Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Alexandre Belloni <alexandre.belloni@bootlin.com>
To:     Dario Binacchi <dariobin@libero.it>, linux-kernel@vger.kernel.org
Cc:     Alexandre Belloni <alexandre.belloni@bootlin.com>,
        Alessandro Zummo <a.zummo@towertech.it>,
        linux-rtc@vger.kernel.org
Subject: Re: [PATCH] rtc: omap: use rtc_write to access OMAP_RTC_OSC_REG
Date:   Thu, 29 Apr 2021 23:29:49 +0200
Message-Id: <161973168395.2582973.9374098143035295509.b4-ty@bootlin.com>
In-Reply-To: <20210425145924.23353-1-dariobin@libero.it>
References: <20210425145924.23353-1-dariobin@libero.it>
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
Precedence: bulk

On Sun, 25 Apr 2021 16:59:23 +0200, Dario Binacchi wrote:
> The RTC_OSC_REG register is 32-bit, but the useful information is found
> in the 7 least significant bits (bits 7-31 are reserved). And in fact,
> as you can see from the code, all read accesses are 8-bit, as well as
> some writes. Let's make sure all writes are 8-bit. Moreover, in contexts
> where consecutive reads / writes after the busy check must take place
> within 15 us, it is better not to waste time on useless accesses.

I really doubt writeb versus writel actually make a difference as the bus will
probably still do a 32bit access.

Applied, thanks!

[1/1] rtc: omap: use rtc_write to access OMAP_RTC_OSC_REG
      commit: 50f2cc7f758cdbc7c8f928edc25b1c7bc09ef3e2

Best regards,
-- 
Alexandre Belloni <alexandre.belloni@bootlin.com>