On Thu, Jan 27, 2022 at 10:59:51AM -0600, Adam Ford wrote:
On Wed, Jan 26, 2022 at 2:58 PM Tommaso Merciai tomm.merciai@gmail.com wrote:
On Wed, Jan 26, 2022 at 12:05:22PM -0600, Adam Ford wrote:
On Sat, Dec 25, 2021 at 2:26 PM Tommaso Merciai tomm.merciai@gmail.com wrote:
Override env_get_location function at board level, previously dropped down from arch/arm/mach-imx/imx8m/soc.c
References:
- commit 37d3e3bb95d7532e2503f115dd6c6762fd3b0262
Signed-off-by: Tommaso Merciai tomm.merciai@gmail.com
Hi Adam,
I am going to re-do the patch for the beacon board, because I talked it over with my colleagues, and we'd like to force the environment to one location or another and not base it on whether or not it was the boot device.
I guess my questions is what's the status of the remaining patch series, and how do I integrate my desired changes into it? I can send you what I want for my board if you're going to resubmit the series.
It's ok for me. Send me your changes. Thanks.
I wonder if a sub-function could be called where the existing env_get_location() is renamed to something else, but declared as weak can called from env_get_location. Anyone who wants a their own function can write their own so we don't have two functions with the same name declared as weak.
What are your thoughts on that?
Right. In this way every boards that use imx8mp/imx8mn can override this function according to their own needs.
Hi Adam,
After trying some different options, I wonder if we can just delete env_get_location() completely from arch/arm/mach-imx/imx8m/soc.c.
For my board, it does exactly what I want when I delete that function since the default from env.c is sufficient. I think it's probably safest to use your original approach to move them to individual board files with the idea that some people may not need it at all. When placed into their board files, they can choose what they want for themselves.
Ack.
I would suggest sending a patch without the RFC and place the function into the various imx8mn and imx8mp boards, but know that you can skip the beacon board. I would also suggest that you add a note that people should test this to see if it's really what they want or if they can live without the code and fall back on the default.
Ok, I'll send v4 of the series (without RFC label) in these days.
Thanks, Tommaso
adam
Tommaso
adam
board/beacon/imx8mn/imx8mn_beacon.c | 35 ++++++++++++++++++++++++++++- 1 file changed, 34 insertions(+), 1 deletion(-)
diff --git a/board/beacon/imx8mn/imx8mn_beacon.c b/board/beacon/imx8mn/imx8mn_beacon.c index 7fe252b262..05ab5613ee 100644 --- a/board/beacon/imx8mn/imx8mn_beacon.c +++ b/board/beacon/imx8mn/imx8mn_beacon.c @@ -6,14 +6,47 @@ #include <common.h> #include <miiphy.h> #include <netdev.h>
+#include <env_internal.h> #include <asm/arch/clock.h> #include <asm/arch/sys_proto.h> +#include <asm/mach-imx/boot_mode.h> #include <asm/global_data.h> #include <asm/io.h>
DECLARE_GLOBAL_DATA_PTR;
+enum env_location env_get_location(enum env_operation op, int prio) +{
enum boot_device dev = get_boot_device();enum env_location env_loc = ENVL_UNKNOWN;if (prio)return env_loc;if (IS_ENABLED(CONFIG_ENV_IS_IN_SPI_FLASH) && dev == QSPI_BOOT) {env_loc = ENVL_SPI_FLASH;} else if (IS_ENABLED(CONFIG_ENV_IS_IN_NAND) && dev == NAND_BOOT) {env_loc = ENVL_NAND;} else if (IS_ENABLED(CONFIG_ENV_IS_IN_MMC)) {switch (dev) {case SD1_BOOT:case SD2_BOOT:case SD3_BOOT:case MMC1_BOOT:case MMC2_BOOT:case MMC3_BOOT:env_loc = ENVL_MMC;break;default:break;}} else if (IS_ENABLED(CONFIG_ENV_IS_NOWHERE)) {env_loc = ENVL_MMC;}return env_loc;+}
#if IS_ENABLED(CONFIG_FEC_MXC) static int setup_fec(void) { -- 2.25.1