Geminacy in Persian; consonant cluster or consonant lengthening?

Geminacy appears in two ways cross-linguistically: either as consonant cluster or as consonant lengthening. Consonant clusters or doublings happen in languages like English in which geminacy is not contrastive tautomorphemically and therefore can't make any minimal pairs; and consonant lengthening belongs to languages like Arabic and Italian in which geminacy is contrastive and make minimal pairs. Based on the typology of geminate inventories (Gordon 2016) and some phonetic, phonological and morphological correlates of consonant lengthening (Blevins 2005), it has been shown that although Persian has true tautomorphemic geminate, geminacy is a kind of consonant cluster in this language and has nothing to do with consonant lengthening. It has been discussed that there are a few minimal pairs of singleton and geminate consonants in this language, but they can be considered as nothing because they are all very old or borrowed forms used rarely by the native speakers.