Messages in this thread | | | Date | Thu, 5 Nov 2015 21:32:15 +0900 | Subject | Re: [RFC PATCH 2/7] spi: sunxi: remove redundant "depends on RESET_CONTROLLER" | From | Masahiro Yamada <> |
| |
2015-11-05 21:20 GMT+09:00 Mark Brown <broonie@kernel.org>: > On Thu, Nov 05, 2015 at 09:11:47PM +0900, Masahiro Yamada wrote: >> 2015-11-05 20:28 GMT+09:00 Mark Brown <broonie@kernel.org>: > >> > The dependency is on ARCH_SUNXI || COMPILE_TEST which is not the same as >> > ARCH_SUNXI. > >> The compile test of this driver should pass even if RESET_CONTROLLER >> is not defined. >> So, it does not matter for COMPILE_TEST. > > If that is the case then the change here is that the dependency on > RESET_CONTROLLER is just redundant which isn't what the commit message > says.
For compile test, right, "depends on RESET_CONTROLLER" is redundant in the first place.
For run-time on real SoCs,the driver failed at the following point without the reset-controller sub-system.
sspi->rstc = devm_reset_control_get(&pdev->dev, NULL); if (IS_ERR(sspi->rstc)) { dev_err(&pdev->dev, "Couldn't get reset controller\n"); ret = PTR_ERR(sspi->rstc); goto err_free_master; }
-- Best Regards Masahiro Yamada
| |