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
Currently, if Rizin needs to flip the sign of the type during the analysis using the rz_type_integral_set_sign() function, it calls internally rz_type_typeclass_get_default_sized() which internally just chooses first type of the selected typeclass and size without any preference, thus it largely depends on the loading order. We should find a way to specify the "defaultness" property of the type somehow. Maybe by introducing a new typeclass Default?
If we introduce new Defaullt typeclass, then the parser and code in librz/type/typeclass.c should be changed to allow the sum of typeclasses, e.g. Signed Integral + Default where it belongs to exactly two typeclasses: Signed Integral and Default.
The text was updated successfully, but these errors were encountered:
We already have unknown_t, it's not a problem. Problem that the default should be either intXX_t or uintXX_t or float depending on the sign and typeclass.
Currently, if Rizin needs to flip the sign of the type during the analysis using the
rz_type_integral_set_sign()
function, it calls internallyrz_type_typeclass_get_default_sized()
which internally just chooses first type of the selected typeclass and size without any preference, thus it largely depends on the loading order. We should find a way to specify the "defaultness" property of the type somehow. Maybe by introducing a new typeclassDefault
?If we introduce new
Defaullt
typeclass, then the parser and code inlibrz/type/typeclass.c
should be changed to allow the sum of typeclasses, e.g.Signed Integral + Default
where it belongs to exactly two typeclasses:Signed Integral
andDefault
.The text was updated successfully, but these errors were encountered: