From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mail-vb0-f43.google.com ([209.85.212.43]:53388 "EHLO mail-vb0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934649Ab3BTG0X (ORCPT ); Wed, 20 Feb 2013 01:26:23 -0500 MIME-Version: 1.0 In-Reply-To: <512437FA.2030105@lwfinger.net> References: <20130219.005206.397289032011003833.davem@davemloft.net> <5123BCFF.5090408@lwfinger.net> <20130219.131553.787630407148880340.davem@davemloft.net> <5123C43D.2020602@lwfinger.net> <512437FA.2030105@lwfinger.net> From: =?ISO-8859-1?Q?G=E1bor_Stefanik?= Date: Wed, 20 Feb 2013 07:26:01 +0100 Message-ID: (sfid-20130220_072627_408369_624BAE44) Subject: Re: [PATCH] b43: Increase number of RX DMA slots To: Larry Finger Cc: Julian Calaby , David Miller , David.Laight@aculab.com, linville@tuxdriver.com, linux-wireless@vger.kernel.org, netdev@vger.kernel.org, bittorf@bluebottle.com, stable@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Wed, Feb 20, 2013 at 3:42 AM, Larry Finger wrote: > On 02/19/2013 06:47 PM, Julian Calaby wrote: >> >> >> Is it be possible to increase the number of slots at runtime? Maybe an >> even better solution would be to keep the existing number of slots, >> and if they run out, reset and increase incrementally to some sensible >> maximum value. > > > The number could be increased a bit, but on systems with 32-bit DMA such as > the BCM4318, the maximum size of the ring buffer is 4KB. Even more > importantly, each slot is allocated an skb of 2390 bytes. Even at 256 slots, > the memory allocation is pretty large. > > Larry > Is this an issue that vendor drivers are also vulnerable to? If it is a firmware issue, I would certainly think so. From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?ISO-8859-1?Q?G=E1bor_Stefanik?= Subject: Re: [PATCH] b43: Increase number of RX DMA slots Date: Wed, 20 Feb 2013 07:26:01 +0100 Message-ID: References: <20130219.005206.397289032011003833.davem@davemloft.net> <5123BCFF.5090408@lwfinger.net> <20130219.131553.787630407148880340.davem@davemloft.net> <5123C43D.2020602@lwfinger.net> <512437FA.2030105@lwfinger.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Cc: Julian Calaby , David Miller , David.Laight-JxhZ9S5GRejQT0dZR+AlfA@public.gmane.org, linville-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org, linux-wireless-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, bittorf-yxHw+vPQXZDZJqsBc5GL+g@public.gmane.org, stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Larry Finger Return-path: In-Reply-To: <512437FA.2030105-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org> Sender: linux-wireless-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: netdev.vger.kernel.org On Wed, Feb 20, 2013 at 3:42 AM, Larry Finger wrote: > On 02/19/2013 06:47 PM, Julian Calaby wrote: >> >> >> Is it be possible to increase the number of slots at runtime? Maybe an >> even better solution would be to keep the existing number of slots, >> and if they run out, reset and increase incrementally to some sensible >> maximum value. > > > The number could be increased a bit, but on systems with 32-bit DMA such as > the BCM4318, the maximum size of the ring buffer is 4KB. Even more > importantly, each slot is allocated an skb of 2390 bytes. Even at 256 slots, > the memory allocation is pretty large. > > Larry > Is this an issue that vendor drivers are also vulnerable to? If it is a firmware issue, I would certainly think so. -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html