White lupin (Lupinus albus L.) is well known to form cluster/proteoid roots in which the exudation of citric acid is increased and P can be mobilised from Fe-phosphates, which is a common form of phosphate in acid soils. Additionally, also phosphatases are leased helping to mobilize organic P. These sources might help:

Jones 1998 Plant Soil 205, 25-44; Dakora and Philips 2002

Sorry, I opened the wrong window for the answer. I'm not working with halophytes from Teberna's Desert. Hoever, I'm working on the effect of salinity on soil microorganisms and related matter and nutrient fluxes, if that is of any use.

I do like and support Florian's answer with a focus on white lupin, which produces carboxylate-releasing cluster/proteoid roots. The reason for adding a comment here is his suggestion that "Fe-phosphates are a common form of phosphate in acid soil." Indeed, a statement like this has slipped in the literature dealing with phosphorus and plant-soil interactions a few decades ago and it continues to persist. However, I don't think there is a single study actually showing this. Instead, there are numerous studies showing that phosphate is associated with oxides and hydroxides of Fe (and Al).

I thought this comment might help stop authors claiming that Fe phosphates are a major form of P in acid soil.