On 2008-02-16, Oleg Verych <[email protected]> wrote:
>> Sorry, I've no idea what you're saying. That it's OK for a
>> compiler to generate incorrect code as long as it runs fast?
>
> Yes, if this is 90% or more cases (experimentally, e.g. how often this
> question was addressed here). Otherwise tools must be expanded or fixed.
I completely disgree. It is _never_ acceptible for a compiler
designer/maintainer to decide that it's OK for the compiler to
generate incorrect code because the correct code would be less
efficient. Generating the wrong answer quickly is worse that
usless.
> I think, former is easier way, but probably not understood to
> evaluate, by way i present it.
>
> And as it seems, latter is problem of developer's views
> (current state of the MSP430 toolchain) or/and money (to pay
> for implementing feature, someone wants/"thinks, is correct").
What all the other targets do is correct.
--
Grant Edwards grante Yow! I am a traffic light,
at and Alan Ginzberg kidnapped
visi.com my laundry in 1927!
