From mboxrd@z Thu Jan 1 00:00:00 1970 From: narmstrong@baylibre.com (Neil Armstrong) Date: Mon, 25 Apr 2016 16:48:24 +0200 Subject: [PATCH 1/2] clocksource: sp804: Add support for OX810SE 24bit timer width In-Reply-To: References: <1459520559-13110-1-git-send-email-narmstrong@baylibre.com> <1459520559-13110-2-git-send-email-narmstrong@baylibre.com> <20160422074645.GA3974@linaro.org> Message-ID: <571E2E38.3060401@baylibre.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 04/22/2016 09:53 AM, Thomas Gleixner wrote: > On Fri, 22 Apr 2016, Daniel Lezcano wrote: > >> On Fri, Apr 01, 2016 at 04:22:38PM +0200, Neil Armstrong wrote: >>> In order to support the Dual-Timer on the Oxford Semiconductor OX810SE SoC, >>> implement variable counter width, keeping 32bit as default width. >>> Add new compatible string oxsemi,ox810se-rps-timer in order to select >>> the 24bit counter width. >>> >>> Signed-off-by: Neil Armstrong >>> --- >>> drivers/clocksource/timer-sp804.c | 107 ++++++++++++++++++++++++-------------- >>> include/clocksource/timer-sp804.h | 42 ++++++++++++--- >>> 2 files changed, 102 insertions(+), 47 deletions(-) >> >> I will take those patches but this driver really deserves a cleanup. > > If it deserves a cleanup, then this should happen _BEFORE_ we add new > functionality to it. > > Thanks, > > tglx > Hi Thomas, Daniel, Sure this driver should have a cleanup, but it still depends on old legacy vexpress calls. The reason I submitted a change and a small cleanup over this driver is because the high similarities of the HW and avoiding adding a brand new driver cloning much of its functionalities. I propose to study an eventual cleanup, but could this change be submitted for 4.7 since it's part of a base platform support serie ? Thanks, Neil