public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jeffrey A Law <law@cygnus.com>
To: Richard Henderson <rth@cygnus.com>
Cc: egcs@cygnus.com
Subject: Re: alpha failure on 920810-1
Date: Wed, 06 May 1998 23:49:00 -0000	[thread overview]
Message-ID: <24630.894510594@hurl.cygnus.com> (raw)
In-Reply-To: <19980427014147.34497@dot.cygnus.com>

  In message <19980427014147.34497@dot.cygnus.com>you write:
  > The correct transformation would be to retain the compare and base
  > the conditional move off of that:
  > 
  >     (insn 9 8 10 (set (reg:DF 69)
  >             (eq:DF (reg/v:DF 68)
  >                 (const_double:DF (cc0) 0 0))) -1 (nil)
  >         (nil))
  > 
  >     (insn 25 23 15 (set (reg/v:DF 68)
  >             (if_then_else:DF (ne (reg:DF 69)
  >                     (const_double:DF (cc0) 0 0))
  >                 (reg:DF 48 $f16)
  >                 (reg:DF 72))) -1 (nil)
  >         (nil))
I wonder if you could effectively do this at insn output time?

The problem is you can end up with redundant compares.  You could
use the machine dependent reorg pass to try and clean up some of
the redundant comparisons.

Just thinking outloud...

jeff

      parent reply	other threads:[~1998-05-06 23:49 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1998-04-27  1:41 Richard Henderson
1998-04-27  8:47 ` Joern Rennecke
1998-04-27 13:59   ` Richard Henderson
1998-04-28 22:14 ` Jim Wilson
1998-04-29  2:20   ` Richard Henderson
1998-04-29 12:27     ` Jim Wilson
1998-04-29 16:08   ` Joern Rennecke
1998-04-29 22:57     ` Richard Henderson
1998-04-30 10:25       ` Joern Rennecke
1998-04-30 13:20         ` Richard Henderson
1998-04-30 22:29       ` Richard Henderson
1998-05-08 16:08         ` Jeffrey A Law
1998-05-09  2:05           ` Richard Henderson
1998-05-06 23:49 ` Jeffrey A Law [this message]

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=24630.894510594@hurl.cygnus.com \
    --to=law@cygnus.com \
    --cc=egcs@cygnus.com \
    --cc=rth@cygnus.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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).