From: "Pottumuttu, Sai Teja" <sai.teja.pottumuttu@intel.com>
To: Matt Roper <matthew.d.roper@intel.com>, <intel-xe@lists.freedesktop.org>
Cc: John Harrison <John.C.Harrison@Intel.com>,
<sai.teja.pottumuttu@intel.com>
Subject: Re: [PATCH] drm/xe: Defer gt->mmio initialization until after multi-tile setup
Date: Wed, 18 Sep 2024 22:54:07 +0530 [thread overview]
Message-ID: <d1a3f23c-a0c8-4f45-a1f4-c7443314b8ba@intel.com> (raw)
In-Reply-To: <20240917221615.875962-2-matthew.d.roper@intel.com>
On 18-09-2024 03:46, Matt Roper wrote:
> With the recent xe_mmio redesign, tiles and GTs each have their own MMIO
> accessor, with the GT inheriting some of the information (such as the
> iomap pointer) from their containing tile. Given that non-root tiles
> get initialized later than the root tile (and currently after the point
> at which GT MMIO is initialized for _all_ GTs), we wind up incorrectly
> inheriting uninitialized pointers for the initialization of GT MMIO for
> GTs that reside on non-root tiles. This causes a driver crash on
> multi-tile PVC platforms.
>
> With the general xe_mmio redesign, it's now only necessary to do the
> GT-level MMIO setup before the point we start reading/writing GT
> registers. Move initialization of gt->mmio out of xe_info_init (which
> runs before non-root tiles are initialized) and to the beginning of
> where we start actually accessing the GTs themselves.
>
> The high-level initialization flow now boils down to:
> - General device init, software-only setup
> - (no register access possible yet)
> - Root tile initialization
> - (access to device/tile0 registers possible via xe_root_tile_mmio())
> - Initialization of non-root tiles
> - (access to any tile's registers possible via tile->mmio)
> - GT MMIO initialization, inheriting iomap from each GT's tile
> - (access to any GT's registers possible via gt->mmio)
>
> Fixes: fa599b8c95a7 ("drm/xe: Populate GT's mmio iomap from tile during init")
> Reported-by: John Harrison <John.C.Harrison@Intel.com>
> Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Looks good to me,
Reviewed-by: Sai Teja Pottumuttu <sai.teja.pottumuttu@intel.com>
> ---
> drivers/gpu/drm/xe/xe_device.c | 7 +++++++
> drivers/gpu/drm/xe/xe_gt.c | 24 ++++++++++++++++++++++++
> drivers/gpu/drm/xe/xe_gt.h | 1 +
> drivers/gpu/drm/xe/xe_pci.c | 12 ------------
> 4 files changed, 32 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
> index 4d3c794f134c..998165962cad 100644
> --- a/drivers/gpu/drm/xe/xe_device.c
> +++ b/drivers/gpu/drm/xe/xe_device.c
> @@ -656,6 +656,13 @@ int xe_device_probe(struct xe_device *xe)
> err = xe_gt_init_early(gt);
> if (err)
> return err;
> +
> + /*
> + * Only after this point can GT-specific MMIO operations
> + * (including things like communication with the GuC)
> + * be performed.
> + */
> + xe_gt_mmio_init(gt);
> }
>
> for_each_tile(tile, xe, id) {
> diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
> index b32fb3fdf80d..9b0218109647 100644
> --- a/drivers/gpu/drm/xe/xe_gt.c
> +++ b/drivers/gpu/drm/xe/xe_gt.c
> @@ -624,6 +624,30 @@ int xe_gt_init(struct xe_gt *gt)
> return 0;
> }
>
> +/**
> + * xe_gt_mmio_init() - Initialize GT's MMIO access
> + * @gt: the GT object
> + *
> + * Initialize GT's MMIO accessor, which will be used to access registers inside
> + * this GT.
