From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lee Jones Subject: Re: unmet direct dependencies in -next Date: Mon, 10 Aug 2015 10:10:04 +0100 Message-ID: <20150810091004.GC3249@x1> References: <55C71FF5.1020902@nod.at> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <55C71FF5.1020902-/L3Ra7n9ekc@public.gmane.org> Sender: linux-i2c-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Richard Weinberger Cc: linux-next-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, sameo-VuQAYsv1563Yd54FQh9/CA@public.gmane.org, Arnd Bergmann , yann.morin.1998-GANU6spQydw@public.gmane.org, linux-kbuild , Wolfram Sang , "open list:I2C SUBSYSTEM" , user-mode-linux-devel List-Id: linux-next.vger.kernel.org On Sun, 09 Aug 2015, Richard Weinberger wrote: > Hi! >=20 > -next faces some build issues on UML because of unmet direct dependen= cies. > Mostly due to HAS_IOMEM and I2C. >=20 > warning: (MEDIA_SUBDRV_AUTOSELECT && VIDEO_CX231XX && INV_MPU6050_IIO= ) selects I2C_MUX which has unmet direct dependencies (I2C && HAS_IOMEM= ) > warning: (ST_IRQCHIP && HIP04_ETH && STMMAC_PLATFORM && DWMAC_IPQ806X= && DWMAC_LPC18XX && DWMAC_ROCKCHIP && DWMAC_SOCFPGA && DWMAC_STI && TI= _CPSW && PINCTRL_ROCKCHIP && > PINCTRL_DOVE && POWER_RESET_KEYSTONE && POWER_RESET_SYSCON && POWER_R= ESET_SYSCON_POWEROFF && S3C2410_WATCHDOG && VIDEO_OMAP3 && VIDEO_S5P_FI= MC && RTC_DRV_AT91SAM9 && VIDEO_OMAP4 && > HWSPINLOCK_QCOM && ATMEL_ST && QCOM_GSBI) selects MFD_SYSCON which ha= s unmet direct dependencies (HAS_IOMEM) > warning: (MEDIA_SUBDRV_AUTOSELECT && VIDEO_CX231XX && INV_MPU6050_IIO= ) selects I2C_MUX which has unmet direct dependencies (I2C && HAS_IOMEM= ) > warning: (ST_IRQCHIP && HIP04_ETH && STMMAC_PLATFORM && DWMAC_IPQ806X= && DWMAC_LPC18XX && DWMAC_ROCKCHIP && DWMAC_SOCFPGA && DWMAC_STI && TI= _CPSW && PINCTRL_ROCKCHIP && > PINCTRL_DOVE && POWER_RESET_KEYSTONE && POWER_RESET_SYSCON && POWER_R= ESET_SYSCON_POWEROFF && S3C2410_WATCHDOG && VIDEO_OMAP3 && VIDEO_S5P_FI= MC && RTC_DRV_AT91SAM9 && VIDEO_OMAP4 && > HWSPINLOCK_QCOM && ATMEL_ST && QCOM_GSBI) selects MFD_SYSCON which ha= s unmet direct dependencies (HAS_IOMEM) >=20 > For example MFD_SYSCON cannot build on UML as it depends on HAS_IOMEM= =2E > While the symbol MFD_SYSCON has correct dependencies some users of MF= D_SYSCON > just issue a "select MFD_SYSCON" and bypass the HAS_IOMEM dependency = and causing the build to fail. >=20 > This brings me to a question on kconfig itself, wouldn't it be better= to just disable a symbol if it has > unmet direct dependencies? If this is a question about "kconfig itself", you should probably have included the correct ML and Maintainer: KCONFIG M: Michal Marek L: linux-kbuild-u79uwXL29TY76Z2rM5mHXA@public.gmane.org S: Odd Fixes =46: Documentation/kbuild/kconfig-language.txt =46: scripts/kconfig/ There are also quite a few interested parties on LKML who might be inclined to submit a fix for you. --=20 Lee Jones Linaro STMicroelectronics Landing Team Lead Linaro.org =E2=94=82 Open source software for ARM SoCs =46ollow Linaro: Facebook | Twitter | Blog