Re: [PATCH 19/27] OMAP: DSS2: Use PM runtime & HWMOD support

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 6/6/2011 3:21 PM, Valkeinen, Tomi wrote:
On Mon, 2011-06-06 at 15:15 +0200, Cousson, Benoit wrote:
On 6/6/2011 3:01 PM, Valkeinen, Tomi wrote:
On Mon, 2011-06-06 at 14:56 +0200, Cousson, Benoit wrote:

In this long term solution, if the dss_fclk is the main_clk, how does
the framework handle the situation when we want to switch from the
standard DSS fclk to the one from DSI PLL?

That part cannot be done by the hwmod fmwk anyway. The goal of the fmwk
is to ensure that the module is accessible by the driver whatever the
PRCM clock used.
Enabling the DSI PLL will require the PRCM clock to be enabled first.

Using the DSI PLL as the fclk is doable, but is it really useful or needed?

Yes, it's useful and needed. It gives us much finer control to the clock
frequencies, and so allows us to go to higher frequencies and also more
exactly to the required pixel clock.

Assuming you need that mode, you will always have to explicitly switch
from DSI to PRCM clock before trying to disable the DSS.
This is something you will have to do inside the DSS driver. It should
be transparent to the hwmod fmwk.

This sounds ok.

I think the main question is how do we disable the standard DSS fclk
from PRCM when using DSI PLL? As far as I know, disabling that clock
will allow some areas of OMAP to be shut down even while DSS is working.
So from power management point of view it sounds a needed feature.

Yes, at least in theory, but considering that any use case that will require the DSI PLL will use a LCD panel + backlight, or an OLED panel that will consume 50 times more than the 186 MHz clock, I do not think it is really needed. Moreover, that clock is generated by the PER DPLL that will be always enabled in most usecase because it does generate the UART, I2C and most basic peripherals clocks. If we cannot gate the PER DPLL, there is no saving to expect from gating the DSS fclk only. Bottom-line is that there is no practical power saving to expect from that mode.

If the clock is main_clk for the HWMOD, it sounds to me it's always
enabled if the HWMOD is enabled?

Yes, but that sounds to me a good trade off to avoid unnecessary complexity in your driver or in the hwmod fmwk.

Regards,
Benoit
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [Linux Arm (vger)]     [ARM Kernel]     [ARM MSM]     [Linux Tegra]     [Linux WPAN Networking]     [Linux Wireless Networking]     [Maemo Users]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux