Skip to contentSkip to conversation

drm/i915/guc: Always initialize action_lock

PATCH In review Arkadiusz Hiler 4 participants 8 messages Nov 22, 2016 – Nov 23, 2016

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drm/i915/guc: Always initialize action_lock

Action lock is not being initialized if the GuC submission is disabled (i.e. i915.guc_submission=0).

host2guc_action(), which uses the action_lock can be used for non-submission purposes, e.g. triggering HuC authentication.

Moving action_lock initialization before enablement check will allow us to use the host2guc_action no matter whether submission is enabled or not.

Cc Anusha Srivatsa <[email protected]>
Cc Jeff McGee <[email protected]>
---
 drivers/gpu/drm/i915/i915_guc_submission.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_guc_submission.c +2 -1
index 4462112..17c06d5 100644
@@ -1484,6 +1484,8 @@ int i915_guc_submission_init(struct drm_i915_private *dev_priv)
 	struct intel_guc *guc = &dev_priv->guc;
 	struct i915_vma *vma;
 
+	mutex_init(&guc->action_lock);
+
 	/* Wipe bitmap & delete client in case of reinitialisation */
 	bitmap_clear(guc->doorbell_bitmap, 0, GUC_MAX_DOORBELLS);
 	i915_guc_submission_disable(dev_priv);
@@ -1500,7 +1502,6 @@ int i915_guc_submission_init(struct drm_i915_private *dev_priv)
 
 	guc->ctx_pool_vma = vma;
 	ida_init(&guc->ctx_ids);
-	mutex_init(&guc->action_lock);
 	guc_log_create(guc);
 	guc_addon_create(guc);
 
Signature · 6 lines
2.7.4

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw
Quoted 6 lines · Arkadiusz Hiler

Action lock is not being initialized if the GuC submission is disabled (i.e. i915.guc_submission=0).

host2guc_action(), which uses the action_lock can be used for non-submission purposes, e.g. triggering HuC authentication.

Quoted 7 lines

Moving action_lock initialization before enablement check will allow us to use the host2guc_action no matter whether submission is enabled or not.

Cc Anusha Srivatsa <[email protected]>
Cc Jeff McGee <[email protected]>

Seems like you want to split uc_send(), uc_recv() out of i915_guc_submission.c
-Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#2 lore raw
Quoted 21 lines

-----Original Message-----
From: Intel-gfx [mailto:[email protected]] On Behalf Of Chris Wilson
Sent: Tuesday, November 22, 2016 9:06 AM
To: Hiler, Arkadiusz <[email protected]>
Cc: [email protected]
Subject: Re: [Intel-gfx] [PATCH] drm/i915/guc: Always initialize action_lock

Quoted 10 lines · Arkadiusz Hiler

Action lock is not being initialized if the GuC submission is disabled (i.e. i915.guc_submission=0).

host2guc_action(), which uses the action_lock can be used for non-submission purposes, e.g. triggering HuC authentication.

Moving action_lock initialization before enablement check will allow us to use the host2guc_action no matter whether submission is enabled or not.

Quoted 6 lines
Quoted 3 lines
Cc Anusha Srivatsa <[email protected]>
Cc Jeff McGee <[email protected]>

Seems like you want to split uc_send(), uc_recv() out of i915_guc_submission.c - Chris

Hiler, this fixes the guc_loading enabled and guc_submission disabled scenario that I was facing.
Anusha

Quoted 6 lines

--
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#3 lore raw
Quoted 17 lines · Anusha
Quoted 17 lines

-----Original Message-----
From: Intel-gfx [mailto:[email protected]] On Behalf Of Chris Wilson
Sent: Tuesday, November 22, 2016 9:06 AM
To: Hiler, Arkadiusz <[email protected]>
Cc: [email protected]
Subject: Re: [Intel-gfx] [PATCH] drm/i915/guc: Always initialize action_lock

Quoted 6 lines · Arkadiusz Hiler

Action lock is not being initialized if the GuC submission is disabled (i.e. i915.guc_submission=0).

host2guc_action(), which uses the action_lock can be used for non-submission purposes, e.g. triggering HuC authentication.

Quoted 8 lines
Quoted 6 lines
Quoted 3 lines

Moving action_lock initialization before enablement check will allow us to use the host2guc_action no matter whether submission is enabled or not.

Seems like you want to split uc_send(), uc_recv() out of i915_guc_submission.c - Chris

Hiler, this fixes the guc_loading enabled and guc_submission disabled scenario that I was facing.
Anusha

As you can see from the discussion in this thread I'll approach this differently - by doing some code reorganization.

It will obsolete the patch (and will create cosy place for functions renamed in "Make the GuC fw loading helper functions general").

Feel free to include this patch in your series for now.

PS. Just call me Arek :)

Signature · 6 lines
Cheers,
Arek
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#4 lore raw
In reply to Chris Wilson
Quoted 7 lines · Chris Wilson
Quoted 6 lines · Arkadiusz Hiler

Action lock is not being initialized if the GuC submission is disabled (i.e. i915.guc_submission=0).

host2guc_action(), which uses the action_lock can be used for non-submission purposes, e.g. triggering HuC authentication.

Quoted 6 lines
Quoted 3 lines

Moving action_lock initialization before enablement check will allow us to use the host2guc_action no matter whether submission is enabled or not.

Seems like you want to split uc_send(), uc_recv() out of i915_guc_submission.c
-Chris

The patch I've shared just addressed issue Anusha had with HuC enablement and allowed her to move further.

I was thinking of the split, as the HuC usage scenario rendered those functions more general.

I would like to do it as a learning exercise.

I thought of two ways of approaching that:
1. rename intel_guc_loader.c to something more general (e.g. intel_guc.c) and move the functions there
2. create intel_guc_comm.c (or similar) for those functions

Since guc_send() and guc_recv() are made up from only a couple of dozens of lines I am more inclined to option number 1.

Any thoughts?

Signature · 6 lines
Cheers,
Arek
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#5 lore raw
Quoted 28 lines · Arkadiusz Hiler
Quoted 7 lines · Chris Wilson
Quoted 6 lines · Arkadiusz Hiler

Action lock is not being initialized if the GuC submission is disabled (i.e. i915.guc_submission=0).

host2guc_action(), which uses the action_lock can be used for non-submission purposes, e.g. triggering HuC authentication.

Quoted 6 lines
Quoted 3 lines

Moving action_lock initialization before enablement check will allow us to use the host2guc_action no matter whether submission is enabled or not.

Seems like you want to split uc_send(), uc_recv() out of i915_guc_submission.c
-Chris

The patch I've shared just addressed issue Anusha had with HuC enablement and allowed her to move further.

I was thinking of the split, as the HuC usage scenario rendered those functions more general.

I would like to do it as a learning exercise.

I thought of two ways of approaching that:
1. rename intel_guc_loader.c to something more general (e.g. intel_guc.c) and move the functions there
2. create intel_guc_comm.c (or similar) for those functions

Quoted 4 lines

Since guc_send() and guc_recv() are made up from only a couple of dozens of lines I am more inclined to option number 1.

Any thoughts?

I was thinking 1 seems ok, but intel_guc_loader.c should be reasonably well defined (wrt what we expect to find there) and has some ugly code... So starting a new file (intel_guc.c) and moving the common functions there is ok. It's ok if it starts small, it'll grow! Anything will be better than the surprise that i915_guc_submission.c exports functions for intel_huc.c.

Should it be intel_guc.c though? intel_agent.c? intel_uc.c? intel_fw.c? -Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#6 lore raw
Quoted 32 lines · Chris Wilson
Quoted 31 lines · Arkadiusz Hiler
Quoted 7 lines · Chris Wilson
Quoted 6 lines · Arkadiusz Hiler

Action lock is not being initialized if the GuC submission is disabled (i.e. i915.guc_submission=0).

host2guc_action(), which uses the action_lock can be used for non-submission purposes, e.g. triggering HuC authentication.

Quoted 6 lines
Quoted 3 lines

Moving action_lock initialization before enablement check will allow us to use the host2guc_action no matter whether submission is enabled or not.

Seems like you want to split uc_send(), uc_recv() out of i915_guc_submission.c
-Chris

The patch I've shared just addressed issue Anusha had with HuC enablement and allowed her to move further.

