Re: [PATCH] ftrace: return error instead of 12 bytes read

[Ingo Molnar <mingo@elte.hu>]

* Andy Whitcroft <apw@canonical.com> wrote:

> >> ? ? ? s = kmalloc(sizeof(*s), GFP_KERNEL);
> >> ? ? ? if (!s)
> >> - ? ? ? ? ? ? return ENOMEM;
> >> + ? ? ? ? ? ? return -ENOMEM;
> >>
> >> ? ? ? trace_seq_init(s);
> >>
> >
> > lol, there we go again.
> >
> > Andy, can we have a checkpatch rule please?
> 
> Thats a tricky one.  Not only do we not really have a sensible way to 
> know if ENOMEM is an errno, we also find a bunch of places that we 
> appear to use positive errno's as return values where we would falsly 
> complain about.  Its particularly common in scsi and filesystems. 
> Admittedly the vast majority are return -EXXX form, so we could add 
> this as a non-default check perhaps.
> 
> Thoughts?

Even in filesystems, ~80% of the cases use proper negative values:

 $ git grep 'return -E' fs/ | wc -l
 4540
 $ git grep 'return E' fs/ | wc -l
 895

For SCSI it's even better, ~97% of the cases use the kernel's standard:

 $ git grep 'return -E' drivers/scsi/ | wc -l
 1448
 $ git grep 'return E' drivers/scsi/ | wc -l
 50

So i'd suggest to make this a default-enabled check. (default disabled 
checks are used only by a small minority) For a _long_ time has this 
been the kernel standard.

	Ingo