Thread (12 messages) 12 messages, 5 authors, 2013-03-11

OMAP4 PM bootloader dependency problems

From: paul@pwsan.com (Paul Walmsley)
Date: 2013-01-31 16:32:56
Also in: linux-omap

Hi,

On Thu, 31 Jan 2013, Santosh Shilimkar wrote:
On Thursday 31 January 2013 09:10 PM, Paul Walmsley wrote:
quoted
On Thu, 31 Jan 2013, Rajendra Nayak wrote:
quoted
Throw a pr_warn() at boot only when CONFIG_CPU_IDLE is enabled. Note
that it isn't enabled by default in omap2plus_defconfig. Also
throw one when a suspend fails, saying bootloader *could be* a possible
cause specifying the right version to be used.
In my view, these steps aren't sufficient, for the reasons described in

http://marc.info/?l=linux-omap&m=135964568904053&w=2

Even with CONFIG_CPU_IDLE=n, it's still possible to attempt to enter
full-chip retention idle on OMAP4 via 'echo mem > /sys/power/state', so it
doesn't seem right to restrict a solution to that case.
I think rajendra also mentioned adding one in suspend path too
when it fails.
" Also throw one when a suspend fails, saying bootloader
*could be* a possible cause specifying the right version to be used."

I find Rajendra's suggestion reasonable because some one might
just miss the boot message but getting the message on failure
case cant' be missed.
Right.  I don't have any problem with adding a message to the suspend path 
also.  But I'd like to see a boot-time message added even if 
CONFIG_CPU_IDLE=n, since it's still possible to enter full-chip retention 
idle with dynamic idle.  In other words, not everyone might use 'echo mem 
/sys/power/state'.

- Paul
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help