Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Ceraolo Spurio, Daniele" <daniele.ceraolospurio@intel.com>
To: "Sundaresan, Sujaritha" <sujaritha.sundaresan@intel.com>,
	"Vivekanandan,
	Balasubramani" <balasubramani.vivekanandan@intel.com>,
	<intel-xe@lists.freedesktop.org>
Cc: Tangudu Tilak Tirumalesh <tilak.tirumalesh.tangudu@intel.com>,
	"Matthew Brost" <matthew.brost@intel.com>,
	Stuart Summers <stuart.summers@intel.com>
Subject: Re: [PATCH] drm/xe/guc: Reset GuC before unmapping its post-hwconfig buffers
Date: Fri, 13 Nov 2026 11:51:19 -0800	[thread overview]
Message-ID: <f4facd87-04f7-47b8-8cfb-dfa42fae965d@intel.com> (raw)
In-Reply-To: <3032c66b-9718-4917-b335-caea298c794e@intel.com>

[-- Attachment #1: Type: text/plain, Size: 5387 bytes --]



On 11/3/2026 11:37 PM, Sundaresan, Sujaritha wrote:
>
>
> On 10/26/2026 7:37 PM, Vivekanandan, Balasubramani wrote:
>> On 19.10.2026 15:34, Sujaritha Sundaresan wrote:
>>> guc_fini_hw() resets the GuC on teardown, but it is registered before
>>> xe_guc_init_post_hwconfig() (re)allocates most of the GuC buffers. As
>>> devm actions run in reverse order, these buffers get unmapped from the
>>> GGTT while the GuC is still running and may still access them. When
>>> unmapped GGTT entries are left invalid, this results in a GuC page
>>> fault on unbind or probe failure.
>>>
>>> Re-register guc_fini_hw() after all the GuC buffer allocations, so the
>>> GuC is reset before any of them is unmapped. Keep GuC submit and RC
>>> teardown running before the reset, as they still need to stop or talk
>>> to the GuC.
>>>
>>> Assisted-by: LLM
>>> Signed-off-by: Sujaritha Sundaresan<sujaritha.sundaresan@intel.com>

Do we need a fixes tag for this?

>>> ---
>>>   drivers/gpu/drm/xe/xe_guc.c | 30 ++++++++++++++++++++++--------
>>>   1 file changed, 22 insertions(+), 8 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
>>> index 6c35474ab843..14ae73844d76 100644
>>> --- a/drivers/gpu/drm/xe/xe_guc.c
>>> +++ b/drivers/gpu/drm/xe/xe_guc.c
>>> @@ -882,9 +882,10 @@ static u32 guc_additional_cache_size(struct xe_device *xe)
>>>    */
>>>   int xe_guc_init_post_hwconfig(struct xe_guc *guc)
>>>   {
>>> +	struct xe_device *xe = guc_to_xe(guc);
>>>   	int ret;
>>>   
>>> -	if (IS_SRIOV_VF(guc_to_xe(guc)))
>>> +	if (IS_SRIOV_VF(xe))
>>>   		return vf_guc_init_post_hwconfig(guc);
>>>   
>>>   	ret = xe_guc_realloc_post_hwconfig(guc);
>>> @@ -897,28 +898,41 @@ int xe_guc_init_post_hwconfig(struct xe_guc *guc)
>>>   
>>>   	guc_init_params_post_hwconfig(guc);
>>>   
>>> -	ret = xe_guc_submit_init(guc);
>> The teardown function guc_submit_fini, registered by xe_guc_submit_init,
>> is invoked before releasing the buffers shared with GuC.
>>
>> guc_submit_fini stops guc submission and also does a soft reset using
>> XE_GUC_ACTION_CLIENT_SOFT_RESET.
>> The GuC specification says, on XE_GUC_ACTION_CLIENT_SOFT_RESET, the GuC
>> revokes the log and CTB buffers.
>>
>> So I believe, after that GuC shouldn't be accessing any buffers shared
>> by XeKMD. It must be safe to release the buffers after sending
>> XE_GUC_ACTION_CLIENT_SOFT_RESET.
>>
>> Did we understand why are seeing this issue now but not earlier?
>> Is it a regresssion from a patch, either in KMD or GuC?
>>
>> Regards,
>> Bala
>
> Hi Bala,
>
>
> Thanks for the review. I did some more digging here.
>
>
> The soft reset in guc_submit_fini() goes through fine. Right after 
> that we unmap the GuC log buffer
>
> (0x1e6a000-0x286b000), and the GuC faults reading 0x1e72138, which is 
> inside it.
>
> So from what I see  the GuC is still touching the log buffer after 
> XE_GUC_ACTION_CLIENT_SOFT_RESET.
>

The CLIENT_SOFT_RESET was initially introduced for use by VFs, to allow 
a VF to reset its own state. Although a PF can also use this command, 
the type of state being reset doesn't change, so global allocations (GuC 
log, ADS) are not cleared by this command. I believe that the "log" in 
the specs refers to extra log buffers that can be registered with the 
GuC to get extra info (like the engine utilization buffers), but I'll 
ask the GuC team to clarify the specs.

I agree with the approach in this patch, we should definitely reset the 
GuC before we start unmapping its buffers. I have not reviewed the code 
changes yet, so no comment on those.

Daniele

>
> ee460b8855d4 ("drm/xe/ggtt: fault on unmapped GGTT for multi-queue 
> platforms") dropped the scratch
>
> page on multi-queue platforms. Before that, this late read just landed 
> on scratch; now it faults.
>
> Thanks,
>
> Suja
>
>
>>> +	ret = xe_guc_pc_init(&guc->pc);
>>>   	if (ret)
>>>   		return ret;
>>>   
>>> -	ret = xe_guc_db_mgr_init(&guc->dbm, ~0);
>>> +	ret = xe_guc_engine_activity_init(guc);
>>>   	if (ret)
>>>   		return ret;
>>>   
>>> -	ret = xe_guc_pc_init(&guc->pc);
>>> +	ret = xe_guc_buf_cache_init_with_size(&guc->buf,
>>> +					      guc_additional_cache_size(xe));
>>>   	if (ret)
>>>   		return ret;
>>>   
>>> -	ret = xe_guc_rc_init(guc);
>>> +	/*
>>> +	 * The GuC keeps accessing its GGTT-mapped buffers until it is reset,
>>> +	 * so move guc_fini_hw() after all GuC allocations above to have the
>>> +	 * reset done before any of them is unpinned and its GGTT PTEs cleared.
>>> +	 * Teardown actions that must run before the reset, either because they
>>> +	 * still talk to the GuC (like GuC RC) or because they stop the CT and
>>> +	 * submission (like GuC submit), must be registered after this point.
>>> +	 */
>>> +	devm_remove_action(xe->drm.dev, guc_fini_hw, guc);
>>> +	ret = devm_add_action_or_reset(xe->drm.dev, guc_fini_hw, guc);
>>>   	if (ret)
>>>   		return ret;
>>>   
>>> -	ret = xe_guc_engine_activity_init(guc);
>>> +	ret = xe_guc_submit_init(guc);
>>>   	if (ret)
>>>   		return ret;
>>>   
>>> -	ret = xe_guc_buf_cache_init_with_size(&guc->buf,
>>> -					      guc_additional_cache_size(guc_to_xe(guc)));
>>> +	ret = xe_guc_db_mgr_init(&guc->dbm, ~0);
>>> +	if (ret)
>>> +		return ret;
>>> +
>>> +	ret = xe_guc_rc_init(guc);
>>>   	if (ret)
>>>   		return ret;
>>>   
>>>
>>> base-commit: 5bc00022b24f8a7bc6630609e9fd93dd977a87d3
>>> -- 
>>> 2.34.1
>>>

[-- Attachment #2: Type: text/html, Size: 7607 bytes --]

  reply	other threads:[~2026-11-13 19:52 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-19 10:04 [PATCH] drm/xe/guc: Reset GuC before unmapping its post-hwconfig buffers Sujaritha Sundaresan
2026-10-19  9:26 ` ✓ CI.KUnit: success for " Patchwork
2026-10-19 10:17 ` ✓ Xe.CI.BAT: " Patchwork
2026-10-19 12:08 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-11-17 23:14   ` Summers, Stuart
2026-10-26 14:07 ` [PATCH] " Vivekanandan, Balasubramani
2026-11-04  7:37   ` Sundaresan, Sujaritha
2026-11-13 19:51     ` Ceraolo Spurio, Daniele [this message]
2026-11-18 14:28       ` Vivekanandan, Balasubramani
2026-11-18  9:47 ` ✗ CI.checkpatch: warning for drm/xe/guc: Reset GuC before unmapping its post-hwconfig buffers (rev2) Patchwork
2026-11-18  9:49 ` ✓ CI.KUnit: success " Patchwork
2026-11-18 10:28 ` ✓ Xe.CI.BAT: " 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=f4facd87-04f7-47b8-8cfb-dfa42fae965d@intel.com \
    --to=daniele.ceraolospurio@intel.com \
    --cc=balasubramani.vivekanandan@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.brost@intel.com \
    --cc=stuart.summers@intel.com \
    --cc=sujaritha.sundaresan@intel.com \
    --cc=tilak.tirumalesh.tangudu@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