Re: [PATCH] acpi, nfit: fix acpi event notifications for nfit
From: Dan Williams
Date: Thu Aug 18 2016 - 21:52:51 EST
On Thu, Aug 18, 2016 at 1:22 PM, Verma, Vishal L
<vishal.l.verma@xxxxxxxxx> wrote:
> On Thu, 2016-08-18 at 12:54 -0700, Dan Williams wrote:
>> On Thu, Aug 18, 2016 at 12:52 PM, Linda Knippers <linda.knippers@xxxxx
>> om> wrote:
>> >
>> >
>> >
>> > On 8/18/2016 3:48 PM, Dan Williams wrote:
>> > >
>> > > On Thu, Aug 18, 2016 at 11:48 AM, Vishal Verma <vishal.l.verma@int
>> > > el.com> wrote:
>> > > >
>> > > > The nfit driver had an acpi event notification handler, but it
>> > > > never
>> > > > would've worked because we weren't setting the
>> > > > ACPI_DRIVER_ALL_NOTIFY_EVENTS flag in acpi_driver.
>> > >
>> > > Let's update the changelog to be helpful for someone implementing
>> > > a
>> > > backport or taking this back to a -stable branch. Something like:
>> > >
>> > > Subject: acpi, nfit: fix event notifications
>> > >
>> > > Commit 209851649dc4 "acpi: nfit: Add support for hot-add" added
>> > > support for _FIT notifications, but it neglected to set the
>> > > ACPI_DRIVER_ALL_NOTIFY_EVENTS flag that acpi_bus_notify() uses to
>> > > gate
>> > > notification delivery.
>> >
>> > While we're at it, should we update the notifier function to
>> > explicitly check
>> > for event 0x80 before re-evaluating the _FIT? I'm thinking about
>> > some time
>> > in the future when there might be more than one event.
>>
>> Yes, good idea.
>
> Sounds good, part of the same patch or separately?
Separate.
In general, if you would write the word "also" in a patch description,
that's an indication to make a separate patch.