public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Peter Bergner <bergner@linux.ibm.com>
To: Qing Zhao <qing.zhao@oracle.com>
Cc: "Richard Biener" <richard.guenther@gmail.com>,
	"GCC Patches" <gcc-patches@gcc.gnu.org>,
	"Martin Liška" <mliska@suse.cz>
Subject: Re: [PATCH] middle-end: Skip initialization of opaque type register variables [PR103127]
Date: Tue, 30 Nov 2021 12:08:37 -0600	[thread overview]
Message-ID: <b8c447ee-dbef-63c9-e36b-979282ac279c@linux.ibm.com> (raw)
In-Reply-To: <D82F4736-B206-4DF2-A865-6227F41E193A@oracle.com>

On 11/30/21 11:51 AM, Qing Zhao wrote:
> So, looks like that the variable with OPAQUE_TYPE cannot be all explicitly initialized even at source code level. 
> 
> The only way to initialize such variable (only __vector_quad, not for __vector_pairs) at source code level is through call to __builtin_mma_xxsetaccz as:
> 
> void
> foo (__vector_quad *dst)
> {
>   __vector_quad acc;
>   __builtin_mma_xxsetaccz(&acc);
>   *dst = acc;
> }
> 
> Is this the correct understanding?

Correct.  Or via...


> Is there way to initialize such variable to other values than zero at source code level?

Not for any constant values.  You can load it from memory though like below,
which is also allowed for __vector_pair:

void
foo (__vector_quad *dst, __vector_quad *src)
{
  __vector_quad acc;
  acc = *src;
  ...
}
void
bar (__vector_pair *dst, __vector_pair *src)
{
  __vector_pair pair;
  pair = *src;
  ...
}

We do not accept things like:

  acc = 0;
  acc = {0, 0, ... };
  etc.

Peter

  reply	other threads:[~2021-11-30 18:08 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-29 21:56 Peter Bergner
2021-11-29 22:56 ` Qing Zhao
2021-11-30  8:37   ` Richard Biener
2021-11-30 15:14     ` Peter Bergner
2021-11-30 17:51       ` Qing Zhao
2021-11-30 18:08         ` Peter Bergner [this message]
2021-11-30 19:50           ` Qing Zhao
2021-11-30 20:07             ` Peter Bergner
2021-11-30 20:44               ` Qing Zhao
2021-11-30 22:35                 ` Peter Bergner
2021-12-01  9:01                   ` Richard Biener
2021-12-01 15:06                     ` Qing Zhao
2021-12-01 16:08                       ` Peter Bergner
2021-12-01 17:42                     ` Peter Bergner
2021-12-01 19:07                       ` Richard Biener
2021-12-01 19:29                         ` Peter Bergner
2021-11-30 17:59     ` Qing Zhao

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=b8c447ee-dbef-63c9-e36b-979282ac279c@linux.ibm.com \
    --to=bergner@linux.ibm.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=mliska@suse.cz \
    --cc=qing.zhao@oracle.com \
    --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).