 ` [PATCH 16/16] drm/xe: Work around media 52-bit limitation on CRI Stuart Summers
2026-11-20 6:11 ` sashiko-bot
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=sashiko-outbox-224538@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=sashiko-reviews@lists.linux.dev \
--cc=stuart.summers@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