From: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
To: GCC Patches <gcc-patches@gcc.gnu.org>,
"richard.sandiford@linaro.org" <richard.sandiford@linaro.org>,
Jackson Woodruff <jackson.woodruff@foss.arm.com>
Cc: nd <nd@arm.com>, James Greenhalgh <James.Greenhalgh@arm.com>,
"Richard Earnshaw" <Richard.Earnshaw@arm.com>
Subject: Re: [AArch64, PATCH] Improve Neon store of zero
Date: Wed, 23 Aug 2017 15:06:00 -0000 [thread overview]
Message-ID: <DB6PR0801MB2053AC0437D34EA5FE8D31CE83850@DB6PR0801MB2053.eurprd08.prod.outlook.com> (raw)
Richard Sandiford wrote:
>
> Sorry for only noticing now, but the call to aarch64_legitimate_address_p
> is asking whether the MEM itself is a legitimate LDP/STP address. Also,
> it might be better to pass false for strict_p, since this can be called
> before RA. So maybe:
>
> if (GET_CODE (operands[0]) == MEM
> && !(aarch64_simd_imm_zero (operands[1], <MODE>mode)
> && aarch64_mem_pair_operand (operands[0], <MODE>mode)))
Is there any reason for doing this check at all (or at least this early during
expand)?
There is a similar issue with this part:
(define_insn "*aarch64_simd_mov<mode>"
[(set (match_operand:VQ 0 "nonimmediate_operand"
- "=w, m, w, ?r, ?w, ?r, w")
+ "=w, Ump, m, w, ?r, ?w, ?r, w")
The Ump causes the instruction to always split off the address offset. Ump
cannot be used in patterns that are generated before register allocation as it
also calls laarch64_legitimate_address_p with strict_p set to true.
Wilco
next reply other threads:[~2017-08-23 14:35 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-23 15:06 Wilco Dijkstra [this message]
2017-09-06 9:03 ` Jackson Woodruff
2017-09-12 16:28 ` James Greenhalgh
2017-09-13 16:35 ` Jackson Woodruff
2017-09-13 16:51 ` James Greenhalgh
-- strict thread matches above, loose matches on Subject: below --
2017-08-10 13:38 Jackson Woodruff
2017-08-11 15:16 ` Richard Earnshaw (lists)
2017-08-16 16:01 ` Jackson Woodruff
2017-08-17 13:56 ` Richard Earnshaw (lists)
2017-08-23 14:46 ` Richard Sandiford
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=DB6PR0801MB2053AC0437D34EA5FE8D31CE83850@DB6PR0801MB2053.eurprd08.prod.outlook.com \
--to=wilco.dijkstra@arm.com \
--cc=James.Greenhalgh@arm.com \
--cc=Richard.Earnshaw@arm.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=jackson.woodruff@foss.arm.com \
--cc=nd@arm.com \
--cc=richard.sandiford@linaro.org \
/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).