Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
To: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>,
	<intel-xe@lists.freedesktop.org>,
	Matthew Brost <matthew.brost@intel.com>
Subject: Re: [PATCH 3/3] drm/xe/pf: Add handling for MLRC adverse event threshold
Date: Mon, 15 Dec 2025 14:46:36 +0100	[thread overview]
Message-ID: <929ef11c-e59d-4237-83e3-665bcc9eb5a1@intel.com> (raw)
In-Reply-To: <84f13b72-7210-47ca-a6ce-78329195ded7@intel.com>



On 12/12/2025 9:17 PM, Daniele Ceraolo Spurio wrote:
> 
> 
> On 12/12/2025 12:15 PM, Daniele Ceraolo Spurio wrote:
>>
>>
>> On 12/12/2025 9:34 AM, Michal Wajdeczko wrote:
>>> From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
>>>
>>> Since it is illegal to register a MLRC context when scheduler groups are
>>> enabled, the GuC consider the VF doing so as an adverse event. Like for
>>> other adverse event, there is a threshold for how many times the event
>>> can happen before the GuC throws an error, which we need to add support
>>> for.
>>>
>>> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
>>> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
>>> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
>>
>> LGTM. Not sure if my r-b counts as the original author, but here it is anyway:
>>
>> Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
> 
> BTW, this patch is different enough from previous revs that you should probably take authorship of it.

it's not different - it's just smaller than original ;)

so I guess it should be my r-b:

Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>

+ Matt to resolve it this

> 
> Daniele
> 
>>
>> Daniele
>>
>>> ---
>>> v2: split from previous patch, add GuC version checking
>>> v3: don't move version code to its own file, call MAKE_GUC_VER from the
>>> caller instead.
>>> v4: cherry-picked from [1] and rebased (Michal)
>>>
>>> [1] https://patchwork.freedesktop.org/series/158142/#rev3
>>> ---
>>>   drivers/gpu/drm/xe/abi/guc_klvs_abi.h                | 9 +++++++++
>>>   drivers/gpu/drm/xe/xe_guc_klv_thresholds_set_types.h | 1 +
>>>   2 files changed, 10 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
>>> index 265a135e7061..89a4f8c504e6 100644
>>> --- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
>>> +++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
>>> @@ -352,6 +352,12 @@ enum  {
>>>    *      :1: NORMAL = schedule VF always, irrespective of whether it has work or not
>>>    *      :2: HIGH = schedule VF in the next time-slice after current active
>>>    *          time-slice completes if it has active work
>>> + *
>>> + * _`GUC_KLV_VF_CFG_THRESHOLD_MULTI_LRC_COUNT` : 0x8A0D
>>> + *      Given that multi-LRC contexts are incompatible with SRIOV scheduler
>>> + *      groups and cause the latter to be turned off when registered with the
>>> + *      GuC, this config allows the PF to set a threshold for multi-LRC context
>>> + *      registrations by VFs to monitor their behavior.
>>>    */
>>>     #define GUC_KLV_VF_CFG_GGTT_START_KEY        0x0001
>>> @@ -410,6 +416,9 @@ enum  {
>>>   #define   GUC_SCHED_PRIORITY_NORMAL        1u
>>>   #define   GUC_SCHED_PRIORITY_HIGH        2u
>>>   +#define GUC_KLV_VF_CFG_THRESHOLD_MULTI_LRC_COUNT_KEY 0x8a0d
>>> +#define GUC_KLV_VF_CFG_THRESHOLD_MULTI_LRC_COUNT_LEN    1u
>>> +
>>>   /*
>>>    * Workaround keys:
>>>    */
>>> diff --git a/drivers/gpu/drm/xe/xe_guc_klv_thresholds_set_types.h b/drivers/gpu/drm/xe/xe_guc_klv_thresholds_set_types.h
>>> index 5f84da3d10d3..45ab5a3b5218 100644
>>> --- a/drivers/gpu/drm/xe/xe_guc_klv_thresholds_set_types.h
>>> +++ b/drivers/gpu/drm/xe/xe_guc_klv_thresholds_set_types.h
>>> @@ -37,6 +37,7 @@
>>>       define(H2G_STORM, guc_time_us)            \
>>>       define(IRQ_STORM, irq_time_us)            \
>>>       define(DOORBELL_STORM, doorbell_time_us)    \
>>> +    define(MULTI_LRC_COUNT, multi_lrc_count, 70, 53)\
>>>       /* end */
>>>     /**
>>
> 


  reply	other threads:[~2025-12-15 13:46 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-12 17:34 [PATCH 0/3] PF: Add handling for new adverse event thresholds Michal Wajdeczko
2025-12-12 17:34 ` [PATCH 1/3] drm/xe: Introduce IF_ARGS macro utility Michal Wajdeczko
2025-12-12 19:47   ` Daniele Ceraolo Spurio
2025-12-12 17:34 ` [PATCH 2/3] drm/xe/pf: Prepare for new threshold KLVs Michal Wajdeczko
2025-12-12 20:12   ` Daniele Ceraolo Spurio
2025-12-12 17:34 ` [PATCH 3/3] drm/xe/pf: Add handling for MLRC adverse event threshold Michal Wajdeczko
2025-12-12 20:15   ` Daniele Ceraolo Spurio
2025-12-12 20:17     ` Daniele Ceraolo Spurio
2025-12-15 13:46       ` Michal Wajdeczko [this message]
2025-12-12 18:02 ` ✗ CI.checkpatch: warning for PF: Add handling for new adverse event thresholds Patchwork
2025-12-12 18:03 ` ✓ CI.KUnit: success " Patchwork
2025-12-12 19:14 ` ✓ Xe.CI.BAT: " Patchwork
2025-12-13 10:55 ` ✓ Xe.CI.Full: " 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=929ef11c-e59d-4237-83e3-665bcc9eb5a1@intel.com \
    --to=michal.wajdeczko@intel.com \
    --cc=daniele.ceraolospurio@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.brost@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