[M] Change in coreboot[master]: lib: Support localized text of memory_training_desc in ux_locales.c

[Hsuan-ting Chen (Code Review)]

Hsuan-ting Chen has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/75330?usp=email )

Change subject: lib: Support localized text of memory_training_desc in ux_locales.c
......................................................................

Patch Set 9:

(2 comments)

Commit Message:

https://review.coreboot.org/c/coreboot/+/75330/comment/3a3668ff_5facddb4 :
PS9, Line 13: The preram_locales file follows the format:
> I have a few questions about the format. […]

First of all, let me list my concerns:
* Make the file as compact as possible
* Make the parsing logic in coreboot as simple as possible
* Make the read-only region compatible with some `string.h` functions

---

For 1. I haven't find an easy logic to parse your structures, let me know if there's one I am not awared of.

I have another idea:
prepend one "type byte" in front of each strings. e.g.

\x00 \x01 string_name_1 \x00 \x02 locale_id_1 \x00 \x03 localized_string_1 ...

Hence our `search_for` will always point to the "type byte"

and we could easily parse them by type enum

```
enum content_type {
  CONTENT_TYPE_STRING_NAME = 0x01,
  CONTENT_TYPE_LOCALE_ID = 0x02,
  CONTENT_TYPE_LOCALIZED = 0x03,
};
```

---

For 2. Adding version byte (maybe two bytes?) seems reasonable, I am not sure if we want a checksum -- If so, we might follow the similiar logic of ELOG? e.g. adding SIGNATURE, VERSION, CHEKCSUM

File src/lib/ux_locales.c:

https://review.coreboot.org/c/coreboot/+/75330/comment/615fe052_047a7046 :
PS9, Line 101: const char *ux_locales_get_text(const char *name)
> Add a unit test for this?

I'll take a look at adding this.