Re: [windbg] How to trace the connected standby procedure in kernel?

DL CU wrote:

What I want to know is the kernel routine that checks the table for
each device’s power state you mentioned,
then I can give that routine a breakpoint and use !poaction or
!running to analyze why the device is still active,
now the HCK report me that there is a SDIO controller that keeps the
system working about 20s~30s during Connected standby test,
and the SDIO controller only has one child device - eMMC, the SDIO
controller driver was provided by Intel, and the eMMC storage driver
was MSFT inbox driver.

You can use Device Manager to examine the current power state of any
device, but I’m not sure it’s possible to answer the question of “why
the device is still active” without having the source code for that
device. Peter has bewailed about the pitfalls of debugging connected
standby issues; perhaps he knows of some tools to dig in more. In my
experience, it either works or it doesn’t, and if it doesn’t, you can’t
force it to happen.


Tim Roberts, xxxxx@probo.com
Providenza & Boekelheide, Inc.