All of lore.kernel.org
 help / color / mirror / Atom feed
From: Felix Fietkau <nbd@nbd.name>
To: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com>,
	John Crispin <john@phrozen.org>,
	Sean Wang <sean.wang@mediatek.com>,
	Mark Lee <Mark-MC.Lee@mediatek.com>,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-mediatek@lists.infradead.org
Subject: Re: [PATCH net-next 05/14] net: ethernet: mtk_eth_soc: reduce MDIO bus access latency
Date: Thu, 22 Apr 2021 14:33:31 +0200	[thread overview]
Message-ID: <d96206db-96e2-1eb7-6b19-47c9596ccfea@nbd.name> (raw)
In-Reply-To: <20210422040914.47788-6-ilya.lipnitskiy@gmail.com>


On 2021-04-22 06:09, Ilya Lipnitskiy wrote:
> From: Felix Fietkau <nbd@nbd.name>
> 
> usleep_range often ends up sleeping much longer than the 10-20us provided
> as a range here. This causes significant latency in mdio bus acceses,
> which easily adds multiple seconds to the boot time on MT7621 when polling
> DSA slave ports.
> 
> Use udelay via readx_poll_timeout_atomic, since the MDIO access does not
> take much time
> 
> Signed-off-by: Felix Fietkau <nbd@nbd.name>
> [Ilya: use readx_poll_timeout_atomic instead of cond_resched]
I still prefer the cond_resched() variant. On a fully loaded system, I'd
prefer to let the MDIO access take longer instead of wasting cycles on
udelay.

- Felix

WARNING: multiple messages have this Message-ID (diff)
From: Felix Fietkau <nbd@nbd.name>
To: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com>,
	John Crispin <john@phrozen.org>,
	Sean Wang <sean.wang@mediatek.com>,
	Mark Lee <Mark-MC.Lee@mediatek.com>,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-mediatek@lists.infradead.org
Subject: Re: [PATCH net-next 05/14] net: ethernet: mtk_eth_soc: reduce MDIO bus access latency
Date: Thu, 22 Apr 2021 14:33:31 +0200	[thread overview]
Message-ID: <d96206db-96e2-1eb7-6b19-47c9596ccfea@nbd.name> (raw)
In-Reply-To: <20210422040914.47788-6-ilya.lipnitskiy@gmail.com>


On 2021-04-22 06:09, Ilya Lipnitskiy wrote:
> From: Felix Fietkau <nbd@nbd.name>
> 
> usleep_range often ends up sleeping much longer than the 10-20us provided
> as a range here. This causes significant latency in mdio bus acceses,
> which easily adds multiple seconds to the boot time on MT7621 when polling
> DSA slave ports.
> 
> Use udelay via readx_poll_timeout_atomic, since the MDIO access does not
> take much time
> 
> Signed-off-by: Felix Fietkau <nbd@nbd.name>
> [Ilya: use readx_poll_timeout_atomic instead of cond_resched]
I still prefer the cond_resched() variant. On a fully loaded system, I'd
prefer to let the MDIO access take longer instead of wasting cycles on
udelay.

- Felix

_______________________________________________
Linux-mediatek mailing list
Linux-mediatek@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-mediatek

WARNING: multiple messages have this Message-ID (diff)
From: Felix Fietkau <nbd@nbd.name>
To: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com>,
	John Crispin <john@phrozen.org>,
	Sean Wang <sean.wang@mediatek.com>,
	Mark Lee <Mark-MC.Lee@mediatek.com>,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-mediatek@lists.infradead.org
Subject: Re: [PATCH net-next 05/14] net: ethernet: mtk_eth_soc: reduce MDIO bus access latency
Date: Thu, 22 Apr 2021 14:33:31 +0200	[thread overview]
Message-ID: <d96206db-96e2-1eb7-6b19-47c9596ccfea@nbd.name> (raw)
In-Reply-To: <20210422040914.47788-6-ilya.lipnitskiy@gmail.com>


On 2021-04-22 06:09, Ilya Lipnitskiy wrote:
> From: Felix Fietkau <nbd@nbd.name>
> 
> usleep_range often ends up sleeping much longer than the 10-20us provided
> as a range here. This causes significant latency in mdio bus acceses,
> which easily adds multiple seconds to the boot time on MT7621 when polling
> DSA slave ports.
> 
> Use udelay via readx_poll_timeout_atomic, since the MDIO access does not
> take much time
> 
> Signed-off-by: Felix Fietkau <nbd@nbd.name>
> [Ilya: use readx_poll_timeout_atomic instead of cond_resched]
I still prefer the cond_resched() variant. On a fully loaded system, I'd
prefer to let the MDIO access take longer instead of wasting cycles on
udelay.

