From mboxrd@z Thu Jan 1 00:00:00 1970 From: Boris Brezillon Subject: Re: [PATCH v6 5/6] media: rkvdec: Add the rkvdec driver Date: Mon, 2 Mar 2020 15:35:29 +0100 Message-ID: <20200302153529.4e2429e7@collabora.com> References: <20200220163016.21708-1-ezequiel@collabora.com> <20200220163016.21708-6-ezequiel@collabora.com> <20200302145746.3e94c1d1@coco.lan> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20200302145746.3e94c1d1@coco.lan> Sender: linux-media-owner@vger.kernel.org To: Mauro Carvalho Chehab Cc: Ezequiel Garcia , linux-media@vger.kernel.org, devicetree@vger.kernel.org, linux-rockchip@lists.infradead.org, linux-kernel@vger.kernel.org, Laurent Pinchart , Rob Herring , Tomasz Figa , Nicolas Dufresne , kernel@collabora.com, Paul Kocialkowski , Jonas Karlman , Heiko Stuebner , Sakari Ailus , Hans Verkuil List-Id: linux-rockchip.vger.kernel.org On Mon, 2 Mar 2020 14:57:46 +0100 Mauro Carvalho Chehab wrote: > > + for (j = 0; j < 3; j++) { > > Again a magic number. I guess this came from that big "M_N" table with 4 > elements, right? Please use a define instead of hardcoding a number > here. Nope, it's unrelated to the cabac_table size, it's related to how the HW accesses the P/B reflists (there are 1 P reflist, and 2 B reflists, hence the 3 here). I guess we can have a '#define NUM_REFLISTS 3'.