public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jeff Law <law@redhat.com>
To: Richard Biener <richard.guenther@gmail.com>
Cc: Jakub Jelinek <jakub@redhat.com>,
	       Tamar Christina <Tamar.Christina@arm.com>,
	       GCC Patches <gcc-patches@gcc.gnu.org>,
	       "rguenther@suse.de" <rguenther@suse.de>, nd <nd@arm.com>
Subject: Re: [PATCH] Optimise the fpclassify builtin to perform integer operations when possible
Date: Thu, 15 Sep 2016 16:02:00 -0000	[thread overview]
Message-ID: <134f0dc5-f4ed-c5a3-d89f-6d79ba9d93f6@redhat.com> (raw)
In-Reply-To: <CAFiYyc0wTmUZJA0n-8K9muCujKSX7og919O2A4Da9XLY+bq2VA@mail.gmail.com>

On 09/14/2016 02:24 AM, Richard Biener wrote:
> On Tue, Sep 13, 2016 at 6:15 PM, Jeff Law <law@redhat.com> wrote:
>> On 09/13/2016 02:41 AM, Jakub Jelinek wrote:
>>>
>>> On Mon, Sep 12, 2016 at 04:19:32PM +0000, Tamar Christina wrote:
>>>>
>>>> This patch adds an optimized route to the fpclassify builtin
>>>> for floating point numbers which are similar to IEEE-754 in format.
>>>>
>>>> The goal is to make it faster by:
>>>> 1. Trying to determine the most common case first
>>>>    (e.g. the float is a Normal number) and then the
>>>>    rest. The amount of code generated at -O2 are
>>>>    about the same +/- 1 instruction, but the code
>>>>    is much better.
>>>> 2. Using integer operation in the optimized path.
>>>
>>>
>>> Is it generally preferable to use integer operations for this instead
>>> of floating point operations?  I mean various targets have quite high
>>> costs
>>> of moving data in between the general purpose and floating point register
>>> file, often it has to go through memory etc.
>>
>> Bit testing/twiddling is obviously a trade-off for a non-addressable object.
>> I don't think there's any reasonable way to always generate the most
>> efficient code as it's going to depend on (for example) register allocation
>> behavior.
>>
>> So what we're stuck doing is relying on the target costing bits to guide
>> this kind of thing.
>
> I think the reason for this patch is to provide a general optimized
> integer version.
And just to be clear, that's fine with me.  While there are cases where 
bit twiddling hurts, I think bit twiddling is generally better.


> I think it asks for a FP (class) propagation pass somewhere (maybe as part of
> complex lowering which already has a similar "coarse" lattice -- not that I like
> its implementation very much) and doing the "lowering" there.
Not a bad idea -- I wonder how much a coarse tracking of the exceptional 
cases would allow later optimization.

>
> Not something that should block this patch though.
Agreed.

jeff

  reply	other threads:[~2016-09-15 15:52 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-12 16:21 Tamar Christina
2016-09-12 22:33 ` Joseph Myers
2016-09-13 12:25   ` Tamar Christina
2016-09-12 22:41 ` Joseph Myers
2016-09-13 12:30   ` Tamar Christina
2016-09-13 12:44     ` Joseph Myers
2016-09-15  9:08       ` Tamar Christina
2016-09-15 11:21         ` Wilco Dijkstra
2016-09-15 12:56           ` Joseph Myers
2016-09-15 13:05         ` Joseph Myers
2016-09-12 22:49 ` Joseph Myers
2016-09-13 12:33   ` Tamar Christina
2016-09-13 12:48     ` Joseph Myers
2016-09-13  8:58 ` Jakub Jelinek
2016-09-13 16:16   ` Jeff Law
2016-09-14  8:31     ` Richard Biener
2016-09-15 16:02       ` Jeff Law [this message]
2016-09-15 16:28         ` Richard Biener
2016-09-16 19:53 ` Jeff Law
2016-09-20 12:14   ` Tamar Christina
2016-09-20 14:52     ` Jeff Law
2016-09-20 17:52       ` Joseph Myers
2016-09-21  7:13       ` Richard Biener
2016-09-19 22:43 ` Michael Meissner
     [not found]   ` <41217f33-3861-dbb8-2f11-950ab30a7021@arm.com>
2016-09-20 21:27     ` Michael Meissner
2016-09-21  2:05       ` Joseph Myers
2016-09-21  8:32         ` Richard Biener
2016-09-12 17:24 Moritz Klammler
2016-09-12 20:08 ` Andrew Pinski
2016-09-13 12:16 Wilco Dijkstra
2016-09-13 16:10 ` Joseph Myers
2016-09-21 14:51 ` Richard Earnshaw (lists)

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=134f0dc5-f4ed-c5a3-d89f-6d79ba9d93f6@redhat.com \
    --to=law@redhat.com \
    --cc=Tamar.Christina@arm.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=jakub@redhat.com \
    --cc=nd@arm.com \
    --cc=rguenther@suse.de \
    --cc=richard.guenther@gmail.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).