On Friday 02 February 2018 07:46 PM, Sean Paul wrote:
On Fri, Feb 02, 2018 at 04:15:16PM +0530, Ramalingam C wrote:
When BKSV is invalid, to mitigate any communication errors,
BKSV is read once again.
Why is the Bksv read any more volatile than other reads? If the channel is
noisy, the retries should be done in the shim implementation (I think there are
already i2c retries).
Sean
True. But compliance spec says that when we determine that bksv is
invalid (not with 20 1s), we should re-read and confirm.
CTS test 1A-05 test expect this behavior.
--Ram
Signed-off-by: Ramalingam C <ramalingam.c@xxxxxxxxx>
---
drivers/gpu/drm/i915/intel_hdcp.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/intel_hdcp.c b/drivers/gpu/drm/i915/intel_hdcp.c
index b3f407cef8b0..fa2e7c727d00 100644
--- a/drivers/gpu/drm/i915/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/intel_hdcp.c
@@ -401,7 +401,7 @@ static int intel_hdcp_auth(struct intel_connector *connector)
struct drm_i915_private *dev_priv;
enum port port;
unsigned long r0_prime_gen_start;
- int ret, i;
+ int ret, i, retry = 1;
union {
u32 reg[2];
u8 shim[DRM_HDCP_AN_LEN];
@@ -443,11 +443,16 @@ static int intel_hdcp_auth(struct intel_connector *connector)
r0_prime_gen_start = jiffies;
memset(&bksv, 0, sizeof(bksv));
- ret = shim->read_bksv(intel_dig_port, bksv.shim);
+
+ do {
+ ret = shim->read_bksv(intel_dig_port, bksv.shim);
+ if (!ret)
+ if (!intel_hdcp_is_ksv_valid(bksv.shim))
+ ret = -ENODEV;
+ } while (ret && retry--);
+
if (ret)
return ret;
- else if (!intel_hdcp_is_ksv_valid(bksv.shim))
- return -ENODEV;
I915_WRITE(PORT_HDCP_BKSVLO(port), bksv.reg[0]);
I915_WRITE(PORT_HDCP_BKSVHI(port), bksv.reg[1]);
--
2.7.4
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
