Re: [PATCH 3/7] inflate pt1: clean up input logic

[Russell King <rmk+lkml@arm.linux.org.uk>]

On Sat, Feb 25, 2006 at 12:51:36AM -0600, Matt Mackall wrote:
> On Fri, Feb 24, 2006 at 10:19:09PM +0000, Russell King wrote:
> > How does this change handle the case where we run out of input data?
> > This condition needs to be handled explicitly because the inflate
> > functions can infinitely loop.
> 
> And if you look at the current users, you'll see that only two of 15
> actually use it.

Sorry, I don't understand the relevance of your comment.

As the code stands in mainline, if we run out of input data, we are
guaranteed to exit from inflate.

With your change in this patch set, we no longer guaranteed to exit,
but will in some circumstances loop indefinitely.

The problem this causes is that if the ramdisk decompression runs out
of data, the kernel will just silently hang.

Please do not back out this fix.

> > Relying on a bit pattern returned by get_byte() is how this code
> > pre-fix used to work, and it caused several confused bug reports.
> 
> Just about everywhere, get_byte prints an error message and halts.

And the cases where it doesn't halt is the important case.

Sorry, but I hope that this code does not get merged as is.  It's
backing out a fix that I was involved in getting in, and therefore
I'm completely opposed to your code as it stands.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 Serial core