2019-12-12 02:40:27

by Kuninori Morimoto

[permalink] [raw]
Subject: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay


From: Kuninori Morimoto <[email protected]>

__delay() is used from kernel module.
We need EXPORT_SYMBOL(), otherwise we will get compile error.

ERROR: "__delay" [drivers/net/phy/mdio-cavium.ko] undefined!

Signed-off-by: Kuninori Morimoto <[email protected]>
---
arch/sh/lib/delay.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c
index dad8e6a..540e670 100644
--- a/arch/sh/lib/delay.c
+++ b/arch/sh/lib/delay.c
@@ -29,6 +29,7 @@ void __delay(unsigned long loops)
: "0" (loops)
: "t");
}
+EXPORT_SYMBOL(__delay);

inline void __const_udelay(unsigned long xloops)
{
--
2.7.4


2019-12-12 07:44:32

by Geert Uytterhoeven

[permalink] [raw]
Subject: Re: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay

Hi Morimoto-san,

On Thu, Dec 12, 2019 at 3:38 AM Kuninori Morimoto
<[email protected]> wrote:
> From: Kuninori Morimoto <[email protected]>
>
> __delay() is used from kernel module.
> We need EXPORT_SYMBOL(), otherwise we will get compile error.
>
> ERROR: "__delay" [drivers/net/phy/mdio-cavium.ko] undefined!
>
> Signed-off-by: Kuninori Morimoto <[email protected]>

Thanks for your patch!

> --- a/arch/sh/lib/delay.c
> +++ b/arch/sh/lib/delay.c
> @@ -29,6 +29,7 @@ void __delay(unsigned long loops)
> : "0" (loops)
> : "t");
> }
> +EXPORT_SYMBOL(__delay);
>
> inline void __const_udelay(unsigned long xloops)
> {

I believe the correct fix is make drivers/net/phy/mdio-cavium.c use one
of [nmu]delay() instead.

Cfr.
https://lore.kernel.org/lkml/CAMuHMdUERaoHLNKi03zCuYi7NevgBFjXrV=pt0Yy=HOeRiL25Q@mail.gmail.com/

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

2019-12-12 23:58:09

by Kuninori Morimoto

[permalink] [raw]
Subject: Re: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay


Hi Geert

Thank you for your feedback

> > --- a/arch/sh/lib/delay.c
> > +++ b/arch/sh/lib/delay.c
> > @@ -29,6 +29,7 @@ void __delay(unsigned long loops)
> > : "0" (loops)
> > : "t");
> > }
> > +EXPORT_SYMBOL(__delay);
> >
> > inline void __const_udelay(unsigned long xloops)
> > {
>
> I believe the correct fix is make drivers/net/phy/mdio-cavium.c use one
> of [nmu]delay() instead.
>
> Cfr.
> https://lore.kernel.org/lkml/CAMuHMdUERaoHLNKi03zCuYi7NevgBFjXrV=pt0Yy=HOeRiL25Q@mail.gmail.com/

OK, make sense.
I did it because some architecture is using EXPORT_SYMBOL() to __delay.

Thank you for your help !!
Best regards
---
Kuninori Morimoto

2020-07-22 02:41:12

by Guenter Roeck

[permalink] [raw]
Subject: Re: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay

On Thu, Dec 12, 2019 at 11:38:43AM +0900, Kuninori Morimoto wrote:
> From: Kuninori Morimoto <[email protected]>
>
> __delay() is used from kernel module.
> We need EXPORT_SYMBOL(), otherwise we will get compile error.
>
> ERROR: "__delay" [drivers/net/phy/mdio-cavium.ko] undefined!
>
> Signed-off-by: Kuninori Morimoto <[email protected]>

I must admit that this patch completely baffles me. __delay was
already exported, only elsewhere in the file. With this patch
in place, it is exported twice, and all sh builds in -next fail
with

In file included from include/linux/linkage.h:7,
from arch/sh/include/asm/bug.h:5,
from include/linux/bug.h:5,
from include/linux/thread_info.h:12,
from include/asm-generic/current.h:5,
from ./arch/sh/include/generated/asm/current.h:1,
from include/linux/sched.h:12,
from arch/sh/lib/delay.c:8:
include/linux/export.h:67:36: error: redefinition of '__ksymtab___delay'

Guenter

> ---
> arch/sh/lib/delay.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c
> index dad8e6a..540e670 100644
> --- a/arch/sh/lib/delay.c
> +++ b/arch/sh/lib/delay.c
> @@ -29,6 +29,7 @@ void __delay(unsigned long loops)
> : "0" (loops)
> : "t");
> }
> +EXPORT_SYMBOL(__delay);
>
> inline void __const_udelay(unsigned long xloops)
> {

2020-07-22 22:54:34

by Rich Felker

[permalink] [raw]
Subject: Re: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay

On Tue, Jul 21, 2020 at 07:38:40PM -0700, Guenter Roeck wrote:
> On Thu, Dec 12, 2019 at 11:38:43AM +0900, Kuninori Morimoto wrote:
> > From: Kuninori Morimoto <[email protected]>
> >
> > __delay() is used from kernel module.
> > We need EXPORT_SYMBOL(), otherwise we will get compile error.
> >
> > ERROR: "__delay" [drivers/net/phy/mdio-cavium.ko] undefined!
> >
> > Signed-off-by: Kuninori Morimoto <[email protected]>
>
> I must admit that this patch completely baffles me. __delay was
> already exported, only elsewhere in the file. With this patch
> in place, it is exported twice, and all sh builds in -next fail
> with
>
> In file included from include/linux/linkage.h:7,
> from arch/sh/include/asm/bug.h:5,
> from include/linux/bug.h:5,
> from include/linux/thread_info.h:12,
> from include/asm-generic/current.h:5,
> from ./arch/sh/include/generated/asm/current.h:1,
> from include/linux/sched.h:12,
> from arch/sh/lib/delay.c:8:
> include/linux/export.h:67:36: error: redefinition of '__ksymtab___delay'
>
> Guenter
>
> > ---
> > arch/sh/lib/delay.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c
> > index dad8e6a..540e670 100644
> > --- a/arch/sh/lib/delay.c
> > +++ b/arch/sh/lib/delay.c
> > @@ -29,6 +29,7 @@ void __delay(unsigned long loops)
> > : "0" (loops)
> > : "t");
> > }
> > +EXPORT_SYMBOL(__delay);
> >
> > inline void __const_udelay(unsigned long xloops)
> > {

I presently have a revert of this commit in queue for next. If it's
sufficiently breaking (and especially if there are other regressions
that need to be fixed, see the pmd_free thing) I could try to get it
in for 5.8 still but that's getting a bit late.

Rich

2020-07-22 23:53:34

by Guenter Roeck

[permalink] [raw]
Subject: Re: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay

Rich,

On 7/22/20 3:52 PM, Rich Felker wrote:
> On Tue, Jul 21, 2020 at 07:38:40PM -0700, Guenter Roeck wrote:
>> On Thu, Dec 12, 2019 at 11:38:43AM +0900, Kuninori Morimoto wrote:
>>> From: Kuninori Morimoto <[email protected]>
>>>
>>> __delay() is used from kernel module.
>>> We need EXPORT_SYMBOL(), otherwise we will get compile error.
>>>
>>> ERROR: "__delay" [drivers/net/phy/mdio-cavium.ko] undefined!
>>>
>>> Signed-off-by: Kuninori Morimoto <[email protected]>
>>
>> I must admit that this patch completely baffles me. __delay was
>> already exported, only elsewhere in the file. With this patch
>> in place, it is exported twice, and all sh builds in -next fail
>> with
>>
>> In file included from include/linux/linkage.h:7,
>> from arch/sh/include/asm/bug.h:5,
>> from include/linux/bug.h:5,
>> from include/linux/thread_info.h:12,
>> from include/asm-generic/current.h:5,
>> from ./arch/sh/include/generated/asm/current.h:1,
>> from include/linux/sched.h:12,
>> from arch/sh/lib/delay.c:8:
>> include/linux/export.h:67:36: error: redefinition of '__ksymtab___delay'
>>
>> Guenter
>>
>>> ---
>>> arch/sh/lib/delay.c | 1 +
>>> 1 file changed, 1 insertion(+)
>>>
>>> diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c
>>> index dad8e6a..540e670 100644
>>> --- a/arch/sh/lib/delay.c
>>> +++ b/arch/sh/lib/delay.c
>>> @@ -29,6 +29,7 @@ void __delay(unsigned long loops)
>>> : "0" (loops)
>>> : "t");
>>> }
>>> +EXPORT_SYMBOL(__delay);
>>>
>>> inline void __const_udelay(unsigned long xloops)
>>> {
>
> I presently have a revert of this commit in queue for next. If it's
> sufficiently breaking (and especially if there are other regressions
> that need to be fixed, see the pmd_free thing) I could try to get it
> in for 5.8 still but that's getting a bit late.
>

The patch in mainline is ok. It appears that it has been applied
again in -next.

"git log --oneline v5.7.. arch/sh/lib/delay.c" on top of next-20200721
reports:

ee0e4f15dfd4 (origin/akpm) sh: add missing EXPORT_SYMBOL() for __delay
d1f56f318d23 sh: add missing EXPORT_SYMBOL() for __delay

Maybe it just needs to be dropped from the akpm tree in -next ?

Thanks,
Guenter

2020-07-24 19:45:39

by Rich Felker

[permalink] [raw]
Subject: Re: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay

On Wed, Jul 22, 2020 at 04:52:56PM -0700, Guenter Roeck wrote:
> Rich,
>
> On 7/22/20 3:52 PM, Rich Felker wrote:
> > On Tue, Jul 21, 2020 at 07:38:40PM -0700, Guenter Roeck wrote:
> >> On Thu, Dec 12, 2019 at 11:38:43AM +0900, Kuninori Morimoto wrote:
> >>> From: Kuninori Morimoto <[email protected]>
> >>>
> >>> __delay() is used from kernel module.
> >>> We need EXPORT_SYMBOL(), otherwise we will get compile error.
> >>>
> >>> ERROR: "__delay" [drivers/net/phy/mdio-cavium.ko] undefined!
> >>>
> >>> Signed-off-by: Kuninori Morimoto <[email protected]>
> >>
> >> I must admit that this patch completely baffles me. __delay was
> >> already exported, only elsewhere in the file. With this patch
> >> in place, it is exported twice, and all sh builds in -next fail
> >> with
> >>
> >> In file included from include/linux/linkage.h:7,
> >> from arch/sh/include/asm/bug.h:5,
> >> from include/linux/bug.h:5,
> >> from include/linux/thread_info.h:12,
> >> from include/asm-generic/current.h:5,
> >> from ./arch/sh/include/generated/asm/current.h:1,
> >> from include/linux/sched.h:12,
> >> from arch/sh/lib/delay.c:8:
> >> include/linux/export.h:67:36: error: redefinition of '__ksymtab___delay'
> >>
> >> Guenter
> >>
> >>> ---
> >>> arch/sh/lib/delay.c | 1 +
> >>> 1 file changed, 1 insertion(+)
> >>>
> >>> diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c
> >>> index dad8e6a..540e670 100644
> >>> --- a/arch/sh/lib/delay.c
> >>> +++ b/arch/sh/lib/delay.c
> >>> @@ -29,6 +29,7 @@ void __delay(unsigned long loops)
> >>> : "0" (loops)
> >>> : "t");
> >>> }
> >>> +EXPORT_SYMBOL(__delay);
> >>>
> >>> inline void __const_udelay(unsigned long xloops)
> >>> {
> >
> > I presently have a revert of this commit in queue for next. If it's
> > sufficiently breaking (and especially if there are other regressions
> > that need to be fixed, see the pmd_free thing) I could try to get it
> > in for 5.8 still but that's getting a bit late.
> >
>
> The patch in mainline is ok. It appears that it has been applied
> again in -next.
>
> "git log --oneline v5.7.. arch/sh/lib/delay.c" on top of next-20200721
> reports:
>
> ee0e4f15dfd4 (origin/akpm) sh: add missing EXPORT_SYMBOL() for __delay
> d1f56f318d23 sh: add missing EXPORT_SYMBOL() for __delay
>
> Maybe it just needs to be dropped from the akpm tree in -next ?

