Hi Am 15. Januar 2026 13:36:15 MEZ schrieb Quentin Schulz <[email protected]>: >Hi Frank, > >On 1/11/26 10:23 AM, Frank Wunderlich wrote: >> From: Frank Wunderlich <[email protected]> >> >> Show name of configuration node which was not found. >> >> current state gives no hint if fit image is wrong or the requested name. >> >> Could not find configuration node >> load of <NULL> failed >> >> After this patch we see name like this: >> >> Could not find configuration node '#ov-test' >> load of <NULL> failed >> >> Signed-off-by: Frank Wunderlich <[email protected]> >> --- >> boot/image-fit.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/boot/image-fit.c b/boot/image-fit.c >> index 2d040e38d97b..92675e98f2eb 100644 >> --- a/boot/image-fit.c >> +++ b/boot/image-fit.c >> @@ -2125,7 +2125,7 @@ int fit_image_load(struct bootm_headers *images, ulong >> addr, >> if (ret < 0 && ret != -EINVAL) >> ret = fit_conf_get_node(fit, fit_uname_config); >> if (ret < 0) { >> - puts("Could not find configuration node\n"); >> + printf("Could not find configuration node '%s'\n", >> fit_uname_config); > >It seems like fit_uname_config can be NULL (see outside of the patch context). >How does printf behave when this is the case? Should we do something like > >fit_uname_config ?: '' > >instead? That does seem like the message could be confusing also as now it >says "Could not find configuration node ''".
Thank you for taking a look. In my current testcase it was an empty string, but yes I could do something like you suggested...maybe call it "null" or "empty" instead of empty string? >Cheers, >Quentin regards Frank
