[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <20160115.113430.2270570646723842525.davem@davemloft.net>
Date: Fri, 15 Jan 2016 11:34:30 -0500 (EST)
From: David Miller <davem@...emloft.net>
To: david.vrabel@...rix.com
Cc: netdev@...r.kernel.org, wei.liu2@...rix.com,
ian.campbell@...rix.com, xen-devel@...ts.xenproject.org
Subject: Re: [Xen-devel] [PATCHv1 net] xen-netback: use skb to determine
number of required guest Rx requests
From: David Vrabel <david.vrabel@...rix.com>
Date: Fri, 15 Jan 2016 10:31:57 +0000
> On 14/01/16 21:54, David Miller wrote:
>> From: David Vrabel <david.vrabel@...rix.com>
>> Date: Thu, 14 Jan 2016 15:18:30 +0000
>>
>>> - needed = xenvif_rx_ring_slots_needed(queue->vif);
>>> + skb = skb_peek(&queue->rx_queue);
>>> + if (!skb)
>>> + return false;
>>> +
>>> + needed = DIV_ROUND_UP(skb->len, XEN_PAGE_SIZE);
>>> + if (skb_is_gso(skb))
>>> + needed++;
>>
>> If I am not mistaken, we moved away from this kind of test exactly because
>> it is inaccurate and may under-estimate the needs.
>>
>> It is possible for an N byte SKB to require N segments. Therefore, the:
>>
>> DIV_ROUND_UP(skb->len, XEN_PAGE_SIZE);
>>
>> calculation doesn't cut it.
>
> After 1650d5455bd2dc6b5ee134bd6fc1a3236c266b5b (xen-netback: always
> fully coalesce guest Rx packets) we always fully pack a packet into its
> guest Rx slots. Each slot has space for XEN_PAGE_SIZE bytes so this
> calculation for the number of slots is correct.
>
> Shall I resend with a more description changelog?
Yeah that would help a lot.
Powered by blists - more mailing lists