Linux MIPS Architecture development
 help / color / mirror / Atom feed
From: "Kevin D. Kissell" <kevink@mips.com>
To: "Florian Lohoff" <flo@rfc822.org>, <linux-mips@oss.sgi.com>
Subject: Re: MIPS32 patches breaking DecStation
Date: Tue, 9 Jan 2001 01:34:47 +0100	[thread overview]
Message-ID: <000501c079d3$fefe1a60$0deca8c0@Ulysses> (raw)
In-Reply-To: 20010109004101.B27674@paradigm.rfc822.org

Florian,

Could you do me a huge favor and try a build that
uses 3 or 4 nops instead of the branch to the instruction
after the delay slot?   There was a reason that I eliminated
the branch construct from the MIPS internal Linux source
base - it's a hack that works perfectly on R4000's, but
it's pretty much a coincidence that it does so.  Yes,
the code fragment in question is R4K-specific, but
we really need to migrate towards the use of consistent
mechanisms that work across the full range of MIPS
CPUs.  Ideally, *all* CP0 hazards should some day be 
padded out with "ssnops" (sll $0,$0,1, if I recall), which 
force a 1 cycle delay per instruction even on superscalar 
MIPS CPUs.

            Kevin K.

WARNING: multiple messages have this Message-ID (diff)
From: "Kevin D. Kissell" <kevink@mips.com>
To: Florian Lohoff <flo@rfc822.org>, linux-mips@oss.sgi.com
Subject: Re: MIPS32 patches breaking DecStation
Date: Tue, 9 Jan 2001 01:34:47 +0100	[thread overview]
Message-ID: <000501c079d3$fefe1a60$0deca8c0@Ulysses> (raw)
Message-ID: <20010109003447.gQcRFGX1ReLVpl8ZBcA9dP4-Itflj1n1MfSfDI7x65w@z> (raw)
In-Reply-To: 20010109004101.B27674@paradigm.rfc822.org

Florian,

Could you do me a huge favor and try a build that
uses 3 or 4 nops instead of the branch to the instruction
after the delay slot?   There was a reason that I eliminated
the branch construct from the MIPS internal Linux source
base - it's a hack that works perfectly on R4000's, but
it's pretty much a coincidence that it does so.  Yes,
the code fragment in question is R4K-specific, but
we really need to migrate towards the use of consistent
mechanisms that work across the full range of MIPS
CPUs.  Ideally, *all* CP0 hazards should some day be 
padded out with "ssnops" (sll $0,$0,1, if I recall), which 
force a 1 cycle delay per instruction even on superscalar 
MIPS CPUs.

            Kevin K.

  parent reply	other threads:[~2001-01-09  0:31 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-01-08 23:41 MIPS32 patches breaking DecStation Florian Lohoff
2001-01-08 23:41 ` Florian Lohoff
2001-01-09  0:34 ` Kevin D. Kissell [this message]
2001-01-09  0:34   ` Kevin D. Kissell
2001-01-09  8:54   ` Florian Lohoff
2001-01-09 17:11     ` Harald Koerfgen
2001-01-09 18:28       ` Ralf Baechle
2001-01-09 19:30         ` Kevin D. Kissell
2001-01-09 19:30           ` Kevin D. Kissell
2001-01-09 19:54           ` Ralf Baechle
2001-01-09 19:54             ` Ralf Baechle
2001-01-09 20:33             ` Kevin D. Kissell
2001-01-09 20:33               ` Kevin D. Kissell
2001-01-09 15:09   ` Ralf Baechle
2001-01-09 15:09     ` Ralf Baechle
2001-01-09 15:50     ` Kevin D. Kissell
2001-01-09 15:50       ` Kevin D. Kissell
2001-01-09 16:43 ` Harald Koerfgen
2001-01-09 17:01   ` Ralf Baechle

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='000501c079d3$fefe1a60$0deca8c0@Ulysses' \
    --to=kevink@mips.com \
    --cc=flo@rfc822.org \
    --cc=linux-mips@oss.sgi.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox