Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>
To: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>,
	 intel-xe@lists.freedesktop.org
Subject: Re: [Intel-xe] [RFC v2 2/6] drm/xe/xe2: Determine bios enablement for flat ccs on igfx.
Date: Fri, 24 Nov 2023 12:05:43 +0100	[thread overview]
Message-ID: <0876b09852ce6391b04be329692a29b161ec5154.camel@linux.intel.com> (raw)
In-Reply-To: <20231121100906.3587649-3-himal.prasad.ghimiray@intel.com>

On Tue, 2023-11-21 at 15:39 +0530, Himal Prasad Ghimiray wrote:
> If bios disables flat ccs on igfx  make has_flat_ccs as 0 and notify
double space above                 ^^
> 
> via drm_info.
> 
> Bspec:59255
> 
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> Signed-off-by: Himal Prasad Ghimiray
> <himal.prasad.ghimiray@intel.com>
> ---
>  drivers/gpu/drm/xe/regs/xe_gt_regs.h |  3 +++
>  drivers/gpu/drm/xe/xe_device.c       | 30
> ++++++++++++++++++++++++++++
>  2 files changed, 33 insertions(+)
> 
> diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> index cc27fe8fc363..b642301947f5 100644
> --- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> +++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> @@ -142,6 +142,9 @@
>  #define XEHP_SLICE_COMMON_ECO_CHICKEN1         XE_REG_MCR(0x731c,
> XE_REG_OPTION_MASKED)
>  #define   MSC_MSAA_REODER_BUF_BYPASS_DISABLE   REG_BIT(14)
>  
> +#define XE2_FLAT_CCS_BASE_RANGE_LOWER          XE_REG_MCR(0x8800)
> +#define   XE2_FLAT_CCS_ENABLE                  REG_BIT(0)
> +
>  #define VF_PREEMPTION                          XE_REG(0x83a4,
> XE_REG_OPTION_MASKED)
>  #define   PREEMPTION_VERTEX_COUNT              REG_GENMASK(15, 0)
>  
> diff --git a/drivers/gpu/drm/xe/xe_device.c
> b/drivers/gpu/drm/xe/xe_device.c
> index 8be765adf702..07a3e4cf48d1 100644
> --- a/drivers/gpu/drm/xe/xe_device.c
> +++ b/drivers/gpu/drm/xe/xe_device.c
> @@ -16,6 +16,7 @@
>  #include <drm/xe_drm.h>
>  
>  #include "regs/xe_regs.h"
> +#include "regs/xe_gt_regs.h"
>  #include "xe_bo.h"
>  #include "xe_debugfs.h"
>  #include "xe_display.h"
> @@ -25,6 +26,7 @@
>  #include "xe_exec_queue.h"
>  #include "xe_exec.h"
>  #include "xe_gt.h"
> +#include "xe_gt_mcr.h"
>  #include "xe_irq.h"
>  #include "xe_mmio.h"
>  #include "xe_module.h"
> @@ -342,6 +344,29 @@ static void xe_device_sanitize(struct drm_device
> *drm, void *arg)
>                 xe_gt_sanitize(gt);
>  }
>  
> +static int xe_device_set_has_flat_ccs(struct  xe_device *xe)
> +{
> +       u32 reg;
> +       int err;
> +
> +       if (IS_DGFX(xe) || GRAPHICS_VER(xe) < 20 || !xe-
> >info.has_flat_ccs)
> +               return 0;

Depending on the outcome of the iommu discussion we might need to check
for iommu present here as well?

> +
> +       struct xe_gt *gt = xe_root_mmio_gt(xe);
> +
> +       err = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
> +       if (err)
> +               return err;
> +
> +       reg = xe_gt_mcr_unicast_read_any(gt,
> XE2_FLAT_CCS_BASE_RANGE_LOWER);
> +       xe->info.has_flat_ccs = (reg & XE2_FLAT_CCS_ENABLE);
> +
> +       if (!xe->info.has_flat_ccs)
> +               drm_info(&xe->drm,
> +                        "Flat CCS has been disabled in bios, May
> lead to performance impact");
> +       return 0;
> +}
> +
>  int xe_device_probe(struct xe_device *xe)
>  {
>         struct xe_tile *tile;
> @@ -352,6 +377,7 @@ int xe_device_probe(struct xe_device *xe)
>         xe_pat_init_early(xe);
>  
>         xe->info.mem_region_mask = 1;
> +

Unrelated change.

>         err = xe_display_init_nommio(xe);
>         if (err)
>                 return err;
> @@ -390,6 +416,10 @@ int xe_device_probe(struct xe_device *xe)
>                         goto err_irq_shutdown;
>         }
>  
> +       err = xe_device_set_has_flat_ccs(xe);
> +       if (err)
> +               return err;
> +
>         err = xe_mmio_probe_vram(xe);
>         if (err)
>                 goto err_irq_shutdown;


  parent reply	other threads:[~2023-11-24 11:06 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-21 10:09 [Intel-xe] [RFC v2 0/6] Enable compression handling on LNL Himal Prasad Ghimiray
2023-11-21 10:09 ` [Intel-xe] [RFC v2 1/6] drm/xe/xe2: Support flat ccs Himal Prasad Ghimiray
2023-11-23 16:28   ` Matthew Auld
2023-11-27  3:04     ` Ghimiray, Himal Prasad
2023-11-24 11:02   ` Thomas Hellström
2023-11-27  3:05     ` Ghimiray, Himal Prasad
2023-11-21 10:09 ` [Intel-xe] [RFC v2 2/6] drm/xe/xe2: Determine bios enablement for flat ccs on igfx Himal Prasad Ghimiray
2023-11-23 16:37   ` Matthew Auld
2023-11-23 17:01     ` Matthew Auld
2023-11-27  3:11       ` Ghimiray, Himal Prasad
2023-11-24 11:05   ` Thomas Hellström [this message]
2023-11-27  3:12     ` Ghimiray, Himal Prasad
2023-11-21 10:09 ` [Intel-xe] [RFC v2 3/6] drm/xe/xe2: Allocate extra pages for ccs during bo create Himal Prasad Ghimiray
2023-11-24 11:09   ` Thomas Hellström
2023-11-27  3:19     ` Ghimiray, Himal Prasad
2023-11-27  9:41       ` Thomas Hellström
2023-11-29  4:53         ` Ghimiray, Himal Prasad
2023-11-21 10:09 ` [Intel-xe] [RFC v2 4/6] drm/xe/xe2: Updates on XY_CTRL_SURF_COPY_BLT Himal Prasad Ghimiray
2023-11-24 11:11   ` Thomas Hellström
2023-11-27  3:21     ` Ghimiray, Himal Prasad
2023-11-21 10:09 ` [Intel-xe] [RFC v2 5/6] drm/xe/xe_migrate.c: Use NULL 1G PTE mapped at 255GiB VA for ccs clear Himal Prasad Ghimiray
2023-11-24 11:19   ` Thomas Hellström
2023-11-27  3:23     ` Ghimiray, Himal Prasad
2023-11-21 10:09 ` [Intel-xe] [RFC v2 6/6] drm/xe/xe2: Handle flat ccs move for igfx Himal Prasad Ghimiray
2023-11-24 15:48   ` Thomas Hellström
2023-11-27  3:25     ` Ghimiray, Himal Prasad
2023-11-28 13:36 ` [Intel-xe] [RFC v2 0/6] Enable compression handling on LNL Thomas Hellström
2023-11-29  4:49   ` Ghimiray, Himal Prasad

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=0876b09852ce6391b04be329692a29b161ec5154.camel@linux.intel.com \
    --to=thomas.hellstrom@linux.intel.com \
    --cc=himal.prasad.ghimiray@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    /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