Rocksolid Light

Welcome to novaBBS (click a section below)

mail  files  register  newsreader  groups  login

Message-ID:  

It is not well to be thought of as one who meekly submits to insolence and intimidation.


devel / comp.arch / Re: Another use for R0

SubjectAuthor
* Another use for R0robf...@gmail.com
`- Re: Another use for R0MitchAlsup

1
Another use for R0

<3cdc9552-fa54-4c8c-86d2-7f01da87f928n@googlegroups.com>

  copy mid

https://www.novabbs.com/devel/article-flat.php?id=28204&group=comp.arch#28204

  copy link   Newsgroups: comp.arch
X-Received: by 2002:a05:620a:4398:b0:6ee:be9f:435c with SMTP id a24-20020a05620a439800b006eebe9f435cmr21680652qkp.35.1666586043714;
Sun, 23 Oct 2022 21:34:03 -0700 (PDT)
X-Received: by 2002:a05:6870:d623:b0:13a:f0ad:804a with SMTP id
a35-20020a056870d62300b0013af0ad804amr13252062oaq.79.1666586043419; Sun, 23
Oct 2022 21:34:03 -0700 (PDT)
Path: i2pn2.org!i2pn.org!usenet.blueworldhosting.com!feed1.usenet.blueworldhosting.com!peer02.iad!feed-me.highwinds-media.com!news.highwinds-media.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: comp.arch
Date: Sun, 23 Oct 2022 21:34:03 -0700 (PDT)
Injection-Info: google-groups.googlegroups.com; posting-host=2607:fea8:1dde:6a00:c31:b344:d1a3:8e47;
posting-account=QId4bgoAAABV4s50talpu-qMcPp519Eb
NNTP-Posting-Host: 2607:fea8:1dde:6a00:c31:b344:d1a3:8e47
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <3cdc9552-fa54-4c8c-86d2-7f01da87f928n@googlegroups.com>
Subject: Another use for R0
From: robfi...@gmail.com (robf...@gmail.com)
Injection-Date: Mon, 24 Oct 2022 04:34:03 +0000
Content-Type: text/plain; charset="UTF-8"
X-Received-Bytes: 1571
 by: robf...@gmail.com - Mon, 24 Oct 2022 04:34 UTC

In the rfPhoenix core any register may be used as a vector mask register.
So, then I hit upon the idea of having R0 act as the value -1 for the vector
mask, meaning the operation is unmasked.

I just switched the ISA to 48-bit instructions from 40. In part to reduce the number of different instruction formats. Having four six-bit register fields in
an instruction, plus a size or precision specifier and a format specifier ate
up a lot of bits. There are now a few extra bits in instructions rather than
having a more complex decoder.

Re: Another use for R0

<d4b468a9-35dd-42ca-b7d1-775c56945c35n@googlegroups.com>

  copy mid

https://www.novabbs.com/devel/article-flat.php?id=28205&group=comp.arch#28205

  copy link   Newsgroups: comp.arch
X-Received: by 2002:ac8:5902:0:b0:39c:e440:6adb with SMTP id 2-20020ac85902000000b0039ce4406adbmr28783029qty.18.1666640046613;
Mon, 24 Oct 2022 12:34:06 -0700 (PDT)
X-Received: by 2002:a05:6870:14cd:b0:13b:bc46:287f with SMTP id
l13-20020a05687014cd00b0013bbc46287fmr3090621oab.118.1666640046249; Mon, 24
Oct 2022 12:34:06 -0700 (PDT)
Path: i2pn2.org!i2pn.org!usenet.blueworldhosting.com!feed1.usenet.blueworldhosting.com!peer02.iad!feed-me.highwinds-media.com!news.highwinds-media.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: comp.arch
Date: Mon, 24 Oct 2022 12:34:06 -0700 (PDT)
In-Reply-To: <3cdc9552-fa54-4c8c-86d2-7f01da87f928n@googlegroups.com>
Injection-Info: google-groups.googlegroups.com; posting-host=2600:1700:291:29f0:a508:692b:c5cc:99f5;
posting-account=H_G_JQkAAADS6onOMb-dqvUozKse7mcM
NNTP-Posting-Host: 2600:1700:291:29f0:a508:692b:c5cc:99f5
References: <3cdc9552-fa54-4c8c-86d2-7f01da87f928n@googlegroups.com>
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <d4b468a9-35dd-42ca-b7d1-775c56945c35n@googlegroups.com>
Subject: Re: Another use for R0
From: MitchAl...@aol.com (MitchAlsup)
Injection-Date: Mon, 24 Oct 2022 19:34:06 +0000
Content-Type: text/plain; charset="UTF-8"
X-Received-Bytes: 1975
 by: MitchAlsup - Mon, 24 Oct 2022 19:34 UTC

On Sunday, October 23, 2022 at 11:34:05 PM UTC-5, robf...@gmail.com wrote:
> In the rfPhoenix core any register may be used as a vector mask register.
> So, then I hit upon the idea of having R0 act as the value -1 for the vector
> mask, meaning the operation is unmasked.
<
a) If you had sign control on register operands--this would be straightforward.
b) if you had your vector compare deliver {1,0} instead of {0,1} it would be
unnecessary.
>
> I just switched the ISA to 48-bit instructions from 40. In part to reduce the number of different instruction formats. Having four six-bit register fields in
> an instruction, plus a size or precision specifier and a format specifier ate
> up a lot of bits. There are now a few extra bits in instructions rather than
> having a more complex decoder.

1
server_pubkey.txt

rocksolid light 0.9.81
clearnet tor