The common clock and several other clock API implementations allow
calling clk_get_rate with a NULL pointer. While not specified as
expected behavior of the API, device drivers have come to rely on that,
causing them to OOPS when run on a platform with a different clock API
implementation.
Fix this by making sure all clk_get_rate implementations handle
NULL clocks instead of OOPSing.
While some custom implementations even allow ERR_PTR()s, I decided
against that as IIRC the usual idea is that errors should be handled and
not silently carried over.
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Jonas Gorski (9):
ARM: ep93xx: allow NULL clock for clk_get_rate
ARM: mmp: allow NULL clock for clk_get_rate
blackfin: bf609: allow NULL clock for clk_get_rate
m68k: allow NULL clock for clk_get_rate
MIPS: AR7: allow NULL clock for clk_get_rate
MIPS: BCM63XX: allow NULL clock for clk_get_rate
MIPS: Loongson 2F: allow NULL clock for clk_get_rate
MIPS: ralink: allow NULL clock for clk_get_rate
unicore32: allow NULL clock for clk_get_rate
arch/arm/mach-ep93xx/clock.c | 3 +++
arch/arm/mach-mmp/clock.c | 4 +++-
arch/blackfin/mach-bf609/clock.c | 2 +-
arch/m68k/coldfire/clk.c | 3 +++
arch/mips/ar7/clock.c | 3 +++
arch/mips/bcm63xx/clk.c | 3 +++
arch/mips/loongson64/lemote-2f/clock.c | 3 +++
arch/mips/ralink/clk.c | 3 +++
arch/unicore32/kernel/clock.c | 3 +++
9 files changed, 25 insertions(+), 2 deletions(-)
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: 1d81eedb8f6c ("[ARM] 3634/1: ep93xx: initial implementation of the clk_* API")
Cc: Hartley Sweeten <[email protected]>
Cc: Alexander Sverdlin <[email protected]>
Cc: Russell King <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/arm/mach-ep93xx/clock.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/arm/mach-ep93xx/clock.c b/arch/arm/mach-ep93xx/clock.c
index 39ef3b613912..f0768befafe8 100644
--- a/arch/arm/mach-ep93xx/clock.c
+++ b/arch/arm/mach-ep93xx/clock.c
@@ -316,6 +316,9 @@ static unsigned long get_uart_rate(struct clk *clk)
unsigned long clk_get_rate(struct clk *clk)
{
+ if (!clk)
+ return 0;
+
if (clk->get_rate)
return clk->get_rate(clk);
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: facdf0ed4f59 ("m68knommu: introduce basic clk infrastructure")
Cc: Greg Ungerer <[email protected]>
Cc: Geert Uytterhoeven <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/m68k/coldfire/clk.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/m68k/coldfire/clk.c b/arch/m68k/coldfire/clk.c
index 1e3c7e9193d1..856069a3196d 100644
--- a/arch/m68k/coldfire/clk.c
+++ b/arch/m68k/coldfire/clk.c
@@ -121,6 +121,9 @@ EXPORT_SYMBOL(clk_put);
unsigned long clk_get_rate(struct clk *clk)
{
+ if (!clk)
+ return 0;
+
return clk->rate;
}
EXPORT_SYMBOL(clk_get_rate);
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: 64909882862e ("unicore32 additional architecture files: pm related files")
Cc: Guan Xuetao <[email protected]>
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/unicore32/kernel/clock.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/unicore32/kernel/clock.c b/arch/unicore32/kernel/clock.c
index b1ca775f6f6e..d867f34fdb74 100644
--- a/arch/unicore32/kernel/clock.c
+++ b/arch/unicore32/kernel/clock.c
@@ -92,6 +92,9 @@ EXPORT_SYMBOL(clk_disable);
unsigned long clk_get_rate(struct clk *clk)
{
+ if (!clk)
+ return 0;
+
return clk->rate;
}
EXPORT_SYMBOL(clk_get_rate);
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: 3f0a06b0368d ("MIPS: ralink: adds clkdev code")
Cc: John Crispin <[email protected]>
Cc: Ralf Baechle <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/mips/ralink/clk.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/mips/ralink/clk.c b/arch/mips/ralink/clk.c
index eb1c61917eb7..1b7df115eb60 100644
--- a/arch/mips/ralink/clk.c
+++ b/arch/mips/ralink/clk.c
@@ -53,6 +53,9 @@ EXPORT_SYMBOL_GPL(clk_disable);
unsigned long clk_get_rate(struct clk *clk)
{
+ if (!clk)
+ return 0;
+
return clk->rate;
}
EXPORT_SYMBOL_GPL(clk_get_rate);
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter, as some device
drivers rely on this.
