I think what abaddon probably means is: you have no idea what a biometric passport actually is.
All "biometric" means in the context of passports is that they contain an NFC chip which contains encrypted information about the biological characteristics of the passport holder. But that of course is entirely and wholly prevalent upon a) people giving permission for certain biological data to be obtained, b) those data themselves being obtained, c) those data being placed onto the chip.
At the moment, the standard EU passport (IIRC) contains nothing more on its biometric chip than a version of the holder's image - taken directly from the very photo that is on the passport. Of course this serves a certain purpose, in that it helps prevent fraudulent use of passports by replacing one photo with another: if the photo on the page doesn't match the biometric chip's photo, then obviously there's a problem. I believe you're wrong to believe that EU passports currently contain embedded fingerprint information* (though that was, and remains, a lofty ideal). And any embedding of DNA information on a passport's biometric chip would not only require an ethically-debatable consent matter to arise, but also (if it did ever become law) it would require an extraordinary exercise in taking samples from every single passport holder or passport applicant.
On top of all this, as others have correctly pointed out, this is entirely moot to the debate here anyhow. In order to have a workable DNA database for "fighting crime", one would need a full profile from a significant majority of the population (bearing in mind also that the minority who avoided/missed/otherwise declined giving a sample might have a fairly strong correlation with the proportion of the population who committed most of the serious crime - thus rendering the whole exercise of extremely limited value from the get-go). And on top of this, experience shows clearly that DNA databases have an extremely limited effect in actually preventing or deterring crime - it just makes the perpetrators easier to identify and catch in some circumstances. Most serious crimes in England & Wales, for example, are committed by repeat offenders - almost all of whom will by now have had to place their DNA profiles on the national database. Yet serious crime levels have had their long-term trends virtually unaffected by the introduction of the criminal DNA database. It's not stopping crimes from happening. And even in terms of identification, DNA databases are only useful where the perp has no connection to the crime scene or victim. If, for example, a wife is killed inside her house, it's laughably obvious that the police will request and obtain a DNA profile from her husband and everybody else close to the victim very quickly anyhow. And the killer is extremely likely to come from within that group.
And when you drill down to the Kercher case in particular, it would have been of very limited value in identifying and apprehending the real culprit (Rudy Guede). Even if Knox, Sollecito and Guede had had their full DNA profiles on an Italian national database, the police could and should have taken DNA from Knox and Sollecito early on in the investigation as a matter of course, and they should have identified Guede within a week or so as a person to whom they needed to speak (and take DNA). Ironically perhaps, at least a national Italy DNA database would have readily provided the DNA profiles of the likes of Laura Mezzetti, Filomena Romanelli, Kercher's boyfriend Silenzi and others of that ilk - whom it appears the incompetent police didn't even obtain reference DNA profiles from.
* ETA: Finnish EU passports may of course obtain fingerprint data by now. UK EU ones do not.