- Felix

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  parent reply	other threads:[~2021-04-22 12:33 UTC|newest]

Thread overview: 69+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-22  4:09 [PATCH net-next 00/14] mtk_eth_soc: fixes and performance improvements Ilya Lipnitskiy
2021-04-22  4:09 ` Ilya Lipnitskiy
2021-04-22  4:09 ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 01/14] net: ethernet: mtk_eth_soc: fix RX VLAN offload Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 02/14] net: ethernet: mtk_eth_soc: unmap RX data before calling build_skb Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 03/14] net: ethernet: mtk_eth_soc: fix build_skb cleanup Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 04/14] net: ethernet: mtk_eth_soc: use napi_consume_skb Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 05/14] net: ethernet: mtk_eth_soc: reduce MDIO bus access latency Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22 12:18   ` Andrew Lunn
2021-04-22 12:18     ` Andrew Lunn
2021-04-22 12:18     ` Andrew Lunn
2021-04-22 12:33   ` Felix Fietkau [this message]
2021-04-22 12:33     ` Felix Fietkau
2021-04-22 12:33     ` Felix Fietkau
2021-04-23  4:16     ` Ilya Lipnitskiy
2021-04-23  4:16       ` Ilya Lipnitskiy
2021-04-23  4:16       ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 06/14] net: ethernet: mtk_eth_soc: remove unnecessary TX queue stops Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 07/14] net: ethernet: mtk_eth_soc: use larger burst size for QDMA TX Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 08/14] net: ethernet: mtk_eth_soc: increase DMA ring sizes Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 09/14] net: ethernet: mtk_eth_soc: implement dynamic interrupt moderation Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 10/14] net: ethernet: mtk_eth_soc: cache HW pointer of last freed TX descriptor Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 11/14] net: ethernet: mtk_eth_soc: only read the full RX descriptor if DMA is done Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 12/14] net: ethernet: mtk_eth_soc: reduce unnecessary interrupts Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22 16:26   ` Jakub Kicinski
2021-04-22 16:26     ` Jakub Kicinski
2021-04-22 16:26     ` Jakub Kicinski
2021-04-23  4:20     ` Ilya Lipnitskiy
2021-04-23  4:20       ` Ilya Lipnitskiy
2021-04-23  4:20       ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 13/14] net: ethernet: mtk_eth_soc: set PPE flow hash as skb hash if present Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09 ` [PATCH net-next 14/14] net: ethernet: mtk_eth_soc: use iopoll.h macro for DMA init Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22  4:09   ` Ilya Lipnitskiy
2021-04-22 12:20   ` Andrew Lunn
2021-04-22 12:20     ` Andrew Lunn
2021-04-22 12:20     ` Andrew Lunn
2021-04-22 12:23 ` [PATCH net-next 00/14] mtk_eth_soc: fixes and performance improvements Andrew Lunn
2021-04-22 12:23   ` Andrew Lunn
2021-04-22 12:23   ` Andrew Lunn
2021-04-23  4:18   ` Ilya Lipnitskiy
2021-04-23  4:18     ` Ilya Lipnitskiy
2021-04-23  4:18     ` Ilya Lipnitskiy

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=d96206db-96e2-1eb7-6b19-47c9596ccfea@nbd.name \
    --to=nbd@nbd.name \
    --cc=Mark-MC.Lee@mediatek.com \
    --cc=davem@davemloft.net \
    --cc=ilya.lipnitskiy@gmail.com \
    --cc=john@phrozen.org \
    --cc=kuba@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=matthias.bgg@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=sean.wang@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.