Not actually sending patches

[Jerry DeLisle]

On 02/08/2011 10:26 PM, Ralf Wildenhues wrote:
> Hello Jerry, all,
>
--- snip ---
>
> Your mail does not contain the actual change that you applied.
> It looks like this is becoming a trend, and I think it needs to stop.
>

OK by me.

> The gcc-patches mailing list is read by an estimated 300+ people.  Most
> of them have very little time, but many of them are quickly able to
> point out that missing set of parentheses, that typo, or that unportable
> construct you may have used in your inline posted patch.  Not reading a
> patch that doesn't interest them maybe costs them three seconds.
>

In this case, they pointed out my cobbled up subject line and missing reference 
to the PR in the Changelog.

Mea culpa

> But going to SVN or git, or just gcc-cvs, and digging out the revision
> number just to read the patch, much less comment on it, easily takes
> more than a minute.  That means, most readers won't do it, unless the
> patch happens to really really interest them.
>
> So please, save 300 people work by investing 10 seconds yourself and
> post the actual patch.  In turn you get something out too: better
> review.
>

Agreed,

Jerry

The patch:

--- trunk/gcc/fortran/io.c	2011/01/08 19:17:03	168600
+++ trunk/gcc/fortran/io.c	2011/02/09 06:02:56	169961
@@ -3091,7 +3091,8 @@
    cp = gfc_get_code ();
    cp->op = EXEC_TRANSFER;
    cp->expr1 = expr;
-  cp->ext.dt = current_dt;
+  if (k != M_INQUIRE)
+    cp->ext.dt = current_dt;

    *cpp = cp;
    return MATCH_YES;