From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wolfgang Grandegger Subject: Re: [PATCH] CAN: Add Flexcan CAN controller driver Date: Thu, 15 Jul 2010 12:21:30 +0200 Message-ID: <4C3EE12A.1030307@grandegger.com> References: <1279144811-12251-1-git-send-email-mkl@pengutronix.de> <4C3ED6F5.7040606@grandegger.com> <4C3EDDB1.5040109@pengutronix.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org, netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Marc Kleine-Budde Return-path: In-Reply-To: <4C3EDDB1.5040109-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: socketcan-core-bounces-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org Errors-To: socketcan-core-bounces-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org List-Id: netdev.vger.kernel.org On 07/15/2010 12:06 PM, Marc Kleine-Budde wrote: > Hey Wolfgang, > > Wolfgang Grandegger wrote: >>> This core is found on some Freescale SoCs and also some Coldfire >>> SoCs. Support for Coldfire is missing though at the moment as >>> they have an older revision of the core which does not have RX FIFO >>> support. >>> >>> Signed-off-by: Sascha Hauer >>> Signed-off-by: Marc Kleine-Budde >>> --- >>> >>> Changes to prev version: >>> * The is now GPLv2 (only) as no one complained. >>> >>> The patch applies to current net-next-2.6/master. >>> If there aren't any objections please consider applying this patch. >>> Wolfgang, can I an Acked-by? >> >> I already had a look to the previous version and I realized that >> accessing reg_esr is racy. I will dig out my notes and come up with a >> full review later today or tomorrow. > > Let me have a look.... > > I think I should remove the read reg_esr in "flexcan_irq_err()" and use > the value read in "flexcan_irq()" > > Without the fix, there's a race window that we loose some error bits > from the interrupt handler to the napi function, because the error bits > are cleared on read. Right, that is my concern. reg_esr should only be read and handled *once*. Currently it's read *3* times! I would read it in the ISR and handle it in the NAPI poll function, including state change notification. But I need a closer look... > Regarding the upcoming improvement of the can error frames upon state > changes, I'd first like to get this driver merged and then discuss a > solution for the error frames. Fine for me. I'm currently preparing a RFC patch for SJA1000 and MSCAN. Wolfgang.