public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Joseph Myers <joseph@codesourcery.com>
To: chenxiaolong <chenxiaolong@loongson.cn>
Cc: <gcc-patches@gcc.gnu.org>, <xry111@xry111.site>, <i@xen0n.name>,
	<xuchenghua@loongson.cn>, <chenglulu@loongson.cn>
Subject: Re: [PATCH v3] LoongArch:Implement 128-bit floating point functions in gcc.
Date: Wed, 16 Aug 2023 13:16:34 +0000	[thread overview]
Message-ID: <ee45b16b-b910-ce9d-427-14e4c76e5489@codesourcery.com> (raw)
In-Reply-To: <d75ca4a887abb545a9548a8189a8267bc2145912.camel@loongson.cn>

On Wed, 16 Aug 2023, chenxiaolong wrote:

> Thanks for the tip! Similar functions (e.g. __builtin_fabsf128
> (_Float128 a) are already supported by the compiler and can be handled
> correctly, but functions that can be implemented on the LoongArch
> architecture directly using the "bstrins" directive (e.g. fabsq,
> copysignq, etc.) are better optimized because they generate fewer
> assembly instructions. copysignq, etc.) on the LoongArch architecture
> are better optimized because they generate fewer assembly instructions.

Then you should make the existing built-in functions for _Float128 or long 
double generate the desired instructions, rather than adding a legacy and 
duplicative API to a new architecture.

-- 
Joseph S. Myers
joseph@codesourcery.com

  reply	other threads:[~2023-08-16 13:16 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-15 10:39 chenxiaolong
2023-08-15 10:48 ` Xi Ruoyao
2023-08-17  6:56   ` chenxiaolong
2023-08-15 20:03 ` Joseph Myers
2023-08-16 10:08   ` chenxiaolong
2023-08-16 13:16     ` Joseph Myers [this message]
2023-08-17  3:44   ` Xi Ruoyao
2023-08-17 15:08     ` Joseph Myers
2023-08-18  6:39       ` chenxiaolong
2023-08-18  6:58         ` Xi Ruoyao
2023-08-18  7:05           ` Xi Ruoyao
2023-08-18  7:19             ` Xi Ruoyao
2023-08-18  7:52               ` chenxiaolong

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=ee45b16b-b910-ce9d-427-14e4c76e5489@codesourcery.com \
    --to=joseph@codesourcery.com \
    --cc=chenglulu@loongson.cn \
    --cc=chenxiaolong@loongson.cn \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=i@xen0n.name \
    --cc=xry111@xry111.site \
    --cc=xuchenghua@loongson.cn \
    /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).