[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAOsiSVXcKc583rEb9hXhtuhpTtrQ+ib_pgxaX9h77v98BWfgKg@mail.gmail.com>
Date: Mon, 25 Mar 2013 17:06:01 +0000
From: Wei Liu <liuw@...w.name>
To: David Vrabel <david.vrabel@...rix.com>
Cc: Wei Liu <wei.liu2@...rix.com>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"annie.li@...cle.com" <annie.li@...cle.com>,
"konrad.wilk@...cle.com" <konrad.wilk@...cle.com>,
Ian Campbell <Ian.Campbell@...rix.com>,
"xen-devel@...ts.xen.org" <xen-devel@...ts.xen.org>
Subject: Re: [Xen-devel] [PATCH 5/6] xen-netback: coalesce slots before copying
On Mon, Mar 25, 2013 at 4:58 PM, Wei Liu <liuw@...w.name> wrote:
>
> So it is a bit tricky here. Let me clarify this, the head field is
> used to indicate the start of a new tx requests queue and the end of
> previous queue.
>
> Imagine a sequence of head fileds(I = INVALID_PENDING_RING_IDX below),
> the number is the starting index of pending ring.
>
> .... 0 I I I 5 I I ...
>
> consume all tx_info but not setting I to 0 (or any number other then
> I) makes the sequence remains the same as before. The in subsequent
> call to process next SKB, which has 3 extra slots, which makes the
Sorry, should be "4 extra slots"
sequence looks like
.... 8 | | | | | | ...
but in fact the correct sequence should be
.... 8 | | | | 0 0 ...
Wei.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists