From: "Alex Bennée" <alex.bennee@linaro.org>
To: Joel Stanley <joel@jms.id.au>
Cc: Alistair Francis <alistair.francis@wdc.com>,
Daniel Henrique Barboza <daniel.barboza@oss.qualcomm.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Peter Maydell <peter.maydell@linaro.org>,
Weiwei Li <liwei1518@gmail.com>,
Liu Zhiwei <zhiwei_liu@linux.alibaba.com>,
Chao Liu <chao.liu@processmission.com>,
Nicholas Piggin <npiggin@gmail.com>,
Anirudh Srinivasan <asrinivasan@oss.tenstorrent.com>,
Michael Ellerman <mpe@kernel.org>,
Portia Stephens <portias@oss.tenstorrent.com>,
qemu-riscv@nongnu.org, qemu-devel@nongnu.org,
qemu-arm@nongnu.org
Subject: Re: [PATCH 09/12] tests/tcg/multiarch: Fix vectorisation of memory test
Date: Tue, 01 Sep 2026 11:06:23 +0100 [thread overview]
Message-ID: <87cxuxwajk.fsf@draig.linaro.org> (raw)
In-Reply-To: <20260827050953.67184-10-joel@jms.id.au> (Joel Stanley's message of "Thu, 27 Aug 2026 14:39:48 +0930")
Joel Stanley <joel@jms.id.au> writes:
> Compilers are not able to vectorise the load and store loops. Building
> with -fopt-info-vec-missed with gcc 16.1:
>
> missed: statement clobbers memory: ml_printf (".");
>
> This can be fixed by doing the printf after the loops.
>
> Also, autovectorisation doesn't like more than one if statement in the
> loop:
>
> missed: not vectorized: unsupported control flow in loop.
>
> By combining them into ternary statements and one if, the loops are able
> to be vectorised with both loads and stores.
>
> The trade off is the dots are gone, and the tests no longer report fine
> grained miscompares. A failure does report the number of bad elements in
> a failing test loop.
>
> Signed-off-by: Joel Stanley <joel@jms.id.au>
Acked-by: Alex Bennée <alex.bennee@linaro.org>
--
Alex Bennée
Virtualisation Tech Lead @ Linaro
next prev parent reply other threads:[~2026-09-01 10:07 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-27 5:09 [PATCH 00/12] tests/tcg: Multiarch tests for riscv64 Joel Stanley
2026-08-27 5:09 ` [PATCH 01/12] tests/tcg/riscv64: Include stdint.h in semicall.h Joel Stanley
2026-08-31 1:07 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 02/12] tests/tcg/riscv64: Clean up semicall.h inline asm Joel Stanley
2026-08-31 1:09 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 03/12] tests/tcg/riscv64: Boot semihost tests with -bios none -kernel Joel Stanley
2026-08-31 1:10 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 04/12] tests/tcg/riscv64: Update semihosting linker script Joel Stanley
2026-08-31 1:14 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 05/12] tests/tcg/riscv64: Rework rules for semihost tests Joel Stanley
2026-08-31 1:17 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 06/12] tests/tcg/riscv64: Enable multiarch system tests Joel Stanley
2026-08-31 1:20 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 07/12] tests/tcg/riscv64: Enable memory record and replay test Joel Stanley
2026-08-31 1:21 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 08/12] tests/tcg/aarch64: Don't leak memory-sve CFLAGS Joel Stanley
2026-09-22 5:39 ` Alistair Francis
2026-09-22 7:17 ` Alex Bennée
2026-08-27 5:09 ` [PATCH 09/12] tests/tcg/multiarch: Fix vectorisation of memory test Joel Stanley
2026-08-31 1:24 ` Alistair Francis
2026-09-01 10:06 ` Alex Bennée [this message]
2026-08-27 5:09 ` [PATCH 10/12] tests/tcg/riscv64: Enable vectorised " Joel Stanley
2026-08-31 1:25 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 11/12] configure: Provide OBJDUMP to the tcg tests Joel Stanley
2026-08-31 1:25 ` Alistair Francis
2026-08-27 5:09 ` [PATCH 12/12] tests/tcg: Check vector memory tests contain vector ops Joel Stanley
2026-08-31 1:26 ` Alistair Francis
2026-09-22 9:50 ` [PATCH 00/12] tests/tcg: Multiarch tests for riscv64 Alistair Francis
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=87cxuxwajk.fsf@draig.linaro.org \
--to=alex.bennee@linaro.org \
--cc=alistair.francis@wdc.com \
--cc=asrinivasan@oss.tenstorrent.com \
--cc=chao.liu@processmission.com \
--cc=daniel.barboza@oss.qualcomm.com \
--cc=joel@jms.id.au \
--cc=liwei1518@gmail.com \
--cc=mpe@kernel.org \
--cc=npiggin@gmail.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=portias@oss.tenstorrent.com \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-riscv@nongnu.org \
--cc=zhiwei_liu@linux.alibaba.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.