When setting the framework's DYLIB_CURRENT_VERSION larger than 65535 Xcode emits a warning:
warning build: Truncating -current_version to fit in 32-bit space used by old mach-o format
Why would Xcode want to truncate versions that are larger than 16-bit max if the current version space is defined as 32-bit integer?
Is this a bug or is there some other setting to tweak in order for this warning to go away?
The Xcode version I am using is 14.0.1.
Because those 32 bits are split up:
uint32_t version; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */
That's from a different load command, but all 32-bit version fields in Mach-Os work that way. The highest version that can be represented by these is 65535.255.255.
Also note that this may be "old" from the perspective of LLVM, but most structs and load commands in Mach-O headers still use 32-bit versions, with no alternative available, including the "current" and "compatibility" versions of dylibs. The only exception seems to be LC_SOURCE_VERSION/struct source_version_command, which uses a 64-bit field:
uint64_t version; /* A.B.C.D.E packed as a24.b10.c10.d10.e10 */