Re: [PATCH v3 03/15] Read CTF by the ctf target

[Yao Qi <yao at codesourcery dot com>]

On 03/14/2013 04:08 AM, Tom Tromey wrote:
> Yao> +if test "x$with_bt_include" != x; then
> Yao> +  LIBBABELTRACE_CFLAGS="-I$with_bt_include "
> Yao> +fi
> Yao> +if test "x$with_bt_lib" != x; then
> Yao> +  LIBBABELTRACE="-L$with_bt_lib -lbabeltrace -lbabeltrace-ctf"
> Yao> +fi
> Yao> +
> Yao> +if test "x$with_babeltrace" != "xno"; then
>
> This only checks for libbabeltrace if --with-babeltrace is specified.
> So, the text above about equivalency isn't really correct, at least the
> way I read it; if you want to give --with-babeltrace-include, you have
> to also give --with-babeltrace.

It works if we only give --with-babeltrace-include and 
--with-babeltrace-lib (without giving --with-babeltrace).  In this case, 
$with_babeltrace is empty, and the condition above is still true.  I 
tried to configure with --with-babeltrace-include=/usr/include and 
--with-babeltrace-lib=/usr/lib, everything looks OK.

> Yao> +static int
> Yao> +ctf_has_all_memory (struct target_ops *ops)
> Yao> +{
> Yao> +  return 0;
> Yao> +}
>
> Unfortunately for you there is plenty of this patch I don't feel
> competent to review.  I don't know much about the target API.
>
> If nobody else looks after a decent interval, ping me and I will ok it.
> I suppose if these methods parallel the existing tfile target, then that
> is pretty good evidence for it being ok.

Most of them are copied from tfile target, and I'll examine them (in 
both tfile target and ctf target) one by one.  I'll give an updated 
version tomorrow.

--
Yao (éå)