Re: [boost] Exception specifications in smart_ptr.hpp

From: David Abrahams (abrahams_at_[hidden])
Date: 2000-07-17 08:38:42

----- Original Message -----
From: "Greg Colvin" <greg_at_[hidden]>

> As Beman noted:
>
> > smart_ptr.hpp has throw specifiers on a number of time-critical inline
> > functions such as dereferencing:

Yes, and on these particular functions the exception-specifications arguably
can only hurt, not help.

Consider:
None of these functions calls other functions; they only manipulate
fundamental data types like pointers
Therefore, no runtime behavior of the exception-specification can ever be
invoked
They are completely exposed to the compiler; indeed they're declared inline
Therefore, a smart compiler can easily deduce that the functions are
incapable of throwing exceptions, and make the same optimizations it would
have made based on the empty exception-specification.
A "dumb" compiler, however, could make all kinds of pessimizations, as we
have seen.

I'm just trying to show once again that although initially appealing, an
exception-specification tends to have consequences that require _very_
careful thought to understand.

-Dave