From: Felix Fietkau <nbd@nbd.name>
To: Ryder Lee <ryder.lee@mediatek.com>,
Johannes Berg <johannes.berg@intel.com>
Cc: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>,
Shayne Chen <shayne.chen@mediatek.com>,
Evelyn Tsai <evelyn.tsai@mediatek.com>,
Bo Jiao <bo.jiao@mediatek.com>,
Sujuan Chen <sujuan.chen@mediatek.com>,
linux-wireless@vger.kernel.org,
linux-mediatek@lists.infradead.org
Subject: Re: [PATCH v2 1/4] mac80211: call ieee80211_tx_h_rate_ctrl() when dequeue
Date: Thu, 17 Jun 2021 19:59:26 +0200 [thread overview]
Message-ID: <4d652f4a-c258-f248-911f-28238531ae4d@nbd.name> (raw)
In-Reply-To: <a89137bf-401b-c9d0-8848-d79ea9e2a9c3@nbd.name>
On 2021-06-17 12:50, Felix Fietkau wrote:
>
> On 2021-05-28 08:05, Ryder Lee wrote:
>> Make ieee80211_tx_h_rate_ctrl() get called on dequeue to improve
>> performance since it reduces the turnaround time for rate control.
>>
>> Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
>> ---
>> change since v2 - roll ieee80211_tx_h_rate_ctrl checks into one condition
> There were some OpenWrt crash reported which appear to be related to
> this patch. I was able to reproduce a deadlock with ath9k, and I'm
> currently looking into it.
Some more information about the crash:
- ath9k calls ieee80211_tx_dequeue with the tx queue lock held
- ieee80211_tx_dequeue calls minstrel get_rate
- get_rate calls minstrel_aggr_check
- minstrel_aggr check calls ieee80211_start_tx_ba_session
- ieee80211_start_tx_ba_session tries to send a frame
- ath9k tries to acquire the tx lock it already holds
I've fixed this in v3 by moving the logic of minstrel_aggr_check into
mac80211
- Felix
WARNING: multiple messages have this Message-ID (diff)
From: Felix Fietkau <nbd@nbd.name>
To: Ryder Lee <ryder.lee@mediatek.com>,
Johannes Berg <johannes.berg@intel.com>
Cc: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>,
Shayne Chen <shayne.chen@mediatek.com>,
Evelyn Tsai <evelyn.tsai@mediatek.com>,
Bo Jiao <bo.jiao@mediatek.com>,
Sujuan Chen <sujuan.chen@mediatek.com>,
linux-wireless@vger.kernel.org,
linux-mediatek@lists.infradead.org
Subject: Re: [PATCH v2 1/4] mac80211: call ieee80211_tx_h_rate_ctrl() when dequeue
Date: Thu, 17 Jun 2021 19:59:26 +0200 [thread overview]
Message-ID: <4d652f4a-c258-f248-911f-28238531ae4d@nbd.name> (raw)
In-Reply-To: <a89137bf-401b-c9d0-8848-d79ea9e2a9c3@nbd.name>
On 2021-06-17 12:50, Felix Fietkau wrote:
>
> On 2021-05-28 08:05, Ryder Lee wrote:
>> Make ieee80211_tx_h_rate_ctrl() get called on dequeue to improve
>> performance since it reduces the turnaround time for rate control.
>>
>> Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
>> ---
>> change since v2 - roll ieee80211_tx_h_rate_ctrl checks into one condition
> There were some OpenWrt crash reported which appear to be related to
> this patch. I was able to reproduce a deadlock with ath9k, and I'm
> currently looking into it.
Some more information about the crash:
- ath9k calls ieee80211_tx_dequeue with the tx queue lock held
- ieee80211_tx_dequeue calls minstrel get_rate
- get_rate calls minstrel_aggr_check
- minstrel_aggr check calls ieee80211_start_tx_ba_session
- ieee80211_start_tx_ba_session tries to send a frame
- ath9k tries to acquire the tx lock it already holds
I've fixed this in v3 by moving the logic of minstrel_aggr_check into
mac80211
- Felix
_______________________________________________
Linux-mediatek mailing list
Linux-mediatek@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-mediatek
next prev parent reply other threads:[~2021-06-17 17:59 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-28 6:05 [PATCH v2 1/4] mac80211: call ieee80211_tx_h_rate_ctrl() when dequeue Ryder Lee
2021-05-28 6:05 ` Ryder Lee
2021-05-28 6:05 ` [PATCH v2 2/4] mac80211: minstrel_ht: fix minstrel_aggr_check for encap offload Ryder Lee
2021-05-28 6:05 ` Ryder Lee
2021-05-28 6:05 ` [PATCH v2 3/4] mac80211: add rate control support " Ryder Lee
2021-05-28 6:05 ` Ryder Lee
2021-05-28 10:46 ` Arend van Spriel
2021-05-28 10:46 ` Arend van Spriel
2021-05-28 6:05 ` [PATCH v2 4/4] mt76: mt7615: add 802.11 encap offload support Ryder Lee
2021-05-28 6:05 ` Ryder Lee
2021-06-10 11:10 ` [PATCH v2 1/4] mac80211: call ieee80211_tx_h_rate_ctrl() when dequeue Felix Fietkau
2021-06-10 11:10 ` Felix Fietkau
2021-06-16 22:45 ` Ryder Lee
2021-06-16 22:45 ` Ryder Lee
2021-06-17 10:50 ` Felix Fietkau
2021-06-17 10:50 ` Felix Fietkau
2021-06-17 17:59 ` Felix Fietkau [this message]
2021-06-17 17:59 ` Felix Fietkau
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=4d652f4a-c258-f248-911f-28238531ae4d@nbd.name \
--to=nbd@nbd.name \
--cc=bo.jiao@mediatek.com \
--cc=evelyn.tsai@mediatek.com \
--cc=johannes.berg@intel.com \
--cc=linux-mediatek@lists.infradead.org \
--cc=linux-wireless@vger.kernel.org \
--cc=lorenzo.bianconi@redhat.com \
--cc=ryder.lee@mediatek.com \
--cc=shayne.chen@mediatek.com \
--cc=sujuan.chen@mediatek.com \
/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.