Re: [PATCH] more helpful WARN_ON and BUG_ON messages

[Dave Jones]

On Sat, Oct 21, 2006 at 12:07:17AM +0200, Adrian Bunk wrote:

 > > Most debugging code makes the kernel bigger, slower... and easier to
 > > debug, no?
 > > 
 > > It's not a question of not being -able- to locate sources; it's a
 > > question of being able to look at a bug report and triage it quickly
 > > without digging around to find the kernel du jour that produced it.  *shrug*
 > 
 > It's not that BUGs were that frequent.

You're not trying hard enough ;) 

 > And with your suggestion "I suppose this could be put under CONFIG_DEBUG", 
 > it would anyway be turned off by nearly everyone.

For better bug reports, 16K is peanuts. We had exactly the same straw-man
come up when kksymoops was first proposed. And now, most people don't run
without it.

I've seen numerous cases where reporters have hand transcribed BUG() reports,
and got the line numbers wrong because they misremembered a 4 digit number.
Words are inherently easier to remember, and even if typoed, usually there's
enough context to figure out what the problem was without another round-trip
to the bug reporter.

If this were optional, I don't see how anyone can argue against it,
and that should be a trivial improvement to Eric's existing patch.

	Dave

-- 
http://www.codemonkey.org.uk
-
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/