[PATCH 1/1] mac80211: [RFC] do not fragment while aggregation is in use

[PATCH 1/1] mac80211: [RFC] do not fragment while aggregation is in use

[Ron Rindjunsky]

This patch denies the use of framentation while ampdu is used.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
---
 net/mac80211/tx.c |   14 ++------------
 1 files changed, 2 insertions(+), 12 deletions(-)

diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 1ad9e66..d4693f6 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -658,17 +658,6 @@ ieee80211_tx_h_fragment(struct ieee80211_tx_data *tx)
 	if (!(tx->flags & IEEE80211_TX_FRAGMENTED))
 		return TX_CONTINUE;
 
-	/*
-	 * Warn when submitting a fragmented A-MPDU frame and drop it.
-	 * This is an error and needs to be fixed elsewhere, but when
-	 * done needs to take care of monitor interfaces (injection)
-	 * etc.
-	 */
-	if (WARN_ON(tx->flags & IEEE80211_TX_CTL_AMPDU ||
-		    skb_get_queue_mapping(tx->skb) >=
-			ieee80211_num_regular_queues(&tx->local->hw)))
-		return TX_DROP;
-
 	first = tx->skb;
 
 	hdrlen = ieee80211_get_hdrlen(tx->fc);
@@ -981,7 +970,8 @@ __ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
 	if (tx->flags & IEEE80211_TX_FRAGMENTED) {
 		if ((tx->flags & IEEE80211_TX_UNICAST) &&
 		    skb->len + FCS_LEN > local->fragmentation_threshold &&
-		    !local->ops->set_frag_threshold)
+		    !local->ops->set_frag_threshold &&
+		    !(info->flags & IEEE80211_TX_CTL_AMPDU))
 			tx->flags |= IEEE80211_TX_FRAGMENTED;
 		else
 			tx->flags &= ~IEEE80211_TX_FRAGMENTED;
-- 
1.5.3.3

---------------------------------------------------------------------
Intel Israel (74) Limited

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

Re: [PATCH 1/1] mac80211: [RFC] do not fragment while aggregation is in use

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 595 bytes --]

Thanks.

> -	/*
> -	 * Warn when submitting a fragmented A-MPDU frame and drop it.
> -	 * This is an error and needs to be fixed elsewhere, but when
> -	 * done needs to take care of monitor interfaces (injection)
> -	 * etc.
> -	 */
> -	if (WARN_ON(tx->flags & IEEE80211_TX_CTL_AMPDU ||
> -		    skb_get_queue_mapping(tx->skb) >=
> -			ieee80211_num_regular_queues(&tx->local->hw)))
> -		return TX_DROP;
> -

Should we keep that, just in case? If it ever happens and the hw rejects
a fragment for some reason, we'll otherwise clobber some state and
probably crash.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

RE: [PATCH 1/1] mac80211: [RFC] do not fragment while aggregationis in use

[Rindjunsky, Ron]

>> -	/*
>> -	 * Warn when submitting a fragmented A-MPDU frame and drop it.
>> -	 * This is an error and needs to be fixed elsewhere, but when
>> -	 * done needs to take care of monitor interfaces (injection)
>> -	 * etc.
>> -	 */
>> -	if (WARN_ON(tx->flags & IEEE80211_TX_CTL_AMPDU ||
>> -		    skb_get_queue_mapping(tx->skb) >=
>> -			ieee80211_num_regular_queues(&tx->local->hw)))
>> -		return TX_DROP;
>> -

> Should we keep that, just in case? If it ever happens and the hw
rejects
> a fragment for some reason, we'll otherwise clobber some state and
> probably crash.

Agree. Should I resend the patch or would you just remove it from the
original patch?

> johannes
---------------------------------------------------------------------
Intel Israel (74) Limited

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

RE: [PATCH 1/1] mac80211: [RFC] do not fragment while aggregationis in use

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 874 bytes --]

On Thu, 2008-06-12 at 12:56 +0300, Rindjunsky, Ron wrote:
> >> -	/*
> >> -	 * Warn when submitting a fragmented A-MPDU frame and drop it.
> >> -	 * This is an error and needs to be fixed elsewhere, but when
> >> -	 * done needs to take care of monitor interfaces (injection)
> >> -	 * etc.
> >> -	 */
> >> -	if (WARN_ON(tx->flags & IEEE80211_TX_CTL_AMPDU ||
> >> -		    skb_get_queue_mapping(tx->skb) >=
> >> -			ieee80211_num_regular_queues(&tx->local->hw)))
> >> -		return TX_DROP;
> >> -
> 
> > Should we keep that, just in case? If it ever happens and the hw
> rejects
> > a fragment for some reason, we'll otherwise clobber some state and
> > probably crash.
> 
> Agree. Should I resend the patch or would you just remove it from the
> original patch?

Well, you probably want to adjust the comment, since you did fix it
elsewhere :)

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]