Message ID | 20170411183725.4183-9-fcooper@ti.com |
---|---|
State | Superseded |
Delegated to: | Heiko Schocher |
Headers | show |
On Tue, Apr 11, 2017 at 01:37:25PM -0500, Franklin S Cooper Jr wrote: > Now that the davinci I2C driver is converted to driver model enable > it in 66AK2Gx defconfig > > Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> > --- > configs/k2e_evm_defconfig | 1 + > configs/k2g_evm_defconfig | 1 + > configs/k2hk_evm_defconfig | 1 + > configs/k2l_evm_defconfig | 1 + > 4 files changed, 4 insertions(+) Should we be imply/select'ing DM_I2C from the Kconfig file now?
On 04/12/2017 02:13 PM, Tom Rini wrote: > On Tue, Apr 11, 2017 at 01:37:25PM -0500, Franklin S Cooper Jr wrote: > >> Now that the davinci I2C driver is converted to driver model enable >> it in 66AK2Gx defconfig >> >> Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> >> --- >> configs/k2e_evm_defconfig | 1 + >> configs/k2g_evm_defconfig | 1 + >> configs/k2hk_evm_defconfig | 1 + >> configs/k2l_evm_defconfig | 1 + >> 4 files changed, 4 insertions(+) > > Should we be imply/select'ing DM_I2C from the Kconfig file now? Is the goal to reduce the size of the defconfig? Or is it to essentially force all K2 boards (including possible future non TI boards) to use the DM so we can deprecate/remove non DM code in the future? If its for the latter then I don't have an issue doing so. >
On Wed, Apr 12, 2017 at 02:26:21PM -0500, Franklin S Cooper Jr wrote: > > > On 04/12/2017 02:13 PM, Tom Rini wrote: > > On Tue, Apr 11, 2017 at 01:37:25PM -0500, Franklin S Cooper Jr wrote: > > > >> Now that the davinci I2C driver is converted to driver model enable > >> it in 66AK2Gx defconfig > >> > >> Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> > >> --- > >> configs/k2e_evm_defconfig | 1 + > >> configs/k2g_evm_defconfig | 1 + > >> configs/k2hk_evm_defconfig | 1 + > >> configs/k2l_evm_defconfig | 1 + > >> 4 files changed, 4 insertions(+) > > > > Should we be imply/select'ing DM_I2C from the Kconfig file now? > > Is the goal to reduce the size of the defconfig? Or is it to essentially > force all K2 boards (including possible future non TI boards) to use the > DM so we can deprecate/remove non DM code in the future? > > If its for the latter then I don't have an issue doing so. Well, both. I think 'imply' is a great way to specify default values in such a way that making new boards/defconfigs is easy and likely to be correct. But also, yes, non-DM will go away at some point.
On 04/12/2017 02:33 PM, Tom Rini wrote: > On Wed, Apr 12, 2017 at 02:26:21PM -0500, Franklin S Cooper Jr wrote: >> >> >> On 04/12/2017 02:13 PM, Tom Rini wrote: >>> On Tue, Apr 11, 2017 at 01:37:25PM -0500, Franklin S Cooper Jr wrote: >>> >>>> Now that the davinci I2C driver is converted to driver model enable >>>> it in 66AK2Gx defconfig >>>> >>>> Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> >>>> --- >>>> configs/k2e_evm_defconfig | 1 + >>>> configs/k2g_evm_defconfig | 1 + >>>> configs/k2hk_evm_defconfig | 1 + >>>> configs/k2l_evm_defconfig | 1 + >>>> 4 files changed, 4 insertions(+) >>> >>> Should we be imply/select'ing DM_I2C from the Kconfig file now? >> >> Is the goal to reduce the size of the defconfig? Or is it to essentially >> force all K2 boards (including possible future non TI boards) to use the >> DM so we can deprecate/remove non DM code in the future? >> >> If its for the latter then I don't have an issue doing so. > > Well, both. I think 'imply' is a great way to specify default values in > such a way that making new boards/defconfigs is easy and likely to be > correct. But also, yes, non-DM will go away at some point. > I'm ok with that. Do you want me to resend this patchset and switch this patch from enabling DM_I2C via defconfig to KConfig? Or you can just drop this patch from the patchset and then I can just send a separate patch enabling DM_I2C via KConfig. I'm ok with what ever you prefer.
On Wed, Apr 12, 2017 at 02:41:50PM -0500, Franklin S Cooper Jr wrote: > > > On 04/12/2017 02:33 PM, Tom Rini wrote: > > On Wed, Apr 12, 2017 at 02:26:21PM -0500, Franklin S Cooper Jr wrote: > >> > >> > >> On 04/12/2017 02:13 PM, Tom Rini wrote: > >>> On Tue, Apr 11, 2017 at 01:37:25PM -0500, Franklin S Cooper Jr wrote: > >>> > >>>> Now that the davinci I2C driver is converted to driver model enable > >>>> it in 66AK2Gx defconfig > >>>> > >>>> Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> > >>>> --- > >>>> configs/k2e_evm_defconfig | 1 + > >>>> configs/k2g_evm_defconfig | 1 + > >>>> configs/k2hk_evm_defconfig | 1 + > >>>> configs/k2l_evm_defconfig | 1 + > >>>> 4 files changed, 4 insertions(+) > >>> > >>> Should we be imply/select'ing DM_I2C from the Kconfig file now? > >> > >> Is the goal to reduce the size of the defconfig? Or is it to essentially > >> force all K2 boards (including possible future non TI boards) to use the > >> DM so we can deprecate/remove non DM code in the future? > >> > >> If its for the latter then I don't have an issue doing so. > > > > Well, both. I think 'imply' is a great way to specify default values in > > such a way that making new boards/defconfigs is easy and likely to be > > correct. But also, yes, non-DM will go away at some point. > > > > I'm ok with that. Do you want me to resend this patchset and switch this > patch from enabling DM_I2C via defconfig to KConfig? Or you can just > drop this patch from the patchset and then I can just send a separate > patch enabling DM_I2C via KConfig. > > I'm ok with what ever you prefer. Do a new 8/8 that does imply instead please, thanks!
diff --git a/configs/k2e_evm_defconfig b/configs/k2e_evm_defconfig index d319705..52e5bd4 100644 --- a/configs/k2e_evm_defconfig +++ b/configs/k2e_evm_defconfig @@ -54,3 +54,4 @@ CONFIG_USB=y CONFIG_USB_XHCI_HCD=y CONFIG_USB_XHCI_DWC3=y CONFIG_USB_STORAGE=y +CONFIG_DM_I2C=y diff --git a/configs/k2g_evm_defconfig b/configs/k2g_evm_defconfig index bbdcc2f..d4cdf61 100644 --- a/configs/k2g_evm_defconfig +++ b/configs/k2g_evm_defconfig @@ -58,3 +58,4 @@ CONFIG_USB=y CONFIG_USB_XHCI_HCD=y CONFIG_USB_XHCI_DWC3=y CONFIG_USB_STORAGE=y +CONFIG_DM_I2C=y diff --git a/configs/k2hk_evm_defconfig b/configs/k2hk_evm_defconfig index b22086d..7d22cd6 100644 --- a/configs/k2hk_evm_defconfig +++ b/configs/k2hk_evm_defconfig @@ -54,3 +54,4 @@ CONFIG_USB=y CONFIG_USB_XHCI_HCD=y CONFIG_USB_XHCI_DWC3=y CONFIG_USB_STORAGE=y +CONFIG_DM_I2C=y diff --git a/configs/k2l_evm_defconfig b/configs/k2l_evm_defconfig index 5a28112..ec2fcbf 100644 --- a/configs/k2l_evm_defconfig +++ b/configs/k2l_evm_defconfig @@ -54,3 +54,4 @@ CONFIG_USB=y CONFIG_USB_XHCI_HCD=y CONFIG_USB_XHCI_DWC3=y CONFIG_USB_STORAGE=y +CONFIG_DM_I2C=y
Now that the davinci I2C driver is converted to driver model enable it in 66AK2Gx defconfig Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> --- configs/k2e_evm_defconfig | 1 + configs/k2g_evm_defconfig | 1 + configs/k2hk_evm_defconfig | 1 + configs/k2l_evm_defconfig | 1 + 4 files changed, 4 insertions(+)