Could it be a bad merge or patch applied twice or something? I don't
see how the symbol is being exported twice. The argument to revert the
patch still seems correct.

Looking at current linux-next, ba722ca780 from akpm is re-adding the
patch after it gets reverted. Andrew, could you drop your copy of this
patch? It's already upstream as of 5.8-rc1 and now just needs to be
reverted upstream.

Rich

2020-07-24 23:27:10

by Andrew Morton

[permalink] [raw]
Subject: Re: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay

On Fri, 24 Jul 2020 15:44:23 -0400 Rich Felker <[email protected]> wrote:

> Andrew, could you drop your copy of this
> patch?

I have done so, thanks.

2020-08-03 10:02:29

by Geert Uytterhoeven

[permalink] [raw]
Subject: Re: [PATCH] sh: add missing EXPORT_SYMBOL() for __delay

On Thu, Jul 23, 2020 at 1:53 AM Guenter Roeck <[email protected]> wrote:
> On 7/22/20 3:52 PM, Rich Felker wrote:
> > On Tue, Jul 21, 2020 at 07:38:40PM -0700, Guenter Roeck wrote:
> >> On Thu, Dec 12, 2019 at 11:38:43AM +0900, Kuninori Morimoto wrote:
> >>> From: Kuninori Morimoto <[email protected]>
> >>>
> >>> __delay() is used from kernel module.
> >>> We need EXPORT_SYMBOL(), otherwise we will get compile error.
> >>>
> >>> ERROR: "__delay" [drivers/net/phy/mdio-cavium.ko] undefined!
> >>>
> >>> Signed-off-by: Kuninori Morimoto <[email protected]>
> >>
> >> I must admit that this patch completely baffles me. __delay was
> >> already exported, only elsewhere in the file. With this patch
> >> in place, it is exported twice, and all sh builds in -next fail
> >> with
> >>
> >> In file included from include/linux/linkage.h:7,
> >> from arch/sh/include/asm/bug.h:5,
> >> from include/linux/bug.h:5,
> >> from include/linux/thread_info.h:12,
> >> from include/asm-generic/current.h:5,
> >> from ./arch/sh/include/generated/asm/current.h:1,
> >> from include/linux/sched.h:12,
> >> from arch/sh/lib/delay.c:8:
> >> include/linux/export.h:67:36: error: redefinition of '__ksymtab___delay'
> >>
> >> Guenter
> >>
> >>> ---
> >>> arch/sh/lib/delay.c | 1 +
> >>> 1 file changed, 1 insertion(+)
> >>>
> >>> diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c
> >>> index dad8e6a..540e670 100644
> >>> --- a/arch/sh/lib/delay.c
> >>> +++ b/arch/sh/lib/delay.c
> >>> @@ -29,6 +29,7 @@ void __delay(unsigned long loops)
> >>> : "0" (loops)
> >>> : "t");
> >>> }
> >>> +EXPORT_SYMBOL(__delay);
> >>>
> >>> inline void __const_udelay(unsigned long xloops)
> >>> {
> >
> > I presently have a revert of this commit in queue for next. If it's
> > sufficiently breaking (and especially if there are other regressions
> > that need to be fixed, see the pmd_free thing) I could try to get it
> > in for 5.8 still but that's getting a bit late.
> >
>
> The patch in mainline is ok. It appears that it has been applied
> again in -next.
>
> "git log --oneline v5.7.. arch/sh/lib/delay.c" on top of next-20200721
> reports:
>
> ee0e4f15dfd4 (origin/akpm) sh: add missing EXPORT_SYMBOL() for __delay
> d1f56f318d23 sh: add missing EXPORT_SYMBOL() for __delay
>
> Maybe it just needs to be dropped from the akpm tree in -next ?

IMHO all of them should be dropped/reverted.
__delay is an internal implementation detail, not to be used by drivers.

See also include/asm-generic/delay.h:

/* Undefined functions to get compile-time errors */
...
extern void __delay(unsigned long loops);

Cfr. '[PATCH] Revert "sh: add missing EXPORT_SYMBOL() for __delay"'
https://lore.kernel.org/lkml/[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