[PATCH v3] drm/ast: DisplayPort edid supports 256 bytes

[PATCH v3] drm/ast: DisplayPort edid supports 256 bytes

[Jammy Huang]

DisplayPort supports EDID up to 256 bytes (blocks 0 and 1). Update the
block check to allow these two blocks. For any block beyond that
(i.e. the monitor's EDID is larger than the 256 bytes ASPEED DP can
hold), report it as an all-zero block instead of a read failure.

Also remove the manual checksum/extension-count patch of block 0's
last four bytes. That code was already dead, so the condition can never
be true.

Signed-off-by: Jammy Huang <jammy_huang@aspeedtech.com>
---
ASPEED DisplayPort's EDID size can be 256 bytes at most. Thus, EDID
blocks fetched can be 0 and 1.
---
Changes in v3:
- For block > 1, write a zeroed buffer. Returning a zeroed block
  instead makes drm_edid's block checker classify it as
  EDID_BLOCK_ZERO, which is treated as a non-fatal invalid block.
- Link to v2: https://lore.kernel.org/r/20260317-upstream_ast_dp_edid-v2-1-31166703c491@aspeedtech.com

Changes in v2:
Becasue drm-edid will handle invalid EDID if happen, we have 2 changes
below.
- Return 0 for the number of block more than 1.
- Remove modification of EDID
- Link to v1: https://lore.kernel.org/r/20260313-upstream_ast_dp_edid-v1-1-2a75b7c091b2@aspeedtech.com
---
 drivers/gpu/drm/ast/ast_dp.c | 24 ++++++++----------------
 1 file changed, 8 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c
index 9d07dad358c..38530cd126c 100644
--- a/drivers/gpu/drm/ast/ast_dp.c
+++ b/drivers/gpu/drm/ast/ast_dp.c
@@ -88,8 +88,14 @@ static int ast_astdp_read_edid_block(void *data, u8 *buf, unsigned int block, si
 	int ret = 0;
 	unsigned int i;
 
-	if (block > 0)
-		return -EIO; /* extension headers not supported */
+	if (block > 1) {
+		/* ASPEED DP's EDID buffer holds blocks 0 and 1 only (256
+		 * bytes). Report anything beyond that as an all-zero block
+		 * rather than a read failure.
+		 */
+		memset(buf, 0, len);
+		return 0;
+	}
 
 	/*
 	 * Protect access to I/O registers from concurrent modesetting
@@ -154,20 +160,6 @@ static int ast_astdp_read_edid_block(void *data, u8 *buf, unsigned int block, si
 		ediddata[2] = ast_get_index_reg(ast, AST_IO_VGACRI, 0xda);
 		ediddata[3] = ast_get_index_reg(ast, AST_IO_VGACRI, 0xdb);
 
-		if (i == 31) {
-			/*
-			 * For 128-bytes EDID_1.3,
-			 * 1. Add the value of Bytes-126 to Bytes-127.
-			 *		The Bytes-127 is Checksum. Sum of all 128bytes should
-			 *		equal 0	(mod 256).
-			 * 2. Modify Bytes-126 to be 0.
-			 *		The Bytes-126 indicates the Number of extensions to
-			 *		follow. 0 represents noextensions.
-			 */
-			ediddata[3] = ediddata[3] + ediddata[2];
-			ediddata[2] = 0;
-		}
-
 		memcpy(buf, ediddata, min((len - i), 4));
 		buf += 4;
 	}

---
base-commit: 5ee8dbf54602dc340d6235b1d6aa17c0f283f48c
change-id: 20260313-upstream_ast_dp_edid-5fe6adf7ad36

Best regards,
-- 
Jammy Huang <jammy_huang@aspeedtech.com>

Re: [PATCH v3] drm/ast: DisplayPort edid supports 256 bytes

[Thomas Zimmermann]

