On Wed, 10 Feb 2021, Daniel Thompson wrote: > > Wrapping the relevant parts of this file into #ifdef MIPS_FP_SUPPORT > > would be as easy though and would qualify as a proper fix given that we > > have no XML description support for the MIPS target (so we need to supply > > the inexistent registers in the protocol; or maybe we can return NULL in > > `dbg_get_reg' to get them padded out in the RSP packet, I haven't checked > > if generic KGDB code supports this feature). > > Returning NULL should be fine. > > The generic code will cope OK. The values in the f.p. registers may > act a little odd if gdb uses a 'G' packet to set them to non-zero values > (since kgdb will cache the values gdb sent it) but the developer > operating the debugger will probably figure out what is going on without > too much pain. Ack, thanks! NB if GDB sees a register padded out (FAOD it means all-x's rather than a hex string placed throughout the respective slot) in a `g' packet, then it will mark the register internally as "unavailable" and present it to the receiver of the information as such rather than giving any specific value. I don't remember offhand what the syntax for the `G' packet is in that case; possibly GDB just sends all-zeros, and in any case you can't make GDB write any specific value to such a register via any user interface. The way the unavailability is shown depends on the interface used, i.e. it will be different between the `info all-registers'/`info register $reg' commands, and the `p $reg' command (or any expression involving `$reg'), and the MI interface. But in any case it will be unambiguous. In no case however there will be user confusion for such registers. Maciej
