Rocksolid Light

Welcome to novaBBS (click a section below)

mail  files  register  newsreader  groups  login

Message-ID:  

The documentation is in Japanese. Good luck. -- Rich $alz


devel / comp.arch / Filling unused instruction bits

SubjectAuthor
* Filling unused instruction bitsrobf...@gmail.com
+- Re: Filling unused instruction bitsStephen Fuld
`* Re: Filling unused instruction bitsJimBrakefield
 `* Re: Filling unused instruction bitsrobf...@gmail.com
  `* Re: Filling unused instruction bitsJohn Levine
   `* Re: Filling unused instruction bitsTerje Mathisen
    +- Re: Filling unused instruction bitsStephen Fuld
    `* Re: Filling unused instruction bitsBGB
     +- Re: Filling unused instruction bitsrobf...@gmail.com
     `* Re: Filling unused instruction bitsMitchAlsup
      `- Re: Filling unused instruction bitsBranimir Maksimovic

1
Filling unused instruction bits

<aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20727&group=comp.arch#20727

 copy link   Newsgroups: comp.arch
X-Received: by 2002:ac8:71cd:: with SMTP id i13mr10359157qtp.159.1633302433330;
Sun, 03 Oct 2021 16:07:13 -0700 (PDT)
X-Received: by 2002:aca:f0d:: with SMTP id 13mr11936580oip.110.1633302433142;
Sun, 03 Oct 2021 16:07:13 -0700 (PDT)
Path: rocksolid2!i2pn.org!weretis.net!feeder6.news.weretis.net!news.misty.com!border2.nntp.dca1.giganews.com!nntp.giganews.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: comp.arch
Date: Sun, 3 Oct 2021 16:07:12 -0700 (PDT)
Injection-Info: google-groups.googlegroups.com; posting-host=2607:fea8:1de5:5800:181:e467:596f:70d4;
posting-account=QId4bgoAAABV4s50talpu-qMcPp519Eb
NNTP-Posting-Host: 2607:fea8:1de5:5800:181:e467:596f:70d4
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
Subject: Filling unused instruction bits
From: robfi...@gmail.com (robf...@gmail.com)
Injection-Date: Sun, 03 Oct 2021 23:07:13 +0000
Content-Type: text/plain; charset="UTF-8"
Lines: 0
 by: robf...@gmail.com - Sun, 3 Oct 2021 23:07 UTC

I am wondering if unused instruction bits should be filled with random values instead of zero? Would it have better entropy characteristics to fill randomly versus zero filled?

Re: Filling unused instruction bits

<sjdgni$geg$1@dont-email.me>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20728&group=comp.arch#20728

 copy link   Newsgroups: comp.arch
Path: rocksolid2!i2pn.org!eternal-september.org!reader02.eternal-september.org!.POSTED!not-for-mail
From: sfu...@alumni.cmu.edu.invalid (Stephen Fuld)
Newsgroups: comp.arch
Subject: Re: Filling unused instruction bits
Date: Sun, 3 Oct 2021 17:09:22 -0700
Organization: A noiseless patient Spider
Lines: 19
Message-ID: <sjdgni$geg$1@dont-email.me>
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Date: Mon, 4 Oct 2021 00:09:22 -0000 (UTC)
Injection-Info: reader02.eternal-september.org; posting-host="7452241f3ffe218f7511993dedf2140d";
logging-data="16848"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1/56bBGr+vToLebhlFvUKhbg2wTV8Tqbuw="
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101
Thunderbird/91.1.2
Cancel-Lock: sha1:RbwAlJ3yG/5qouFCTxFvIFV9V5I=
In-Reply-To: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
Content-Language: en-US
 by: Stephen Fuld - Mon, 4 Oct 2021 00:09 UTC

On 10/3/2021 4:07 PM, robf...@gmail.com wrote:
> I am wondering if unused instruction bits should be filled with random values instead of zero? Would it have better entropy characteristics to fill randomly versus zero filled?
>

What a remarkable idea! Not only does it have better entropy, but it
prevents some other damn fool architect from messing up your carefully
designed ISA by defining extensions using those bits.

You could even take this a step further, buy allowing the destination
register field to contain a random value. Not only does that provide
better entropy, but, since it probably isn't the register where you need
the value, you add a register to register move instruction, which, since
it only uses one of the two source fields, allows you to fill the other
source field with random bits. This further betters entropy.