Hi

Am 19.08.26 um 07:58 schrieb Jammy Huang:
> DisplayPort supports EDID up to 256 bytes (blocks 0 and 1). Update the
> block check to allow these two blocks. For any block beyond that
> (i.e. the monitor's EDID is larger than the 256 bytes ASPEED DP can
> hold), report it as an all-zero block instead of a read failure.
>
> Also remove the manual checksum/extension-count patch of block 0's
> last four bytes. That code was already dead, so the condition can never
> be true.
>
> Signed-off-by: Jammy Huang <jammy_huang@aspeedtech.com>

I know it's controversial, but there appears to be no better solution 
now. Therefore

Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>

Best regards
Thomas

> ---
> ASPEED DisplayPort's EDID size can be 256 bytes at most. Thus, EDID
> blocks fetched can be 0 and 1.
> ---
> Changes in v3:
> - For block > 1, write a zeroed buffer. Returning a zeroed block
>    instead makes drm_edid's block checker classify it as
>    EDID_BLOCK_ZERO, which is treated as a non-fatal invalid block.
> - Link to v2: https://lore.kernel.org/r/20260317-upstream_ast_dp_edid-v2-1-31166703c491@aspeedtech.com
>
> Changes in v2:
> Becasue drm-edid will handle invalid EDID if happen, we have 2 changes
> below.
> - Return 0 for the number of block more than 1.
> - Remove modification of EDID
> - Link to v1: https://lore.kernel.org/r/20260313-upstream_ast_dp_edid-v1-1-2a75b7c091b2@aspeedtech.com
> ---
>   drivers/gpu/drm/ast/ast_dp.c | 24 ++++++++----------------
>   1 file changed, 8 insertions(+), 16 deletions(-)
>
> diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c
> index 9d07dad358c..38530cd126c 100644
> --- a/drivers/gpu/drm/ast/ast_dp.c
> +++ b/drivers/gpu/drm/ast/ast_dp.c
> @@ -88,8 +88,14 @@ static int ast_astdp_read_edid_block(void *data, u8 *buf, unsigned int block, si
>   	int ret = 0;
>   	unsigned int i;
>   
> -	if (block > 0)
> -		return -EIO; /* extension headers not supported */
> +	if (block > 1) {
> +		/* ASPEED DP's EDID buffer holds blocks 0 and 1 only (256
> +		 * bytes). Report anything beyond that as an all-zero block
> +		 * rather than a read failure.
> +		 */
> +		memset(buf, 0, len);
> +		return 0;
> +	}
>   
>   	/*
>   	 * Protect access to I/O registers from concurrent modesetting
> @@ -154,20 +160,6 @@ static int ast_astdp_read_edid_block(void *data, u8 *buf, unsigned int block, si
>   		ediddata[2] = ast_get_index_reg(ast, AST_IO_VGACRI, 0xda);
>   		ediddata[3] = ast_get_index_reg(ast, AST_IO_VGACRI, 0xdb);
>   
> -		if (i == 31) {
> -			/*
> -			 * For 128-bytes EDID_1.3,
> -			 * 1. Add the value of Bytes-126 to Bytes-127.
> -			 *		The Bytes-127 is Checksum. Sum of all 128bytes should
> -			 *		equal 0	(mod 256).
> -			 * 2. Modify Bytes-126 to be 0.
> -			 *		The Bytes-126 indicates the Number of extensions to
> -			 *		follow. 0 represents noextensions.
> -			 */
> -			ediddata[3] = ediddata[3] + ediddata[2];
> -			ediddata[2] = 0;
> -		}
> -
>   		memcpy(buf, ediddata, min((len - i), 4));
>   		buf += 4;
>   	}
>
> ---
> base-commit: 5ee8dbf54602dc340d6235b1d6aa17c0f283f48c
> change-id: 20260313-upstream_ast_dp_edid-5fe6adf7ad36
>
> Best regards,

-- 
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, (HRB 36809, AG Nürnberg)