public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Ramalingam C <ramalingam.c@intel.com>
To: Sean Paul <seanpaul@chromium.org>
Cc: daniel.vetter@ffwll.ch, intel-gfx@lists.freedesktop.org,
	rodrigo.vivi@intel.com
Subject: Re: [PATCH 5/8] drm/i915: Optimize HDCP key load
Date: Fri, 2 Feb 2018 20:03:01 +0530	[thread overview]
Message-ID: <2c16f245-e6c4-cb65-e3e5-95faf61d2140@intel.com> (raw)
In-Reply-To: <20180202141812.fjoyh7tfcqhzlbns@art_vandelay>



On Friday 02 February 2018 07:48 PM, Sean Paul wrote:
> On Fri, Feb 02, 2018 at 04:15:17PM +0530, Ramalingam C wrote:
>> HDCP key need not be cleared on each hdcp disable. And HDCP key Load
>> is skipped if key is already loaded.
>>
> I had previously encountered issues without clearing the key in my testing.
> IIRC, without clearing the keys things acted differently. How much time are we
> saving by optimizing this?
>
> Sean
Time profiling is not done. As per the Bspec, Once Keys are loaded, they 
will be cleared only when PG1/PG0 is off.
So on Resume we need to load the keys. Since we have the status register 
for key load state, I feel we could rely on them.

Now with our code state, I am not seeing the need for reloading the 
keys. I have tested on KBL.
I think this is worth an attempt as no failures are observed.

--Ram
>
>
>> Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
>> ---
>>   drivers/gpu/drm/i915/intel_hdcp.c | 6 ++++--
>>   1 file changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/intel_hdcp.c b/drivers/gpu/drm/i915/intel_hdcp.c
>> index fa2e7c727d00..5de9afd275b2 100644
>> --- a/drivers/gpu/drm/i915/intel_hdcp.c
>> +++ b/drivers/gpu/drm/i915/intel_hdcp.c
>> @@ -51,6 +51,10 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv)
>>   	int ret;
>>   	u32 val;
>>   
>> +	val = I915_READ(HDCP_KEY_STATUS);
>> +	if ((val & HDCP_KEY_LOAD_DONE) && (val & HDCP_KEY_LOAD_STATUS))
>> +		return 0;
>> +
>>   	/*
>>   	 * On HSW and BDW HW loads the HDCP1.4 Key when Display comes
>>   	 * out of reset. So if Key is not already loaded, its an error state.
>> @@ -542,8 +546,6 @@ static int _intel_hdcp_disable(struct intel_connector *connector)
>>   		return -ETIMEDOUT;
>>   	}
>>   
>> -	intel_hdcp_clear_keys(dev_priv);
>> -
>>   	ret = connector->hdcp_shim->toggle_signalling(intel_dig_port, false);
>>   	if (ret) {
>>   		DRM_ERROR("Failed to disable HDCP signalling\n");
>> -- 
>> 2.7.4
>>

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

  reply	other threads:[~2018-02-02 14:38 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-02 10:45 [PATCH 0/8] Adhering to HDCP1.4 Compliance Test Spec Ramalingam C
2018-02-02 10:45 ` [PATCH 1/8] drm/i915: Handle failure from 2nd stage HDCP auth Ramalingam C
2018-02-02 14:09   ` Sean Paul
2018-02-02 14:22     ` Ramalingam C
2018-02-02 14:45       ` Sean Paul
2018-02-02 14:51         ` Ramalingam C
2018-02-02 15:22           ` Sean Paul
2018-02-02 15:26             ` Ramalingam C
2018-02-02 10:45 ` [PATCH 2/8] drm/i915: Stop encryption for repeater with no sink Ramalingam C
2018-02-02 14:13   ` Sean Paul
2018-02-02 14:12     ` Ramalingam C
2018-02-02 14:48       ` Sean Paul
2018-02-02 15:03         ` Ramalingam C
2018-02-02 15:19           ` Sean Paul
2018-02-02 10:45 ` [PATCH 3/8] drm/i915: Connector info in HDCP debug msgs Ramalingam C
2018-02-02 14:15   ` Sean Paul
2018-02-02 10:45 ` [PATCH 4/8] drm/i915: Retry HDCP BKSV read Ramalingam C
2018-02-02 14:16   ` Sean Paul
2018-02-02 14:26     ` Ramalingam C
2018-02-02 14:44       ` Sean Paul
2018-02-02 10:45 ` [PATCH 5/8] drm/i915: Optimize HDCP key load Ramalingam C
2018-02-02 14:18   ` Sean Paul
2018-02-02 14:33     ` Ramalingam C [this message]
2018-02-02 15:24       ` Sean Paul
2018-02-02 10:45 ` [PATCH 6/8] drm/i915: Detect panel's hdcp capability Ramalingam C
2018-02-02 14:24   ` Sean Paul
2018-02-02 14:38     ` Ramalingam C
2018-02-02 15:40       ` Sean Paul
2018-02-02 10:45 ` [PATCH 7/8] drm/i915: Reauthenticate HDCP on failure Ramalingam C
2018-02-02 14:37   ` Sean Paul
2018-02-02 15:05     ` Ramalingam C
2018-02-02 10:45 ` [PATCH 8/8] drm/i915: fix misalignment in HDCP register def Ramalingam C
2018-02-02 14:38   ` Sean Paul
2018-02-02 11:10 ` ✓ Fi.CI.BAT: success for Adhering to HDCP1.4 Compliance Test Spec Patchwork
2018-02-02 12:50 ` ✗ Fi.CI.IGT: failure " 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=2c16f245-e6c4-cb65-e3e5-95faf61d2140@intel.com \
    --to=ramalingam.c@intel.com \
    --cc=daniel.vetter@ffwll.ch \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=rodrigo.vivi@intel.com \
    --cc=seanpaul@chromium.org \
    /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