From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from metis.ext.pengutronix.de ([2001:6f8:1178:4:290:27ff:fe1d:cc33]) by merlin.infradead.org with esmtps (Exim 4.76 #1 (Red Hat Linux)) id 1TOk7U-0007xK-Hz for barebox@lists.infradead.org; Thu, 18 Oct 2012 07:02:18 +0000 Date: Thu, 18 Oct 2012 09:02:11 +0200 From: Sascha Hauer Message-ID: <20121018070211.GO27665@pengutronix.de> References: <1350410685-46202-1-git-send-email-carlo.caione@gmail.com> <1350410685-46202-5-git-send-email-carlo.caione@gmail.com> <20121016212416.GC27665@pengutronix.de> <7273C8F9-F3BD-4606-8C35-B370AA04C98C@gmail.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <7273C8F9-F3BD-4606-8C35-B370AA04C98C@gmail.com> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: barebox-bounces@lists.infradead.org Errors-To: barebox-bounces+u.kleine-koenig=pengutronix.de@lists.infradead.org Subject: Re: [PATCH 4/5] BCM2835: add support (arch) To: Carlo Caione Cc: "barebox@lists.infradead.org" On Thu, Oct 18, 2012 at 12:27:14AM +0200, Carlo Caione wrote: > > On Oct 16, 2012, at 11:24 PM, Sascha Hauer wrote: > > > On Tue, Oct 16, 2012 at 08:04:44PM +0200, Carlo Caione wrote: > >> + > >> +static struct clk ref_3_clk = { > >> + .rate = 3 * 1000 * 1000, > >> +}; > >> + > >> +static struct clk ref_1_clk = { > >> + .rate = 1 * 1000 * 1000, > >> +}; > > > > Add a 'select COMMON_CLK' to your Kconfig and do a: > > > > clk_dummy = clk_fixed("dummy", 0); > > clk_ref_3 = clk_fixed("ref3", 3 * 1000 * 1000); > > clk_ref_1 = clk_fixed("ref1", 1 * 1000 * 1000); > > > > Then you can drop your clk code and get a clk_dump command > > which you can use to see what clocks you have. > > RFC > > ok I ask here before submitting the whole set because I'm not sure if this is a convenient solution. > I'm not sure if you are suggesting to have two different clock definition one for clk_fixed and the other one for clkdev, > so I was thinking something like: > > enum brcm_clks { > dummy, clk_ref_3, clk_ref_1, clks_max > }; > > static struct clk *clks[clks_max]; > > int brmc_clk_create(struct clk *clk, const char *con_id, const char *dev_id) > { > struct clk_lookup *clkdev; > > clkdev = clkdev_alloc(clk, con_id, dev_id); > if (!clkdev) > return -ENOMEM; > clkdev_add(clkdev); > return 0; > } > > static int bcm2835_clk_init(void) > { > int ret; > > clks[dummy] = clk_fixed("dummy", 0); > ret = brmc_clk_create(clks[dummy], "apb_pclk", NULL); > if (ret) > goto clk_err; > > clks[clk_ref_3] = clk_fixed("ref3", 3 * 1000 * 1000); > ret = brmc_clk_create(clks[clk_ref_3], NULL, "uart0-pl0110"); > if (ret) > goto clk_err; > > clks[clk_ref_1] = clk_fixed("ref1", 1 * 1000 * 1000); > ret = brmc_clk_create(clks[clk_ref_1], NULL, "bcm2835-cs"); > if (ret) > goto clk_err; The above is basically what I was thinking of. The only thing I see that this *looks* like there is exactly one lookup per clock. This indeed is the case currenty here, but this does not have to be true. A clock may have multiple lookups, for example the ref3 clk may also be used by the lcd controller or something else. So I suggest to first register all clocks and the lookups afterwards. What you have as brmc_clk_create could well be a generic helper function, it will be useful for others aswell. BTW error checking here is a good idea, but a rollback of the allocated resources will be unnecessary as you are doomed anyway when something fails here. Sascha -- Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | _______________________________________________ barebox mailing list barebox@lists.infradead.org http://lists.infradead.org/mailman/listinfo/barebox