--
- Stephen Fuld
(e-mail address disguised to prevent spam)

Re: Filling unused instruction bits

<97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20729&group=comp.arch#20729

 copy link   Newsgroups: comp.arch
X-Received: by 2002:a05:620a:409:: with SMTP id 9mr7984794qkp.76.1633306825543;
Sun, 03 Oct 2021 17:20:25 -0700 (PDT)
X-Received: by 2002:a9d:764c:: with SMTP id o12mr7733537otl.129.1633306825303;
Sun, 03 Oct 2021 17:20:25 -0700 (PDT)
Path: rocksolid2!i2pn.org!weretis.net!feeder6.news.weretis.net!news.misty.com!border2.nntp.dca1.giganews.com!nntp.giganews.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: comp.arch
Date: Sun, 3 Oct 2021 17:20:25 -0700 (PDT)
In-Reply-To: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
Injection-Info: google-groups.googlegroups.com; posting-host=136.50.182.0; posting-account=AoizIQoAAADa7kQDpB0DAj2jwddxXUgl
NNTP-Posting-Host: 136.50.182.0
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com>
Subject: Re: Filling unused instruction bits
From: jim.brak...@ieee.org (JimBrakefield)
Injection-Date: Mon, 04 Oct 2021 00:20:25 +0000
Content-Type: text/plain; charset="UTF-8"
Lines: 3
 by: JimBrakefield - Mon, 4 Oct 2021 00:20 UTC

On Sunday, October 3, 2021 at 6:07:14 PM UTC-5, robf...@gmail.com wrote:
> I am wondering if unused instruction bits should be filled with random values instead of zero? Would it have better entropy characteristics to fill randomly versus zero filled?

Do the random values vary with time or memory address?

Re: Filling unused instruction bits

<390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20730&group=comp.arch#20730

 copy link   Newsgroups: comp.arch
X-Received: by 2002:ac8:5788:: with SMTP id v8mr10903081qta.372.1633307796410;
Sun, 03 Oct 2021 17:36:36 -0700 (PDT)
X-Received: by 2002:a9d:332:: with SMTP id 47mr7422413otv.333.1633307796147;
Sun, 03 Oct 2021 17:36:36 -0700 (PDT)
Path: rocksolid2!i2pn.org!weretis.net!feeder6.news.weretis.net!news.misty.com!border2.nntp.dca1.giganews.com!nntp.giganews.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: comp.arch
Date: Sun, 3 Oct 2021 17:36:35 -0700 (PDT)
In-Reply-To: <97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com>
Injection-Info: google-groups.googlegroups.com; posting-host=2607:fea8:1de5:5800:181:e467:596f:70d4;
posting-account=QId4bgoAAABV4s50talpu-qMcPp519Eb
NNTP-Posting-Host: 2607:fea8:1de5:5800:181:e467:596f:70d4
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com> <97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com>
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
Subject: Re: Filling unused instruction bits
From: robfi...@gmail.com (robf...@gmail.com)
Injection-Date: Mon, 04 Oct 2021 00:36:36 +0000
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Lines: 26
 by: robf...@gmail.com - Mon, 4 Oct 2021 00:36 UTC

On Sunday, October 3, 2021 at 8:20:26 PM UTC-4, JimBrakefield wrote:
> On Sunday, October 3, 2021 at 6:07:14 PM UTC-5, robf...@gmail.com wrote:
> > I am wondering if unused instruction bits should be filled with random values instead of zero? Would it have better entropy characteristics to fill randomly versus zero filled?
> Do the random values vary with time or memory address?

I think they would be filled in when the program is assembled by the assembler. So they would vary by memory address but not in time. To get them to vary in time would require the code to be updated at run-time.

>What a remarkable idea! Not only does it have better entropy, but it
>prevents some other damn fool architect from messing up your carefully
>designed ISA by defining extensions using those bits.

This would be for instructions where there is little likely-hood of them being extended. A couple of instructions studied are instructions requiring only one source register like ABS, NOT or NEG. Since the instruction is a fixed size there are about seven bits unused in the R1 type instructions. Conditional branch to register instruction has a whopping 16 bits unused. I think it would be safe to set 15 of them to random values. Leaving one bit for future extensions. But I am wondering if it would make a difference to power consumption for instance. Having the bus more random might lead to it averaging at the half-way point electrically?
There are lots of unused opcodes which would not be set randomly, so there is room for future extensions.

Re: Filling unused instruction bits

<sjdl44$2n92$1@gal.iecc.com>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20731&group=comp.arch#20731

 copy link   Newsgroups: comp.arch
Path: rocksolid2!i2pn.org!weretis.net!feeder6.news.weretis.net!news.misty.com!news.iecc.com!.POSTED.news.iecc.com!not-for-mail
From: joh...@taugh.com (John Levine)
Newsgroups: comp.arch
Subject: Re: Filling unused instruction bits
Date: Mon, 4 Oct 2021 01:24:20 -0000 (UTC)
Organization: Taughannock Networks
Message-ID: <sjdl44$2n92$1@gal.iecc.com>
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com> <97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com> <390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
Injection-Date: Mon, 4 Oct 2021 01:24:20 -0000 (UTC)
Injection-Info: gal.iecc.com; posting-host="news.iecc.com:2001:470:1f07:1126:0:676f:7373:6970";
logging-data="89378"; mail-complaints-to="abuse@iecc.com"
In-Reply-To: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com> <97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com> <390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
Cleverness: some
X-Newsreader: trn 4.0-test77 (Sep 1, 2010)
Originator: johnl@iecc.com (John Levine)
 by: John Levine - Mon, 4 Oct 2021 01:24 UTC

According to robf...@gmail.com <robfi680@gmail.com>:
>>What a remarkable idea! Not only does it have better entropy, but it
>>prevents some other damn fool architect from messing up your carefully
>>designed ISA by defining extensions using those bits.
>
>This would be for instructions where there is little likely-hood of them being extended.

So you use the bit pattern for some totally unrelated instruction you forgot to include
the first time around.

I think you should take the hint -- this is not a good idea.

--
Regards,
John Levine, johnl@taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly

Re: Filling unused instruction bits

<sjerp9$rde$1@gioia.aioe.org>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20732&group=comp.arch#20732

 copy link   Newsgroups: comp.arch
Path: rocksolid2!news.neodome.net!news.mixmin.net!aioe.org!rd9pRsUZyxkRLAEK7e/Uzw.user.46.165.242.91.POSTED!not-for-mail
From: terje.ma...@tmsw.no (Terje Mathisen)
Newsgroups: comp.arch
Subject: Re: Filling unused instruction bits
Date: Mon, 4 Oct 2021 14:24:08 +0200
Organization: Aioe.org NNTP Server
Message-ID: <sjerp9$rde$1@gioia.aioe.org>
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
<97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com>
<390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
<sjdl44$2n92$1@gal.iecc.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Info: gioia.aioe.org; logging-data="28078"; posting-host="rd9pRsUZyxkRLAEK7e/Uzw.user.gioia.aioe.org"; mail-complaints-to="abuse@aioe.org";
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:60.0) Gecko/20100101
Firefox/60.0 SeaMonkey/2.53.9
X-Notice: Filtered by postfilter v. 0.9.2
 by: Terje Mathisen - Mon, 4 Oct 2021 12:24 UTC

John Levine wrote:
> According to robf...@gmail.com <robfi680@gmail.com>:
>>> What a remarkable idea! Not only does it have better entropy, but it
>>> prevents some other damn fool architect from messing up your carefully
>>> designed ISA by defining extensions using those bits.
>>
>> This would be for instructions where there is little likely-hood of them being extended.
>
> So you use the bit pattern for some totally unrelated instruction you forgot to include
> the first time around.
>
> I think you should take the hint -- this is not a good idea.
>
I was hoping the initial "idea" post was meant as a joke, if there is
one thing all instruction set architects strive for it is to make the
resulting code as compact as possible.

It is right up there with making it testable, as well as fast to decode
on superscalar versions.

Terje

--
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"

Re: Filling unused instruction bits

<sjf4e7$anv$1@dont-email.me>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20733&group=comp.arch#20733

 copy link   Newsgroups: comp.arch
Path: rocksolid2!i2pn.org!eternal-september.org!reader02.eternal-september.org!.POSTED!not-for-mail
From: sfu...@alumni.cmu.edu.invalid (Stephen Fuld)
Newsgroups: comp.arch
Subject: Re: Filling unused instruction bits
Date: Mon, 4 Oct 2021 07:51:49 -0700
Organization: A noiseless patient Spider
Lines: 39
Message-ID: <sjf4e7$anv$1@dont-email.me>
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
<97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com>
<390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
<sjdl44$2n92$1@gal.iecc.com> <sjerp9$rde$1@gioia.aioe.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Date: Mon, 4 Oct 2021 14:51:51 -0000 (UTC)
Injection-Info: reader02.eternal-september.org; posting-host="7452241f3ffe218f7511993dedf2140d";
logging-data="11007"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1+n2ybsk/bJRIzoZobAeDGQ2TkZ1KzdyNM="
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101
Thunderbird/91.1.2
Cancel-Lock: sha1:J8maL0H3GDl7ZByT78/hj8s2rQc=
In-Reply-To: <sjerp9$rde$1@gioia.aioe.org>
Content-Language: en-US
 by: Stephen Fuld - Mon, 4 Oct 2021 14:51 UTC

On 10/4/2021 5:24 AM, Terje Mathisen wrote:
> John Levine wrote:
>> According to robf...@gmail.com <robfi680@gmail.com>:
>>>> What a remarkable idea! Not only does it have better entropy, but it
>>>> prevents some other damn fool architect from messing up your carefully
>>>> designed ISA by defining extensions using those bits.
>>>
>>> This would be for instructions where there is little likely-hood of
>>> them being extended.
>>
>> So you use the bit pattern for some totally unrelated instruction you
>> forgot to include
>> the first time around.
>>
>> I think you should take the hint -- this is not a good idea.
>>
> I was hoping the initial "idea" post was meant as a joke,

In my initial response, I assumed it was a homework assignment and tried
to respond in the traditional comp.arch manner.

> if there is
> one thing all instruction set architects strive for it is to make the
> resulting code as compact as possible.
>
> It is right up there with making it testable, as well as fast to decode
> on superscalar versions.

If it was a real question, then, of course, I agree with the group
consensus that the proposal provides essentially zero advantages and
some non-trivial disadvantages, so shouldn't be implemented.

--
- Stephen Fuld
(e-mail address disguised to prevent spam)

Re: Filling unused instruction bits

<sjf7h6$499$1@dont-email.me>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20734&group=comp.arch#20734

 copy link   Newsgroups: comp.arch
Path: rocksolid2!news.neodome.net!news.mixmin.net!eternal-september.org!reader02.eternal-september.org!.POSTED!not-for-mail
From: cr88...@gmail.com (BGB)
Newsgroups: comp.arch
Subject: Re: Filling unused instruction bits
Date: Mon, 4 Oct 2021 10:44:37 -0500
Organization: A noiseless patient Spider
Lines: 31
Message-ID: <sjf7h6$499$1@dont-email.me>
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
<97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com>
<390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
<sjdl44$2n92$1@gal.iecc.com> <sjerp9$rde$1@gioia.aioe.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Date: Mon, 4 Oct 2021 15:44:38 -0000 (UTC)
Injection-Info: reader02.eternal-september.org; posting-host="f9cbd5831bf664e86e4af1faf63e77ba";
logging-data="4393"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX194nZYYylioo495QjSzHX3Q"
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101
Thunderbird/91.1.2
Cancel-Lock: sha1:6Nc8MqB/+Xes5KjNb8wiNtdd23M=
In-Reply-To: <sjerp9$rde$1@gioia.aioe.org>
Content-Language: en-US
 by: BGB - Mon, 4 Oct 2021 15:44 UTC

On 10/4/2021 7:24 AM, Terje Mathisen wrote:
> John Levine wrote:
>> According to robf...@gmail.com <robfi680@gmail.com>:
>>>> What a remarkable idea! Not only does it have better entropy, but it
>>>> prevents some other damn fool architect from messing up your carefully
>>>> designed ISA by defining extensions using those bits.
>>>
>>> This would be for instructions where there is little likely-hood of
>>> them being extended.
>>
>> So you use the bit pattern for some totally unrelated instruction you
>> forgot to include
>> the first time around.
>>
>> I think you should take the hint -- this is not a good idea.
>>
> I was hoping the initial "idea" post was meant as a joke, if there is
> one thing all instruction set architects strive for it is to make the
> resulting code as compact as possible.
>
> It is right up there with making it testable, as well as fast to decode
> on superscalar versions.
>

Agreed. The OP was a pretty much terrible idea.
Bad for extensibility;
Bad for compression;
...

My recent idea, while kinda hackish and awful, at least has a point.

Re: Filling unused instruction bits

<2ef0f0d1-6cea-4a29-8f40-2c754a0def33n@googlegroups.com>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20737&group=comp.arch#20737

 copy link   Newsgroups: comp.arch
X-Received: by 2002:ac8:40d5:: with SMTP id f21mr3192162qtm.265.1633591935741;
Thu, 07 Oct 2021 00:32:15 -0700 (PDT)
X-Received: by 2002:aca:f0d:: with SMTP id 13mr11133488oip.110.1633591935555;
Thu, 07 Oct 2021 00:32:15 -0700 (PDT)
Path: rocksolid2!i2pn2.org!i2pn.org!weretis.net!feeder6.news.weretis.net!news.misty.com!border2.nntp.dca1.giganews.com!border1.nntp.dca1.giganews.com!nntp.giganews.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: comp.arch
Date: Thu, 7 Oct 2021 00:32:15 -0700 (PDT)
In-Reply-To: <sjf7h6$499$1@dont-email.me>
Injection-Info: google-groups.googlegroups.com; posting-host=2607:fea8:1de5:5800:29c1:3bfc:dccf:5fbe;
posting-account=QId4bgoAAABV4s50talpu-qMcPp519Eb
NNTP-Posting-Host: 2607:fea8:1de5:5800:29c1:3bfc:dccf:5fbe
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
<97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com> <390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
<sjdl44$2n92$1@gal.iecc.com> <sjerp9$rde$1@gioia.aioe.org> <sjf7h6$499$1@dont-email.me>
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <2ef0f0d1-6cea-4a29-8f40-2c754a0def33n@googlegroups.com>
Subject: Re: Filling unused instruction bits
From: robfi...@gmail.com (robf...@gmail.com)
Injection-Date: Thu, 07 Oct 2021 07:32:15 +0000
Content-Type: text/plain; charset="UTF-8"
Lines: 2
 by: robf...@gmail.com - Thu, 7 Oct 2021 07:32 UTC

Thanks for the feedback. I did not think it was a good idea, but was not sure.

Right up there with bus balancing NOPs. Inserting special NOP instructions in the instruction stream to help average out the bus DC level.

Re: Filling unused instruction bits

<c7ec1eaa-88e8-4756-a0d3-a252dd08a68dn@googlegroups.com>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20744&group=comp.arch#20744

 copy link   Newsgroups: comp.arch
X-Received: by 2002:a0c:fc43:: with SMTP id w3mr7215035qvp.46.1633652704912;
Thu, 07 Oct 2021 17:25:04 -0700 (PDT)
X-Received: by 2002:aca:b5c3:: with SMTP id e186mr14445435oif.51.1633652704719;
Thu, 07 Oct 2021 17:25:04 -0700 (PDT)
Path: rocksolid2!i2pn.org!weretis.net!feeder6.news.weretis.net!news.misty.com!border2.nntp.dca1.giganews.com!nntp.giganews.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: comp.arch
Date: Thu, 7 Oct 2021 17:25:04 -0700 (PDT)
In-Reply-To: <sjf7h6$499$1@dont-email.me>
Injection-Info: google-groups.googlegroups.com; posting-host=2600:1700:291:29f0:ec77:5d4:3f50:5858;
posting-account=H_G_JQkAAADS6onOMb-dqvUozKse7mcM
NNTP-Posting-Host: 2600:1700:291:29f0:ec77:5d4:3f50:5858
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
<97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com> <390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
<sjdl44$2n92$1@gal.iecc.com> <sjerp9$rde$1@gioia.aioe.org> <sjf7h6$499$1@dont-email.me>
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <c7ec1eaa-88e8-4756-a0d3-a252dd08a68dn@googlegroups.com>
Subject: Re: Filling unused instruction bits
From: MitchAl...@aol.com (MitchAlsup)
Injection-Date: Fri, 08 Oct 2021 00:25:04 +0000
Content-Type: text/plain; charset="UTF-8"
Lines: 40
 by: MitchAlsup - Fri, 8 Oct 2021 00:25 UTC

