Fixed factor clock has two initializations at of_clk_init() time
and during platform driver probe. Before of_clk_init() call,
node is marked as populated and so its probe never gets called.
During of_clk_init() fixed factor clock registration may fail if
any of its parent clock is not registered. In this case, it doesn't
get chance to retry registration from probe. Clear OF_POPULATED
flag if fixed factor clock registration fails so that clock
registration is attempted again from probe.
Signed-off-by: Rajan Vaja <[email protected]>
---
drivers/clk/clk-fixed-factor.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/clk/clk-fixed-factor.c b/drivers/clk/clk-fixed-factor.c
index d72ef2d..f3ae4ff 100644
--- a/drivers/clk/clk-fixed-factor.c
+++ b/drivers/clk/clk-fixed-factor.c
@@ -177,8 +177,15 @@ static struct clk *_of_fixed_factor_clk_setup(struct device_node *node)
clk = clk_register_fixed_factor(NULL, clk_name, parent_name, flags,
mult, div);
- if (IS_ERR(clk))
+ if (IS_ERR(clk)) {
+ /*
+ * If parent clock is not registered, registration would fail.
+ * Clear OF_POPULATED flag so that clock registration can be
+ * attempted again from probe function.
+ */
+ of_node_clear_flag(node, OF_POPULATED);
return clk;
+ }
ret = of_clk_add_provider(node, of_clk_src_simple_get, clk);
if (ret) {
--
2.7.4
This email and any attachments are intended for the sole use of the named recipient(s) and contain(s) confidential information that may be proprietary, privileged or copyrighted under applicable law. If you are not the intended recipient, do not read, copy, or forward this email message or any attachments. Delete this email message and any attachments immediately.
Quoting Rajan Vaja (2018-07-17 06:17:00)
> Fixed factor clock has two initializations at of_clk_init() time
> and during platform driver probe. Before of_clk_init() call,
> node is marked as populated and so its probe never gets called.
>
> During of_clk_init() fixed factor clock registration may fail if
> any of its parent clock is not registered. In this case, it doesn't
> get chance to retry registration from probe. Clear OF_POPULATED
> flag if fixed factor clock registration fails so that clock
> registration is attempted again from probe.
>
> Signed-off-by: Rajan Vaja <[email protected]>
> ---
Applied to clk-next
Quoting Rajan Vaja (2018-07-17 06:17:00)
> Fixed factor clock has two initializations at of_clk_init() time
> and during platform driver probe. Before of_clk_init() call,
> node is marked as populated and so its probe never gets called.
>
> During of_clk_init() fixed factor clock registration may fail if
> any of its parent clock is not registered. In this case, it doesn't
> get chance to retry registration from probe. Clear OF_POPULATED
> flag if fixed factor clock registration fails so that clock
> registration is attempted again from probe.
>
> Signed-off-by: Rajan Vaja <[email protected]>
> ---
> drivers/clk/clk-fixed-factor.c | 9 ++++++++-
> 1 file changed, 8 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/clk/clk-fixed-factor.c b/drivers/clk/clk-fixed-factor.c
> index d72ef2d..f3ae4ff 100644
> --- a/drivers/clk/clk-fixed-factor.c
> +++ b/drivers/clk/clk-fixed-factor.c
> @@ -177,8 +177,15 @@ static struct clk *_of_fixed_factor_clk_setup(struct device_node *node)
>
> clk = clk_register_fixed_factor(NULL, clk_name, parent_name, flags,
> mult, div);
> - if (IS_ERR(clk))
> + if (IS_ERR(clk)) {
> + /*
> + * If parent clock is not registered, registration would fail.
> + * Clear OF_POPULATED flag so that clock registration can be
> + * attempted again from probe function.
> + */
> + of_node_clear_flag(node, OF_POPULATED);
> return clk;
> + }
BTW, this patch was seriously mangled for me so I had to apply it by
hand. Please be more careful next time and consider using something like
git-send-email and git-format-patch to send patches.
Hi Stephen,
> -----Original Message-----
> From: Stephen Boyd [mailto:[email protected]]
> Sent: 25 July 2018 10:11 PM
> To: Rajan Vaja <[email protected]>; [email protected]
> Cc: [email protected]; [email protected]; Rajan Vaja
> <[email protected]>
> Subject: Re: [PATCH] clk: clk-fixed-factor: Clear OF_POPULATED flag in case of
> failure
>
> Quoting Rajan Vaja (2018-07-17 06:17:00)
> > Fixed factor clock has two initializations at of_clk_init() time
> > and during platform driver probe. Before of_clk_init() call,
> > node is marked as populated and so its probe never gets called.
> >
> > During of_clk_init() fixed factor clock registration may fail if
> > any of its parent clock is not registered. In this case, it doesn't
> > get chance to retry registration from probe. Clear OF_POPULATED
> > flag if fixed factor clock registration fails so that clock
> > registration is attempted again from probe.
> >
> > Signed-off-by: Rajan Vaja <[email protected]>
> > ---
> > drivers/clk/clk-fixed-factor.c | 9 ++++++++-
> > 1 file changed, 8 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/clk/clk-fixed-factor.c b/drivers/clk/clk-fixed-factor.c
> > index d72ef2d..f3ae4ff 100644
> > --- a/drivers/clk/clk-fixed-factor.c
> > +++ b/drivers/clk/clk-fixed-factor.c
> > @@ -177,8 +177,15 @@ static struct clk *_of_fixed_factor_clk_setup(struct
> device_node *node)
> >
> > clk = clk_register_fixed_factor(NULL, clk_name, parent_name, flags,
> > mult, div);
> > - if (IS_ERR(clk))
> > + if (IS_ERR(clk)) {
> > + /*
> > + * If parent clock is not registered, registration would fail.
> > + * Clear OF_POPULATED flag so that clock registration can be
> > + * attempted again from probe function.
> > + */
> > + of_node_clear_flag(node, OF_POPULATED);
> > return clk;
> > + }
>
> BTW, this patch was seriously mangled for me so I had to apply it by
> hand. Please be more careful next time and consider using something like
> git-send-email and git-format-patch to send patches.
[Rajan] I used git-send-email only and sent same way as I used to send other patches. Not sure where was issue.
Sorry for the inconvenience.
Quoting Rajan Vaja (2018-07-26 06:03:16)
> Hi Stephen,
>
> > -----Original Message-----
> > From: Stephen Boyd [mailto:[email protected]]
> > Sent: 25 July 2018 10:11 PM
> > To: Rajan Vaja <[email protected]>; [email protected]
> > Cc: [email protected]; [email protected]; Rajan Vaja
> > <[email protected]>
> > Subject: Re: [PATCH] clk: clk-fixed-factor: Clear OF_POPULATED flag in case of
> > failure
> >
> > Quoting Rajan Vaja (2018-07-17 06:17:00)
> > > Fixed factor clock has two initializations at of_clk_init() time
> > > and during platform driver probe. Before of_clk_init() call,
> > > node is marked as populated and so its probe never gets called.
> > >
> > > During of_clk_init() fixed factor clock registration may fail if
> > > any of its parent clock is not registered. In this case, it doesn't
> > > get chance to retry registration from probe. Clear OF_POPULATED
> > > flag if fixed factor clock registration fails so that clock
> > > registration is attempted again from probe.
> > >
> > > Signed-off-by: Rajan Vaja <[email protected]>
> > > ---
> > > drivers/clk/clk-fixed-factor.c | 9 ++++++++-
> > > 1 file changed, 8 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/clk/clk-fixed-factor.c b/drivers/clk/clk-fixed-factor.c
> > > index d72ef2d..f3ae4ff 100644
> > > --- a/drivers/clk/clk-fixed-factor.c
> > > +++ b/drivers/clk/clk-fixed-factor.c
> > > @@ -177,8 +177,15 @@ static struct clk *_of_fixed_factor_clk_setup(struct
> > device_node *node)
> > >
> > > clk = clk_register_fixed_factor(NULL, clk_name, parent_name, flags,
> > > mult, div);
> > > - if (IS_ERR(clk))
> > > + if (IS_ERR(clk)) {
> > > + /*
> > > + * If parent clock is not registered, registration would fail.
> > > + * Clear OF_POPULATED flag so that clock registration can be
> > > + * attempted again from probe function.
> > > + */
> > > + of_node_clear_flag(node, OF_POPULATED);
> > > return clk;
> > > + }
> >
> > BTW, this patch was seriously mangled for me so I had to apply it by
> > hand. Please be more careful next time and consider using something like
> > git-send-email and git-format-patch to send patches.
> [Rajan] I used git-send-email only and sent same way as I used to send other patches. Not sure where was issue.
> Sorry for the inconvenience.
>
It must be your Outlook server messing things up. I see
Content-Transfer-Encoding: quoted-printable in the headers, which
probably caused my MUA to choke and line break things in the wrong
place.