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: f8ede0f700f5 ("MIPS: Loongson 2F: Add CPU frequency scaling support")
Cc: Ralf Baechle <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/mips/loongson64/lemote-2f/clock.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/mips/loongson64/lemote-2f/clock.c b/arch/mips/loongson64/lemote-2f/clock.c
index a78fb657068c..8281334df9c8 100644
--- a/arch/mips/loongson64/lemote-2f/clock.c
+++ b/arch/mips/loongson64/lemote-2f/clock.c
@@ -80,6 +80,9 @@ EXPORT_SYMBOL(clk_disable);
unsigned long clk_get_rate(struct clk *clk)
{
+ if (!clk)
+ return 0;
+
return (unsigned long)clk->rate;
}
EXPORT_SYMBOL(clk_get_rate);
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: 969003152aa9 ("blackfin: bf60x: add clock support")
Cc: Steven Miao <[email protected]>
Cc: Masahiro Yamada <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/blackfin/mach-bf609/clock.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/blackfin/mach-bf609/clock.c b/arch/blackfin/mach-bf609/clock.c
index 392a59b9a504..21a0ec18829f 100644
--- a/arch/blackfin/mach-bf609/clock.c
+++ b/arch/blackfin/mach-bf609/clock.c
@@ -109,7 +109,7 @@ EXPORT_SYMBOL(clk_disable);
unsigned long clk_get_rate(struct clk *clk)
{
unsigned long ret = 0;
- if (clk->ops && clk->ops->get_rate)
+ if (clk && clk->ops && clk->ops->get_rate)
ret = clk->ops->get_rate(clk);
return ret;
}
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: e7300d04bd08 ("MIPS: BCM63xx: Add support for the Broadcom BCM63xx family of SOCs.")
Cc: Ralf Baechle <[email protected]>
Cc: Florian Fainelli <[email protected]>
Cc: [email protected]
Cc: James Hogan <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/mips/bcm63xx/clk.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/mips/bcm63xx/clk.c b/arch/mips/bcm63xx/clk.c
index 73626040e4d6..19577f771c1f 100644
--- a/arch/mips/bcm63xx/clk.c
+++ b/arch/mips/bcm63xx/clk.c
@@ -339,6 +339,9 @@ EXPORT_SYMBOL(clk_disable);
unsigned long clk_get_rate(struct clk *clk)
{
+ if (!clk)
+ return 0;
+
return clk->rate;
}
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: 780019ddf02f ("MIPS: AR7: Implement clock API")
Cc: Ralf Baechle <[email protected]>
Cc: Paul Gortmaker <[email protected]>
Cc: James Hogan <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/mips/ar7/clock.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/mips/ar7/clock.c b/arch/mips/ar7/clock.c
index dda422a0f36c..0137656107a9 100644
--- a/arch/mips/ar7/clock.c
+++ b/arch/mips/ar7/clock.c
@@ -430,6 +430,9 @@ EXPORT_SYMBOL(clk_disable);
unsigned long clk_get_rate(struct clk *clk)
{
+ if (!clk)
+ return 0;
+
return clk->rate;
}
EXPORT_SYMBOL(clk_get_rate);
--
2.11.0
Make the behaviour of clk_get_rate consistent with common clk's
clk_get_rate by accepting NULL clocks as parameter. Some device
drivers rely on this, and will cause an OOPS otherwise.
Fixes: 49cbe78637eb ("[ARM] pxa: add base support for Marvell's PXA168 processor line")
Cc: Eric Miao <[email protected]>
Cc: Haojian Zhuang <[email protected]>
Cc: Russell King <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reported-by: Mathias Kresin <[email protected]>
Signed-off-by: Jonas Gorski <[email protected]>
---
arch/arm/mach-mmp/clock.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/arch/arm/mach-mmp/clock.c b/arch/arm/mach-mmp/clock.c
index 28fe64c6e2f5..bdfb113431ec 100644
--- a/arch/arm/mach-mmp/clock.c
+++ b/arch/arm/mach-mmp/clock.c
@@ -83,7 +83,9 @@ unsigned long clk_get_rate(struct clk *clk)
{
unsigned long rate;
- if (clk->ops->getrate)
+ if (!clk)
+ rate = 0;
+ else if (clk->ops->getrate)
rate = clk->ops->getrate(clk);
else
rate = clk->rate;
--
2.11.0
Hi Jonas,
On Tue, Jul 18, 2017 at 12:17 PM, Jonas Gorski <[email protected]> wrote:
> The common clock and several other clock API implementations allow
> calling clk_get_rate with a NULL pointer. While not specified as
> expected behavior of the API, device drivers have come to rely on that,
> causing them to OOPS when run on a platform with a different clock API
> implementation.
>
> Fix this by making sure all clk_get_rate implementations handle
> NULL clocks instead of OOPSing.
>
> While some custom implementations even allow ERR_PTR()s, I decided
> against that as IIRC the usual idea is that errors should be handled and
> not silently carried over.
>
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
>
> Jonas Gorski (9):
> ARM: ep93xx: allow NULL clock for clk_get_rate
> ARM: mmp: allow NULL clock for clk_get_rate
> blackfin: bf609: allow NULL clock for clk_get_rate
> m68k: allow NULL clock for clk_get_rate
> MIPS: AR7: allow NULL clock for clk_get_rate
> MIPS: BCM63XX: allow NULL clock for clk_get_rate
> MIPS: Loongson 2F: allow NULL clock for clk_get_rate
> MIPS: ralink: allow NULL clock for clk_get_rate
> unicore32: allow NULL clock for clk_get_rate
>
> arch/arm/mach-ep93xx/clock.c | 3 +++
> arch/arm/mach-mmp/clock.c | 4 +++-
> arch/blackfin/mach-bf609/clock.c | 2 +-
> arch/m68k/coldfire/clk.c | 3 +++
> arch/mips/ar7/clock.c | 3 +++
> arch/mips/bcm63xx/clk.c | 3 +++
> arch/mips/loongson64/lemote-2f/clock.c | 3 +++
> arch/mips/ralink/clk.c | 3 +++
> arch/unicore32/kernel/clock.c | 3 +++
> 9 files changed, 25 insertions(+), 2 deletions(-)
For the whole series:
Reviewed-by: Geert Uytterhoeven <[email protected]>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi Jonas,
On 18/07/17 20:17, Jonas Gorski wrote:
> Make the behaviour of clk_get_rate consistent with common clk's
> clk_get_rate by accepting NULL clocks as parameter. Some device
> drivers rely on this, and will cause an OOPS otherwise.
>
> Fixes: facdf0ed4f59 ("m68knommu: introduce basic clk infrastructure")
> Cc: Greg Ungerer <[email protected]>
Acked-by: Greg Ungerer <[email protected]>
Do you want me to push this via the m68knommu git tree?
Or are you (or someone) taking the series as a whole?
Regards
Greg
> Cc: Geert Uytterhoeven <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Reported-by: Mathias Kresin <[email protected]>
> Signed-off-by: Jonas Gorski <[email protected]>
> ---
> arch/m68k/coldfire/clk.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/arch/m68k/coldfire/clk.c b/arch/m68k/coldfire/clk.c
> index 1e3c7e9193d1..856069a3196d 100644
> --- a/arch/m68k/coldfire/clk.c
> +++ b/arch/m68k/coldfire/clk.c
> @@ -121,6 +121,9 @@ EXPORT_SYMBOL(clk_put);
>
> unsigned long clk_get_rate(struct clk *clk)
> {
> + if (!clk)
> + return 0;
> +
> return clk->rate;
> }
> EXPORT_SYMBOL(clk_get_rate);
>
2017-07-18 19:17 GMT+09:00 Jonas Gorski <[email protected]>:
> Make the behaviour of clk_get_rate consistent with common clk's
> clk_get_rate by accepting NULL clocks as parameter. Some device
> drivers rely on this, and will cause an OOPS otherwise.
>
> Fixes: 969003152aa9 ("blackfin: bf60x: add clock support")
> Cc: Steven Miao <[email protected]>
> Cc: Masahiro Yamada <[email protected]>
This Cc can be replaced with
Reviewed-by: Masahiro Yamada <[email protected]>
Thank you.
> Cc: Andrew Morton <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Reported-by: Mathias Kresin <[email protected]>
> Signed-off-by: Jonas Gorski <[email protected]>
> ---
> arch/blackfin/mach-bf609/clock.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/blackfin/mach-bf609/clock.c b/arch/blackfin/mach-bf609/clock.c
> index 392a59b9a504..21a0ec18829f 100644
> --- a/arch/blackfin/mach-bf609/clock.c
> +++ b/arch/blackfin/mach-bf609/clock.c
> @@ -109,7 +109,7 @@ EXPORT_SYMBOL(clk_disable);
> unsigned long clk_get_rate(struct clk *clk)
> {
> unsigned long ret = 0;
> - if (clk->ops && clk->ops->get_rate)
> + if (clk && clk->ops && clk->ops->get_rate)
> ret = clk->ops->get_rate(clk);
> return ret;
> }
> --
> 2.11.0
>
--
Best Regards
Masahiro Yamada
On 07/18/2017 03:17 AM, Jonas Gorski wrote:
> Make the behaviour of clk_get_rate consistent with common clk's
> clk_get_rate by accepting NULL clocks as parameter. Some device
> drivers rely on this, and will cause an OOPS otherwise.