I was thinking of the split, as the HuC usage scenario rendered those functions more general.

I would like to do it as a learning exercise.

I thought of two ways of approaching that:
1. rename intel_guc_loader.c to something more general (e.g. intel_guc.c) and move the functions there
2. create intel_guc_comm.c (or similar) for those functions

Since guc_send() and guc_recv() are made up from only a couple of dozens of lines I am more inclined to option number 1.

Quoted 8 lines
Quoted 1 line

Any thoughts?

I was thinking 1 seems ok, but intel_guc_loader.c should be reasonably well defined (wrt what we expect to find there) and has some ugly code... So starting a new file (intel_guc.c) and moving the common functions there is ok. It's ok if it starts small, it'll grow! Anything will be better than the surprise that i915_guc_submission.c exports functions for intel_huc.c.

I am convinced. I'll start slowly by moving the GuC communication there.

Quoted 2 lines

Should it be intel_guc.c though? intel_agent.c? intel_uc.c? intel_fw.c? -Chris

Browsing through already mentioned HuC series you'll see there's a big patch reusing lot's of GuC code for HuC. It's done by making funtions and structs more generic and renaming them to include just the 'uc'. Most of it could end up in the new file as well.

I am in favor of intel_uc.c.

Signature · 6 lines
Cheers,
Arek
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#7 lore raw

✗ Fi.CI.BAT: failure for drm/i915/guc: Always initialize action_lock

In reply to Arkadiusz Hiler
== Series Details ==

== Series Details ==

Series: drm/i915/guc: Always initialize action_lock
URL : https://patchwork.freedesktop.org/series/15755/
State : failure

== Summary ==

Series 15755v1 drm/i915/guc: Always initialize action_lock
https://patchwork.freedesktop.org/api/1.0/series/15755/revisions/1/mbox/

Test drv_module_reload_basic:

                dmesg-warn -> PASS       (fi-skl-6770hq)
Test kms_setmode:
        Subgroup basic-clone-single-crtc:
                pass       -> INCOMPLETE (fi-skl-6700k)
fi-bdw-5557u     total:244  pass:229  dwarn:0   dfail:0   fail:0   skip:15 
fi-bsw-n3050     total:244  pass:204  dwarn:0   dfail:0   fail:0   skip:40 
fi-bxt-t5700     total:244  pass:216  dwarn:0   dfail:0   fail:0   skip:28 
fi-byt-j1900     total:244  pass:216  dwarn:0   dfail:0   fail:0   skip:28 
fi-byt-n2820     total:244  pass:212  dwarn:0   dfail:0   fail:0   skip:32 
fi-hsw-4770      total:244  pass:224  dwarn:0   dfail:0   fail:0   skip:20 
fi-hsw-4770r     total:244  pass:224  dwarn:0   dfail:0   fail:0   skip:20 
fi-ilk-650       total:244  pass:191  dwarn:0   dfail:0   fail:0   skip:53 
fi-ivb-3520m     total:244  pass:222  dwarn:0   dfail:0   fail:0   skip:22 
fi-ivb-3770      total:244  pass:222  dwarn:0   dfail:0   fail:0   skip:22 
fi-kbl-7200u     total:244  pass:222  dwarn:0   dfail:0   fail:0   skip:22 
fi-skl-6260u     total:244  pass:230  dwarn:0   dfail:0   fail:0   skip:14 
fi-skl-6700hq    total:244  pass:223  dwarn:0   dfail:0   fail:0   skip:21 
fi-skl-6700k     total:209  pass:187  dwarn:1   dfail:0   fail:0   skip:20 
fi-skl-6770hq    total:244  pass:230  dwarn:0   dfail:0   fail:0   skip:14 
fi-snb-2520m     total:244  pass:212  dwarn:0   dfail:0   fail:0   skip:32 
fi-snb-2600      total:244  pass:211  dwarn:0   dfail:0   fail:0   skip:33 

eeec5e7742b23082dd20523c8baa08fe495175e4 drm-intel-nightly: 2016y-11m-21d-18h-22m-22s UTC integration manifest 6500d86 drm/i915/guc: Always initialize action_lock

== Logs ==

For more details see: https://intel-gfx-ci.01.org/CI/Patchwork_3083/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx