linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Bjørn Mork" <bjorn@mork.no>
To: Jim Baxter <jim_baxter@mentor.com>
Cc: <linux-usb@vger.kernel.org>, <netdev@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>, Oliver Neukum <oliver@neukum.org>
Subject: Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low
Date: Tue, 16 May 2017 20:24:10 +0200	[thread overview]
Message-ID: <87shk4fynp.fsf@miraculix.mork.no> (raw)
In-Reply-To: <1494956480-6127-2-git-send-email-jim_baxter@mentor.com> (Jim Baxter's message of "Tue, 16 May 2017 18:41:20 +0100")

Jim Baxter <jim_baxter@mentor.com> writes:

> The CDC-NCM driver can require large amounts of memory to create
> skb's and this can be a problem when the memory becomes fragmented.
>
> This especially affects embedded systems that have constrained
> resources but wish to maximise the throughput of CDC-NCM with 16KiB
> NTB's.
>
> The issue is after running for a while the kernel memory can become
> fragmented and it needs compacting.
> If the NTB allocation is needed before the memory has been compacted
> the atomic allocation can fail which can cause increased latency,
> large re-transmissions or disconnections depending upon the data
> being transmitted at the time.
> This situation occurs for less than a second until the kernel has
> compacted the memory but the failed devices can take a lot longer to
> recover from the failed TX packets.
>
> To ease this temporary situation I modified the CDC-NCM TX path to
> temporarily switch into a reduced memory mode which allocates an NTB
> that will fit into a USB_CDC_NCM_NTB_MIN_OUT_SIZE (default 2048 Bytes)
> sized memory block and only transmit NTB's with a single network frame
> until the memory situation is resolved.
> Once the memory is compacted the CDC-NCM data can resume transmitting
> at the normal tx_max rate once again.

I must say that I don't like the additional complexity added here.  If
there are memory issues and you can reduce the buffer size to
USB_CDC_NCM_NTB_MIN_OUT_SIZE, then why don't you just set a lower tx_max
buffer size in the first place?

  echo 2048 > /sys/class/net/wwan0/cdc_ncm/tx_max




Bjørn

  reply	other threads:[~2017-05-16 18:24 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-16 17:41 [RFC V1 0/1] Reduce cdc_ncm memory use when kernel memory low Jim Baxter
2017-05-16 17:41 ` [RFC V1 1/1] net: cdc_ncm: Reduce " Jim Baxter
2017-05-16 18:24   ` Bjørn Mork [this message]
2017-05-17  7:44     ` Oliver Neukum
2017-05-17 10:56       ` Baxter, Jim
2017-05-17 18:18     ` David Miller
2017-05-18 10:01       ` Oliver Neukum
2017-05-22 15:45       ` Baxter, Jim
2017-05-22 15:54         ` David Miller
2017-05-23  8:42           ` Oliver Neukum
2017-05-23 15:26             ` David Miller
2017-05-23 19:06               ` Baxter, Jim
     [not found]                 ` <1497263047.15677.13.camel@suse.com>
2017-06-12 12:32                   ` Baxter, Jim
2017-05-19 11:10   ` David Laight
2017-05-19 13:55     ` Bjørn Mork
2017-05-19 14:46       ` David Laight
2017-05-22 13:27         ` Oliver Neukum

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=87shk4fynp.fsf@miraculix.mork.no \
    --to=bjorn@mork.no \
    --cc=jim_baxter@mentor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=oliver@neukum.org \
    /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).