From: Andrew Lunn Subject: Re: crypto: marvell/CESA: Issues with non cache-line aligned buffers Date: Fri, 3 Jul 2015 16:41:16 +0200 Message-ID: <20150703144116.GK13481@lunn.ch> References: <20150703114305.15611807@bbrezillon> <20150703131059.GX7557@n2100.arm.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Boris Brezillon , Thomas Petazzoni , Herbert Xu , Arnaud Ebalard , Will Deacon , "linux-crypto@vger.kernel.org" , Gregory CLEMENT , "linux-arm-kernel@lists.infradead.org" To: Russell King - ARM Linux Return-path: Received: from vps0.lunn.ch ([178.209.37.122]:33675 "EHLO vps0.lunn.ch" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932206AbbGCOrc (ORCPT ); Fri, 3 Jul 2015 10:47:32 -0400 Content-Disposition: inline In-Reply-To: <20150703131059.GX7557@n2100.arm.linux.org.uk> Sender: linux-crypto-owner@vger.kernel.org List-ID: > A few other things I notice when looking at this code: > > /* Not all platforms can gate the clock, so it is not > an error if the clock does not exists. */ > cp->clk = clk_get(&pdev->dev, NULL); > if (!IS_ERR(cp->clk)) > clk_prepare_enable(cp->clk); > > So, if clk_get() returns PTR_ERR(-EPROBE_DEFER) we treat that clock as > missing? Is that really the behaviour you want there? Hi Russell I probably added that to the older driver and it got copied into the newer one. The clock it is using is added to the system using CLK_OF_DECLARE() This results in an entry in the __clk_of_table table. That table is walked in of_clk_init(), which is called by time_init() in arch/arc/kernel/time.c, early in the boot. If we get -EPROBE_DEFER, something is seriously wrong. Andrew