[PATCH] drm/i915: Kick the rps worker when changing the boost frequency

[PATCH] drm/i915: Kick the rps worker when changing the boost frequency

[Chris Wilson]

The boost frequency is only applied from the RPS worker while someone is
waiting on a request and requested a boost. As such, when the user
wishes to change the frequency, we have to kick the worker in order to
re-evaluate whether to apply the boost frequency.

Fixes: 29ecd78d3b79 ("drm/i915: Define a separate variable and control for RPS waitboost frequency")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
---
 drivers/gpu/drm/i915/i915_sysfs.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index b33d2158c234..d97463bb69ad 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -304,8 +304,9 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
 {
 	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
 	struct intel_rps *rps = &dev_priv->gt_pm.rps;
-	u32 val;
+	bool boost = false;
 	ssize_t ret;
+	u32 val;
 
 	ret = kstrtou32(buf, 0, &val);
 	if (ret)
@@ -317,8 +318,13 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
 		return -EINVAL;
 
 	mutex_lock(&dev_priv->pcu_lock);
-	rps->boost_freq = val;
+	if (val != rps->boost_freq) {
+		rps->boost_freq = val;
+		boost = true;
+	}
 	mutex_unlock(&dev_priv->pcu_lock);
+	if (boost)
+		schedule_work(&rps->work);
 
 	return count;
 }
-- 
2.16.2

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915: Kick the rps worker when changing the boost frequency

[Mika Kuoppala]

Chris Wilson <chris@chris-wilson.co.uk> writes:

> The boost frequency is only applied from the RPS worker while someone is
> waiting on a request and requested a boost. As such, when the user
> wishes to change the frequency, we have to kick the worker in order to
> re-evaluate whether to apply the boost frequency.
>
> Fixes: 29ecd78d3b79 ("drm/i915: Define a separate variable and control for RPS waitboost frequency")
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Mika Kuoppala <mika.kuoppala@intel.com>
> ---
>  drivers/gpu/drm/i915/i915_sysfs.c | 10 ++++++++--
>  1 file changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
> index b33d2158c234..d97463bb69ad 100644
> --- a/drivers/gpu/drm/i915/i915_sysfs.c
> +++ b/drivers/gpu/drm/i915/i915_sysfs.c
> @@ -304,8 +304,9 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
>  {
>  	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
>  	struct intel_rps *rps = &dev_priv->gt_pm.rps;
> -	u32 val;
> +	bool boost = false;
>  	ssize_t ret;
> +	u32 val;
>  
>  	ret = kstrtou32(buf, 0, &val);
>  	if (ret)
> @@ -317,8 +318,13 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
>  		return -EINVAL;
>  
>  	mutex_lock(&dev_priv->pcu_lock);
> -	rps->boost_freq = val;
> +	if (val != rps->boost_freq) {
> +		rps->boost_freq = val;
> +		boost = true;
> +	}
>  	mutex_unlock(&dev_priv->pcu_lock);
> +	if (boost)
> +		schedule_work(&rps->work);
>

Can the rps work handler handle a situation
where the powers are off? It bails out if we dont
have iir nor boost, which might be enough of a
safeguard.

If it can, why not just kick the rps work
unconditionally?

Further, on all other freq store entries in sysfs
update val and schedule unconditionally?

Then the handler would be the single entry to
re-evaluation and setting the hardware.

-Mika

>  	return count;
>  }
> -- 
> 2.16.2
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH v2] drm/i915: Kick the rps worker when changing the boost frequency

[Chris Wilson]

The boost frequency is only applied from the RPS worker while someone is
waiting on a request and requested a boost. As such, when the user
wishes to change the frequency, we have to kick the worker in order to
re-evaluate whether to apply the boost frequency.

v2: Check num_waiters to decide if we should kick the worker to handle
boosting.

Fixes: 29ecd78d3b79 ("drm/i915: Define a separate variable and control for RPS waitboost frequency")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
---
 drivers/gpu/drm/i915/i915_sysfs.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index b33d2158c234..e5e6f6bb2b05 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -304,8 +304,9 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
 {
 	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
 	struct intel_rps *rps = &dev_priv->gt_pm.rps;
-	u32 val;
+	bool boost = false;
 	ssize_t ret;
+	u32 val;
 
 	ret = kstrtou32(buf, 0, &val);
 	if (ret)
@@ -317,8 +318,13 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
 		return -EINVAL;
 
 	mutex_lock(&dev_priv->pcu_lock);
-	rps->boost_freq = val;
+	if (val != rps->boost_freq) {
+		rps->boost_freq = val;
+		boost = atomic_read(&rps->num_waiters);
+	}
 	mutex_unlock(&dev_priv->pcu_lock);
+	if (boost)
+		schedule_work(&rps->work);
 
 	return count;
 }
-- 
2.16.2

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915: Kick the rps worker when changing the boost frequency

[Chris Wilson]

Quoting Mika Kuoppala (2018-03-08 09:53:01)
> Chris Wilson <chris@chris-wilson.co.uk> writes:
> 
> > The boost frequency is only applied from the RPS worker while someone is
> > waiting on a request and requested a boost. As such, when the user
> > wishes to change the frequency, we have to kick the worker in order to
> > re-evaluate whether to apply the boost frequency.
> >
> > Fixes: 29ecd78d3b79 ("drm/i915: Define a separate variable and control for RPS waitboost frequency")
> > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> > Cc: Mika Kuoppala <mika.kuoppala@intel.com>
> > ---
> >  drivers/gpu/drm/i915/i915_sysfs.c | 10 ++++++++--
> >  1 file changed, 8 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
> > index b33d2158c234..d97463bb69ad 100644
> > --- a/drivers/gpu/drm/i915/i915_sysfs.c
> > +++ b/drivers/gpu/drm/i915/i915_sysfs.c
> > @@ -304,8 +304,9 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
> >  {
> >       struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
> >       struct intel_rps *rps = &dev_priv->gt_pm.rps;
> > -     u32 val;
> > +     bool boost = false;
> >       ssize_t ret;
> > +     u32 val;
> >  
> >       ret = kstrtou32(buf, 0, &val);
> >       if (ret)
> > @@ -317,8 +318,13 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
> >               return -EINVAL;
> >  
> >       mutex_lock(&dev_priv->pcu_lock);
> > -     rps->boost_freq = val;
> > +     if (val != rps->boost_freq) {
> > +             rps->boost_freq = val;
> > +             boost = true;
> > +     }
> >       mutex_unlock(&dev_priv->pcu_lock);
> > +     if (boost)
> > +             schedule_work(&rps->work);
> >
> 
> Can the rps work handler handle a situation
> where the powers are off? It bails out if we dont
> have iir nor boost, which might be enough of a
> safeguard.

Yes, we check whether the interrupts are enabled and do nothing if they
are not. We disable the interrupts when idling.
 
> If it can, why not just kick the rps work
> unconditionally?

That's the plan. I'm pushing this because I've a small series waiting
for the conflicts to resolve...
 
> Further, on all other freq store entries in sysfs
> update val and schedule unconditionally?

Coming soon :)
 
> Then the handler would be the single entry to
> re-evaluation and setting the hardware.

\o/
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH v2] drm/i915: Kick the rps worker when changing the boost frequency

[Mika Kuoppala]

Chris Wilson <chris@chris-wilson.co.uk> writes:

> The boost frequency is only applied from the RPS worker while someone is
> waiting on a request and requested a boost. As such, when the user
> wishes to change the frequency, we have to kick the worker in order to
> re-evaluate whether to apply the boost frequency.
>
> v2: Check num_waiters to decide if we should kick the worker to handle
> boosting.
>
> Fixes: 29ecd78d3b79 ("drm/i915: Define a separate variable and control for RPS waitboost frequency")
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Mika Kuoppala <mika.kuoppala@intel.com>

Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

> ---
>  drivers/gpu/drm/i915/i915_sysfs.c | 10 ++++++++--
>  1 file changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
> index b33d2158c234..e5e6f6bb2b05 100644
> --- a/drivers/gpu/drm/i915/i915_sysfs.c
> +++ b/drivers/gpu/drm/i915/i915_sysfs.c
> @@ -304,8 +304,9 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
>  {
>  	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
>  	struct intel_rps *rps = &dev_priv->gt_pm.rps;
> -	u32 val;
> +	bool boost = false;
>  	ssize_t ret;
> +	u32 val;
>  
>  	ret = kstrtou32(buf, 0, &val);
>  	if (ret)
> @@ -317,8 +318,13 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
>  		return -EINVAL;
>  
>  	mutex_lock(&dev_priv->pcu_lock);
> -	rps->boost_freq = val;
> +	if (val != rps->boost_freq) {
> +		rps->boost_freq = val;
> +		boost = atomic_read(&rps->num_waiters);
> +	}
>  	mutex_unlock(&dev_priv->pcu_lock);
> +	if (boost)
> +		schedule_work(&rps->work);
>  
>  	return count;
>  }
> -- 
> 2.16.2
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✓ Fi.CI.BAT: success for drm/i915: Kick the rps worker when changing the boost frequency (rev2)

[Patchwork]

== Series Details ==

Series: drm/i915: Kick the rps worker when changing the boost frequency (rev2)
URL   : https://patchwork.freedesktop.org/series/39586/
State : success

== Summary ==

Series 39586v2 drm/i915: Kick the rps worker when changing the boost frequency
https://patchwork.freedesktop.org/api/1.0/series/39586/revisions/2/mbox/

fi-bdw-5557u     total:288  pass:267  dwarn:0   dfail:0   fail:0   skip:21  time:424s
fi-bdw-gvtdvm    total:288  pass:264  dwarn:0   dfail:0   fail:0   skip:24  time:421s
fi-blb-e6850     total:288  pass:223  dwarn:1   dfail:0   fail:0   skip:64  time:373s
fi-bsw-n3050     total:288  pass:242  dwarn:0   dfail:0   fail:0   skip:46  time:505s
fi-bwr-2160      total:288  pass:183  dwarn:0   dfail:0   fail:0   skip:105 time:277s
fi-bxt-dsi       total:288  pass:258  dwarn:0   dfail:0   fail:0   skip:30  time:489s
fi-bxt-j4205     total:288  pass:259  dwarn:0   dfail:0   fail:0   skip:29  time:493s
fi-byt-j1900     total:288  pass:253  dwarn:0   dfail:0   fail:0   skip:35  time:478s
fi-byt-n2820     total:288  pass:249  dwarn:0   dfail:0   fail:0   skip:39  time:467s
fi-cfl-8700k     total:288  pass:260  dwarn:0   dfail:0   fail:0   skip:28  time:403s
fi-cfl-s2        total:288  pass:262  dwarn:0   dfail:0   fail:0   skip:26  time:583s
fi-cfl-u         total:288  pass:262  dwarn:0   dfail:0   fail:0   skip:26  time:503s
fi-elk-e7500     total:288  pass:229  dwarn:0   dfail:0   fail:0   skip:59  time:419s
fi-gdg-551       total:288  pass:179  dwarn:0   dfail:0   fail:1   skip:108 time:291s
fi-glk-1         total:288  pass:260  dwarn:0   dfail:0   fail:0   skip:28  time:525s
fi-hsw-4770      total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:396s
fi-ilk-650       total:288  pass:227  dwarn:0   dfail:0   fail:1   skip:60  time:407s
fi-ivb-3520m     total:288  pass:259  dwarn:0   dfail:0   fail:0   skip:29  time:468s
fi-ivb-3770      total:288  pass:255  dwarn:0   dfail:0   fail:0   skip:33  time:418s
fi-kbl-7500u     total:288  pass:263  dwarn:1   dfail:0   fail:0   skip:24  time:466s
fi-kbl-7567u     total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:463s
fi-kbl-r         total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:507s
fi-pnv-d510      total:288  pass:222  dwarn:1   dfail:0   fail:0   skip:65  time:583s
fi-skl-6260u     total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:432s
fi-skl-6600u     total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:523s
fi-skl-6700hq    total:288  pass:262  dwarn:0   dfail:0   fail:0   skip:26  time:533s
fi-skl-6700k2    total:288  pass:264  dwarn:0   dfail:0   fail:0   skip:24  time:498s
fi-skl-6770hq    total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:480s
fi-skl-guc       total:288  pass:260  dwarn:0   dfail:0   fail:0   skip:28  time:419s
fi-skl-gvtdvm    total:288  pass:265  dwarn:0   dfail:0   fail:0   skip:23  time:426s
fi-snb-2520m     total:288  pass:248  dwarn:0   dfail:0   fail:0   skip:40  time:512s
fi-snb-2600      total:288  pass:248  dwarn:0   dfail:0   fail:0   skip:40  time:397s
Blacklisted hosts:
fi-cnl-drrs      total:288  pass:257  dwarn:3   dfail:0   fail:0   skip:19  time:525s

