Re: libstdc++/7926

[Phil Edwards <phil at jaj dot com>]

On Fri, Nov 01, 2002 at 10:44:10PM -0500, Daniel Jacobowitz wrote:
> On Fri, Nov 01, 2002 at 07:41:29PM -0600, Benjamin Kosnik wrote:
> > 
> > tested x86/linux
> > 
> > Stops pretending that i386's atomicity.h ever existed, use generic. Arm,
> > same. Use i486 for x86-64, and athlon. Remove extraneous bits.
> 
> Well, OK.  Given that a lot of vendors (well, maybe not a lot, but I'm
> sure Debian isn't the only one!) ship i386-linux compilers... is there
> any plan to make the generic ones _correct_ by using mutexes where
> available?

It's been on my back burner, yes.


> Given that std::string's default constructor uses these (a design
> decision I'm not really fond of - why on earth is it necessary?  Is
> handling having no representation really that much of a hit?)

Ask Nathan on the libstdc++ list.


Phil

-- 
I would therefore like to posit that computing's central challenge, viz. "How
not to make a mess of it," has /not/ been met.
                                                 - Edsger Dijkstra, 1930-2002