From: Tom Musta <tommusta@gmail.com>
To: Richard Henderson <rth@twiddle.net>,
QEMU Developers <qemu-devel@nongnu.org>
Cc: "qemu-ppc@nongnu.org" <qemu-ppc@nongnu.org>
Subject: Re: [Qemu-devel] [PATCH 15/19] Add VSX xmax/xmin Instructions
Date: Thu, 24 Oct 2013 16:07:52 -0500 [thread overview]
Message-ID: <52698C28.6040001@gmail.com> (raw)
In-Reply-To: <526986CF.8090503@twiddle.net>
On 10/24/2013 3:45 PM, Richard Henderson wrote:
> On 10/24/2013 09:26 AM, Tom Musta wrote:
>> Because of the Power ISA definitions of maximum and minimum
>> on various boundary cases, the standard softfloat comparison
>> routines (e.g. float64_lt) do not work as well as one might
>> think. Therefore specific routines for comparing 64 and 32
>> bit floating point numbers are implemented in the PowerPC
>> helper code.
>
> Really? All I see in the document is ">fp", used both here
> in the minmax insn and in the cmp insn.
>
> If the softfloat compare isn't good enough for minmax, how
> can it be good enough for cmp?
Example:
The ISA is very explicit that max(-0.0, +0.0) = +0.0.
But the comparison operations (and instructions) both consider
-0.0 == +0.0. Because of this, I do not see how it is possible
to implement max using float*_eq, float*_lt and float*_le.
See, for example, table 58 (Actions for xsmaxdp) on p. 369 of the
V2.06 ISA.
next prev parent reply other threads:[~2013-10-24 21:08 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-24 16:16 [Qemu-devel] [PATCH 00/19] PowerPC VSX Stage 3 Tom Musta
2013-10-24 16:17 ` [Qemu-devel] [PATCH 01/19] Add New softfloat Routines for VSX Tom Musta
2013-10-24 18:34 ` Richard Henderson
2013-10-25 11:34 ` Alex Bennée
2013-10-25 11:44 ` Peter Maydell
2013-10-25 13:09 ` Alex Bennée
2013-10-25 13:24 ` Tom Musta
2013-10-25 11:55 ` Peter Maydell
2013-10-25 13:01 ` Tom Musta
2013-10-25 13:37 ` Peter Maydell
2013-10-24 16:18 ` [Qemu-devel] [PATCH 02/19] Add set_fprf Argument to fload_invalid_op_excp() Tom Musta
2013-10-24 16:19 ` [Qemu-devel] [PATCH 03/19] General Support for VSX Helpers Tom Musta
2013-10-24 18:51 ` Richard Henderson
2013-10-24 20:42 ` Tom Musta
2013-10-24 21:00 ` Richard Henderson
2013-10-24 16:20 ` [Qemu-devel] [PATCH 04/19] Add VSX ISA2.06 xadd Instructions Tom Musta
2013-10-24 19:44 ` Richard Henderson
2013-10-24 16:20 ` [Qemu-devel] [PATCH 05/19] Add VSX ISA2.06 xsub Instructions Tom Musta
2013-10-24 19:48 ` Richard Henderson
2013-10-24 16:21 ` [Qemu-devel] [PATCH 06/19] Add VSX ISA2.06 xmul Instructions Tom Musta
2013-10-24 20:07 ` Richard Henderson
2013-10-24 16:21 ` [Qemu-devel] [PATCH 07/19] Add VSX ISA2.06 xdiv Instructions Tom Musta
2013-10-24 20:08 ` Richard Henderson
2013-10-24 16:22 ` [Qemu-devel] [PATCH 08/19] Add VSX ISA2.06 xre Instructions Tom Musta
2013-10-24 20:11 ` Richard Henderson
2013-10-24 16:22 ` [Qemu-devel] [PATCH 09/19] Add VSX ISA2.06 xsqrt Instructions Tom Musta
2013-10-24 20:23 ` Richard Henderson
2013-10-24 16:23 ` [Qemu-devel] [PATCH 10/19] Add VSX ISA2.06 xrsqrte Instructions Tom Musta
2013-10-24 20:25 ` Richard Henderson
2013-10-24 16:23 ` [Qemu-devel] [PATCH 11/19] Add VSX ISA2.06 xtdiv Instructions Tom Musta
2013-10-24 20:30 ` Richard Henderson
2013-10-24 16:24 ` [Qemu-devel] [PATCH 12/19] Add VSX ISA2.06 xtsqrt Instructions Tom Musta
2013-10-24 20:34 ` Richard Henderson
2013-10-24 16:25 ` [Qemu-devel] [PATCH 13/19] Add VSX ISA2.06 Multiply Add Instructions Tom Musta
2013-10-24 20:38 ` Richard Henderson
2013-10-25 13:49 ` Tom Musta
2013-10-25 16:25 ` Tom Musta
2013-10-25 16:42 ` Richard Henderson
2013-10-25 17:13 ` Tom Musta
2013-10-25 17:29 ` Richard Henderson
2013-10-25 17:20 ` Peter Maydell
2013-10-25 17:34 ` Richard Henderson
2013-10-24 16:25 ` [Qemu-devel] [PATCH 14/19] Add VSX xscmp*dp Instructions Tom Musta
2013-10-24 20:39 ` Richard Henderson
2013-10-24 16:26 ` [Qemu-devel] [PATCH 15/19] Add VSX xmax/xmin Instructions Tom Musta
2013-10-24 20:45 ` Richard Henderson
2013-10-24 21:07 ` Tom Musta [this message]
2013-10-24 21:18 ` Richard Henderson
2013-10-24 22:10 ` Peter Maydell
2013-10-25 13:52 ` Tom Musta
2013-10-25 13:55 ` Peter Maydell
2013-10-24 16:26 ` [Qemu-devel] [PATCH 16/19] Add VSX Vector Compare Instructions Tom Musta
2013-10-24 16:27 ` [Qemu-devel] [PATCH 17/19] Add VSX Floating Point to Floating Point Conversion Instructions Tom Musta
2013-10-24 20:49 ` Richard Henderson
2013-10-24 16:27 ` [Qemu-devel] [PATCH 18/19] Add VSX ISA2.06 Integer " Tom Musta
2013-10-24 20:51 ` Richard Henderson
2013-10-24 16:28 ` [Qemu-devel] [PATCH 19/19] Add VSX Rounding Instructions Tom Musta
2013-10-24 20:54 ` Richard Henderson
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=52698C28.6040001@gmail.com \
--to=tommusta@gmail.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
--cc=rth@twiddle.net \
/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 NNTP newsgroup(s).