From: Craig Burley <burley@gnu.org>
To: jbuck@Synopsys.COM
Cc: burley@gnu.org
Subject: Re: FLOATING-POINT CONSISTENCY, -FFLOAT-STORE, AND X86
Date: Tue, 15 Dec 1998 00:04:00 -0000 [thread overview]
Message-ID: <199812150804.DAA02265@melange.gnu.org> (raw)
In-Reply-To: <199812141723.JAA02195@atrus.synopsys.com>
>What would the performance cost be if we spilled ix86 FP registers
>as 80 bits? Unfortunately I'm not familiar enough with the ix86
>instruction set to answer this question for myself. How difficult
>would it be to change gcc to do 80 bit spills, without changing
>the size of float and double?
I think the performance cost is likely to be pretty minor, mostly
not noticeable, since it doesn't affect any code that isn't already
spilling 80-bit values by chopping them down. (Though I guess
any code that makes certain uses of functions returning FP values
are likely to do such spills...but a bit of extra cache usage
around a procedure call isn't usually a noticeable hit against
performance, right?)
It's the difficulty of changing gcc that I think is the biggest
problem.
>It seems to me that it should cost a lot less than -ffloat-store
>and should get rid of the unpredictable behavior. (Yes, it still
>doesn't match IEEE unless more is done, but it stops weirdness like
>root-finding algorithms that fail to converge if you're unlucky).
What I've come to believe is that, without this change, -ffloat-store
is useless except for the lucky and the very, very careful (who
also accept comparatively poor performance). With this change,
-ffloat-store adds reasonable stability on top of reasonable
stability.
I don't feel comfortable saying that even the combination of this
change and -ffloat-store results in complete stability/predictability,
however, because I just don't know enough to say that, and I tend
to be pessimistic about such things.
tq vm, (burley)
next prev parent reply other threads:[~1998-12-15 0:04 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
1998-12-13 6:19 Stephen L Moshier
1998-12-13 10:49 ` Craig Burley
1998-12-13 15:18 ` Stephen L Moshier
1998-12-14 8:49 ` Craig Burley
1998-12-14 9:25 ` Joe Buck
1998-12-14 14:30 ` Edward Jason Riedy
1998-12-15 0:04 ` Craig Burley [this message]
-- strict thread matches above, loose matches on Subject: below --
1998-12-22 13:30 Toon Moene
1998-12-22 11:07 John Wehle
1998-12-21 23:30 N8TM
1998-12-19 15:17 Geert Bosch
1998-12-20 8:09 ` Toon Moene
1998-12-22 4:17 ` Dave Love
1998-12-19 14:26 N8TM
1998-12-19 14:23 N8TM
1998-12-20 13:51 ` Marc Lehmann
1998-12-20 13:52 ` Marc Lehmann
1998-12-19 13:00 N8TM
1998-12-19 9:05 N8TM
1998-12-19 12:39 ` Toon Moene
1998-12-19 14:42 ` Dave Love
1998-12-18 23:07 N8TM
1998-12-19 13:39 ` Marc Lehmann
1998-12-18 21:58 N8TM
1998-12-18 22:36 ` Richard Henderson
1998-12-19 13:41 ` Marc Lehmann
1998-12-17 1:43 N8TM
1998-12-17 12:35 ` Marc Lehmann
1998-12-18 12:14 ` Dave Love
1998-12-18 14:25 ` Gerald Pfeifer
1998-12-19 13:50 ` Dave Love
1998-12-18 18:37 ` Marc Lehmann
1998-12-19 14:03 ` Dave Love
1998-12-16 6:10 N8TM
1998-12-15 0:05 N8TM
1998-12-15 10:01 ` Joe Buck
1998-12-03 6:34 N8TM
1998-12-04 15:23 ` Craig Burley
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=199812150804.DAA02265@melange.gnu.org \
--to=burley@gnu.org \
--cc=jbuck@Synopsys.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).