On Monday, October 4, 2021 at 10:44:41 AM UTC-5, BGB wrote:
> On 10/4/2021 7:24 AM, Terje Mathisen wrote:
> > John Levine wrote:
> >> According to robf...@gmail.com <robf...@gmail.com>:
> >>>> What a remarkable idea! Not only does it have better entropy, but it
> >>>> prevents some other damn fool architect from messing up your carefully
> >>>> designed ISA by defining extensions using those bits.
> >>>
> >>> This would be for instructions where there is little likely-hood of
> >>> them being extended.
> >>
> >> So you use the bit pattern for some totally unrelated instruction you
> >> forgot to include
> >> the first time around.
> >>
> >> I think you should take the hint -- this is not a good idea.
> >>
> > I was hoping the initial "idea" post was meant as a joke, if there is
> > one thing all instruction set architects strive for it is to make the
> > resulting code as compact as possible.
> >
> > It is right up there with making it testable, as well as fast to decode
> > on superscalar versions.
> >
> Agreed. The OP was a pretty much terrible idea.
> Bad for extensibility;
> Bad for compression;
> ...
>
> My recent idea, while kinda hackish and awful, at least has a point.
<
I have been "out of reach of the net" for the last week.
I am glad this topic reached its "obvious" and "determative" conclusion.
<
Although one might achieve a trifling of entropy advantage.
<
You end up through out the entire ability to add to the original instruction
set -- simple because you consumed all that entropy in a way that makes
enhancements exceedingly difficult.
<
Please don't do this to yourself.

Re: Filling unused instruction bits

<sfN7J.53058$3p3.45502@fx16.iad>

 copy mid

https://www.novabbs.com/devel/article-flat.php?id=20749&group=comp.arch#20749

 copy link   Newsgroups: comp.arch
Path: rocksolid2!i2pn.org!weretis.net!feeder8.news.weretis.net!feeder1.feed.usenet.farm!feed.usenet.farm!newsfeed.xs4all.nl!newsfeed8.news.xs4all.nl!news-out.netnews.com!news.alt.net!fdc2.netnews.com!peer01.ams1!peer.ams1.xlned.com!news.xlned.com!peer01.iad!feed-me.highwinds-media.com!news.highwinds-media.com!fx16.iad.POSTED!not-for-mail
Newsgroups: comp.arch
From: branimir...@icloud.com (Branimir Maksimovic)
Subject: Re: Filling unused instruction bits
References: <aa5dffc9-9c6f-4c79-ae23-b904b0f81634n@googlegroups.com>
<97552c83-a6d4-4932-882e-7466f73a7b37n@googlegroups.com>
<390d9eed-7a7b-478b-b05a-0dfaa276f991n@googlegroups.com>
<sjdl44$2n92$1@gal.iecc.com> <sjerp9$rde$1@gioia.aioe.org>
<sjf7h6$499$1@dont-email.me>
<c7ec1eaa-88e8-4756-a0d3-a252dd08a68dn@googlegroups.com>
User-Agent: slrn/1.0.3 (Darwin)
Lines: 13
Message-ID: <sfN7J.53058$3p3.45502@fx16.iad>
X-Complaints-To: abuse@usenet-news.net
NNTP-Posting-Date: Fri, 08 Oct 2021 01:35:20 UTC
Organization: usenet-news.net
Date: Fri, 08 Oct 2021 01:35:20 GMT
X-Received-Bytes: 1288
 by: Branimir Maksimovic - Fri, 8 Oct 2021 01:35 UTC

On 2021-10-08, MitchAlsup <MitchAlsup@aol.com> wrote:
><
> I have been "out of reach of the net" for the last week.
You mean, you were somewhere without internet, no access to usenet,
or simply you hadn't had internet connection?
Regarding recent internet outage, I believe it is
because Sun large Protuberances :P

--

7-77-777
Evil Sinner!
with software, you repeat same experiment, expecting different results...

1
server_pubkey.txt

rocksolid light 0.9.7
clearnet tor