You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Mach-O objects can contain a data structure pointed to by the LC_DYLD_CHAINED_FIXUPS linkedit-type load command that, uh... does things. To be honest, I really don't know what its purpose truly is, but it seems to describe optional modifications to apply to import symbols upon loading.
In any case, having it supported in object would be nice as there are currently no implementation of it publicly available, at least I haven't found any. The start of it would be porting the fixup-chains.h definitions. Then, the parsing algorithm would need to be implemented as well if it is desirable to have it.
A few questions about the contribution then:
Does this even have its place in object?
What would go in machovs.read::macho? There is a bunch of struct definitions similar to the rest of macho and also some non-trivial parsing logic.
Would it be desirable to add it to a higher-level API?
If so, should it be added directly onto MachOFile or to the Object trait and its implementation of it? Having it outside of the trait makes more sense to me because it would not be behavior common to all formats, but I'm just checking anyways.
The text was updated successfully, but these errors were encountered:
Mach-O objects can contain a data structure pointed to by the
LC_DYLD_CHAINED_FIXUPS
linkedit-type load command that, uh... does things. To be honest, I really don't know what its purpose truly is, but it seems to describe optional modifications to apply to import symbols upon loading.In any case, having it supported in
object
would be nice as there are currently no implementation of it publicly available, at least I haven't found any. The start of it would be porting thefixup-chains.h
definitions. Then, the parsing algorithm would need to be implemented as well if it is desirable to have it.A few questions about the contribution then:
object
?macho
vs.read::macho
? There is a bunch of struct definitions similar to the rest ofmacho
and also some non-trivial parsing logic.MachOFile
or to theObject
trait and its implementation of it? Having it outside of the trait makes more sense to me because it would not be behavior common to all formats, but I'm just checking anyways.The text was updated successfully, but these errors were encountered: