public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Peter Bergner <bergner@linux.ibm.com>
To: acsawdey@linux.ibm.com, gcc-patches@gcc.gnu.org
Cc: segher@kernel.crashing.org, wschmidt@linux.ibm.com
Subject: Re: [PATCH,rs6000] Make MMA builtins use opaque modes [v2]
Date: Thu, 19 Nov 2020 14:08:16 -0600	[thread overview]
Message-ID: <1f1e7da8-597e-ab82-97e5-61e374dc0f59@linux.ibm.com> (raw)
In-Reply-To: <20201119185847.703536-1-acsawdey@linux.ibm.com>

On 11/19/20 12:58 PM, acsawdey@linux.ibm.com wrote:
> +(define_expand "mma_disassemble_pair"
> +  [(match_operand:V16QI 0 "mma_disassemble_output_operand")
> +   (match_operand:OO 1 "input_operand")
> +   (match_operand 2 "const_0_to_1_operand")]

Maybe we should use vsx_register_operand instead of input_operand here?



> +(define_insn_and_split "*mma_disassemble_pair"
> +  [(set (match_operand:V16QI 0 "mma_disassemble_output_operand" "=mwa")
> +       (unspec:V16QI [(match_operand:OO 1 "input_operand" "wa")
> +                      (match_operand 2 "const_0_to_1_operand")]
> +		      UNSPEC_MMA_EXTRACT))]

Likewise?



> +  "TARGET_MMA
> +   && fpr_reg_operand (operands[1], OOmode)"

pairs can be assigned to any vsx register, so I think we want
vsx_register_operand here too.




> +(define_expand "mma_disassemble_acc"
> +  [(match_operand:V16QI 0 "mma_disassemble_output_operand")
> +   (match_operand:XO 1 "input_operand")
> +   (match_operand 2 "const_0_to_3_operand")]

Likewise as above, do we want to use the fpr_reg_operand predicate here
instead of input_operand?



> +(define_insn_and_split "*mma_disassemble_acc"
> +  [(set (match_operand:V16QI 0 "mma_disassemble_output_operand" "=mwa")
> +       (unspec:V16QI [(match_operand:XO 1 "input_operand" "d")
> +                      (match_operand 2 "const_0_to_3_operand")]

Likewise?


Peter



  reply	other threads:[~2020-11-19 20:08 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-17 17:48 [PATCH,rs6000] Make MMA builtins use opaque modes acsawdey
2020-11-17 18:41 ` Peter Bergner
2020-11-17 23:01   ` Segher Boessenkool
2020-11-17 23:24     ` Peter Bergner
2020-11-17 22:42 ` Segher Boessenkool
2020-11-19 18:58   ` [PATCH,rs6000] Make MMA builtins use opaque modes [v2] acsawdey
2020-11-19 20:08     ` Peter Bergner [this message]
2020-11-20  1:02     ` Aaron Sawdey
2020-11-20 19:10     ` Segher Boessenkool

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=1f1e7da8-597e-ab82-97e5-61e374dc0f59@linux.ibm.com \
    --to=bergner@linux.ibm.com \
    --cc=acsawdey@linux.ibm.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=segher@kernel.crashing.org \
    --cc=wschmidt@linux.ibm.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).