netdev - Re: [PATCH net-next v3 5/6] net: qualcomm: rmnet: don't use C bit-fields in rmnet checksum trailer

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 9 Mar 2021 18:13:16 -0600
From:   Alex Elder <elder@...aro.org>
To:     subashab@...eaurora.org, stranche@...eaurora.org,
        davem@...emloft.net, kuba@...nel.org
Cc:     sharathv@...eaurora.org, bjorn.andersson@...aro.org,
        evgreen@...omium.org, cpratapa@...eaurora.org,
        David.Laight@...LAB.COM, elder@...nel.org, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v3 5/6] net: qualcomm: rmnet: don't use C
 bit-fields in rmnet checksum trailer

On 3/9/21 6:48 AM, Alex Elder wrote:
> Replace the use of C bit-fields in the rmnet_map_dl_csum_trailer
> structure with a single one-byte field, using constant field masks
> to encode or get at embedded values.
> 
> Signed-off-by: Alex Elder <elder@...aro.org>
> Reviewed-by: Bjorn Andersson <bjorn.andersson@...aro.org>
> ---
> v3: Use BIT(x) and don't use u8_get_bits() for the checksum valid flag
> 
>   .../ethernet/qualcomm/rmnet/rmnet_map_data.c    |  2 +-
>   include/linux/if_rmnet.h                        | 17 +++++++----------
>   2 files changed, 8 insertions(+), 11 deletions(-)
> 
> diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_map_data.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_map_data.c
> index 3291f252d81b0..72dbbe2c27bd7 100644
> --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_map_data.c
> +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_map_data.c
> @@ -365,7 +365,7 @@ int rmnet_map_checksum_downlink_packet(struct sk_buff *skb, u16 len)
>   
>   	csum_trailer = (struct rmnet_map_dl_csum_trailer *)(skb->data + len);
>   
> -	if (!csum_trailer->valid) {
> +	if (csum_trailer->flags & MAP_CSUM_DL_VALID_FLAG) {

This was the actual problem.  The result of the new
test is the reverse of what it should be.

In any case, I'm taking a break from this for a while.

					-Alex

>   		priv->stats.csum_valid_unset++;
>   		return -EINVAL;
>   	}
> diff --git a/include/linux/if_rmnet.h b/include/linux/if_rmnet.h
> index 22ccc89bb5d8e..a848bb2e0dad0 100644
> --- a/include/linux/if_rmnet.h
> +++ b/include/linux/if_rmnet.h
> @@ -19,21 +19,18 @@ struct rmnet_map_header {
>   #define MAP_PAD_LEN_FMASK		GENMASK(5, 0)
>   
>   struct rmnet_map_dl_csum_trailer {
> -	u8  reserved1;
> -#if defined(__LITTLE_ENDIAN_BITFIELD)
> -	u8  valid:1;
> -	u8  reserved2:7;
> -#elif defined (__BIG_ENDIAN_BITFIELD)
> -	u8  reserved2:7;
> -	u8  valid:1;
> -#else
> -#error	"Please fix <asm/byteorder.h>"
> -#endif
> +	u8 reserved1;
> +	u8 flags;			/* MAP_CSUM_DL_*_FMASK */
>   	__be16 csum_start_offset;
>   	__be16 csum_length;
>   	__be16 csum_value;
>   } __aligned(1);
>   
> +/* rmnet_map_dl_csum_trailer flags field:
> + *  VALID:	1 = checksum and length valid; 0 = ignore them
> + */
> +#define MAP_CSUM_DL_VALID_FLAG		BIT(0)
> +
>   struct rmnet_map_ul_csum_header {
>   	__be16 csum_start_offset;
>   #if defined(__LITTLE_ENDIAN_BITFIELD)
>