e4a2b905cbe1a129e1252b6f80a97add0dfbff40 drm-tip: 2018y-03m-08d-10h-27m-13s UTC integration manifest
eb5fc14618f9 drm/i915: Kick the rps worker when changing the boost frequency

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_8269/issues.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✗ Fi.CI.IGT: warning for drm/i915: Kick the rps worker when changing the boost frequency (rev2)

[Patchwork]

== Series Details ==

Series: drm/i915: Kick the rps worker when changing the boost frequency (rev2)
URL   : https://patchwork.freedesktop.org/series/39586/
State : warning

== Summary ==

---- Possible new issues:

Test drv_suspend:
        Subgroup sysfs-reader:
                pass       -> SKIP       (shard-snb)
Test kms_chv_cursor_fail:
        Subgroup pipe-a-128x128-top-edge:
                pass       -> DMESG-WARN (shard-hsw)

---- Known issues:

Test gem_eio:
        Subgroup in-flight-contexts:
                incomplete -> PASS       (shard-apl) fdo#105341
Test kms_chv_cursor_fail:
        Subgroup pipe-b-256x256-bottom-edge:
                dmesg-warn -> PASS       (shard-snb) fdo#105185 +2
Test kms_cursor_crc:
        Subgroup cursor-256x256-suspend:
                pass       -> INCOMPLETE (shard-hsw) fdo#103375
Test kms_flip:
        Subgroup dpms-vs-vblank-race-interruptible:
                pass       -> FAIL       (shard-hsw) fdo#103060
        Subgroup flip-vs-expired-vblank:
                fail       -> PASS       (shard-hsw) fdo#102887
        Subgroup flip-vs-panning-vs-hang:
                pass       -> DMESG-WARN (shard-snb) fdo#103821
Test kms_sysfs_edid_timing:
                warn       -> PASS       (shard-apl) fdo#100047
Test pm_lpsp:
        Subgroup screens-disabled:
                pass       -> FAIL       (shard-hsw) fdo#104941

fdo#105341 https://bugs.freedesktop.org/show_bug.cgi?id=105341
fdo#105185 https://bugs.freedesktop.org/show_bug.cgi?id=105185
fdo#103375 https://bugs.freedesktop.org/show_bug.cgi?id=103375
fdo#103060 https://bugs.freedesktop.org/show_bug.cgi?id=103060
fdo#102887 https://bugs.freedesktop.org/show_bug.cgi?id=102887
fdo#103821 https://bugs.freedesktop.org/show_bug.cgi?id=103821
fdo#100047 https://bugs.freedesktop.org/show_bug.cgi?id=100047
fdo#104941 https://bugs.freedesktop.org/show_bug.cgi?id=104941

shard-apl        total:3467 pass:1825 dwarn:1   dfail:0   fail:9   skip:1632 time:12283s
shard-hsw        total:3383 pass:1719 dwarn:3   dfail:0   fail:3   skip:1656 time:11407s
shard-snb        total:3467 pass:1360 dwarn:3   dfail:0   fail:2   skip:2102 time:6878s
Blacklisted hosts:
shard-kbl        total:3381 pass:1858 dwarn:29  dfail:1   fail:9   skip:1483 time:9078s

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_8269/shards.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx