1. with — de Vaan
with 'chariot' is not clear; some therefore regard it as a loanword from Etruscan curtus (office terminology). Pit. *-or- is reflect both as or and as ur in Latin; no phonetic conditioning has been found. We may therefore accept the traditional etymology *kys> *kors- > *kurs- > curr-. It is somewhat*troublesome that only Latin attests a verb, but since curro is of a primary derivation, it cannot be derived from the … — [de Vaan, s.v. with, p. 171]
2. with — de Vaan
with 'haste' is hardly compelling, so this etymology remains gratuitous. BibL: WH I: 259, 488, EM 231, IEW 143, Leumann 1977: 327, Schrijver 1990, 1995:410, Vine 1999c. -► mfestus — [de Vaan, s.v. with, p. 230]
3. with — de Vaan
with 'spoken praise', 'reputation by hearsay'. Against the assumed etymology speak the phonetics. A dissimilation *gnor- (> *gror-l) > *glor- is contradicted by gnarus and igndrare. On the other hand, gnarus kept its gn- on the model of ignarus, so that the retention of gn-r* in gnarus and ignorare may be due to non-initial (= VOLat. unstressed) position of the syllables in question. In *gnoria, the stress may have … — [de Vaan, s.v. with, p. 280]