From: Peng Fan <[email protected]>
V2:
Include Leonard's patch to fix build break after enable compile test
Add Leonard's R-b tag
Rename soc-imx8.c to soc-imx8m.c which is for i.MX8M family
Add SOC_IMX8M for build gate soc-imx8m.c
Increase build coverage for i.MX SoC driver
Leonard Crestez (1):
soc: imx: gpcv2: include linux/sizes.h
Peng Fan (4):
soc: imx: Kconfig: add SOC_IMX8M entry
arm64: defconfig: Enable CONFIG_SOC_IMX8M by default
soc: Makefile: increase build coverage for i.MX
soc: imx: Use CONFIG_SOC_IMX8M as build gate
arch/arm64/configs/defconfig | 1 +
drivers/soc/Makefile | 2 +-
drivers/soc/imx/Kconfig | 8 ++++++++
drivers/soc/imx/Makefile | 2 +-
drivers/soc/imx/gpcv2.c | 1 +
drivers/soc/imx/{soc-imx8.c => soc-imx8m.c} | 0
6 files changed, 12 insertions(+), 2 deletions(-)
rename drivers/soc/imx/{soc-imx8.c => soc-imx8m.c} (100%)
--
2.16.4
From: Leonard Crestez <[email protected]>
This header is included indirectly on arm/arm64 but not on x86 so
CONFIG_COMPILE_TEST breaks. Fix by including <linux/sizes.h> directly.
Signed-off-by: Leonard Crestez <[email protected]>
---
drivers/soc/imx/gpcv2.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/soc/imx/gpcv2.c b/drivers/soc/imx/gpcv2.c
index b0dffb06c05d..6cf8a7a412bd 100644
--- a/drivers/soc/imx/gpcv2.c
+++ b/drivers/soc/imx/gpcv2.c
@@ -14,6 +14,7 @@
#include <linux/pm_domain.h>
#include <linux/regmap.h>
#include <linux/regulator/consumer.h>
+#include <linux/sizes.h>
#include <dt-bindings/power/imx7-power.h>
#include <dt-bindings/power/imx8mq-power.h>
--
2.16.4
From: Peng Fan <[email protected]>
Increase build coverage for i.MX SoC drivers.
Reviewed-by: Leonard Crestez <[email protected]>
Signed-off-by: Peng Fan <[email protected]>
---
drivers/soc/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/soc/Makefile b/drivers/soc/Makefile
index 2ec355003524..614986cd1713 100644
--- a/drivers/soc/Makefile
+++ b/drivers/soc/Makefile
@@ -11,7 +11,7 @@ obj-$(CONFIG_ARCH_DOVE) += dove/
obj-$(CONFIG_MACH_DOVE) += dove/
obj-y += fsl/
obj-$(CONFIG_ARCH_GEMINI) += gemini/
-obj-$(CONFIG_ARCH_MXC) += imx/
+obj-y += imx/
obj-$(CONFIG_ARCH_IXP4XX) += ixp4xx/
obj-$(CONFIG_SOC_XWAY) += lantiq/
obj-y += mediatek/
--
2.16.4
On Mon, Jan 27, 2020 at 10:44 AM Peng Fan <[email protected]> wrote:
>
> From: Peng Fan <[email protected]>
>
>
> V2:
> Include Leonard's patch to fix build break after enable compile test
> Add Leonard's R-b tag
>
> Rename soc-imx8.c to soc-imx8m.c which is for i.MX8M family
> Add SOC_IMX8M for build gate soc-imx8m.c
> Increase build coverage for i.MX SoC driver
The changes all look good to me, but I'd just do it all in one combined
patch, as the changes are all logically part of the same thing. You
can leave Leonard's fix as a [PATCH 1/2] if you want, but the rest
should clearly be a single change.
Arnd
Hi Arnd,
> Subject: Re: [PATCH V2 0/5] soc: imx: increase build coverage for imx8 soc
> driver
>
> On Mon, Jan 27, 2020 at 10:44 AM Peng Fan <[email protected]> wrote:
> >
> > From: Peng Fan <[email protected]>
> >
> >
> > V2:
> > Include Leonard's patch to fix build break after enable compile test
> > Add Leonard's R-b tag
> >
> > Rename soc-imx8.c to soc-imx8m.c which is for i.MX8M family Add
> > SOC_IMX8M for build gate soc-imx8m.c Increase build coverage for i.MX
> > SoC driver
>
> The changes all look good to me, but I'd just do it all in one combined patch, as
> the changes are all logically part of the same thing. You can leave Leonard's fix
> as a [PATCH 1/2] if you want, but the rest should clearly be a single change.
There is a arm64 defconfig change, should it be also included in the single change?
Thanks,
Peng.
>
> Arnd
On Mon, Jan 27, 2020 at 1:33 PM Peng Fan <[email protected]> wrote:
>
> Hi Arnd,
>
> > Subject: Re: [PATCH V2 0/5] soc: imx: increase build coverage for imx8 soc
> > driver
> >
> > On Mon, Jan 27, 2020 at 10:44 AM Peng Fan <[email protected]> wrote:
> > >
> > > From: Peng Fan <[email protected]>
> > >
> > >
> > > V2:
> > > Include Leonard's patch to fix build break after enable compile test
> > > Add Leonard's R-b tag
> > >
> > > Rename soc-imx8.c to soc-imx8m.c which is for i.MX8M family Add
> > > SOC_IMX8M for build gate soc-imx8m.c Increase build coverage for i.MX
> > > SoC driver
> >
> > The changes all look good to me, but I'd just do it all in one combined patch, as
> > the changes are all logically part of the same thing. You can leave Leonard's fix
> > as a [PATCH 1/2] if you want, but the rest should clearly be a single change.
>
> There is a arm64 defconfig change, should it be also included in the single change?
Good point, that one is probably better left separate indeed.
Arnd
> Subject: Re: [PATCH V2 0/5] soc: imx: increase build coverage for imx8 soc
> driver
>
> On Mon, Jan 27, 2020 at 1:33 PM Peng Fan <[email protected]> wrote:
> >
> > Hi Arnd,
> >
> > > Subject: Re: [PATCH V2 0/5] soc: imx: increase build coverage for
> > > imx8 soc driver
> > >
> > > On Mon, Jan 27, 2020 at 10:44 AM Peng Fan <[email protected]> wrote:
> > > >
> > > > From: Peng Fan <[email protected]>
> > > >
> > > >
> > > > V2:
> > > > Include Leonard's patch to fix build break after enable compile
> > > > test Add Leonard's R-b tag
> > > >
> > > > Rename soc-imx8.c to soc-imx8m.c which is for i.MX8M family Add
> > > > SOC_IMX8M for build gate soc-imx8m.c Increase build coverage for
> > > > i.MX SoC driver
> > >
> > > The changes all look good to me, but I'd just do it all in one
> > > combined patch, as the changes are all logically part of the same
> > > thing. You can leave Leonard's fix as a [PATCH 1/2] if you want, but the
> rest should clearly be a single change.
> >
> > There is a arm64 defconfig change, should it be also included in the single
> change?
>
> Good point, that one is probably better left separate indeed.
Since the defconfig change needs stay alone in a patch,
merge other patches into one might not be good. The patchset
I did is to make sure the soc-imx8m.c could always be built. If
I merge the others into one, without the defconfig patch set CONFIG
option to y, soc-imx8m.c will not be built. This might break git bisect
to check the soc-imx8m.c
So I prefer not to merge the others into one patch. Do you agree?
Thanks,
Peng.
>
> Arnd
On Mon, Jan 27, 2020 at 2:22 PM Peng Fan <[email protected]> wrote:
>
> > Subject: Re: [PATCH V2 0/5] soc: imx: increase build coverage for imx8 soc
> > driver
> >
> > On Mon, Jan 27, 2020 at 1:33 PM Peng Fan <[email protected]> wrote:
> > >
> > > > Subject: Re: [PATCH V2 0/5] soc: imx: increase build coverage for
> > > > imx8 soc driver
> > > >
> > > > On Mon, Jan 27, 2020 at 10:44 AM Peng Fan <[email protected]> wrote:
> > > > >
> > > > > From: Peng Fan <[email protected]>
> > > > >
> > > > >
> > > > > V2:
> > > > > Include Leonard's patch to fix build break after enable compile
> > > > > test Add Leonard's R-b tag
> > > > >
> > > > > Rename soc-imx8.c to soc-imx8m.c which is for i.MX8M family Add
> > > > > SOC_IMX8M for build gate soc-imx8m.c Increase build coverage for
> > > > > i.MX SoC driver
> > > >
> > > > The changes all look good to me, but I'd just do it all in one
> > > > combined patch, as the changes are all logically part of the same
> > > > thing. You can leave Leonard's fix as a [PATCH 1/2] if you want, but the
> > rest should clearly be a single change.
> > >
> > > There is a arm64 defconfig change, should it be also included in the single
> > change?
> >
> > Good point, that one is probably better left separate indeed.
>
> Since the defconfig change needs stay alone in a patch,
> merge other patches into one might not be good. The patchset
> I did is to make sure the soc-imx8m.c could always be built. If
> I merge the others into one, without the defconfig patch set CONFIG
> option to y, soc-imx8m.c will not be built. This might break git bisect
> to check the soc-imx8m.c
>
> So I prefer not to merge the others into one patch. Do you agree?
I'm generally not too worried about intermittent defconfig breaks during
bisection, as the defconfig is not use all that much in practice. Splitting
the other changes into separate patches wouldn't help here either
unless you want to spread it out over multiple merge windows.
I'd probably just put it all in one patch (including the defconfig change)
in this case, alternatively you could add a 'default ARCH_MXC && ARM64'
to the Kconfig symbol.
Arnd