Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: Matt Roper <matthew.d.roper@intel.com>
Cc: <intel-xe@lists.freedesktop.org>
Subject: Re: [PATCH v2 18/43] drm/xe/pci: Convert register access to use xe_mmio
Date: Tue, 10 Sep 2024 14:40:04 -0400	[thread overview]
Message-ID: <ZuCShHJy_rJpGfCH@intel.com> (raw)
In-Reply-To: <20240907000748.2614020-63-matthew.d.roper@intel.com>

On Fri, Sep 06, 2024 at 05:08:07PM -0700, Matt Roper wrote:
> Stop using GT pointers for register access.
> 
> Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_pci.c | 13 ++++++++-----
>  1 file changed, 8 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
> index aa0c29450b4a..e0188ef9995e 100644
> --- a/drivers/gpu/drm/xe/xe_pci.c
> +++ b/drivers/gpu/drm/xe/xe_pci.c
> @@ -473,13 +473,15 @@ enum xe_gmdid_type {
>  
>  static void read_gmdid(struct xe_device *xe, enum xe_gmdid_type type, u32 *ver, u32 *revid)
>  {
> -	struct xe_gt *gt = xe_root_mmio_gt(xe);
> +	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
>  	struct xe_reg gmdid_reg = GMD_ID;
>  	u32 val;
>  
>  	KUNIT_STATIC_STUB_REDIRECT(read_gmdid, xe, type, ver, revid);
>  
>  	if (IS_SRIOV_VF(xe)) {
> +		struct xe_gt *gt = xe_root_mmio_gt(xe);
> +
>  		/*
>  		 * To get the value of the GMDID register, VFs must obtain it
>  		 * from the GuC using MMIO communication.
> @@ -515,14 +517,15 @@ static void read_gmdid(struct xe_device *xe, enum xe_gmdid_type type, u32 *ver,
>  		gt->info.type = XE_GT_TYPE_UNINITIALIZED;
>  	} else {
>  		/*
> -		 * We need to apply the GSI offset explicitly here as at this
> -		 * point the xe_gt is not fully uninitialized and only basic
> -		 * access to MMIO registers is possible.
> +		 * We need to apply the GSI offset explicitly here since we're
> +		 * reading through the tile-level MMIO accessor (at this
> +		 * point the GT is not fully uninitialized and only basic
> +		 * access to MMIO registers is possible).

it is kind of strange to mention GT initialization now that gt struct is
not visible and that we are trying to detach mmio from gt.

But well, perhaps the strange part is this entire manual offset thing and
this change in the comment is needed now that xe_gt struct is not used
and this keep the meaning, so


Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

>  		 */
>  		if (type == GMDID_MEDIA)
>  			gmdid_reg.addr += MEDIA_GT_GSI_OFFSET;
>  
> -		val = xe_mmio_read32(gt, gmdid_reg);
> +		val = xe_mmio_read32(mmio, gmdid_reg);
>  	}
>  
>  	*ver = REG_FIELD_GET(GMD_ID_ARCH_MASK, val) * 100 + REG_FIELD_GET(GMD_ID_RELEASE_MASK, val);
> -- 
> 2.45.2
> 

