public inbox for gcc-prs@sourceware.org
help / color / mirror / Atom feed
From: nathan@rtfm.net
To: gcc-gnats@gcc.gnu.org
Subject: target/2732: Interaction with Solaris assembler and SPARCV9 instructions
Date: Thu, 03 May 2001 13:06:00 -0000	[thread overview]
Message-ID: <20010503200135.20189.qmail@sourceware.cygnus.com> (raw)

>Number:         2732
>Category:       target
>Synopsis:       Interaction with Solaris assembler and SPARCV9 instructions
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu May 03 13:06:00 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Nathan Dorfman <nathan@rtfm.net>
>Release:        2.95.2, 2.95.3, and gcc version 3.1 20010502 (experimental)
>Organization:
>Environment:
SunOS kiev 5.7 Generic_106541-15 sun4u sparc SUNW,Ultra-5_10
>Description:
This appears to be a minor oversight, but exists on
2.95.2 and 2.95.3 release versions as well as yesterday's
cvs checkout, which I did just to test this. 

If you use -mcpu=v9 (which should mean "allow SPARCV9
instruction set"), gcc will generate V9 instructions,
but will fail to tell the assembler to assemble them
(as -xarch=v9), generating the below message:

/usr/ccs/bin/as: "/var/tmp//cc9CdJR3.s", line 11: error: cannot use v8plus instructions in a non-v8plus target binary

This is the case for both C source files for which
V9 instructions are generated as well as ASM source files
which already include them.

HOWEVER: if -mcpu=ultrasparc is used, everything works
fine (as is called with -xarch=v9)...it's just that
-mcpu=v9 *should* give you the V9 instruction set.
>How-To-Repeat:
Use the attached file. Compile it with -mcpu=v9 should give
you a 'mulx' instruction. Alternatively, contact me for
a simple asm 3-liner that just uses mulx directly.

Anyway, -mcpu=v9 and -mcpu=ultrasparc will produce the
same assembler output (containing the V9 instruction mulx).
However, with -mcpu=v9 the assembly stage will fail because
as is not told to accept V9 instructions.
>Fix:
as -xarch=v9
>Release-Note:
>Audit-Trail:
>Unformatted:
----gnatsweb-attachment----
Content-Type: application/octet-stream; name="multest.c"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="multest.c"

aW50IG1haW4oKSB7CglyZWdpc3RlciBpbnQgaSA9IDUsIGogPSAzOwoJaSAqPSBqOwp9Cg==


             reply	other threads:[~2001-05-03 13:06 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-05-03 13:06 nathan [this message]
2002-01-03 21:37 rodrigc
2002-01-27 16:06 Nathan Dorfman
2002-01-27 17:45 rodrigc

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=20010503200135.20189.qmail@sourceware.cygnus.com \
    --to=nathan@rtfm.net \
    --cc=gcc-gnats@gcc.gnu.org \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).