Re: [patch 3/3] Implement qXfer:libraries for Linux/gdbserver #2

[Eli Zaretskii <eliz at gnu dot org>]

> Date: Mon, 3 Oct 2011 23:55:30 +0200
> From: Jan Kratochvil <jan.kratochvil@redhat.com>
> Cc: Paul Pluzhnikov <ppluzhnikov@google.com>
> 
> --- a/gdb/doc/gdb.texinfo
> +++ b/gdb/doc/gdb.texinfo
> @@ -36412,15 +36412,21 @@ are loaded.
>  
>  The @samp{qXfer:libraries:read} packet returns an XML document which
>  lists loaded libraries and their offsets.  Each library has an
> -associated name and one or more segment or section base addresses,
> -which report where the library was loaded in memory.
> +associated name and one or more segment, section base addresses or SVR4 linker
> +parameters, which report where the library was loaded in memory.

Please reword:

 ... one or more segment addresses, section base addresses, or SVR4
 linker parameters.  These describe where the library was loaded into
 memory.

> +depend on the library's link-time base addresses.  For SVR4 systems
> +are reported parameters @{lm} with address of @code{struct link_map}
> +used for TLS (Thread Local Storage) access, @code{l_addr} specifying
> +bias of the mapped memory address minus the prelinked base address
> +and also @code{l_ld} which is memory address of the @code{PT_DYNAMIC}
> +segment.  SVR4 systems additionally specify @code{struct link_map}
> +address of the main executable in the @code{<library-list>} element.

This is too much for a single sentence.  I suggest an itemized list
instead:

 For SVR4 systems, the following parameters are reported:

 @itemize @minus
 @item
 @code{lm} with address of @code{struct link_map} used ...
 @item
 @code{l_ld}, which is memory address of ...
 ...
 @end itemize

And btw, are we sure that sayin "SVR4 systems" is enough to
unambiguously identify those systems?  How about at least a partial
list in parentheses?

> +  if (*phdr_memaddr == 0 || *num_phdr == 0)
> +    {
> +      warning ("Unexpected missing AT_PHDR and/or AT_PHNUM: "
> +	       "phdr_memaddr = %ld, phdr_num = %d",

No _() ?  Doesn't gdbserver support translations of messages?