On Wed, May 6, 2020 at 10:57 PM Grygorii Strashko
<grygorii.strashko@xxxxxx> wrote:
> On 06/05/2020 09:51, Clay McClure wrote:
> > On Tue, May 05, 2020 at 10:41:26AM +0300, Grygorii Strashko wrote:
> >
>
> Ok. After some thinking and hence you commit b6d49cab44b5 ("net: Make PTP-specific drivers depend on PTP_1588_CLOCK")
> seems was merged in net (not net-next)
> 1) for-net: defconfig changes can be separated to fix build fail, but add change for multi_v7_defconfig
> 2) for-net-next: rest of changes plus below diff on top of it
>
> diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig
> index f3f8bb724294..62f809b67469 100644
> --- a/drivers/net/ethernet/ti/Kconfig
> +++ b/drivers/net/ethernet/ti/Kconfig
> @@ -49,6 +49,7 @@ config TI_CPSW_PHY_SEL
> config TI_CPSW
> tristate "TI CPSW Switch Support"
> depends on ARCH_DAVINCI || ARCH_OMAP2PLUS || COMPILE_TEST
> + depends on TI_CPTS || !TI_CPTS
> select TI_DAVINCI_MDIO
> select MFD_SYSCON
> select PAGE_POOL
> @@ -64,6 +65,7 @@ config TI_CPSW_SWITCHDEV
> tristate "TI CPSW Switch Support with switchdev"
> depends on ARCH_DAVINCI || ARCH_OMAP2PLUS || COMPILE_TEST
> depends on NET_SWITCHDEV
> + depends on TI_CPTS || !TI_CPTS
> select PAGE_POOL
> select TI_DAVINCI_MDIO
> select MFD_SYSCON
> @@ -78,11 +80,9 @@ config TI_CPSW_SWITCHDEV
>
> config TI_CPTS
> tristate "TI Common Platform Time Sync (CPTS) Support"
> - depends on TI_CPSW || TI_KEYSTONE_NETCP || TI_CPSW_SWITCHDEV || COMPILE_TEST
> + depends on ARCH_OMAP2PLUS || ARCH_KEYSTONE || COMPILE_TEST
> depends on COMMON_CLK
> depends on PTP_1588_CLOCK
> - default y if TI_CPSW=y || TI_KEYSTONE_NETCP=y || TI_CPSW_SWITCHDEV=y
> - default m
> ---help---
> This driver supports the Common Platform Time Sync unit of
> the CPSW Ethernet Switch and Keystone 2 1g/10g Switch Subsystem.
> @@ -109,6 +109,7 @@ config TI_KEYSTONE_NETCP
> select TI_DAVINCI_MDIO
> depends on OF
> depends on KEYSTONE_NAVIGATOR_DMA && KEYSTONE_NAVIGATOR_QMSS
> + depends on TI_CPTS || !TI_CPTS
> ---help---
> This driver supports TI's Keystone NETCP Core.
>
> It should properly resolve "M" vs "Y" dependencies between
> PTP_1588_CLOCK->TI_CPTS->TI_CPSW
>
> On thing, TI_CPTS can be selected without TI_CPSW, but it's probably ok.
I think that solution is reasonable. I had come up with a different
for when I ran
into the build failure, but I like yours better. Here is my patch, for
reference:
commit 94233d78655876f735189890eb40ef33c49e8523 (HEAD -> randconfig)
Author: Arnd Bergmann <arnd@xxxxxxxx>
Date: Thu May 7 21:25:59 2020 +0200
cpsw: fix cpts link failure
When CONFIG_PTP_1588_CLOCK is a loadable module, trying to build cpts
support into the built-in cpsw driver results in a link error:
arm-linux-gnueabi-ld: drivers/net/ethernet/ti/cpsw.o: in function
`cpsw_probe':
cpsw.c:(.text+0x918): undefined reference to `cpts_release'
arm-linux-gnueabi-ld: drivers/net/ethernet/ti/cpsw.o: in function
`cpsw_remove':
cpsw.c:(.text+0x1048): undefined reference to `cpts_release'
arm-linux-gnueabi-ld: drivers/net/ethernet/ti/cpsw.o: in function
`cpsw_rx_handler':
cpsw.c:(.text+0x12c0): undefined reference to `cpts_rx_timestamp'
arm-linux-gnueabi-ld: drivers/net/ethernet/ti/cpsw.o: in function
`cpsw_ndo_open':
Add a dependency for CPTS that only allows it to be enabled when
doing so does not cause link-time probles.
Fixes: b6d49cab44b5 ("net: Make PTP-specific drivers depend on
PTP_1588_CLOCK")
Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig
index 4ab35ce7b451..571caf4192c5 100644
--- a/drivers/net/ethernet/ti/Kconfig
+++ b/drivers/net/ethernet/ti/Kconfig
@@ -79,6 +79,9 @@ config TI_CPSW_SWITCHDEV
config TI_CPTS
bool "TI Common Platform Time Sync (CPTS) Support"
depends on TI_CPSW || TI_KEYSTONE_NETCP || TI_CPSW_SWITCHDEV
|| COMPILE_TEST
+ depends on TI_CPSW=n || TI_CPSW=PTP_1588_CLOCK || PTP_1588_CLOCK=y
+ depends on TI_KEYSTONE_NETCP=n ||
TI_KEYSTONE_NETCP=PTP_1588_CLOCK || PTP_1588_CLOCK=y
+ depends on TI_CPSW_SWITCHDEV=n ||
TI_CPSW_SWITCHDEV=PTP_1588_CLOCK || PTP_1588_CLOCK=y
depends on COMMON_CLK
depends on POSIX_TIMERS
---help---
