From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752753AbdKHVkc (ORCPT ); Wed, 8 Nov 2017 16:40:32 -0500 Received: from fldsmtpe03.verizon.com ([140.108.26.142]:43699 "EHLO fldsmtpe03.verizon.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752503AbdKHVka (ORCPT ); Wed, 8 Nov 2017 16:40:30 -0500 X-Host: pioneer.tdc.vzwcorp.com From: "Levin, Alexander (Sasha Levin)" To: Mark Brown CC: "Levin, Alexander (Sasha Levin)" , "linux-kernel@vger.kernel.org" , "stable@vger.kernel.org" , Jiada Wang Subject: Re: [PATCH AUTOSEL for-4.9 04/53] spi: imx: adjust watermark level according to transfer length Thread-Topic: [PATCH AUTOSEL for-4.9 04/53] spi: imx: adjust watermark level according to transfer length Thread-Index: AQHTWNoEURjsGDAKL0unM5p56rBNTg== Date: Wed, 8 Nov 2017 21:39:11 +0000 Message-ID: <20171108213909.mm4mnzga3vic3sbc@sasha-lappy> References: <20171108204940.27321-1-alexander.levin@verizon.com> <20171108204940.27321-4-alexander.levin@verizon.com> <20171108205752.awsx5rzo4aecfl7i@sirena.co.uk> In-Reply-To: <20171108205752.awsx5rzo4aecfl7i@sirena.co.uk> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: NeoMutt/20170113 (1.7.2) x-ms-exchange-messagesentrepresentingtype: 1 x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.144.60.250] Content-Type: text/plain; charset="us-ascii" Content-ID: <55016500692E984E93B6C4CFE451AD2A@vzwcorp.com> MIME-Version: 1.0 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 quoted-printable to 8bit by nfs id vA8LebFP004738 On Wed, Nov 08, 2017 at 08:57:52PM +0000, Mark Brown wrote: >On Wed, Nov 08, 2017 at 08:49:55PM +0000, Levin, Alexander (Sasha Levin) wrote: >> From: Jiada Wang >> >> [ Upstream commit 66459c5a50a787c474b734b586328f7111ab6df0 ] >> >> Previously DMA watermark level is configured to fifosize/2, >> DMA mode can be used only when transfer length can be divided >> by 'watermark level * bpw', which makes DMA mode not pratical. >> >> This patch adjusts watermark level to largest number (no bigger >> than fifosize/2) which can divide 'tranfer length / bpw' for >> each transfer. > >This seems like an optimization not a bug fix... Hm, is it? I read it as "DMA is not being used at all even though we thought we're using it". Yes, the impact is "just" performance, but doesn't it result in quite a significat impact? -- Thanks, Sasha