[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20240619.130337.398996009719520372.fujita.tomonori@gmail.com>
Date: Wed, 19 Jun 2024 13:03:37 +0900 (JST)
From: FUJITA Tomonori <fujita.tomonori@...il.com>
To: kuba@...nel.org
Cc: fujita.tomonori@...il.com, netdev@...r.kernel.org, andrew@...n.ch,
horms@...nel.org, jiri@...nulli.us, pabeni@...hat.com,
linux@...linux.org.uk, hfdevel@....net, naveenm@...vell.com,
jdamato@...tly.com
Subject: Re: [PATCH net-next v11 5/7] net: tn40xx: add basic Rx handling
On Tue, 18 Jun 2024 18:52:10 -0700
Jakub Kicinski <kuba@...nel.org> wrote:
> On Tue, 18 Jun 2024 14:16:06 +0900 FUJITA Tomonori wrote:
>> + skb = napi_build_skb(page_address(dm->page), PAGE_SIZE);
>> + if (!skb) {
>> + netdev_err(priv->ndev, "napi_build_skb() failed\n");
>
> memory pressure happens a lot in real world scenarios,
> allocations will fail, and you don't want to spam the logs
> in such cases.
>
> In general prints on the datapath can easily turn into a DoS vector.
> You're better off using appropriate statistics (here struct
> netdev_queue_stats_rx :: alloc_fail)
Removed prints on the datapath.
I'll add supporting get_queue_stats_tx/rx to the to-do list after
merged.
Powered by blists - more mailing lists