[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20230726012430.GA7072@hu-mdtipton-lv.qualcomm.com>
Date: Tue, 25 Jul 2023 18:24:30 -0700
From: Mike Tipton <quic_mdtipton@...cinc.com>
To: Pavan Kondeti <quic_pkondeti@...cinc.com>
CC: <djakov@...nel.org>, <gregkh@...uxfoundation.org>,
<rafael@...nel.org>, <linux-kernel@...r.kernel.org>,
<linux-pm@...r.kernel.org>, <quic_okukatla@...cinc.com>,
<quic_viveka@...cinc.com>
Subject: Re: [PATCH 2/3] interconnect: Reintroduce icc_get()
On Tue, Jul 25, 2023 at 11:18:58AM +0530, Pavan Kondeti wrote:
> On Mon, Jul 24, 2023 at 06:28:58PM -0700, Mike Tipton wrote:
> > The original icc_get() that took integer node IDs was removed due to
> > lack of users. Reintroduce a new version that takes string node names,
> > which is needed for the debugfs client.
> >
[..]
>
> Is this API meant to be public()? As you pointed out, it was removed
> recently as there were no users. Since debugfs client is part of ICC
> frarmwork, should it be made private?
With debugfs as the only user, it could technically be made private.
Generally speaking, people should use of_icc_get() instead of this. We
have some other downstream users of icc_get(), but we'll likely push
them to switch to of_icc_get() instead. There are some potential future
use cases that would require exporting this, but none immediate.
Powered by blists - more mailing lists