Re: POSIX message queues, libmqueue: mq_open, mq_unlink

[Alex Riesen]

> > Looking over the code in libmqueue-4.31, I noticed the checks for the
> > name validity in the mq_open and mq_unlink. Why are they needed?  They
> > are pointless if the code in kernel depends on the valid name,
>
> You are contradicting yourself.
>
> Anyway, non-absolute path names passed to the functions mean the
> behavior is unspecified.  No portable application must ever do this.  It
> is enforced for this reason plus if there comes a time when we want to
> do something special which doesn't conflict with standard-compliant
> behavior we have a possibility for that.  Unlike wh6at you think, the
> tests *are* useful.

My concern is that the tests are rather pointing that something in
kernel is not implemented correctly. _The_ checks in particular.
Because if they _are_ implemented correctly, you don't need to patch the
functionality in the user space.

And if the kernel code does check the incoming arguments correctly,
what is the point to check them again? Just to make the point, that
passing in not an absolute path is not portable?


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/