On Thursday, November 17, 2016 10:30:10 AM CET kbuild test robot wrote:
>
> url:
> https://github.com/0day-ci/linux/commits/Florian-Fainelli/net-fsl-Allow-most-drivers-to-be-built-with-COMPILE_TEST/20161116-094841
> config: m32r-allmodconfig (attached as .config)
> compiler: m32r-linux-gcc (GCC) 6.2.0
> reproduce:
> wget
> https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross
> -O ~/bin/make.cross
> chmod +x ~/bin/make.cross
> # save the attached .config to linux build tree
> make.cross ARCH=m32r
>
> All errors (new ones prefixed by >>):
>
> ERROR: "bad_dma_ops" [sound/core/snd-pcm.ko] undefined!
> ERROR: "dma_common_mmap" [sound/core/snd-pcm.ko] undefined!
> >> ERROR: "bad_dma_ops" [drivers/net/ethernet/marvell/mvpp2.ko] undefined!
> ERROR: "mvebu_mbus_get_dram_win_info"
> [drivers/net/ethernet/marvell/mvneta_bm.ko] undefined!
> >> ERROR: "bad_dma_ops" [drivers/net/ethernet/marvell/mvneta_bm.ko] undefined!
> >> ERROR: "bad_dma_ops" [drivers/net/ethernet/marvell/mvneta.ko] undefined!
> >> ERROR: "bad_dma_ops" [drivers/net/ethernet/marvell/mv643xx_eth.ko]
> >> undefined!
> ERROR: "bad_dma_ops" [drivers/net/ethernet/freescale/gianfar_driver.ko]
> undefined!
>
>
This can be avoided by adding 'depends on HAS_DMA'. We should probably just
fix m32r instead to not require that and allow building all drivers even when
they can't work in the end.
Arnd
