Bernhard Walle wrote:
* Dave Anderson [2008-05-14 11:11]:
I suppose we could go with 5 instead of 4, and have dump_trace()
skip the first one, presuming that this anomoly is not architecture-
or compiler-dependent. Or maybe make it macro?
Did you compile with some optimisation? I think I remember that gcc
only inlines code with optimisation turned on.
No -- the Makefile is used as is -- you're the one modifying things... ;-)
I think it makes sense to use
`always_inline'
Generally, functions are not inlined unless optimization is
specified. For functions declared inline, this attribute inlines
the function even if no optimization level was specified.
here, i.e. __attribute__((always_inline)). I personally prefer inline
functions over macros.
Bernhard
Well, I don't have that personal preference, ;-) and after making it a macro,
it works as expected. At least with the macro, I *know* what's going to
happen...
Dave
--
Crash-utility mailing list
Crash-utility@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/crash-utility
