Re: [PATCH 3/7] Clarify doc about memory read/write and non-8-bits bytes

[Eli Zaretskii <eliz at gnu dot org>]

> From: Simon Marchi <simon.marchi@ericsson.com>
> CC: Simon Marchi <simon.marchi@ericsson.com>
> Date: Wed, 8 Apr 2015 15:56:15 -0400
> 
> This patch modifies the manual to clarify the MI, RSP and Python APIs in
> regard to reading/writing memory on architectures with non-8-bits bytes.
> 
> Care is taken to use the word byte when referring to one piece of the
> smallest addressable size on the current architecture and the word octet
> when referring to an 8-bits data piece. I try to avoid "word", because
> it can be ambiguous.

Thanks.  However, I think we need to be more explicit about this
issue.  Just using "target memory bytes" without ever explaining what
that means, or how it is different from the host bytes, makes these
changes more accurate, but not more clear to the reader.

So please add some short discussion of this issue (the node "Data"
sounds pertinent, or maybe "Memory"), and introduce there the "target
memory byte" term.

> +Optional argument indicating the number of target bytes to be written.
                                              ^^^^^^^^^^^^
Let's be consistent and use "target memory bytes" everywhere.

> -Read @var{length} bytes of memory starting at address @var{addr}.
> +Read @var{length} octets of memory starting at address @var{addr}.

"Octet" is not defined anywhere, so I think it should be part of the
above-mentioned introduction.  Here, I would add a cross-reference to
that place.

> -Read @var{length} bytes of memory from the inferior, starting at
> +Read @var{length} target bytes of memory from the inferior, starting at
                     ^^^^^^^^^^^^^^^^^^^^^^
"target memory bytes"

Thanks.