Re: [Scheme-reports] fresh empty strings Alaric Snell-Pym (23 Jan 2012 13:28 UTC)
Re: [Scheme-reports] fresh empty strings Andy Wingo (23 Jan 2012 15:23 UTC)
Re: [Scheme-reports] fresh empty strings Alaric Snell-Pym (23 Jan 2012 15:34 UTC)
Re: [Scheme-reports] fresh empty strings Ray Dillinger (23 Jan 2012 20:17 UTC)
Re: [Scheme-reports] fresh empty strings Andy Wingo (23 Jan 2012 20:52 UTC)
Re: [Scheme-reports] fresh empty strings John Cowan (23 Jan 2012 23:05 UTC)
Re: [Scheme-reports] fresh empty strings Vincent Manis (24 Jan 2012 01:02 UTC)
Re: [Scheme-reports] fresh empty strings John Cowan (24 Jan 2012 02:06 UTC)
Re: [Scheme-reports] fresh empty strings Alaric Snell-Pym (24 Jan 2012 09:37 UTC)
Re: [Scheme-reports] fresh empty strings John Cowan (24 Jan 2012 15:10 UTC)
Re: [Scheme-reports] fresh empty strings xacc.ide@gmail.com (24 Jan 2012 16:31 UTC)
Re: [Scheme-reports] fresh empty strings Per Bothner (24 Jan 2012 17:21 UTC)
Re: [Scheme-reports] fresh empty strings Andy Wingo (24 Jan 2012 17:00 UTC)
Re: [Scheme-reports] fresh empty strings Aubrey Jaffer (24 Jan 2012 02:16 UTC)

Re: [Scheme-reports] fresh empty strings Ray Dillinger 23 Jan 2012 20:16 UTC

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 01/23/2012 07:03 AM, Andy Wingo wrote:
> On Mon 23 Jan 2012 14:27, Alaric Snell-Pym <alaric@snell-pym.org.uk> writes:
>
>> Short strings might be represented in a single SSE2 vector register or
>> something
>
> The point of this nit-pick is that this is impossible -- strings of one
> character *cannot* be immediates.
>
> Regards,

That's just not true.

The size of immediate values varies from implementation to
implementation.  Even allowing a full byte for a typetag,
in an implementation where immediates are  64-bits it's
possible to have immediate strings of up to 7 UTF8 values.

That's very short for a string, but worth it if you're
handling a lot of strings which are one word (in the language
sense).

You don't have to heap-allocate strings until can no longer
prove that you have only one reference to the string.

				Bear

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJPHcAuAAoJEAOzWkqOibfNNGUH/2sElzwKBoVaDTR/cLZ+vbJ1
tWYhl9cm5FD+2LQdMTVHYafOLe6gypSyABS9lZqCJyUhvtQ0Wk1RDDkd0gjDHHcL
vujMvTpSgVye1stet0z8IUHLtOUvNJ3m/jqwLP4QrH58DQNycvWSFM64apcfdOek
fB+4nfkG4psuymXnwvu0ORcxsDg+5DYaLjIaBSyt7mfOWtMhAIr+/hwOT6/Nhbvl
/dfXaTJpmYmei6yia6fDTSO684p+5zPzIWEHzBlUI5eB8M2u9NnQTG+s3RKrkdVn
OTMhCZGt52I4xMJPynMw+ZHON8L9kqD9TkMTwSTI6fMFEu65MnOtj6yuowUI4ko=
=/Zts
-----END PGP SIGNATURE-----

_______________________________________________
Scheme-reports mailing list
Scheme-reports@scheme-reports.org
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports