Aw: Re: [BUG] MTK SoC Ethernet throughput TX only ~620Mbit/s since 6.2-rc1

[Felix Fietkau]

On 18.03.23 14:07, Frank Wunderlich wrote:
>> Gesendet: Freitag, 17. März 2023 um 19:23 Uhr
>> Von: "Frank Wunderlich" <frank-w at public-files.de>
>> An: "Daniel Golle" <daniel at makrotopia.org>
>> Cc: "Felix Fietkau" <nbd at nbd.name>, "Mark Lee" <Mark-MC.Lee at mediatek.com>, "Sean Wang" <sean.wang at mediatek.com>, "Lorenzo Bianconi" <lorenzo at kernel.org>, "David S. Miller" <davem at davemloft.net>, "Eric Dumazet" <edumazet at google.com>, "Jakub Kicinski" <kuba at kernel.org>, "Paolo Abeni" <pabeni at redhat.com>, "Matthias Brugger" <matthias.bgg at gmail.com>, "John Crispin" <john at phrozen.org>, "AngeloGioacchino Del Regno" <angelogioacchino.delregno at collabora.com>, netdev at vger.kernel.org, linux-mediatek at lists.infradead.org, linux-arm-kernel at lists.infradead.org
>> Betreff: Aw: Re: [BUG] MTK SoC Ethernet throughput TX only ~620Mbit/s since 6.2-rc1
>>
>> Hi,
>> > Gesendet: Dienstag, 14. März 2023 um 13:47 Uhr
>> > Von: "Daniel Golle" <daniel at makrotopia.org>
>> > An: "Felix Fietkau" <nbd at nbd.name>
>> > Cc: "Frank Wunderlich" <frank-w at public-files.de>, "Mark Lee" <Mark-MC.Lee at mediatek.com>, "Sean Wang" <sean.wang at mediatek.com>, "Lorenzo Bianconi" <lorenzo at kernel.org>, "David S. Miller" <davem at davemloft.net>, "Eric Dumazet" <edumazet at google.com>, "Jakub Kicinski" <kuba at kernel.org>, "Paolo Abeni" <pabeni at redhat.com>, "Matthias Brugger" <matthias.bgg at gmail.com>, "John Crispin" <john at phrozen.org>, "AngeloGioacchino Del Regno" <angelogioacchino.delregno at collabora.com>, netdev at vger.kernel.org, linux-mediatek at lists.infradead.org, linux-arm-kernel at lists.infradead.org
>> > Betreff: Re: [BUG] MTK SoC Ethernet throughput TX only ~620Mbit/s since 6.2-rc1
>> >
>> > Hi Felix,
>> > 
>> > On Tue, Mar 14, 2023 at 11:30:53AM +0100, Felix Fietkau wrote:
>> > > On 07.03.23 19:32, Frank Wunderlich wrote:
>> > > > Hi,
>> > > > 
>> > > > i've noticed that beginning on 6.2-rc1 the throughput on my Bananapi-R2 and Bananapi-R3 goes from 940Mbit/s down do 620Mbit/s since 6.2-rc1.
>> > > > Only TX (from SBC PoV) is affected, RX is still 940Mbit/s.
>> > > > 
>> > > > i bisected this to this commit:
>> > > > 
>> > > > f63959c7eec3151c30a2ee0d351827b62e742dcb ("net: ethernet: mtk_eth_soc: implement multi-queue support for per-port queues")
>> > > > 
>> > > > Daniel reported me that this is known so far and they need assistance from MTK and i should report it officially.
>> > > > 
>> > > > As far as i understand it the commit should fix problems with clients using non-GBE speeds (10/100 Mbit/s) on the Gbit-capable dsa
>> > > > interfaces (mt753x connected) behind the mac, but now the Gigabit speed is no more reached.
>> > > > I see no CRC/dropped packets, retransmitts or similar.
>> > > > 
>> > > > after reverting the commit above i get 940Mbit like in rx direction, but this will introduce the problems mentioned above so this not a complete fix.
>> > > I don't have a BPI-R2, but I tested on BPI-R3 and can't reproduce this
>> > > issue. Do you see it on all ports, or only on specific ones?
>> > 
>> > I also can't reproduce this if unsing any of the gigE ports wired via
>> > MT7531 on the R3. However, I can reproduce the issue if using a 1 GBit/s
>> > SFP module in slot SFP1 of the R3 (connected directly to GMAC2/eth1).
>> > 
>> > Users have reported this also to be a problem also on MT7622 on devices
>> > directly connecting a PHY (and not using MT7531).
>> > 
>> > In all cases, reverting the commit above fixes the issue.
>> 
>> 
>> made quick test with 6.3-rc1 on r3 without reverting the patch above and can confirm daniels test
>> 
>> it seems the problem is no more on switch-ports, but on eth1 i have massive packet loss...seems this caused by the same patch because i tested with reverted version and have no issue there.
> 
> on BPI-R2 the eth0/gmac0 (tested with wan-port) is affected. here i have in TX-Direction only 620Mbit.
> 
> I have no idea yet why there the gmac0 is affected and on r3 only gmac1.
> 
> But it looks differently...on r3 the gmac1 is nearly completely broken.
Please try this patch and let me know if it resolves the regression:

---
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -718,8 +718,6 @@ static void mtk_mac_link_up(struct phyli
  		break;
  	}
  
-	mtk_set_queue_speed(mac->hw, mac->id, speed);
-
  	/* Configure duplex */
  	if (duplex == DUPLEX_FULL)
  		mcr |= MAC_MCR_FORCE_DPX;