linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] ARM: shmobile: fix build regressions
@ 2018-12-10 21:00 Arnd Bergmann
  2018-12-11  8:22 ` Geert Uytterhoeven
  2018-12-11  8:58 ` Geert Uytterhoeven
  0 siblings, 2 replies; 6+ messages in thread
From: Arnd Bergmann @ 2018-12-10 21:00 UTC (permalink / raw)
  To: Simon Horman, Magnus Damm
  Cc: Arnd Bergmann, Geert Uytterhoeven, Biju Das, Fabrizio Castro,
	Yoshihiro Shimoda, Jacopo Mondi, Sergei Shtylyov,
	linux-renesas-soc, linux-kernel

A number of Kconfig options have become available now to random ARM
platforms outside of ARCH_MULTIPLATFORM, which now causes Kconfig
warnings, and other build errors when those select options that
lack additonal dependencies, e.g.:

WARNING: unmet direct dependencies detected for HAVE_ARM_ARCH_TIMER
  Depends on [n]: CPU_V7 [=n]
  Selected by [y]:
  - ARCH_RCAR_GEN2 [=y] && SOC_RENESAS [=y]
  - ARCH_R8A73A4 [=y] && SOC_RENESAS [=y] && ARM [=y]

WARNING: unmet direct dependencies detected for SYS_SUPPORTS_EM_STI
  Depends on [n]: GENERIC_CLOCKEVENTS [=n]
  Selected by [y]:
  - ARCH_EMEV2 [=y] && SOC_RENESAS [=y] && ARM [=y]

Put the old dependency on ARCH_RENESAS back for the moment to restore
the previous behavior.

Fixes: 062887bf5ef7 ("ARM: shmobile: Move SoC Kconfig symbols to drivers/soc/renesas/")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 drivers/soc/renesas/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/soc/renesas/Kconfig b/drivers/soc/renesas/Kconfig
index 4d8012e1205c..68bfca6f20dd 100644
--- a/drivers/soc/renesas/Kconfig
+++ b/drivers/soc/renesas/Kconfig
@@ -44,7 +44,7 @@ config ARCH_RZN1
 	bool
 	select ARM_AMBA
 
-if ARM
+if ARM && ARCH_RENESAS
 
 #comment "Renesas ARM SoCs System Type"
 
-- 
2.20.0


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] ARM: shmobile: fix build regressions
  2018-12-10 21:00 [PATCH] ARM: shmobile: fix build regressions Arnd Bergmann
@ 2018-12-11  8:22 ` Geert Uytterhoeven
  2018-12-11  8:37   ` Arnd Bergmann
  2018-12-11  8:58 ` Geert Uytterhoeven
  1 sibling, 1 reply; 6+ messages in thread
From: Geert Uytterhoeven @ 2018-12-11  8:22 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: Simon Horman, Magnus Damm, Geert Uytterhoeven, Biju Das,
	Fabrizio Castro, Yoshihiro Shimoda, Jacopo Mondi,
	Sergei Shtylyov, Linux-Renesas, Linux Kernel Mailing List

Hi Arnd,

On Mon, Dec 10, 2018 at 10:00 PM Arnd Bergmann <arnd@arndb.de> wrote:
> A number of Kconfig options have become available now to random ARM
> platforms outside of ARCH_MULTIPLATFORM, which now causes Kconfig
> warnings, and other build errors when those select options that
> lack additonal dependencies, e.g.:
>
> WARNING: unmet direct dependencies detected for HAVE_ARM_ARCH_TIMER
>   Depends on [n]: CPU_V7 [=n]
>   Selected by [y]:
>   - ARCH_RCAR_GEN2 [=y] && SOC_RENESAS [=y]
>   - ARCH_R8A73A4 [=y] && SOC_RENESAS [=y] && ARM [=y]
>
> WARNING: unmet direct dependencies detected for SYS_SUPPORTS_EM_STI
>   Depends on [n]: GENERIC_CLOCKEVENTS [=n]
>   Selected by [y]:
>   - ARCH_EMEV2 [=y] && SOC_RENESAS [=y] && ARM [=y]
>
> Put the old dependency on ARCH_RENESAS back for the moment to restore
> the previous behavior.

