[Bug breakpoints/34381] Emit an error message when CONFIG_PROC_MEM_NO_FORCE=y causes attach failure

https://sourceware.org/bugzilla/show_bug.cgi?id=34381

--- Comment #6 from Adrian Ratiu <adrian.ratiu at collabora dot com> ---
@Tom

Yes, now that you mention it, after reading that file, I remember one of the
missing pieces in GDB was that gdbserver has no ptrace fallback. 

That is one of the main GDB blockers (gdbserver).

However falling back to ptrace is very inefficient: there are more modern
kernel interfaces which could be used, like process_vm_readv() and
process_vm_writev().

So I think the correct long-term way forward is to port GDB away from both
ptrace and writing /proc/mem (which in turn relies on the problematic
FOLL_FORCE flag).

I don't know of anyone to be working on this. :)

-- 
You are receiving this mail because:
You are on the CC list for the bug.