drm/amd/display: Simplify hdcp validate_bksv

* drm/amd/display: Simplify hdcp validate_bksv
@ 2021-07-11 19:46 Joe Perches
  2021-07-11 23:02 ` kernel test robot
  2021-07-12  3:34 ` kernel test robot
  0 siblings, 2 replies; 4+ messages in thread
From: Joe Perches @ 2021-07-11 19:46 UTC (permalink / raw)
  To: Bhawanpreet Lakha; +Cc: Kees Cook, Alex Deucher, amd-gfx, dri-devel, LKML

commit 06888d571b51 ("drm/amd/display: Avoid HDCP over-read and corruption")
fixed an overread with an invalid buffer length but added an unnecessary
buffer and copy.

Simplify the code by using a single uint64_t and __builtin_popcountll to
count the number of bits set in the original bksv buffer instead of a loop.

This also avoid a possible unaligned access of the temporary bksv.

Signed-off-by: Joe Perches <joe@perches.com>
---

It seems quite odd 20 bits set is a magic number here.
Should it be a specific be/le value instead?

 drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c
index de872e7958b06..78a4c6dd95d99 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c
@@ -28,17 +28,10 @@
 static inline enum mod_hdcp_status validate_bksv(struct mod_hdcp *hdcp)
 {
 	uint64_t n = 0;
-	uint8_t count = 0;
-	u8 bksv[sizeof(n)] = { };
 
-	memcpy(bksv, hdcp->auth.msg.hdcp1.bksv, sizeof(hdcp->auth.msg.hdcp1.bksv));
-	n = *(uint64_t *)bksv;
+	memcpy(&n, hdcp->auth.msg.hdcp1.bksv, sizeof(hdcp->auth.msg.hdcp1.bksv));
 
-	while (n) {
-		count++;
-		n &= (n - 1);
-	}
-	return (count == 20) ? MOD_HDCP_STATUS_SUCCESS :
+	return (__builtin_popcountll(n) == 20) ? MOD_HDCP_STATUS_SUCCESS :
 			MOD_HDCP_STATUS_HDCP1_INVALID_BKSV;
 }
 



^ permalink raw reply related	[flat|nested] 4+ messages in thread