[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <497267937.1868067.1416501088558.JavaMail.zimbra@redhat.com>
Date: Thu, 20 Nov 2014 11:31:28 -0500 (EST)
From: Pankaj Gupta <pagupta@...hat.com>
To: Cong Wang <cwang@...pensource.com>
Cc: linux-kernel@...r.kernel.org, netdev <netdev@...r.kernel.org>,
David Miller <davem@...emloft.net>,
Jason Wang <jasowang@...hat.com>,
"Michael S. Tsirkin" <mst@...hat.com>, dgibson@...hat.com,
vfalico@...il.com, Eric Dumazet <edumazet@...gle.com>,
Vladislav Yasevich <vyasevic@...hat.com>,
Jerry Chu <hkchu@...gle.com>, wuzhy@...hat.com,
Pavel Emelyanov <xemul@...allels.com>,
Tom Herbert <therbert@...gle.com>, bhutchings@...arflare.com,
xii@...gle.com, Stephen Hemminger <stephen@...workplumber.org>,
Jiří Pírko <jiri@...nulli.us>,
sergei shtylyov <sergei.shtylyov@...entembedded.com>
Subject: Re: [PATCH net-next 1/4] net: allow large number of rx queues
> On Tue, Nov 18, 2014 at 8:22 AM, Pankaj Gupta <pagupta@...hat.com> wrote:
> >
> > As vmalloc() adds overhead on a critical network path,
> > __GFP_REPEAT flag is used with kzalloc() to do this fallback
> > only when really needed.
> >
>
> Are you sure we need __GFP_REPEAT? We have vmalloc() as
> fallback of kmalloc() in many places of networking (grep kvfree),
> none of those I checked has this flag set.
Its there in netif_alloc_netdev_queues() function in same file.
Also, I found it some other places as well.
I think its good to have.
>
--
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