Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Matt Roper <matthew.d.roper@intel.com>
To: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Cc: <intel-xe@lists.freedesktop.org>,
	Lucas De Marchi <lucas.demarchi@intel.com>,
	Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>,
	Akshata Jahagirdar <akshata.jahagirdar@intel.com>,
	Matthew Auld <matthew.auld@intel.com>
Subject: Re: [PATCH v3 06/11] drm/xe/xe2hpg: Determine flat ccs offset for vram
Date: Mon, 8 Apr 2024 14:44:14 -0700	[thread overview]
Message-ID: <20240408214414.GC958162@mdroper-desk1.amr.corp.intel.com> (raw)
In-Reply-To: <20240408170545.3769566-7-balasubramani.vivekanandan@intel.com>

On Mon, Apr 08, 2024 at 10:35:40PM +0530, Balasubramani Vivekanandan wrote:
> From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> 
> on Xe2 dgfx platform determine the offset using Flat CCS size
> bitfield of XE2_FLAT_CCS_BASE_RANGE_[UPPER/LOWER] mcr registers.
> 
> Bspec: 68023

We generally treat the bspec references as trailers, so this should be
moved down with the other trailers below, not separated by the
changelog.

Aside from that,

        Reviewed-by: Matt Roper <matthew.d.roper@intel.com>

> 
> v2: function argument tile_size changed from pass by reference to pass
> by value
> 
> Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> Signed-off-by: Akshata Jahagirdar <akshata.jahagirdar@intel.com>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
> ---
>  drivers/gpu/drm/xe/regs/xe_gt_regs.h |  5 ++++
>  drivers/gpu/drm/xe/xe_mmio.c         | 39 ++++++++++++++++++++++++++--
>  2 files changed, 42 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> index 6617c86a096b..d404f211bc36 100644
> --- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> +++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> @@ -69,6 +69,7 @@
>  
>  #define XEHP_TILE_ADDR_RANGE(_idx)		XE_REG_MCR(0x4900 + (_idx) * 4)
>  #define XEHP_FLAT_CCS_BASE_ADDR			XE_REG_MCR(0x4910)
> +#define XEHP_FLAT_CCS_PTR			REG_GENMASK(31, 8)
>  
>  #define WM_CHICKEN3				XE_REG_MCR(0x5588, XE_REG_OPTION_MASKED)
>  #define   HIZ_PLANE_COMPRESSION_DIS		REG_BIT(10)
> @@ -142,6 +143,10 @@
>  
>  #define XE2_FLAT_CCS_BASE_RANGE_LOWER		XE_REG_MCR(0x8800)
>  #define   XE2_FLAT_CCS_ENABLE			REG_BIT(0)
> +#define XE2_FLAT_CCS_BASE_LOWER_ADDR_MASK	REG_GENMASK(31, 6)
> +
> +#define XE2_FLAT_CCS_BASE_RANGE_UPPER		XE_REG_MCR(0x8804)
> +#define XE2_FLAT_CCS_BASE_UPPER_ADDR_MASK	REG_GENMASK(7, 0)
>  
>  #define GSCPSMI_BASE				XE_REG(0x880c)
>  
> diff --git a/drivers/gpu/drm/xe/xe_mmio.c b/drivers/gpu/drm/xe/xe_mmio.c
> index 5d13fc7cb9d2..d66da1a9f165 100644
> --- a/drivers/gpu/drm/xe/xe_mmio.c
> +++ b/drivers/gpu/drm/xe/xe_mmio.c
> @@ -163,6 +163,42 @@ static int xe_determine_lmem_bar_size(struct xe_device *xe)
>  	return 0;
>  }
>  
> +static inline u64 get_flat_ccs_offset(struct xe_gt *gt, u64 tile_size)
> +{
> +	struct xe_device *xe = gt_to_xe(gt);
> +	u64 offset;
> +	u32 reg;
> +
> +	if (GRAPHICS_VER(xe) >= 20) {
> +		u64 ccs_size = tile_size / 512;
> +		u64 offset_hi, offset_lo;
> +		u32 nodes, num_enabled;
> +
> +		reg = xe_mmio_read32(gt, MIRROR_FUSE3);
> +		nodes = REG_FIELD_GET(XE2_NODE_ENABLE_MASK, reg);
> +		num_enabled = hweight32(nodes); /* Number of enabled l3 nodes */
> +
> +		reg = xe_gt_mcr_unicast_read_any(gt, XE2_FLAT_CCS_BASE_RANGE_LOWER);
> +		offset_lo = REG_FIELD_GET(XE2_FLAT_CCS_BASE_LOWER_ADDR_MASK, reg);
> +
> +		reg = xe_gt_mcr_unicast_read_any(gt, XE2_FLAT_CCS_BASE_RANGE_UPPER);
> +		offset_hi = REG_FIELD_GET(XE2_FLAT_CCS_BASE_UPPER_ADDR_MASK, reg);
> +
> +		offset = offset_hi << 32; /* HW view bits 39:32 */
> +		offset |= offset_lo << 6; /* HW view bits 31:6 */
> +		offset *= num_enabled; /* convert to SW view */
> +
> +		/* We don't expect any holes */
> +		xe_assert_msg(xe, offset == (xe_mmio_read64_2x32(gt, GSMBASE) - ccs_size),
> +			      "Hole between CCS and GSM.\n");
> +	} else {
> +		reg = xe_gt_mcr_unicast_read_any(gt, XEHP_FLAT_CCS_BASE_ADDR);
> +		offset = (u64)REG_FIELD_GET(XEHP_FLAT_CCS_PTR, reg) * SZ_64K;
> +	}
> +
> +	return offset;
> +}
> +
>  /**
>   * xe_mmio_tile_vram_size() - Collect vram size and offset information
>   * @tile: tile to get info for
> @@ -207,8 +243,7 @@ static int xe_mmio_tile_vram_size(struct xe_tile *tile, u64 *vram_size,
>  
>  	/* minus device usage */
>  	if (xe->info.has_flat_ccs) {
> -		reg = xe_gt_mcr_unicast_read_any(gt, XEHP_FLAT_CCS_BASE_ADDR);
> -		offset = (u64)REG_FIELD_GET(GENMASK(31, 8), reg) * SZ_64K;
> +		offset = get_flat_ccs_offset(gt, *tile_size);
>  	} else {
>  		offset = xe_mmio_read64_2x32(gt, GSMBASE);
>  	}
> -- 
> 2.25.1
> 

-- 
Matt Roper
Graphics Software Engineer
Linux GPU Platform Enablement
Intel Corporation

  reply	other threads:[~2024-04-08 21:44 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-08 17:05 [PATCH v3 00/11] Add Battlemage support Balasubramani Vivekanandan
2024-04-08 17:05 ` [PATCH v3 01/11] drm/xe/xe2: Recognize Xe2_HPG IP Balasubramani Vivekanandan
2024-04-08 17:05 ` [PATCH v3 02/11] drm/xe/xe2: Recognize Xe2_HPM IP Balasubramani Vivekanandan
2024-04-08 17:05 ` [PATCH v3 03/11] drm/xe/bmg: Add BMG platform definition Balasubramani Vivekanandan
2024-04-08 19:09   ` Ghimiray, Himal Prasad
2024-04-09 15:46   ` Lucas De Marchi
2024-04-08 17:05 ` [PATCH v3 04/11] drm/xe/bmg: Add BMG mocs table Balasubramani Vivekanandan
2024-04-08 17:05 ` [PATCH v3 05/11] drm/xe/bmg: Program an additional discrete-specific PAT setting Balasubramani Vivekanandan
2024-04-08 19:05   ` Ghimiray, Himal Prasad
2024-04-08 17:05 ` [PATCH v3 06/11] drm/xe/xe2hpg: Determine flat ccs offset for vram Balasubramani Vivekanandan
2024-04-08 21:44   ` Matt Roper [this message]
2024-04-08 17:05 ` [PATCH v3 07/11] drm/xe/xe2hpg: Remove extra allocation of CCS pages for dgfx Balasubramani Vivekanandan
2024-04-08 18:57   ` Ghimiray, Himal Prasad
2024-04-08 21:46   ` Matt Roper
2024-04-22 10:15   ` Thomas Hellström
2024-04-08 17:05 ` [PATCH v3 08/11] drm/xe/xe2: Add workaround 18034896535 Balasubramani Vivekanandan
2024-04-26 21:02   ` Lucas De Marchi
2024-04-29 13:51     ` Gustavo Sousa
2024-04-08 17:05 ` [PATCH v3 09/11] drm/xe/xe2hpg: Add initial GT workarounds Balasubramani Vivekanandan
2024-04-08 17:05 ` [PATCH v3 10/11] drm/xe/xe2hpg: Introduce performance tuning changes for Xe2_HPG Balasubramani Vivekanandan
2024-04-08 17:05 ` [PATCH v3 11/11] drm/xe/xe2hpm: Add initial set of workarounds Balasubramani Vivekanandan
2024-04-08 18:33 ` ✓ CI.Patch_applied: success for Add Battlemage support (rev3) Patchwork
2024-04-08 18:33 ` ✗ CI.checkpatch: warning " Patchwork
2024-04-08 18:34 ` ✓ CI.KUnit: success " Patchwork
2024-04-08 18:46 ` ✓ CI.Build: " Patchwork
2024-04-08 18:48 ` ✓ CI.Hooks: " Patchwork
2024-04-08 18:50 ` ✓ CI.checksparse: " Patchwork
2024-04-08 19:12 ` ✓ CI.BAT: " Patchwork
2024-04-09  0:34 ` ✓ CI.FULL: " Patchwork
2024-04-09 21:25 ` [PATCH v3 00/11] Add Battlemage support Matt Roper

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=20240408214414.GC958162@mdroper-desk1.amr.corp.intel.com \
    --to=matthew.d.roper@intel.com \
    --cc=akshata.jahagirdar@intel.com \
    --cc=balasubramani.vivekanandan@intel.com \
    --cc=himal.prasad.ghimiray@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=lucas.demarchi@intel.com \
    --cc=matthew.auld@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