I hope you agree that the verb form ir- 'to be' [DEDR 480] either as copula or independent verb cannot be attested in Telugu. This verb, so important in South Dravidian, is remarkably absent in Central and North Dravidian languages, and in many of South-Central languages. In Telugu, you may find noun forms such as iravu 'place', irugu 'neighbour, neighbourhood' etc. which can possibly be argued as recent borrowings from Kannada or Tamil.
The words ṟēvu (ఱేవు), ṟēvaḍu (ఱేవడు), ṟēvadi (ఱేవది) etc. are not derived from 'iravu' (<ir-) (note the alveolar trill, instead of tap). They are related to [DEDR 516] *iṯ-/iṟ- 'to descend', 'to go beyond' iṟaṅku (iṟaṅki-) to descend, alight, fall (as rain), disembark; iṟaṅkal place of descent, of debarkation etc.
If you believe in my proposed theory of external origin of Dravidian, then the whole question of what constitutes Proto-Dravidian becomes problematic. Under the usual historical linguistic principles, words and features found only in one branch (South Dravidian, in this case) do not provide evidence for parent language (Proto-Dravidian). But Dravidian linguists have been wrongly reconstructing words and features found only in South Dravidian (esp. Tamil) to the parent language under the assumption that Tamil represents the archaic form of Dravidian. It is a classic catch-22 situation.
I believe there is an urgent need to reevaluate the Dravidian languages considering the possibility of pre-Dravidian substrata in various branches of the Dravidian languages. Suspending the belief that Tamil represents the most archaic form of Dravidian would be a starting point in this endeavour.
Regards,
Suresh.