netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Govindarajulu Varadarajan <_govind@gmx.com>
To: David Miller <davem@davemloft.net>
Cc: _govind@gmx.com, netdev@vger.kernel.org, ssujith@cisco.com,
	benve@cisco.com
Subject: Re: [PATCH net-next v2 8/8] enic: do tx cleanup in napi poll
Date: Sun, 22 Jun 2014 12:55:30 +0530 (IST)	[thread overview]
Message-ID: <alpine.LNX.2.03.1406221250230.1311@ws.cisco> (raw)
In-Reply-To: <20140621.152440.2152576710542860161.davem@davemloft.net>



On Sat, 21 Jun 2014, David Miller wrote:
> It is recommended that TX work not count against the quota.
>
> The cost of TX packet liberation is a minute percentage of what it
> costs to process an RX frame.  Furthermore, that SKB freeing makes
> memory available for other paths in the stack.
>
> So please, just unconditionally purge the entire set of pending TX
> queue completion entries without regards to the passed in NAPI quota
> value at all.
>

So I should ignore the budget value and clean all packets possible. And what
should the NAPIs return value should be? I can not be > 64 right?

Or should I netif_napi_add() with highest value possible? For enic its 4096.

Thanks

  reply	other threads:[~2014-06-22  7:26 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-06-18  7:38 [PATCH net-next v2 0/8] enic updates Govindarajulu Varadarajan
2014-06-18  7:38 ` [PATCH net-next v2 1/8] flow_keys: Record IP layer protocol in skb_flow_dissect() Govindarajulu Varadarajan
2014-06-18  7:38 ` [PATCH net-next v2 2/8] enic: fix return value in _vnic_dev_cmd Govindarajulu Varadarajan
2014-06-18  7:38 ` [PATCH net-next v2 3/8] enic: devcmd for adding IP 5 tuple hardware filters Govindarajulu Varadarajan
2014-06-18  7:38 ` [PATCH net-next v2 4/8] enic: alloc/free rx_cpu_rmap Govindarajulu Varadarajan
2014-06-18  7:38 ` [PATCH net-next v2 5/8] enic: Add Accelerated RFS support Govindarajulu Varadarajan
2014-06-18  7:38 ` [PATCH net-next v2 6/8] enic: fix lockdep around devcmd_lock Govindarajulu Varadarajan
2014-06-18  7:38 ` [PATCH net-next v2 7/8] enic: add low latency socket busy_poll support Govindarajulu Varadarajan
2014-06-18  7:38 ` [PATCH net-next v2 8/8] enic: do tx cleanup in napi poll Govindarajulu Varadarajan
2014-06-21 22:24   ` David Miller
2014-06-22  7:25     ` Govindarajulu Varadarajan [this message]
2014-06-22 20:16       ` David Miller

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=alpine.LNX.2.03.1406221250230.1311@ws.cisco \
    --to=_govind@gmx.com \
    --cc=benve@cisco.com \
    --cc=davem@davemloft.net \
    --cc=netdev@vger.kernel.org \
    --cc=ssujith@cisco.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).