[PATCH v2 0/4] GDB fixes for the remote end having gone astray

["Maciej W. Rozycki" <macro@wdc.com>]

Hi Pedro,

> So something like adding a thread_info parameter to
> clear_step_over_info, and then calling clear_step_over_info
> from clear_thread_inferior_resources.  clear_step_over_info
> would only clear the info if the thread matched, or if NULL
> is passed.  Would that work?

 Thank you for your review.

 As issues from software stepping having gone astray seem to have started 
piling up, and with your suggested modifications there is now a dependency 
between the changes, I have decided to make this patch submission a small 
series with 1/4 and 3/4 corresponding to original individual submissions 
and 2/4 and 4/4 being entrely new fixes.

 All these changes were regression-tested with the `x86_64-linux-gnu' 
native target and, now that RISC-V/Linux `gdbserver' I have been working 
on has become usable enough, also the `riscv64-linux-gnu' remote target 
driven by a `x86_64-linux-gnu' host.

 See individual change descriptions and any associated discussions for 
details.

  Maciej