Hello Sascha, On 4/14/25 11:07, Sascha Hauer wrote: > On Mon, Apr 14, 2025 at 08:23:03AM +0200, Ahmad Fatoum wrote: >> From: Ahmad Fatoum <a.fat...@barebox.org> >> >> devm_request_mem_region doesn't return NULL on error on Linux and thus >> it should neither in barebox. Fix this. >> >> Signed-off-by: Ahmad Fatoum <a.fat...@pengutronix.de> >> --- >> drivers/base/driver.c | 2 ++ >> include/linux/device.h | 2 +- >> 2 files changed, 3 insertions(+), 1 deletion(-) >> >> diff --git a/drivers/base/driver.c b/drivers/base/driver.c >> index edba037fa2dd..4e361a96c9ee 100644 >> --- a/drivers/base/driver.c >> +++ b/drivers/base/driver.c >> @@ -587,6 +587,8 @@ void __iomem *dev_request_mem_region(struct device *dev, >> int num) >> struct resource *res; >> >> res = dev_request_mem_resource(dev, num); >> + if (!IS_ERR(res) && WARN_ON(IS_ERR_VALUE(res->start))) >> + return IOMEM_ERR_PTR(res->start); > > Which code can set res->start to an error value?
It can end up being an error value, because there is actual a MMIO region at a high address. In that case, I'd rather we return an error, so dev_request_mem_resource would need to be used instead. Thanks, Ahmad > > Sascha > >> if (IS_ERR(res)) >> return ERR_CAST(res); >> >> diff --git a/include/linux/device.h b/include/linux/device.h >> index 66294910abb3..661d8b24730e 100644 >> --- a/include/linux/device.h >> +++ b/include/linux/device.h >> @@ -39,7 +39,7 @@ static inline void __iomem >> *dev_platform_ioremap_resource(struct device *dev, >> * so we don't need to do anything besides requesting the regions >> * and can leave the memory attributes unchanged. >> */ >> - return dev_request_mem_region_err_null(dev, resource); >> + return dev_request_mem_region(dev, resource); >> } >> >> static inline void __iomem *devm_ioremap(struct device *dev, >> -- >> 2.39.5 >> >> >> > -- Pengutronix e.K. | | Steuerwalder Str. 21 | http://www.pengutronix.de/ | 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
