I suspect that since -flto was added listing generation wasn't working.
-ffat-lto-objects allows tools like objdump to still work by including
both the internal compiler representation (for LTO) and generated code
in the .o files. It increases the compile time from 12s to 17s on my
machine but I think it is small enough cost for the benefit.
Signed-off-by: Jakub Sztandera <kubuxu@protonmail.ch>
License: MIT
* [Build script] Detect languages with translation_*.json files
* [Build script] Swap model and language in the build loop to avoid full rebuild
* [Build script] Cleaning potential bugs according ShellCheck
* [Build script] Rewrite AVAILABLE_LANGUAGES calculation with @vomindoraan help
* Tweak Serbian translations
* Add "cyrillicGlyphs": true to translations that use Cyrillic
* translation_cs_cz.json → translation_cs.json
The correct language code for Czech is CS; CZ is the country code
* Add "cyrillicGlyphs": false to other translations, move "languageLocalName" to top
Also change BG and HU "localLanguageName" to start with a capital letter
* Add missing "languageLocalName" field for Slovak
* Rearrange a few fields so they're in the same order for all languages
* Regenerate translations source file
* Portugues → Português
* translation_dk.json → translation_da.json
DA is the ISO 639-1 code for Danish
* translation_ua.json → translation_uk.json
UK is the ISO 639-1 code for Ukrainian
* Update language codes in build.sh
* Update translation_vl.json
Deze vertaling is geen wijziging van de Nederlandse vertaling; ze is bedoeld voor Vlaamse gebruikers en wordt aangeboden als een nieuwe vertaling. Dit is de eerste versie die nog moet worden getest op de TS-100.
* Restore NL translation & generate NL_BE source