>
> Fixes: e7300d04bd08 ("MIPS: BCM63xx: Add support for the Broadcom BCM63xx family of SOCs.")
> Cc: Ralf Baechle <[email protected]>
> Cc: Florian Fainelli <[email protected]>
> Cc: [email protected]
> Cc: James Hogan <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Reported-by: Mathias Kresin <[email protected]>
> Signed-off-by: Jonas Gorski <[email protected]>
Reviewed-by: Florian Fainelli <[email protected]>
--
Florian
On 18/07/17 12:17, Jonas Gorski wrote:
> Make the behaviour of clk_get_rate consistent with common clk's
> clk_get_rate by accepting NULL clocks as parameter. Some device
> drivers rely on this, and will cause an OOPS otherwise.
>
> Fixes: 1d81eedb8f6c ("[ARM] 3634/1: ep93xx: initial implementation of the clk_* API")
> Cc: Hartley Sweeten <[email protected]>
> Cc: Alexander Sverdlin <[email protected]>
> Cc: Russell King <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Reported-by: Mathias Kresin <[email protected]>
> Signed-off-by: Jonas Gorski <[email protected]>
Acked-by: Alexander Sverdlin <[email protected]>
> ---
> arch/arm/mach-ep93xx/clock.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/arch/arm/mach-ep93xx/clock.c b/arch/arm/mach-ep93xx/clock.c
> index 39ef3b613912..f0768befafe8 100644
> --- a/arch/arm/mach-ep93xx/clock.c
> +++ b/arch/arm/mach-ep93xx/clock.c
> @@ -316,6 +316,9 @@ static unsigned long get_uart_rate(struct clk *clk)
>
> unsigned long clk_get_rate(struct clk *clk)
> {
> + if (!clk)
> + return 0;
> +
> if (clk->get_rate)
> return clk->get_rate(clk);
Hi Greg,
On 18 July 2017 at 14:03, Greg Ungerer <[email protected]> wrote:
> Hi Jonas,
>
> On 18/07/17 20:17, Jonas Gorski wrote:
>>
>> Make the behaviour of clk_get_rate consistent with common clk's
>> clk_get_rate by accepting NULL clocks as parameter. Some device
>> drivers rely on this, and will cause an OOPS otherwise.
>>
>> Fixes: facdf0ed4f59 ("m68knommu: introduce basic clk infrastructure")
>> Cc: Greg Ungerer <[email protected]>
>
>
> Acked-by: Greg Ungerer <[email protected]>
>
> Do you want me to push this via the m68knommu git tree?
> Or are you (or someone) taking the series as a whole?
Please take it through your tree. I totally forgot mentioning in the
cover letter that I'm just a simple patch submitter and don't have my
own tree. Too long ago that I sent a multi-tree patch series last time
... .
Regards
Jonas
Hi Jonas,
On 19/07/17 19:06, Jonas Gorski wrote:
> Hi Greg,
>
> On 18 July 2017 at 14:03, Greg Ungerer <[email protected]> wrote:
>> Hi Jonas,
>>
>> On 18/07/17 20:17, Jonas Gorski wrote:
>>>
>>> Make the behaviour of clk_get_rate consistent with common clk's
>>> clk_get_rate by accepting NULL clocks as parameter. Some device
>>> drivers rely on this, and will cause an OOPS otherwise.
>>>
>>> Fixes: facdf0ed4f59 ("m68knommu: introduce basic clk infrastructure")
>>> Cc: Greg Ungerer <[email protected]>
>>
>>
>> Acked-by: Greg Ungerer <[email protected]>
>>
>> Do you want me to push this via the m68knommu git tree?
>> Or are you (or someone) taking the series as a whole?
>
> Please take it through your tree. I totally forgot mentioning in the
> cover letter that I'm just a simple patch submitter and don't have my
> own tree. Too long ago that I sent a multi-tree patch series last time
No problem. Added to the m68knommu git tree, for-next branch.
Thanks
Greg