From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Return-Path: Message-ID: <55658201.5090208@linaro.org> Date: Wed, 27 May 2015 09:36:17 +0100 From: Daniel Thompson MIME-Version: 1.0 To: Maxime Coquelin CC: Mike Turquette , Stephen Boyd , Rob Herring , Pawel Moll , Mark Rutland , Ian Campbell , Kumar Gala , Russell King , Kamil Lulko , Andreas Farber , linux-clk@vger.kernel.org, "devicetree@vger.kernel.org" , "linux-arm-kernel@lists.infradead.org" , "linux-kernel@vger.kernel.org" , patches@linaro.org, linaro-kernel@lists.linaro.org Subject: Re: [RFC PATCH 0/3] clk: stm32: Add clock driver for STM32F4[23]xxx devices References: <1432327273-6810-1-git-send-email-daniel.thompson@linaro.org> In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed List-ID: On 26/05/15 17:41, Maxime Coquelin wrote: >> 2. Allocating ~80 clock components appears to fragment memory enough >> to prevent busybox (.text is ~300K, non-XIP) from running getty. >> I have to use "init=/bin/sh" to avoid OOM problems. > > Yes, I reproduced the same issue while testing your work. > > Could you try with: > CONFIG_NOMMU_INITIAL_TRIM_EXCESS=0 > > It solves (hides) the problem on my side. > You can eventually setting it once booted to 1 via procfs. That works for me too. Thanks. Incidentally switching to SLOB also reduced the problem for me. It was OK whilst I had ~20 clocks (and debugfs) but not enough for me to run with all the clocks registered. Daniel.