Re: Bug related to stdin/always/jobcontrol

[Daniel Shahaf <d.s@xxxxxxxxxxxxxxxxxx>]

Bart Schaefer wrote on Thu, Sep 22, 2016 at 18:18:47 -0700:
> However, in attempting to figure it out, I found some sixteen-year-old
> code that is clearly wrong:
> 
> diff --git a/Src/exec.c b/Src/exec.c
> index d924148..bf97b5c 100644
> --- a/Src/exec.c
> +++ b/Src/exec.c
> @@ -1702,7 +1702,7 @@ execpline(Estate state, wordcode slcode, int how, int last1)
>  			    jobtab[list_pipe_job].other = newjob;
>  			    jobtab[list_pipe_job].stat |= STAT_SUPERJOB;
>  			    jn->stat |= STAT_SUBJOB | STAT_NOPRINT;
> -			    jn->other = pid;
> +			    jn->other = list_pipe_job;
>  			    jn->gleader = jobtab[list_pipe_job].gleader;
>  			}
>  			if ((list_pipe || last1) && hasprocs(list_pipe_job))
> 
> jn->other is a job table index, not a process ID.  However, (a) I don't
> know if list_pip_job is the right value (although a few lines earlier
> list_pipe_pid = pid is assigned, so it would seem to makes sense) and
> (b) that this has never caused a wild pointer dereference in all this
> time seems to indicate that this code is never actually reached.

The code does get reached: if you do
.
    % f() vim
    % f
    ^Z
.
at a new shell, it executes the line you changed.  Perhaps nothing reads
the value assigned to jn->other, though?

> I don't have any more time to look at this now; hoping this information
> is helpful to PWS if he has a chance to look when he wakes up on Friday.
>