Fixes for 'Implicit conversion loses integer precision' warnings.

[amubiera]

Since lfs_size_t is defined as uint32_t and not size_t, there are a few warnings for conversion from unsigned long (64-bit) to unsigned int (32-bit) on 64-bit platforms.

lfs.c:288:12 Implicit conversion loses integer precision: 'unsigned long' to 'lfs_size_t' (aka 'unsigned int')
lfs.c:1760:29 Implicit conversion loses integer precision: 'unsigned long' to 'lfs_tag_t' (aka 'unsigned int')

Perhaps lfs_size_t should be defined as size_t instead?

[geky-bot]

Tests passed ✓, Code: 17128 B (+0.0%), Stack: 1448 B (+0.0%), Structs: 812 B (+0.0%)

	Code	Stack	Structs		Coverage
Default	17128 B (+0.0%)	1448 B (+0.0%)	812 B (+0.0%)	Lines	2438/2599 lines (-0.0%)
Readonly	6234 B (+0.0%)	448 B (+0.0%)	812 B (+0.0%)	Branches	1288/1624 branches (-0.0%)
Threadsafe	17980 B (+0.0%)	1448 B (+0.0%)	820 B (+0.0%)		Benchmarks
Multiversion	17200 B (+0.0%)	1448 B (+0.0%)	816 B (+0.0%)	Readed	29000746676 B (+0.0%)
Migrate	18792 B (+0.0%)	1752 B (+0.0%)	816 B (+0.0%)	Proged	1482895246 B (+0.0%)
Error-asserts	17952 B (+0.0%)	1440 B (+0.0%)	812 B (+0.0%)	Erased	1568921600 B (+0.0%)

[geky]

Hi @amubiera, thanks for creating a PR, and sorry for the late response.

These casts look reasonable to me, just one nit on the tag cast.

[geky]

Perhaps lfs_size_t should be defined as size_t instead?

That's an interesting question. Not without significant changes as least. We write lfs_size_ts to disk and that would break the on-disk format on 64-bit machines. This is probably fixable (I think it's fixed in littlefs3 (#1111, unstable)), but there's still an assumption sizeof(lfs_size_t) < sizeof(lfs_off_t) everywhere.

Even on 64-bit machines, keeping lfs_size_t uint32_t would be ideal to minimize the RAM consumption, but maybe we leave that up to users to decide.

Unfortunately changing the codebase to support 64-bit lfs_size_t, even just to error on >2GiB inputs, would require a significant amount of work.

[geky-bot]

Tests passed ✓, Code: 17128 B (+0.0%), Stack: 1448 B (+0.0%), Structs: 812 B (+0.0%)

	Code	Stack	Structs		Coverage
Default	17128 B (+0.0%)	1448 B (+0.0%)	812 B (+0.0%)	Lines	2438/2599 lines (-0.0%)
Readonly	6234 B (+0.0%)	448 B (+0.0%)	812 B (+0.0%)	Branches	1288/1624 branches (-0.0%)
Threadsafe	17980 B (+0.0%)	1448 B (+0.0%)	820 B (+0.0%)		Benchmarks
Multiversion	17200 B (+0.0%)	1448 B (+0.0%)	816 B (+0.0%)	Readed	29000746676 B (+0.0%)
Migrate	18792 B (+0.0%)	1752 B (+0.0%)	816 B (+0.0%)	Proged	1482895246 B (+0.0%)
Error-asserts	17952 B (+0.0%)	1440 B (+0.0%)	812 B (+0.0%)	Erased	1568921600 B (+0.0%)

[geky]

Looks good here, thanks for this. Will bring this in on the next patch release.

[geky]

Sorry for the delay, this is finally coming in. Thanks again for the pr!