Hey all,
On 4/2/2018 3:11 PM, Timur Tabi wrote:
On Thu, Mar 29, 2018 at 12:33 AM, Bjorn Andersson
<bjorn.andersson@xxxxxxxxxx> wrote:
And as you say above, the i2c-qup driver does reset the core so this
should be "less of an issue", not merging this patch means that if there
is any issues left we expose these to more testing.
I understand that leaving an interrupt pending doesn't really have any
side-effect since the driver reset the device in the probe, although I
still think it's proper for a driver to clear this interrupt when it
shuts down. I don't want to belabor that point.
However, I don't understand what you mean by "any issues left"? What
issues? What value could there be to leaving resources hanging when
the driver shuts down?
After reviewing the patch and dependencies within the kernel, I'm
withdrawing this change.
There is a loose coupling between PCIe and the I2C driver via the PCIe
hotplug feature. During shutdown or kexec the PCIe driver shuts down the
hotplug driver which issues I2C commands to turn off slot power. There
is no explicit dependency between hotplug and I2C, so the I2C driver is
already gone when this takes place. This can result in the kernel being
stuck in a spinning loop of unhandled ACPI function.
Based on this and Bjorn's comments, I think the change should be dropped.
Austin
--
Qualcomm Datacenter Technologies as an affiliate of Qualcomm
Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the
Code Aurora Forum, a Linux Foundation Collaborative Project.
