From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Dooks Date: Tue, 01 Apr 2014 11:25:10 +0000 Subject: Re: [PATCH 3/6] mmc: sh-mmcif: use devm_ for clock management Message-Id: <533AA216.20504@codethink.co.uk> List-Id: References: <1396347954-13740-1-git-send-email-ben.dooks@codethink.co.uk> <1396347954-13740-4-git-send-email-ben.dooks@codethink.co.uk> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Geert Uytterhoeven Cc: Linux MMC List , Linux-sh list , Laurent Pinchart , Ulf Hansson , Chris Ball , Guennadi Liakhovetski , magnus.damm@opensource.se, linux-kernel@lists.codethink.co.uk On 01/04/14 11:59, Geert Uytterhoeven wrote: > On Tue, Apr 1, 2014 at 12:25 PM, Ben Dooks wrote: >> - host->hclk = clk_get(&pdev->dev, NULL); >> + host->hclk = devm_clk_get(&pdev->dev, NULL); >> if (IS_ERR(host->hclk)) { >> ret = PTR_ERR(host->hclk); >> dev_err(&pdev->dev, "cannot get clock: %d\n", ret); >> - goto eclkget; >> + goto err_pm; >> } >> ret = sh_mmcif_clk_update(host); >> if (ret < 0) >> - goto eclkupdate; >> + goto err_clk; > > This goto doesn't look correct to me. If sh_mmcif_clk_update() failed, > the clock was not enabled nor prepared. "goto err_pm"? Yes, I think the clk_prepare_enable() call there is the only failure mode. As a note, we should probably check all systems that use it and see if we can remove the clock management code in here as the pm_runtime will also be doing this. -- Ben Dooks http://www.codethink.co.uk/ Senior Engineer Codethink - Providing Genius