From mboxrd@z Thu Jan 1 00:00:00 1970 From: daniel.lezcano@linaro.org (Daniel Lezcano) Date: Sun, 18 Nov 2018 02:48:32 +0100 Subject: [PATCH] arm/twd: avoid waking deeply sleeping CPUs for rate change notifier In-Reply-To: <1540812888.2582.4.camel@pengutronix.de> References: <20180417104627.29643-1-l.stach@pengutronix.de> <1528733761.2842.8.camel@pengutronix.de> <1540812888.2582.4.camel@pengutronix.de> Message-ID: To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 29/10/2018 12:34, Lucas Stach wrote: > Hi again, > > Am Montag, den 11.06.2018, 18:16 +0200 schrieb Lucas Stach: >> Hi all, >> >> Am Dienstag, den 17.04.2018, 12:46 +0200 schrieb Lucas Stach: >>> The current clock notifier sends an IPI to all CPUs, even if they are in >>> deep sleep state with the local timer disabled and switched to tick >>> broadcast. This needlessly cuts the CPU sleep times, as nothing is gained >>> from updating a disabled TWDs rate. >>> >>> Keep track of the enabled TWDs and only send an IPI to those CPUs with an >>> active local timer. As disabled TWDs may now miss a CPU frequency update >>> we need to make sure to refresh the rate on re-enabling of the timer. >>> >>> Signed-off-by: Lucas Stach >> >> I would appreciate some feedback to this patch. >> FWIW, I've been running some systems with this patch applied for quite >> some time now with no issues spotted so far. > > Can I persuade anyone to take a look at this? > It's been some months since I've posted this and haven't got any > feedback so far. I'm running this patch on a bunch of devices ever > since posting it and could not spot any adverse effects. Hi Lucas, can you resend this patch (keeping the current recipient) but adding Thomas Gleixner and John Stultz, the maintainer of the time core framework. Thanks -- Daniel -- Linaro.org ? Open source software for ARM SoCs Follow Linaro: Facebook | Twitter | Blog