linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Brad Griffis <bgriffis@nvidia.com>
To: Furong Xu <0x1207@gmail.com>, Jon Hunter <jonathanh@nvidia.com>
Cc: netdev@vger.kernel.org, linux-stm32@st-md-mailman.stormreply.com,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org,
	Alexander Lobakin <aleksander.lobakin@intel.com>,
	Joe Damato <jdamato@fastly.com>,
	Andrew Lunn <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	Maxime Coquelin <mcoquelin.stm32@gmail.com>,
	xfr@outlook.com,
	"linux-tegra@vger.kernel.org" <linux-tegra@vger.kernel.org>
Subject: Re: [PATCH net-next v3 1/4] net: stmmac: Switch to zero-copy in non-XDP RX path
Date: Thu, 23 Jan 2025 11:53:21 -0800	[thread overview]
Message-ID: <e6305e71-5633-48bf-988d-fa2886e16aae@nvidia.com> (raw)
In-Reply-To: <20250124003501.5fff00bc@orangepi5-plus>

On 1/23/25 08:35, Furong Xu wrote:
> What is the MTU of Tegra234 and NFS server? Are they both 1500?

I see the same issue.  Yes, both are 1500.

> Could you please try attached patch to confirm if this regression is
> fixed?

Patch fixes the issue.

> If the attached patch fixes this regression, and so it seems to be a
> cache coherence issue specific to Tegra234, since this patch avoid
> memcpy and the page buffers may be modified by upper network stack of
> course, then cache lines of page buffers may become dirty. But by
> reverting this patch, cache lines of page buffers never become dirty,
> this is the core difference.

Thanks for these insights. I don't have specific experience in this 
driver, but I see we have dma-coherent turned on for this driver in our 
downstream device tree files (i.e. dtbs that coincide with our 
out-of-tree implementation of this driver).  I went back to the original 
code and verified that the issue was there. I did a new test where I 
added dma-coherent to this ethernet node in the dtb and retested. It worked!

Just to clarify, the patch that you had us try was not intended as an 
actual fix, correct? It was only for diagnostic purposes, i.e. to see if 
there is some kind of cache coherence issue, which seems to be the case? 
  So perhaps the only fix needed is to add dma-coherent to our device tree?


  reply	other threads:[~2025-01-23 19:56 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-15  3:27 [PATCH net-next v3 0/4] net: stmmac: RX performance improvement Furong Xu
2025-01-15  3:27 ` [PATCH net-next v3 1/4] net: stmmac: Switch to zero-copy in non-XDP RX path Furong Xu
2025-01-15 16:58   ` Larysa Zaremba
2025-01-16  2:05   ` Yanteng Si
2025-01-23 14:06   ` Jon Hunter
2025-01-23 16:35     ` Furong Xu
2025-01-23 19:53       ` Brad Griffis [this message]
2025-01-23 21:48         ` Andrew Lunn
2025-01-24  2:42           ` Furong Xu
2025-01-24 13:15             ` Thierry Reding
2025-01-28 20:04               ` Lucas Stach
2025-01-25 10:20             ` Ido Schimmel
2025-01-25 14:43               ` Furong Xu
2025-01-26  8:41                 ` Ido Schimmel
2025-01-26 10:37                   ` Furong Xu
2025-01-26 11:35                     ` Ido Schimmel
2025-01-26 12:56                       ` Furong Xu
2025-01-25 15:03               ` Furong Xu
2025-01-25 19:08                 ` Andrew Lunn
2025-01-26  2:39                   ` Furong Xu
2025-01-27 13:28                 ` Thierry Reding
2025-01-29 14:51                   ` Jon Hunter
2025-02-07  9:07                     ` Furong Xu
2025-02-07 13:42                       ` Jon Hunter
2025-01-24  1:53         ` Furong Xu
2025-01-24 15:14           ` Andrew Lunn
2025-01-15  3:27 ` [PATCH net-next v3 2/4] net: stmmac: Set page_pool_params.max_len to a precise size Furong Xu
2025-01-15 10:07   ` Yanteng Si
2025-01-15  3:27 ` [PATCH net-next v3 3/4] net: stmmac: Optimize cache prefetch in RX path Furong Xu
2025-01-15 16:24   ` Yanteng Si
2025-01-15  3:27 ` [PATCH net-next v3 4/4] net: stmmac: Convert prefetch() to net_prefetch() for received frames Furong Xu
2025-01-15 16:33   ` Yanteng Si
2025-01-15 16:35   ` Larysa Zaremba
2025-01-15 17:35   ` Joe Damato
2025-01-16 11:40 ` [PATCH net-next v3 0/4] net: stmmac: RX performance improvement patchwork-bot+netdevbpf

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=e6305e71-5633-48bf-988d-fa2886e16aae@nvidia.com \
    --to=bgriffis@nvidia.com \
    --cc=0x1207@gmail.com \
    --cc=aleksander.lobakin@intel.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=jdamato@fastly.com \
    --cc=jonathanh@nvidia.com \
    --cc=kuba@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-stm32@st-md-mailman.stormreply.com \
    --cc=linux-tegra@vger.kernel.org \
    --cc=mcoquelin.stm32@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=xfr@outlook.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).