[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <154395650817.88331.14817295166592104160@swboyd.mtv.corp.google.com>
Date: Tue, 04 Dec 2018 12:48:28 -0800
From: Stephen Boyd <sboyd@...nel.org>
To: Jerome Brunet <jbrunet@...libre.com>,
Neil Armstrong <narmstrong@...libre.com>
Cc: linux-clk@...r.kernel.org, linux-amlogic@...ts.infradead.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] clk: meson-gxbb: switch video clocks mux tables to static
Quoting Jerome Brunet (2018-12-04 12:05:02)
> On Tue, 2018-12-04 at 20:33 +0100, Neil Armstrong wrote:
> > Le 04/12/2018 18:43, Stephen Boyd a écrit :
> > > Quoting Neil Armstrong (2018-12-04 04:58:39)
> > > > @@ -1983,7 +1983,7 @@ static struct clk_fixed_factor gxbb_vclk2_div12 =
> > > > {
> > > > };
> > > >
> > > > static u32 mux_table_cts_sel[] = { 0, 1, 2, 3, 4, 8, 9, 10, 11, 12 };
> > >
> > > I was talking about this mux_table_cts_sel array. Can this be const?
> > >
> >
> > Other mux tables aren't const, Jerome so you see a reason why they aren't
> > const ?
>
> I'm surpised we missed that. The PLL params tables of amlogic are all 'static
> const' but all the mux tables are just 'static' , humm
>
> [...] 5 min later [...]
>
> Actually the table field of clk_mux is not const, so giving it a const table
> would produce a warning about discarding const qualifier
>
> I guess there should no problem adding const to the table field of struct
> clk_mux. What do you think Stephen ?
Yes it would need to be marked const in the struct clk_mux as well. Go
for it to move more things to the RO section.
Powered by blists - more mailing lists