All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sriram R <srirrama@codeaurora.org>
To: ath11k@lists.infradead.org
Subject: Re: [PATCH] ath11k: Fix skb_panic observed during msdu coalescing
Date: Wed, 14 Aug 2019 10:45:47 +0530	[thread overview]
Message-ID: <ae33c58f7d58ad01a185b40143aec5e2@codeaurora.org> (raw)
In-Reply-To: <1565755661-811-1-git-send-email-srirrama@codeaurora.org>

Please drop this patch. Will send V2 with minor changes to variable 
names
On 2019-08-14 09:37, Sriram R wrote:
> skb_panic is hit during msdu coalescing whenever
> enough tailroom is not allocated based on the remaining
> msdu length which is spread across in different rx buffers.
> 
> Compute the extra length for resizing the skb based on
> the total msdu length and the msdu length of the first buffer.
> 
> Signed-off-by: Sriram R <srirrama@codeaurora.org>
> ---
>  drivers/net/wireless/ath/ath11k/dp_rx.c | 28 
> +++++++++++++++++-----------
>  1 file changed, 17 insertions(+), 11 deletions(-)
> 
> diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c
> b/drivers/net/wireless/ath/ath11k/dp_rx.c
> index 7d6b0c3..27d526b 100644
> --- a/drivers/net/wireless/ath/ath11k/dp_rx.c
> +++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
> @@ -1362,25 +1362,32 @@ void ath11k_dp_htt_htc_t2h_msg_handler(struct
> ath11k_base *ab,
>  static int ath11k_dp_rx_msdu_coalesce(struct ath11k *ar,
>  				      struct sk_buff_head *msdu_list,
>  				      struct sk_buff *first, struct sk_buff *last,
> -				      u8 l3pad_bytes, int msdu_len)
> +				      u8 l3pad_bytes, int msdu_total_len)
>  {
>  	struct sk_buff *skb;
>  	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(first);
> +	int msdu_first_offset, msdu_first_len;
>  	struct hal_rx_desc *ldesc;
>  	int space_extra;
>  	int rem_len;
>  	int buf_len;
> 
> +	/* As the msdu is spread across multiple rx buffers,
> +	 * find the offset to the start of msdu for computing
> +	 * the length of the msdu in the first buffer.
> +	 */
> +	msdu_first_offset = HAL_RX_DESC_SIZE + l3pad_bytes;
> +	msdu_first_len = DP_RX_BUFFER_SIZE - msdu_first_offset;
> +
>  	if (!rxcb->is_continuation) {
> -		skb_put(first, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
> -		skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
> +		skb_put(first, msdu_first_offset + msdu_total_len);
> +		skb_pull(first, msdu_first_offset);
>  		return 0;
>  	}
> 
> -	if (WARN_ON_ONCE(msdu_len <= (DP_RX_BUFFER_SIZE -
> -			 (HAL_RX_DESC_SIZE + l3pad_bytes)))) {
> -		skb_put(first, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
> -		skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
> +	if (WARN_ON_ONCE(msdu_total_len <= msdu_first_len)) {
> +		skb_put(first, msdu_first_offset + msdu_total_len);
> +		skb_pull(first, msdu_first_offset);
>  		return 0;
>  	}
> 
> @@ -1393,9 +1400,9 @@ static int ath11k_dp_rx_msdu_coalesce(struct 
> ath11k *ar,
>  	 * in the first buf is of length DP_RX_BUFFER_SIZE - 
> HAL_RX_DESC_SIZE.
>  	 */
>  	skb_put(first, DP_RX_BUFFER_SIZE);
> -	skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
> +	skb_pull(first, msdu_first_offset);
> 
> -	space_extra = msdu_len - (DP_RX_BUFFER_SIZE + skb_tailroom(first));
> +	space_extra = msdu_total_len - (msdu_first_len + 
> skb_tailroom(first));
>  	if (space_extra > 0 &&
>  	    (pskb_expand_head(first, 0, space_extra, GFP_ATOMIC) < 0)) {
>  		/* Free up all buffers of the MSDU */
> @@ -1415,8 +1422,7 @@ static int ath11k_dp_rx_msdu_coalesce(struct 
> ath11k *ar,
>  	 */
>  	ath11k_dp_rx_desc_end_tlv_copy(rxcb->rx_desc, ldesc);
> 
> -	rem_len = msdu_len -
> -		  (DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE - l3pad_bytes);
> +	rem_len = msdu_total_len - msdu_first_len;
>  	while ((skb = __skb_dequeue(msdu_list)) != NULL && rem_len > 0) {
>  		rxcb = ATH11K_SKB_RXCB(skb);
>  		if (rxcb->is_continuation)

_______________________________________________
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k

      reply	other threads:[~2019-08-14  5:15 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-14  4:07 [PATCH] ath11k: Fix skb_panic observed during msdu coalescing Sriram R
2019-08-14  5:15 ` Sriram R [this message]

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=ae33c58f7d58ad01a185b40143aec5e2@codeaurora.org \
    --to=srirrama@codeaurora.org \
    --cc=ath11k@lists.infradead.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.