[U-Boot] [PATCH v2 1/5] Migrate CONFIG_DRIVER_TI_CPSW to Kconfig
Alex Kiernan
alex.kiernan at gmail.com
Tue Apr 3 06:29:47 UTC 2018
On Mon, Apr 2, 2018 at 4:11 PM, Felix Brack <fb at ltec.ch> wrote:
> On 02.04.2018 17:43, Alex Kiernan wrote:
>> On Mon, Apr 2, 2018 at 12:13 PM, Felix Brack <fb at ltec.ch> wrote:
>>> Hi Alex,
>>>
>>> On 01.04.2018 11:22, Alex Kiernan wrote:
>>>> This converts CONFIG_DRIVER_TI_CPSW to Kconfig
>>>>
>>>> Signed-off-by: Alex Kiernan <alex.kiernan at gmail.com>
>>>> Acked-by: Joe Hershberger <joe.hershberger at ni.com>
>>>> ---
>>>>
>>>> Changes in v2:
>>>> - Move DRIVER_TI_CPSW outside of the NETDEVICES guard
>>>> - Don't mark DRIVER_TI_CPSW default if ARCH_OMAP2PLUS to fix mistranslations
>>>> by moveconfig
>>>>
>>>
>>> [..]
>>>
>>>> diff --git a/configs/am335x_pdu001_defconfig b/configs/am335x_pdu001_defconfig
>>>> index cb75ec0..87ae88c 100644
>>>> --- a/configs/am335x_pdu001_defconfig
>>>> +++ b/configs/am335x_pdu001_defconfig
>>>> @@ -9,13 +9,13 @@ CONFIG_SPL_SERIAL_SUPPORT=y
>>>> CONFIG_SPL_LIBDISK_SUPPORT=y
>>>> # CONFIG_SPL_NAND_SUPPORT is not set
>>>> CONFIG_SPL_WATCHDOG_SUPPORT=y
>>>> +CONFIG_SPL=y
>>>> CONFIG_SPL_FAT_SUPPORT=y
>>>> CONFIG_DEFAULT_DEVICE_TREE="am335x-pdu001"
>>>> CONFIG_LOCALVERSION="-EETS-1.0.0"
>>>> CONFIG_DISTRO_DEFAULTS=y
>>>> CONFIG_BOOTDELAY=1
>>>> # CONFIG_USE_BOOTCOMMAND is not set
>>>> -CONFIG_SPL=y
>>>> # CONFIG_SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR is not set
>>>> CONFIG_SPL_I2C_SUPPORT=y
>>>> CONFIG_SPL_YMODEM_SUPPORT=y
>>>> @@ -39,6 +39,7 @@ CONFIG_DM_GPIO=y
>>>> CONFIG_DM_I2C=y
>>>> CONFIG_MMC_OMAP_HS=y
>>>> CONFIG_MMC_SDHCI=y
>>>> +CONFIG_DRIVER_TI_CPSW=y
>>>
>>> Applying this patch series generates the following warning while
>>> creating the default configuration for board PDU001:
>>>
>>> warning: (BOARD_SPECIFIC_OPTIONS && BOARD_SPECIFIC_OPTIONS &&
>>> BOARD_SPECIFIC_OPTIONS && BOARD_SPECIFIC_OPTIONS && DRIVER_TI_CPSW &&
>>> AG7XXX && ALTERA_TSE && BCM_SF2_ETH && DWC_ETH_QOS && ETH_DESIGNWARE &&
>>> MVNETA && MVPP2 && MACB && PCH_GBE && SUN4I_EMAC && SUN8I_EMAC &&
>>> SH_ETHER && XILINX_AXIEMAC && XILINX_EMACLITE && ZYNQ_GEM && PIC32_ETH
>>> && RENESAS_RAVB) selects PHYLIB which has unmet direct dependencies (NET)
>>>
>>> This is due to the patch enabling CONFIG_DRIVER_TI_CPSW while leaving
>>> CONFIG_NET disabled.
>>> This board does not require/have network support for U-Boot so there is
>>> no need or benefit activating CONFIG_DRIVER_TI_CPSW here. Leaving the
>>> file configs/am355x_pdu001_defconfig without any modifications will make
>>> your patch work properly and result in a clean, warning and error free,
>>> build for the PDU001 board.
>>>
>>>> CONFIG_PINCTRL=y
>>>> CONFIG_PINCTRL_SINGLE=y
>>>> CONFIG_DM_PMIC=y
>>>
>>
>> Oh bother, thanks for trying it. I think in fixing the opposite
>> problem from v1 I've missed a depends on NET. Is your board covered by
>> the Travis tests as I did push v2 and got a green build?
>>
>
> From what I can see it is not, but I'm not sure. Using buildman with
> am335x would include it. I have verified that.
>
>> Unfortunately my laptop decided to die overnight and I'm away from the
>> office for a couple of weeks so I'll have to pick it up when I'm back.
>>
>> Alex
>>
>
> regards, Felix
It is included, just found it, but looks like it doesn't fail because
of that issue:
https://travis-ci.org/akiernan/u-boot/jobs/360830565#L962
I clearly need to go and understand the Travis build a whole lot better.
--
Alex Kiernan
More information about the U-Boot
mailing list