Re: [PATCH v4] PM / devfreq: add PM-QoS support

2012-10-16 Thread Rafael J. Wysocki
On Monday 08 of October 2012 19:59:13 MyungJoo Ham wrote: > Even if the performance of a device is controlled properly with devfreq, > sometimes, we still need to get PM-QoS inputs in order to meet the > required performance. > > In our testbed of Exynos4412, which has on-chip various DMA

Re: [PATCH v4] PM / devfreq: add PM-QoS support

2012-10-16 Thread Rafael J. Wysocki
On Monday 08 of October 2012 19:59:13 MyungJoo Ham wrote: Even if the performance of a device is controlled properly with devfreq, sometimes, we still need to get PM-QoS inputs in order to meet the required performance. In our testbed of Exynos4412, which has on-chip various DMA devices, the

Re: [PATCH v4] PM / devfreq: add PM-QoS support

2012-10-09 Thread mark gross
On Mon, Oct 08, 2012 at 07:59:13PM +0900, MyungJoo Ham wrote: > Even if the performance of a device is controlled properly with devfreq, > sometimes, we still need to get PM-QoS inputs in order to meet the > required performance. > > In our testbed of Exynos4412, which has on-chip various DMA

Re: [PATCH v4] PM / devfreq: add PM-QoS support

2012-10-09 Thread mark gross
On Mon, Oct 08, 2012 at 07:59:13PM +0900, MyungJoo Ham wrote: Even if the performance of a device is controlled properly with devfreq, sometimes, we still need to get PM-QoS inputs in order to meet the required performance. In our testbed of Exynos4412, which has on-chip various DMA devices,

Re: [PATCH v4] PM / devfreq: add PM-QoS support

2012-10-08 Thread MyungJoo Ham
Hello, I didn't include the pm-qos support with the 0-100 (percent) concept though one might still implement that with per-dev qos support with this patch. However, for now, I don't think supporting 0-100 QoS is more appropriate than global QoS at least for some devices such as memory and

[PATCH v4] PM / devfreq: add PM-QoS support

2012-10-08 Thread MyungJoo Ham
Even if the performance of a device is controlled properly with devfreq, sometimes, we still need to get PM-QoS inputs in order to meet the required performance. In our testbed of Exynos4412, which has on-chip various DMA devices, the memory interface and system bus are controlled according to

[PATCH v4] PM / devfreq: add PM-QoS support

2012-10-08 Thread MyungJoo Ham
Even if the performance of a device is controlled properly with devfreq, sometimes, we still need to get PM-QoS inputs in order to meet the required performance. In our testbed of Exynos4412, which has on-chip various DMA devices, the memory interface and system bus are controlled according to

Re: [PATCH v4] PM / devfreq: add PM-QoS support

2012-10-08 Thread MyungJoo Ham
Hello, I didn't include the pm-qos support with the 0-100 (percent) concept though one might still implement that with per-dev qos support with this patch. However, for now, I don't think supporting 0-100 QoS is more appropriate than global QoS at least for some devices such as memory and