Hi, On 18.05.25 19:31, Johannes Roith wrote: > Am Fri, May 16, 2025 at 10:05:49PM +0200 schrieb Ahmad Fatoum: >>> +static void cora_z7_ps7_init(void) >>> +{ >>> + /* >>> + * Read OCM mapping configuration, if only the upper 64 KByte are >>> + * mapped to the high address, it's very likely that we just got control >>> + * from the BootROM. If the mapping is changed something other than the >>> + * BootROM was running before us. Skip PS7 init to avoid cutting the >>> + * branch we are sitting on in that case. >>> + */ >>> + if ((readl(0xf8000910) & 0xf) != 0x8) >> >> Please use a macro for the address and wrap it in IOMEM(). >> Eventually, we will want to enforce addresses to be pointers, so it would >> be cool for new code to already use the correct types. If you want type >> checking, just add >> >> #include <linux/io.h> at the top of your includes. >> > > So, it should look like this, right: > > void __iomem *ocm_config = IOMEM(OCM_CFG);
Or just #define OCM_CFG IOMEM(0xf8000910). > Hmm, good question. In theory most of the code is generic or could be > implemted in functions in mach-zynq and the lowlevel board just calls > these functions with the correct arguments, e.g. the DDR init. But > that way it is quite time consuming to implement. This here is basically > just copy paste from a Vitis generated code adopted to the coding style of > barebox. Ok. If the code is generated anyways, then it's fine to duplicate it IMO. You could also keep the generated code in a separate file and define some macros to make changes to it minimal, but this is fine too. >>> +config MACH_CORA_Z7 >>> + bool "Digilent Cora Z7" >>> + select ARCH_ZYNQ7000 >> >> Can you enable this in zynq_defconfig? > > Yes, it is selectable in the config. Add it to the defconfigs (arch/arm/configs) that enable CONFIG_MACH_ZEDBOARD. Cheers, Ahmad -- 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 |