With two peers using different HCAs, modify QP to RTS on ib_read_lat fails due to out_reads exceeding max_qp_rd_atom

[bryantan-vmware]

I'm using running ib_read_lat between two different HCAs, and modify QP to RTS fails on the side with a lower max_qp_rd_atom because it appears that the code uses the out_reads of the peer when modifying the QP (I think in ctx_modify_qp_to_rts, where we set attr->max_rd_atomic = dest->out_reads;). Shouldn't this just be the device's max_rd_atomic?

[existosar]

I have the same issue as yours. I got this error when I use default num of outstanding read/atom, testing mellanox cx4 and broadcom p225p. I turn to manually set it with -o option and it just works fine.