Re: short read from /dev/urandom

[Matt Mackall <mpm@selenic.com>]

On Sat, Jan 15, 2005 at 07:58:23PM -0800, Ulrich Drepper wrote:
> Matt Mackall wrote:
> >_Neither_ case mentions signals and the "and will return as many bytes
> >as requested" is clearly just a restatement of "does not have this
> >limit". Whoever copied this comment to the manpage was a bit sloppy
> >and dropped the first clause rather than the second:
> 
> It still means the documented API says there are no short reads.

I maintain that it's ambiguous. And read(2) makes it clear that short
reads can happen any time, any where. Further, your interpretation
makes for a nonsensical API as it implies being uninterruptible for
arbitrary lengths of time. 

Changing the longstanding, sensible code to match a silly and highly
non-standard interpretation of the documentation doesn't fix the
problem in apps either. Presumably they'll still be running on kernels
older than 2.6.11 and I believe most *BSDs have /dev/urandom as well.

> >So anyone doing a read() can expect a short read regardless of the fd
> >and is quite clear that reads can be interrupted by signals. "It is
> >not an error". Ever.
> 
> Of course are signal interruptions wrong if the signal uses SA_RESTART.

That's a separate problem. I'll take a look at fixing that.

-- 
Mathematics is the supreme nostalgia of our time.