Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Matthew Brost <matthew.brost@intel.com>
To: Varun Gupta <varun.gupta@intel.com>
Cc: <intel-xe@lists.freedesktop.org>, <matthew.d.roper@intel.com>,
	<himal.prasad.ghimiray@intel.com>, <priyanka.dandamudi@intel.com>,
	<stuart.summers@intel.com>
Subject: Re: [PATCH v5 2/2] drm/xe: Add prefetch fault support for Xe3p
Date: Sun, 22 Feb 2026 20:34:45 -0800	[thread overview]
Message-ID: <aZvY5dty7Xk2B/mV@lstrano-desk.jf.intel.com> (raw)
In-Reply-To: <20260223040729.1157739-3-varun.gupta@intel.com>

On Mon, Feb 23, 2026 at 09:37:29AM +0530, Varun Gupta wrote:
> Xe3p hardware prefetches memory ranges and notifies software via an
> additional bit (bit 11) in the page fault descriptor that the fault
> was caused by prefetch.
> 
> Extract the prefetch bit from the fault descriptor and echo it in the
> response (bit 6) only when the page fault handling fails. This allows
> the HW to suppress CAT errors for unsuccessful prefetch faults.
> 
> For prefetch faults that fail, increment stats counter without verbose
> logging to avoid spamming the log. The prefetch flag is packed into
> BIT(7) of the access_type field to avoid growing the consumer struct.
> 
> Based on original patches by Brian Welty <brian.welty@intel.com> and
> Priyanka Dandamudi <priyanka.dandamudi@intel.com>.
> 
> Bspec: 59311
> Originally-by: Lucas De Marchi <lucas.demarchi@intel.com>
> Cc: Matthew Brost <matthew.brost@intel.com>

Reviewed-by: Matthew Brost <matthew.brost@intel.com>

> Cc: Priyanka Dandamudi <priyanka.dandamudi@intel.com>
> Cc: Matt Roper <matthew.d.roper@intel.com>
> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
> Signed-off-by: Varun Gupta <varun.gupta@intel.com>
> 
> ---
> v5:
>  - Read prefetch from producer msg in guc_ack_fault, not consumer field (Matt Brost)
>  - Pack prefetch flag into BIT(7) of access_type instead of separate u8 field (Matt Brost)
> 
> v4:
>  - Downgrade prefetch error log to xe_gt_dbg to avoid spam (Matt Brost/ Stuart)
> 
> v3:
>  - Drop the rename patch, keep xe_pagefault_print() unchanged (Matt Brost)
>  - Move prefetch check to caller instead of inside print function (Matt Brost)
>  - Remove XE3P_ prefix from prefetch bit defines and add platform comment (Matt Brost)
>  - Show prefetch bit in error messages for debugging (Matt Brost)
>  - Split stats counter into separate patch (Matt Brost)
> 
> v2:
>  - Changed comment wording from "repairs" to "handling" for clarity (Matt Roper)
> ---
>  drivers/gpu/drm/xe/xe_guc_fwif.h        |  5 +++--
>  drivers/gpu/drm/xe/xe_guc_pagefault.c   |  6 +++++-
>  drivers/gpu/drm/xe/xe_pagefault.c       | 16 +++++++++++-----
>  drivers/gpu/drm/xe/xe_pagefault_types.h |  6 ++++--
>  4 files changed, 23 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
> index a33ea288b907..bb8f71d38611 100644
> --- a/drivers/gpu/drm/xe/xe_guc_fwif.h
> +++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
> @@ -261,7 +261,8 @@ struct xe_guc_pagefault_desc {
>  #define PFD_ACCESS_TYPE		GENMASK(1, 0)
>  #define PFD_FAULT_TYPE		GENMASK(3, 2)
>  #define PFD_VFID		GENMASK(9, 4)
> -#define PFD_RSVD_1		GENMASK(11, 10)
> +#define PFD_RSVD_1		BIT(10)
> +#define PFD_PREFETCH		BIT(11) /* Only valid on Xe3+, reserved on prior platforms */
>  #define PFD_VIRTUAL_ADDR_LO	GENMASK(31, 12)
>  #define PFD_VIRTUAL_ADDR_LO_SHIFT 12
>  
> @@ -281,7 +282,7 @@ struct xe_guc_pagefault_reply {
>  
>  	u32 dw1;
>  #define PFR_VFID		GENMASK(5, 0)
> -#define PFR_RSVD_1		BIT(6)
> +#define PFR_PREFETCH		BIT(6)  /* Only valid on Xe3+, reserved on prior platforms */
>  #define PFR_ENG_INSTANCE	GENMASK(12, 7)
>  #define PFR_ENG_CLASS		GENMASK(15, 13)
>  #define PFR_PDATA		GENMASK(31, 16)
> diff --git a/drivers/gpu/drm/xe/xe_guc_pagefault.c b/drivers/gpu/drm/xe/xe_guc_pagefault.c
> index d48f6ed103bb..607e32392f46 100644
> --- a/drivers/gpu/drm/xe/xe_guc_pagefault.c
> +++ b/drivers/gpu/drm/xe/xe_guc_pagefault.c
> @@ -8,10 +8,12 @@
>  #include "xe_guc_ct.h"
>  #include "xe_guc_pagefault.h"
>  #include "xe_pagefault.h"
> +#include "xe_pagefault_types.h"
>  
>  static void guc_ack_fault(struct xe_pagefault *pf, int err)
>  {
>  	u32 vfid = FIELD_GET(PFD_VFID, pf->producer.msg[2]);
> +	u32 prefetch = FIELD_GET(PFD_PREFETCH, pf->producer.msg[2]);
>  	u32 engine_instance = FIELD_GET(PFD_ENG_INSTANCE, pf->producer.msg[0]);
>  	u32 engine_class = FIELD_GET(PFD_ENG_CLASS, pf->producer.msg[0]);
>  	u32 pdata = FIELD_GET(PFD_PDATA_LO, pf->producer.msg[0]) |
> @@ -28,6 +30,7 @@ static void guc_ack_fault(struct xe_pagefault *pf, int err)
>  		FIELD_PREP(PFR_ASID, asid),
>  
>  		FIELD_PREP(PFR_VFID, vfid) |
> +		FIELD_PREP(PFR_PREFETCH, err ? prefetch : 0) |
>  		FIELD_PREP(PFR_ENG_INSTANCE, engine_instance) |
>  		FIELD_PREP(PFR_ENG_CLASS, engine_class) |
>  		FIELD_PREP(PFR_PDATA, pdata),
> @@ -76,7 +79,8 @@ int xe_guc_pagefault_handler(struct xe_guc *guc, u32 *msg, u32 len)
>  		(FIELD_GET(PFD_VIRTUAL_ADDR_LO, msg[2]) <<
>  		 PFD_VIRTUAL_ADDR_LO_SHIFT);
>  	pf.consumer.asid = FIELD_GET(PFD_ASID, msg[1]);
> -	pf.consumer.access_type = FIELD_GET(PFD_ACCESS_TYPE, msg[2]);
> +	pf.consumer.access_type = FIELD_GET(PFD_ACCESS_TYPE, msg[2]) |
> +		(FIELD_GET(PFD_PREFETCH, msg[2]) ? XE_PAGEFAULT_ACCESS_PREFETCH : 0);
>  	if (FIELD_GET(XE2_PFD_TRVA_FAULT, msg[0]))
>  		pf.consumer.fault_type_level = XE_PAGEFAULT_TYPE_LEVEL_NACK;
>  	else
> diff --git a/drivers/gpu/drm/xe/xe_pagefault.c b/drivers/gpu/drm/xe/xe_pagefault.c
> index 72f589fd2b64..a9a20c4a6b2b 100644
> --- a/drivers/gpu/drm/xe/xe_pagefault.c
> +++ b/drivers/gpu/drm/xe/xe_pagefault.c
> @@ -136,7 +136,7 @@ static int xe_pagefault_handle_vma(struct xe_gt *gt, struct xe_vma *vma,
>  static bool
>  xe_pagefault_access_is_atomic(enum xe_pagefault_access_type access_type)
>  {
> -	return access_type == XE_PAGEFAULT_ACCESS_TYPE_ATOMIC;
> +	return (access_type & XE_PAGEFAULT_ACCESS_TYPE_MASK) == XE_PAGEFAULT_ACCESS_TYPE_ATOMIC;
>  }
>  
>  static struct xe_vm *xe_pagefault_asid_to_vm(struct xe_device *xe, u32 asid)