Thanks, the build bots must be lazy or overloaded these days...
The ARM selects are indeed more troublesome than the ARM64 ones.

BTW, did you see issues with the ARM64 Kconfig symbol move too,
or is that mitigated by ARM64 always selecting GENERIC_CLOCKEVENTS?

> Fixes: 062887bf5ef7 ("ARM: shmobile: Move SoC Kconfig symbols to drivers/soc/renesas/")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

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

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] ARM: shmobile: fix build regressions
  2018-12-11  8:22 ` Geert Uytterhoeven
@ 2018-12-11  8:37   ` Arnd Bergmann
  0 siblings, 0 replies; 6+ messages in thread
From: Arnd Bergmann @ 2018-12-11  8:37 UTC (permalink / raw)
  To: Geert Uytterhoeven
  Cc: Simon Horman, Magnus Damm, Geert Uytterhoeven, biju.das,
	fabrizio.castro, yoshihiro.shimoda.uh, Jacopo Mondi,
	Sergei Shtylyov, Linux-Renesas, Linux Kernel Mailing List

On Tue, Dec 11, 2018 at 9:22 AM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>
> Hi Arnd,
>
> On Mon, Dec 10, 2018 at 10:00 PM Arnd Bergmann <arnd@arndb.de> wrote:
> > A number of Kconfig options have become available now to random ARM
> > platforms outside of ARCH_MULTIPLATFORM, which now causes Kconfig
> > warnings, and other build errors when those select options that
> > lack additonal dependencies, e.g.:
> >
> > WARNING: unmet direct dependencies detected for HAVE_ARM_ARCH_TIMER
> >   Depends on [n]: CPU_V7 [=n]
> >   Selected by [y]:
> >   - ARCH_RCAR_GEN2 [=y] && SOC_RENESAS [=y]
> >   - ARCH_R8A73A4 [=y] && SOC_RENESAS [=y] && ARM [=y]
> >
> > WARNING: unmet direct dependencies detected for SYS_SUPPORTS_EM_STI
> >   Depends on [n]: GENERIC_CLOCKEVENTS [=n]
> >   Selected by [y]:
> >   - ARCH_EMEV2 [=y] && SOC_RENESAS [=y] && ARM [=y]
> >
> > Put the old dependency on ARCH_RENESAS back for the moment to restore
> > the previous behavior.
>
> Thanks, the build bots must be lazy or overloaded these days...
> The ARM selects are indeed more troublesome than the ARM64 ones.
>
> BTW, did you see issues with the ARM64 Kconfig symbol move too,
> or is that mitigated by ARM64 always selecting GENERIC_CLOCKEVENTS?

I restarted my randconfig builder yesterday, and after a few hundred builds
on each architecture, I did not see any arm64 failures, so it's
probably ok there.
I have a clean build now (about to rebase to today's linux-next, which will
surely change that).

> > Fixes: 062887bf5ef7 ("ARM: shmobile: Move SoC Kconfig symbols to drivers/soc/renesas/")
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>
> Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>

Thanks.

     Arnd

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] ARM: shmobile: fix build regressions
  2018-12-10 21:00 [PATCH] ARM: shmobile: fix build regressions Arnd Bergmann
  2018-12-11  8:22 ` Geert Uytterhoeven
@ 2018-12-11  8:58 ` Geert Uytterhoeven
  2018-12-12 10:30   ` Simon Horman
  1 sibling, 1 reply; 6+ messages in thread
From: Geert Uytterhoeven @ 2018-12-11  8:58 UTC (permalink / raw)
  To: Arnd Bergmann
  Cc: Simon Horman, Magnus Damm, Geert Uytterhoeven, Biju Das,
	Fabrizio Castro, Yoshihiro Shimoda, Jacopo Mondi,
	Sergei Shtylyov, Linux-Renesas, Linux Kernel Mailing List

