[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: scm_bits_t / scm_ubits_t
From: |
Dirk Herrmann |
Subject: |
Re: scm_bits_t / scm_ubits_t |
Date: |
Thu, 14 Jun 2001 17:56:17 +0200 (MEST) |
On 14 Jun 2001, Marius Vollmer wrote:
> Marius Vollmer <address@hidden> writes:
>
> > and make scm_t_bits an unsigned type.
>
> Err, what was the reason again for making scm_t_bits unsigned?
The reason is, that it is basically a raw bit structure, which does not
have any sign information. Signedness is an issue that only comes up
_after_ the type information has been extracted.
> I just realized that we need a signed version anyway for fixnum
> fiddling.
I think the point is different: If a SCM appears to be a fixnum, the raw
bit structure has to be converted into a signed integral value. The type
scm_t_bits itself should not be signed or unsigned. It is all a matter of
clean type conversions. In our case, the macros SCM_INUM and SCM_MAKINUM
would do the casting.
Best regards,
Dirk Herrmann
- Re: scm_bits_t / scm_ubits_t, (continued)
- Re: scm_bits_t / scm_ubits_t, Dirk Herrmann, 2001/06/11
- Re: scm_bits_t / scm_ubits_t, Jacques A. Vidrine, 2001/06/11
- Re: scm_bits_t / scm_ubits_t, Marius Vollmer, 2001/06/11
- Re: scm_bits_t / scm_ubits_t, Marius Vollmer, 2001/06/13
- Re: scm_bits_t / scm_ubits_t, Jacques A. Vidrine, 2001/06/13
- Re: scm_bits_t / scm_ubits_t, Marius Vollmer, 2001/06/13
- Re: scm_bits_t / scm_ubits_t, Rob Browning, 2001/06/14
- Re: scm_bits_t / scm_ubits_t,
Dirk Herrmann <=
- Re: scm_bits_t / scm_ubits_t, Marius Vollmer, 2001/06/14