> @@ -235,7 +235,7 @@ static void xe_pagefault_print(struct xe_pagefault *pf)
>  		   lower_32_bits(pf->consumer.page_addr),
>  		   FIELD_GET(XE_PAGEFAULT_TYPE_MASK,
>  			     pf->consumer.fault_type_level),
> -		   pf->consumer.access_type,
> +		   pf->consumer.access_type & XE_PAGEFAULT_ACCESS_TYPE_MASK,
>  		   FIELD_GET(XE_PAGEFAULT_LEVEL_MASK,
>  			     pf->consumer.fault_type_level),
>  		   pf->consumer.engine_class,
> @@ -261,9 +261,15 @@ static void xe_pagefault_queue_work(struct work_struct *w)
>  
>  		err = xe_pagefault_service(&pf);
>  		if (err) {
> -			xe_pagefault_print(&pf);
> -			xe_gt_info(pf.gt, "Fault response: Unsuccessful %pe\n",
> -				   ERR_PTR(err));
> +			if (!(pf.consumer.access_type & XE_PAGEFAULT_ACCESS_PREFETCH)) {
> +				xe_pagefault_print(&pf);
> +				xe_gt_info(pf.gt, "Fault response: Unsuccessful %pe\n",
> +					   ERR_PTR(err));
> +			} else {
> +				xe_gt_stats_incr(pf.gt, XE_GT_STATS_ID_INVALID_PREFETCH_PAGEFAULT_COUNT, 1);
> +				xe_gt_dbg(pf.gt, "Prefetch Fault response: Unsuccessful %pe\n",
> +					  ERR_PTR(err));
> +			}
>  		}
>  
>  		pf.producer.ops->ack_fault(&pf, err);
> diff --git a/drivers/gpu/drm/xe/xe_pagefault_types.h b/drivers/gpu/drm/xe/xe_pagefault_types.h
> index 0e378f41ede6..b3289219b1be 100644
> --- a/drivers/gpu/drm/xe/xe_pagefault_types.h
> +++ b/drivers/gpu/drm/xe/xe_pagefault_types.h
> @@ -68,10 +68,12 @@ struct xe_pagefault {
>  		/** @consumer.asid: address space ID */
>  		u32 asid;
>  		/**
> -		 * @consumer.access_type: access type, u8 rather than enum to
> -		 * keep size compact
> +		 * @consumer.access_type: access type and prefetch flag packed
> +		 * into a u8.
>  		 */
>  		u8 access_type;
> +#define XE_PAGEFAULT_ACCESS_TYPE_MASK	GENMASK(1, 0)
> +#define XE_PAGEFAULT_ACCESS_PREFETCH	BIT(7)
>  		/**
>  		 * @consumer.fault_type_level: fault type and level, u8 rather
>  		 * than enum to keep size compact
> -- 
> 2.43.0
> 

  reply	other threads:[~2026-02-23  4:34 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-23  4:07 [PATCH v5 0/2] drm/xe: Add prefetch pagefault support for Xe3p Varun Gupta
2026-02-23  4:07 ` [PATCH v5 1/2] drm/xe: Add counter for invalid prefetch pagefaults Varun Gupta
2026-02-23  4:35   ` Matthew Brost
2026-02-23  4:07 ` [PATCH v5 2/2] drm/xe: Add prefetch fault support for Xe3p Varun Gupta
2026-02-23  4:34   ` Matthew Brost [this message]
2026-02-23  4:41     ` Matthew Brost
2026-02-23  4:14 ` ✗ CI.checkpatch: warning for drm/xe: Add prefetch pagefault support for Xe3p (rev5) Patchwork
2026-02-23  4:15 ` ✓ CI.KUnit: 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=aZvY5dty7Xk2B/mV@lstrano-desk.jf.intel.com \
    --to=matthew.brost@intel.com \
    --cc=himal.prasad.ghimiray@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.d.roper@intel.com \
    --cc=priyanka.dandamudi@intel.com \
    --cc=stuart.summers@intel.com \
    --cc=varun.gupta@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