Hi Arnd,

On Mon, Dec 10, 2018 at 10:00 PM Arnd Bergmann <arnd@arndb.de> wrote:
> A number of Kconfig options have become available now to random ARM
> platforms outside of ARCH_MULTIPLATFORM, which now causes Kconfig
> warnings, and other build errors when those select options that
> lack additonal dependencies, e.g.:
>
> WARNING: unmet direct dependencies detected for HAVE_ARM_ARCH_TIMER
>   Depends on [n]: CPU_V7 [=n]
>   Selected by [y]:
>   - ARCH_RCAR_GEN2 [=y] && SOC_RENESAS [=y]
>   - ARCH_R8A73A4 [=y] && SOC_RENESAS [=y] && ARM [=y]
>
> WARNING: unmet direct dependencies detected for SYS_SUPPORTS_EM_STI
>   Depends on [n]: GENERIC_CLOCKEVENTS [=n]
>   Selected by [y]:
>   - ARCH_EMEV2 [=y] && SOC_RENESAS [=y] && ARM [=y]
>
> Put the old dependency on ARCH_RENESAS back for the moment to restore
> the previous behavior.
>
> Fixes: 062887bf5ef7 ("ARM: shmobile: Move SoC Kconfig symbols to drivers/soc/renesas/")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  drivers/soc/renesas/Kconfig | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/soc/renesas/Kconfig b/drivers/soc/renesas/Kconfig
> index 4d8012e1205c..68bfca6f20dd 100644
> --- a/drivers/soc/renesas/Kconfig
> +++ b/drivers/soc/renesas/Kconfig
> @@ -44,7 +44,7 @@ config ARCH_RZN1
>         bool
>         select ARM_AMBA
>
> -if ARM
> +if ARM && ARCH_RENESAS

After a closer look, I think restricting this from ARM to ARCH_MULTI_V7
should work, too (ARCH_MULTI_V7 selects CPU_V7, and ARCH_MULTIPLATFORM
selects GENERIC_CLOCKEVENTS).

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

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

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] ARM: shmobile: fix build regressions
  2018-12-11  8:58 ` Geert Uytterhoeven
@ 2018-12-12 10:30   ` Simon Horman
  2018-12-17 12:47     ` Simon Horman
  0 siblings, 1 reply; 6+ messages in thread
From: Simon Horman @ 2018-12-12 10:30 UTC (permalink / raw)
  To: Geert Uytterhoeven
  Cc: Arnd Bergmann, Magnus Damm, Geert Uytterhoeven, Biju Das,
	Fabrizio Castro, Yoshihiro Shimoda, Jacopo Mondi,
	Sergei Shtylyov, Linux-Renesas, Linux Kernel Mailing List

On Tue, Dec 11, 2018 at 09:58:05AM +0100, Geert Uytterhoeven wrote:
> Hi Arnd,
> 
> On Mon, Dec 10, 2018 at 10:00 PM Arnd Bergmann <arnd@arndb.de> wrote:
> > A number of Kconfig options have become available now to random ARM
> > platforms outside of ARCH_MULTIPLATFORM, which now causes Kconfig
> > warnings, and other build errors when those select options that
> > lack additonal dependencies, e.g.:
> >
> > WARNING: unmet direct dependencies detected for HAVE_ARM_ARCH_TIMER
> >   Depends on [n]: CPU_V7 [=n]
> >   Selected by [y]:
> >   - ARCH_RCAR_GEN2 [=y] && SOC_RENESAS [=y]
> >   - ARCH_R8A73A4 [=y] && SOC_RENESAS [=y] && ARM [=y]
> >
> > WARNING: unmet direct dependencies detected for SYS_SUPPORTS_EM_STI
> >   Depends on [n]: GENERIC_CLOCKEVENTS [=n]
> >   Selected by [y]:
> >   - ARCH_EMEV2 [=y] && SOC_RENESAS [=y] && ARM [=y]
> >
> > Put the old dependency on ARCH_RENESAS back for the moment to restore
> > the previous behavior.
> >
> > Fixes: 062887bf5ef7 ("ARM: shmobile: Move SoC Kconfig symbols to drivers/soc/renesas/")
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > ---
> >  drivers/soc/renesas/Kconfig | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/soc/renesas/Kconfig b/drivers/soc/renesas/Kconfig
> > index 4d8012e1205c..68bfca6f20dd 100644
> > --- a/drivers/soc/renesas/Kconfig
> > +++ b/drivers/soc/renesas/Kconfig
> > @@ -44,7 +44,7 @@ config ARCH_RZN1
> >         bool
> >         select ARM_AMBA
> >
> > -if ARM
> > +if ARM && ARCH_RENESAS
> 
> After a closer look, I think restricting this from ARM to ARCH_MULTI_V7
> should work, too (ARCH_MULTI_V7 selects CPU_V7, and ARCH_MULTIPLATFORM
> selects GENERIC_CLOCKEVENTS).

Hi Arnd,

what is your preferred way forwards here?
I'm tempted to take your patch as-is and then work on
refining the dependency as a follow-up.

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] ARM: shmobile: fix build regressions
  2018-12-12 10:30   ` Simon Horman
@ 2018-12-17 12:47     ` Simon Horman
  0 siblings, 0 replies; 6+ messages in thread
From: Simon Horman @ 2018-12-17 12:47 UTC (permalink / raw)
  To: Geert Uytterhoeven
  Cc: Arnd Bergmann, Magnus Damm, Geert Uytterhoeven, Biju Das,
	Fabrizio Castro, Yoshihiro Shimoda, Jacopo Mondi,
	Sergei Shtylyov, Linux-Renesas, Linux Kernel Mailing List

On Wed, Dec 12, 2018 at 11:30:33AM +0100, Simon Horman wrote:
> On Tue, Dec 11, 2018 at 09:58:05AM +0100, Geert Uytterhoeven wrote:
> > Hi Arnd,
> > 
> > On Mon, Dec 10, 2018 at 10:00 PM Arnd Bergmann <arnd@arndb.de> wrote:
> > > A number of Kconfig options have become available now to random ARM
> > > platforms outside of ARCH_MULTIPLATFORM, which now causes Kconfig
> > > warnings, and other build errors when those select options that
> > > lack additonal dependencies, e.g.:
> > >
> > > WARNING: unmet direct dependencies detected for HAVE_ARM_ARCH_TIMER
> > >   Depends on [n]: CPU_V7 [=n]
> > >   Selected by [y]:
> > >   - ARCH_RCAR_GEN2 [=y] && SOC_RENESAS [=y]
> > >   - ARCH_R8A73A4 [=y] && SOC_RENESAS [=y] && ARM [=y]
> > >
> > > WARNING: unmet direct dependencies detected for SYS_SUPPORTS_EM_STI
> > >   Depends on [n]: GENERIC_CLOCKEVENTS [=n]
> > >   Selected by [y]:
> > >   - ARCH_EMEV2 [=y] && SOC_RENESAS [=y] && ARM [=y]
> > >
> > > Put the old dependency on ARCH_RENESAS back for the moment to restore
> > > the previous behavior.
> > >
> > > Fixes: 062887bf5ef7 ("ARM: shmobile: Move SoC Kconfig symbols to drivers/soc/renesas/")
> > > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > > ---
> > >  drivers/soc/renesas/Kconfig | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/soc/renesas/Kconfig b/drivers/soc/renesas/Kconfig
> > > index 4d8012e1205c..68bfca6f20dd 100644
> > > --- a/drivers/soc/renesas/Kconfig
> > > +++ b/drivers/soc/renesas/Kconfig
> > > @@ -44,7 +44,7 @@ config ARCH_RZN1
> > >         bool
> > >         select ARM_AMBA
> > >
> > > -if ARM
> > > +if ARM && ARCH_RENESAS
> > 
> > After a closer look, I think restricting this from ARM to ARCH_MULTI_V7
> > should work, too (ARCH_MULTI_V7 selects CPU_V7, and ARCH_MULTIPLATFORM
> > selects GENERIC_CLOCKEVENTS).
> 
> Hi Arnd,
> 
> what is your preferred way forwards here?
> I'm tempted to take your patch as-is and then work on
> refining the dependency as a follow-up.

I have gone ahead and applied Arnd's patch as a fix for v4.21.
It seems like the safest option for a fix.

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2018-12-17 12:47 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-12-10 21:00 [PATCH] ARM: shmobile: fix build regressions Arnd Bergmann
2018-12-11  8:22 ` Geert Uytterhoeven
2018-12-11  8:37   ` Arnd Bergmann
2018-12-11  8:58 ` Geert Uytterhoeven
2018-12-12 10:30   ` Simon Horman
2018-12-17 12:47     ` Simon Horman

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).