From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752645AbeA3UCf (ORCPT ); Tue, 30 Jan 2018 15:02:35 -0500 Received: from mx07-00178001.pphosted.com ([62.209.51.94]:13731 "EHLO mx07-00178001.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751405AbeA3UCe (ORCPT ); Tue, 30 Jan 2018 15:02:34 -0500 From: Philippe CORNU To: Benjamin Gaignard , Yannick FERTRE , Vincent ABRIOU , "laurent.pinchart@ideasonboard.com" CC: "airlied@linux.ie" , "dri-devel@lists.freedesktop.org" , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH] drm/stm: check pitch and size calculations even if !CONFIG_MMU Thread-Topic: [PATCH] drm/stm: check pitch and size calculations even if !CONFIG_MMU Thread-Index: AQHTmd5fCyBFp4Znuki8jEbkNMFw66OMxfSA Date: Tue, 30 Jan 2018 20:02:21 +0000 Message-ID: <49d20fa9-de53-97b4-df9c-24defcef27ee@st.com> References: <20180130152400.28972-1-benjamin.gaignard@linaro.org> In-Reply-To: <20180130152400.28972-1-benjamin.gaignard@linaro.org> Accept-Language: fr-FR, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.5.0 x-ms-exchange-messagesentrepresentingtype: 1 x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.75.127.44] Content-Type: text/plain; charset="utf-8" Content-ID: <826DC2E52FB4D245A61F9054FBCAFC75@st.com> MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:,, definitions=2018-01-30_08:,, signatures=0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from base64 to 8bit by mail.home.local id w0UK2dKV023939 Hi Benjamin, Working on a coming patch (based on the new has_alpha field in fourcc), I realized that in particular cases (2 layers + "odd" layer sizes) memory transfers on MCU are greatly improved too if drm buffers are aligned on 128 bytes : ) So I suggest to cancel this patch. Then you can post a new one removing ifdef CONFIG_MMU endif, or I can do it, it is up to you. Many thanks for your support, Philippe :-) On 01/30/2018 04:24 PM, Benjamin Gaignard wrote: > In all cases we have to check pitch and size calculations. > Rely on drm_gem_cma_dumb_create for no MMU cases. > > Fixes: 21f815bf773c ("drm/stm: drv: Improve data transfers") > > Signed-off-by: Benjamin Gaignard > --- > drivers/gpu/drm/stm/drv.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/stm/drv.c b/drivers/gpu/drm/stm/drv.c > index 8bc7e8418b8d..3e0fb5f34b8e 100644 > --- a/drivers/gpu/drm/stm/drv.c > +++ b/drivers/gpu/drm/stm/drv.c > @@ -44,9 +44,11 @@ static int stm_gem_cma_dumb_create(struct drm_file *file, > */ > args->pitch = roundup(min_pitch, 128); > args->height = roundup(args->height, 4); > -#endif > > return drm_gem_cma_dumb_create_internal(file, dev, args); > +#else > + return drm_gem_cma_dumb_create(file, dev, args); > +#endif > } > > DEFINE_DRM_GEM_CMA_FOPS(drv_driver_fops); >