  reply	other threads:[~2024-09-10 18:40 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-07  0:07 [PATCH v2 00/43] Stop using xe_gt as a register MMIO target Matt Roper
2024-09-07  0:07 ` [PATCH v2 01/43] drm/xe: Move forcewake to 'gt.pm' substructure Matt Roper
2024-09-07  0:07 ` [PATCH v2 02/43] drm/xe: Create dedicated xe_mmio structure Matt Roper
2024-09-07  0:07 ` [PATCH v2 03/43] drm/xe: Clarify size of MMIO region Matt Roper
2024-09-07  0:07 ` [PATCH v2 04/43] drm/xe: Move GSI offset adjustment fields into 'struct xe_mmio' Matt Roper
2024-09-10 18:02   ` Rodrigo Vivi
2024-09-12 13:30     ` Jani Nikula
2024-09-07  0:07 ` [PATCH v2 05/43] drm/xe: Populate GT's mmio iomap from tile during init Matt Roper
2024-09-07  0:07 ` [PATCH v2 06/43] drm/xe: Switch mmio_ext to use 'struct xe_mmio' Matt Roper
2024-09-07  0:07 ` [PATCH v2 07/43] drm/xe: Add xe_tile backpointer to xe_mmio Matt Roper
2024-09-07  0:07 ` [PATCH v2 08/43] drm/xe: Adjust mmio code to pass VF substructure to SRIOV code Matt Roper
2024-09-07  0:07 ` [PATCH v2 09/43] drm/xe: Switch MMIO interface to take xe_mmio instead of xe_gt Matt Roper
2024-09-07  0:07 ` [PATCH v2 10/43] drm/xe/irq: Convert register access to use xe_mmio Matt Roper
2024-09-07  0:08 ` [PATCH v2 11/43] drm/xe/pcode: " Matt Roper
2024-09-07  0:08 ` [PATCH v2 12/43] drm/xe/hwmon: " Matt Roper
2024-09-07  0:08 ` [PATCH v2 13/43] drm/xe/vram: " Matt Roper
2024-09-07  0:08 ` [PATCH v2 14/43] drm/xe/compat-i915: " Matt Roper
2024-09-10 18:15   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 15/43] drm/xe/lmtt: " Matt Roper
2024-09-07  0:08 ` [PATCH v2 16/43] drm/xe/stolen: " Matt Roper
2024-09-07  0:08 ` [PATCH v2 17/43] drm/xe/device: " Matt Roper
2024-09-10 18:05   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 18/43] drm/xe/pci: " Matt Roper
2024-09-10 18:40   ` Rodrigo Vivi [this message]
2024-09-07  0:08 ` [PATCH v2 19/43] drm/xe/wa: " Matt Roper
2024-09-10 18:07   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 20/43] drm/xe/uc: " Matt Roper
2024-09-10 18:42   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 21/43] drm/xe/guc: " Matt Roper
2024-09-10 18:48   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 22/43] drm/xe/huc: " Matt Roper
2024-09-10 18:44   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 23/43] drm/xe/gsc: " Matt Roper
2024-09-10 18:08   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 24/43] drm/xe/query: " Matt Roper
2024-09-10 18:44   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 25/43] drm/xe/mcr: " Matt Roper
2024-09-10 18:11   ` Rodrigo Vivi
2024-09-10 18:49     ` Matt Roper
2024-09-07  0:08 ` [PATCH v2 26/43] drm/xe/mocs: " Matt Roper
2024-09-10 18:41   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 27/43] drm/xe/hw_engine: " Matt Roper
2024-09-10 18:42   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 28/43] drm/xe/gt_throttle: " Matt Roper
2024-09-10 18:07   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 29/43] drm/xe/pat: " Matt Roper
2024-09-10 18:12   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 30/43] drm/xe/wopcm: " Matt Roper
2024-09-10 18:12   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 31/43] drm/xe/oa: " Matt Roper
2024-09-10 18:34   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 32/43] drm/xe/topology: " Matt Roper
2024-09-10 18:11   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 33/43] drm/xe/execlist: " Matt Roper
2024-09-10 18:13   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 34/43] drm/xe/gt_clock: " Matt Roper
2024-09-10 18:44   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 35/43] drm/xe/reg_sr: " Matt Roper
2024-09-10 18:15   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 36/43] drm/xe/gt: " Matt Roper
2024-09-10 18:11   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 37/43] drm/xe/sriov: " Matt Roper
2024-09-10 18:47   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 38/43] drm/xe/tlb: " Matt Roper
2024-09-10 18:45   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 39/43] drm/xe/gt_idle: " Matt Roper
2024-09-10 18:12   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 40/43] drm/xe/forcewake: " Matt Roper
2024-09-10 18:42   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 41/43] drm/xe/ggtt: " Matt Roper
2024-09-10 18:09   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 42/43] drm/xe/ccs_mode: " Matt Roper
2024-09-10 18:46   ` Rodrigo Vivi
2024-09-07  0:08 ` [PATCH v2 43/43] drm/xe/mmio: Drop compatibility macros Matt Roper
2024-09-07  3:10 ` ✓ CI.Patch_applied: success for Stop using xe_gt as a register MMIO target (rev2) Patchwork
2024-09-07  3:11 ` ✗ CI.checkpatch: warning " Patchwork
2024-09-07  3:12 ` ✓ CI.KUnit: success " Patchwork
2024-09-07  3:26 ` ✓ CI.Build: " Patchwork
2024-09-07  3:31 ` ✗ CI.Hooks: failure " Patchwork
2024-09-07  3:34 ` ✓ CI.checksparse: success " Patchwork
2024-09-07  4:22 ` ✗ CI.BAT: failure " Patchwork
2024-09-09 17:04   ` Matt Roper
2024-09-09 16:59 ` ✓ CI.FULL: success " 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=ZuCShHJy_rJpGfCH@intel.com \
    --to=rodrigo.vivi@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