> + */
> +void xe_gt_mmio_init(struct xe_gt *gt)
> +{
> + struct xe_tile *tile = gt_to_tile(gt);
> +
> + gt->mmio.regs = tile->mmio.regs;
> + gt->mmio.regs_size = tile->mmio.regs_size;
> + gt->mmio.tile = tile;
> +
> + if (gt->info.type == XE_GT_TYPE_MEDIA) {
> + gt->mmio.adj_offset = MEDIA_GT_GSI_OFFSET;
> + gt->mmio.adj_limit = MEDIA_GT_GSI_LENGTH;
> + }
> +
> + if (IS_SRIOV_VF(gt_to_xe(gt)))
> + gt->mmio.sriov_vf_gt = gt;
> +}
> +
> void xe_gt_record_user_engines(struct xe_gt *gt)
> {
> struct xe_hw_engine *hwe;
> diff --git a/drivers/gpu/drm/xe/xe_gt.h b/drivers/gpu/drm/xe/xe_gt.h
> index 97def44afa4c..05c8351d0f39 100644
> --- a/drivers/gpu/drm/xe/xe_gt.h
> +++ b/drivers/gpu/drm/xe/xe_gt.h
> @@ -37,6 +37,7 @@ struct xe_gt *xe_gt_alloc(struct xe_tile *tile);
> int xe_gt_init_hwconfig(struct xe_gt *gt);
> int xe_gt_init_early(struct xe_gt *gt);
> int xe_gt_init(struct xe_gt *gt);
> +void xe_gt_mmio_init(struct xe_gt *gt);
> void xe_gt_declare_wedged(struct xe_gt *gt);
> int xe_gt_record_default_lrcs(struct xe_gt *gt);
>
> diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
> index 5ba4ec229494..2afa05799b8d 100644
> --- a/drivers/gpu/drm/xe/xe_pci.c
> +++ b/drivers/gpu/drm/xe/xe_pci.c
> @@ -721,11 +721,6 @@ static int xe_info_init(struct xe_device *xe,
> gt->info.type = XE_GT_TYPE_MAIN;
> gt->info.has_indirect_ring_state = graphics_desc->has_indirect_ring_state;
> gt->info.engine_mask = graphics_desc->hw_engine_mask;
> - gt->mmio.regs = tile->mmio.regs;
> - gt->mmio.regs_size = tile->mmio.regs_size;
> - gt->mmio.tile = tile;
> - if (IS_SRIOV_VF(xe))
> - gt->mmio.sriov_vf_gt = gt;
>
> if (MEDIA_VER(xe) < 13 && media_desc)
> gt->info.engine_mask |= media_desc->hw_engine_mask;
> @@ -745,13 +740,6 @@ static int xe_info_init(struct xe_device *xe,
> gt->info.type = XE_GT_TYPE_MEDIA;
> gt->info.has_indirect_ring_state = media_desc->has_indirect_ring_state;
> gt->info.engine_mask = media_desc->hw_engine_mask;
> - gt->mmio.regs = tile->mmio.regs;
> - gt->mmio.regs_size = tile->mmio.regs_size;
> - gt->mmio.adj_offset = MEDIA_GT_GSI_OFFSET;
> - gt->mmio.adj_limit = MEDIA_GT_GSI_LENGTH;
> - gt->mmio.tile = tile;
> - if (IS_SRIOV_VF(xe))
> - gt->mmio.sriov_vf_gt = gt;
>
> /*
> * FIXME: At the moment multi-tile and standalone media are
prev parent reply other threads:[~2024-09-18 17:24 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-17 22:16 [PATCH] drm/xe: Defer gt->mmio initialization until after multi-tile setup Matt Roper
2024-09-17 22:21 ` ✓ CI.Patch_applied: success for " Patchwork
2024-09-17 22:22 ` ✗ CI.checkpatch: warning " Patchwork
2024-09-17 22:23 ` ✓ CI.KUnit: success " Patchwork
2024-09-17 22:34 ` ✓ CI.Build: " Patchwork
2024-09-17 22:37 ` ✓ CI.Hooks: " Patchwork
2024-09-17 22:38 ` ✓ CI.checksparse: " Patchwork
2024-09-17 22:56 ` ✓ CI.BAT: " Patchwork
2024-09-18 0:04 ` ✗ CI.FULL: failure " Patchwork
2024-09-18 19:58 ` Matt Roper
2024-09-18 17:24 ` Pottumuttu, Sai Teja [this message]
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=d1a3f23c-a0c8-4f45-a1f4-c7443314b8ba@intel.com \
--to=sai.teja.pottumuttu@intel.com \
--cc=John.C.Harrison@Intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.d.roper@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