Hi
On 1/18/19 6:27 PM, Randy Dunlap wrote:
On 1/17/19 8:24 PM, Stephen Rothwell wrote:
Hi all,
Changes since 20190117:
on i386 or x86_64:
ld: drivers/i2c/busses/i2c-designware-platdrv.o: in function `dw_i2c_plat_resume':
i2c-designware-platdrv.c:(.text+0x4b): undefined reference to `i2c_dw_prepare_clk'
ld: drivers/i2c/busses/i2c-designware-platdrv.o: in function `dw_i2c_plat_suspend':
i2c-designware-platdrv.c:(.text+0x95): undefined reference to `i2c_dw_prepare_clk'
ld: drivers/i2c/busses/i2c-designware-platdrv.o: in function `dw_i2c_plat_probe':
i2c-designware-platdrv.c:(.text+0x63f): undefined reference to `i2c_dw_prepare_clk'
ld: i2c-designware-platdrv.c:(.text+0x686): undefined reference to `i2c_dw_read_comp_param'
ld: i2c-designware-platdrv.c:(.text+0x7b7): undefined reference to `i2c_dw_probe_slave'
ld: i2c-designware-platdrv.c:(.text+0x7c1): undefined reference to `i2c_dw_probe'
probably related to:
WARNING: unmet direct dependencies detected for I2C_DESIGNWARE_PLATFORM
Depends on [n]: I2C [=y] && HAS_IOMEM [=y] && (ACPI [=y] && COMMON_CLK [=n] || !ACPI [=y])
Selected by [y]:
- MFD_TPS68470 [=y] && HAS_IOMEM [=y] && ACPI [=y] && I2C [=y]=y
Thanks, this is due CONFIG_MFD_TPS68470 which selects the
I2C_DESIGNWARE_PLATFORM but not the dependencies. There are two others
in drivers/mfd/Kconfig. I'll send a patch changing them to depend on
I2C_DESIGNWARE_PLATFORM=y.
--
Jarkko
