chicken-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Chicken-users] chicken-primitive-inlines.scm


From: felix winkelmann
Subject: Re: [Chicken-users] chicken-primitive-inlines.scm
Date: Fri, 6 Feb 2009 15:45:57 +0100

On Thu, Feb 5, 2009 at 7:34 AM, Jim Ursetto <address@hidden> wrote:
> On Mon, Feb 2, 2009 at 12:07 PM, Kon Lovett <address@hidden> wrote:
>>  (##sys#slot l 1) -> (%cdr l)
>>  (##sys#setslot l 1 x)  -> (%set-cdr l x)
>
> I was wondering.  The compiler already inlines primitives when
> usual-integrations is in effect.  Would it not be possible to tell it,
> with a new declaration, to inline the *unsafe* versions of certain
> primitives?  Rather than manually specifying the % versions.

These primitive inlines are mainly intended for use in the core
libraries, where exact control over what is inlined and what not is crucial.
These libraries can't be compiled in unsafe mode, but certain
primitives can or must be coded with unsafe operations.


cheers,
felix




reply via email to

[Prev in Thread] Current Thread [Next in Thread]