Results from 22 in-situ tests performed on nine large-scale stone masonry panels are reported. Test panels were selected as part of shear walls in six different old masonry buildings so as to represent a reliable sample of the most common masonry types in the region. The tests were carried out under vertical compression, shear compression, and diagonal compression. After the first test, several panels were repaired and reinforced by means of cement mortar injections (full cement grouting) or reinforced concrete (RC) jackets, and then tested again to determine the effectiveness of the intervention. Particular attention was then devoted to evaluate the referential shear strength of these masonry assemblages in the original state. To this purpose, a fitting process for experimental data was used, adopting two different schemes for interpretation of the shear strength of masonry (the Coulomb and the Turnsek-Cacovic equations). The results show that the Turnsek-Cacovic equation gives a better fit to experimental data than the Coulomb equation, especially for medium- and poor-quality masonry walls. Finally, conclusions are presented noting the difference between shear strength values calculated from fitting the data from test results and the values suggested by European and Italian standards. 22 refs.