Re: ieee80211 phy0: rt2x00queue_write_tx_frame: Error - Dropping frame due to full tx queue...?

From: Stanislaw Gruszka <sgruszka@redhat.com>
To: Kofi Agor <nenekofi@gmail.com>, Daniel Golle <daniel@makrotopia.org>
Cc: Enrico Mioso <mrkiko.rs@gmail.com>,
	Craig Matsuura <cmatsuura@vivint.com>,
	Mathias Kresin <dev@kresin.me>,
	Tom Psyborg <pozega.tomislav@gmail.com>,
	linux-wireless <linux-wireless@vger.kernel.org>,
	John Crispin <john@phrozen.org>, Felix Fietkau <nbd@nbd.name>,
	Jamie Stuart <jamie@onebillion.org>
Subject: Re: ieee80211 phy0: rt2x00queue_write_tx_frame: Error - Dropping frame due to full tx queue...?
Date: Wed, 15 Aug 2018 13:40:30 +0200	[thread overview]
Message-ID: <20180815114029.GA1862@redhat.com> (raw)
In-Reply-To: <20180528125039.GA22122@redhat.com>

On Mon, May 28, 2018 at 02:50:39PM +0200, Stanislaw Gruszka wrote:
> I have some updates here.

And now more updates. I have 5 patches to test here:
https://github.com/sgruszka/wireless-drivers-next/commits/rt2800-draft-v2

as reported by T-Bone here:
https://bugzilla.kernel.org/show_bug.cgi?id=82751
first 3 patches made Netgear WN3000RPv3 router workable.
Another 2 should be further improvement, but were not tested broadly
and can have some bugs.

Daniel, could you apply this on your staging tree for testing?
Please also remove:
600-23-rt2x00-rt2800mmio-add-a-workaround-for-spurious-TX_F.patch
991-rt2800_change_rx_ampdu_factor.patch
992-rt2800_change_ba_size.patch
993-rt2800_change_rx_ampdu_density.patch
Those are not needed and can be harmful with the test patches,
(especially spurious interrupt one, patches will not apply cleanly 
with it).

Patches should make "Dropping frame due to full tx queue" and
"Queue 2 failed to flush" errors gone. However if device is somewhat
wrongly configured by driver at low level (i.e. wrong MAC/BBP/RF
registers programing) there still will be troubles to have stable
and fast wireless connection.

On TODO site, the proper watchdog should be added, but I'm not sure
how to exactly detect device/firmware hung and how exactly device should
be rested (by some HW reset by register or by doing full reinitialization).

Another thing is fixing RATE_PROBE frames which are aggregated with
other frames and not sent at requested rate. I implemented qsel queue patch
similar to mt76, but this not work as expected on older Ralink chips.  
https://github.com/sgruszka/wireless-drivers-next/commit/846d205edd8c36d1b7828fee54bf4cf40bf8cb1a

Any help with those 2 problems are welcome.

Thanks
Stanislaw