Kevin Hilman <[email protected]> writes:
> Juha Kuikka <[email protected]> writes:
>
>> On Tue, Oct 6, 2009 at 4:16 PM, Kevin Hilman
>> <[email protected]> wrote:
>>>
>>> Kevin Hilman <[email protected]> writes:
>>>
>>> > Hello,
>>> >
>>> > I've rebased/updated the PM branch based on current linux-omap master
>>> > branch (2.6.32-rc1 based.)
>>> >
>>> > I've also updated the OMAP Power Management wiki, and the 'Current
>>> > version' section highlights the changes, supported platforms as well
>>> > as the features that have made it into mainline.
>>> >
>>> > http://elinux.org/OMAP_Power_Management#Current_version
>>> >
>>>
>>> FYI... I just rebased the PM branch onto current omap/master.
>>>
>>> Some important user-visible changes: The /sys/power/* options are
>>> moving into debugfs. For mainline, these primarily debug options need
>>> to move to debugfs so I've moved them there. So, if you're missing
>>> some flag under /sys/power, it's now under /debug/pm_debug.
>>>
>>> I've updated the 'Using the PM branch' seciont of PM wiki[1] with
>>> the new filenames.
>>>
>>> Kevin
>>>
>>> [1] http://elinux.org/OMAP_Power_Management
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
>>> the body of a message to [email protected]
>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>
>> Hi,
>
> Hi Juha,
>
>> I build and tried this latest on Beagleboard (C2). Booting from MMC
>> using ramdisk image. I changed the LL console to the 3rd UART.
>>
>> The issue I have is that after the system boots up it works for as
>> long as I keep actively using it but if I leave it idle for a short
>> time (~5s) it stops responding. I have tried waking it up using both
>> UART and "USER" button on the board. This is with the default PM
>> configuration.
>
> Thanks for testing, and reporting your problem. I have noticed
> exactly the same problem and am still investigating it.
>
>> If I put the board into suspend (retention) it wakes up without any issues.
>>
>> Also, if I enable "/debug/pm_debug/sleep_while_idle" board wakes up
>> without issues. Well the first couple characers on the console seem to
>> go missing but I guess without CTS/RTS that is expected?
>
> Yes, the loss of the first character is expected.
>
>> Any pointers would be appreciated.
>
> I also noticed that it works when 'sleep_while_idle' is enabled, and
> this has led me to suspect a bug in that serial timout path, but I
> have yet to debug further.
FYI...
I haven't figured out exactly why yet, but this patch gets things
working again.
Basically, I'm just leaving smart-idle disabled all the time.
Kevin
diff --git a/arch/arm/mach-omap2/serial.c b/arch/arm/mach-omap2/serial.c
index 492dd02..90a284c 100644
--- a/arch/arm/mach-omap2/serial.c
+++ b/arch/arm/mach-omap2/serial.c
@@ -305,7 +305,7 @@ static void omap_uart_block_sleep(struct omap_uart_state
*uart)
{
omap_uart_enable_clocks(uart);
- omap_uart_smart_idle_enable(uart, 0);
+/* omap_uart_smart_idle_enable(uart, 0); */
uart->can_sleep = 0;
if (uart->timeout)
mod_timer(&uart->timer, jiffies + uart->timeout);
@@ -323,7 +323,7 @@ static void omap_uart_allow_sleep(struct omap_uart_state
*uart)
if (!uart->clocked)
return;
- omap_uart_smart_idle_enable(uart, 1);
+/* omap_uart_smart_idle_enable(uart, 1); */
uart->can_sleep = 1;
del_timer(&uart->timer);
}
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
