qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: Paolo Savini <paolo.savini@embecosm.com>,
	qemu-devel@nongnu.org, qemu-riscv@nongnu.org
Cc: Palmer Dabbelt <palmer@dabbelt.com>,
	Alistair Francis <alistair.francis@wdc.com>,
	Bin Meng <bmeng.cn@gmail.com>, Weiwei Li <liwei1518@gmail.com>,
	Daniel Henrique Barboza <dbarboza@ventanamicro.com>,
	Liu Zhiwei <zhiwei_liu@linux.alibaba.com>,
	Helene Chelin <helene.chelin@embecosm.com>,
	Max Chou <max.chou@sifive.com>
Subject: Re: [RFC 2/2] target/riscv: rvv: improve performance of RISC-V vector loads and stores on large amounts of data.
Date: Sat, 27 Jul 2024 17:15:55 +1000	[thread overview]
Message-ID: <8e0c2afd-4c31-47f8-ade9-60a83ca20859@linaro.org> (raw)
In-Reply-To: <20240717153040.11073-3-paolo.savini@embecosm.com>

On 7/18/24 01:30, Paolo Savini wrote:
> This patch optimizes the emulation of unit-stride load/store RVV instructions
> when the data being loaded/stored per iteration amounts to 64 bytes or more.
> The optimization consists of calling __builtin_memcpy on chunks of data of 128
> and 256 bytes between the memory address of the simulated vector register and
> the destination memory address and vice versa.
> This is done only if we have direct access to the RAM of the host machine.
> 
> Signed-off-by: Paolo Savini <paolo.savini@embecosm.com>
> ---
>   target/riscv/vector_helper.c | 17 ++++++++++++++++-
>   1 file changed, 16 insertions(+), 1 deletion(-)
> 
> diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
> index 4b444c6bc5..7674972784 100644
> --- a/target/riscv/vector_helper.c
> +++ b/target/riscv/vector_helper.c
> @@ -486,7 +486,22 @@ vext_group_ldst_host(CPURISCVState *env, void *vd, uint32_t byte_end,
>       }
>   
>       fn = fns[is_load][group_size];
> -    fn(vd, byte_offset, host + byte_offset);
> +
> +    if (byte_offset + 32 < byte_end) {
> +      group_size = MO_256;
> +      if (is_load)
> +        __builtin_memcpy((uint8_t *)(vd + byte_offset), (uint8_t *)(host + byte_offset), 32);
> +      else
> +        __builtin_memcpy((uint8_t *)(host + byte_offset), (uint8_t *)(vd + byte_offset), 32);
> +    } else if (byte_offset + 16 < byte_end) {
> +      group_size = MO_128;
> +      if (is_load)
> +        __builtin_memcpy((uint8_t *)(vd + byte_offset), (uint8_t *)(host + byte_offset), 16);
> +      else
> +        __builtin_memcpy((uint8_t *)(host + byte_offset), (uint8_t *)(vd + byte_offset), 16);
> +    } else {
> +      fn(vd, byte_offset, host + byte_offset);
> +    }
>   

This will not work for big-endian hosts.

This may have atomicity issues, depending on the spec, the compiler options, and the host 
capabilities.


r~



  parent reply	other threads:[~2024-07-27  7:16 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-17 15:30 [RFC 0/2] Improve the performance of unit-stride RVV ld/st on Paolo Savini
2024-07-17 15:30 ` [RFC 1/2] target/riscv: rvv: reduce the overhead for simple RISC-V vector unit-stride loads and stores Paolo Savini
2024-07-26 12:22   ` Daniel Henrique Barboza
2024-07-27  7:13   ` Richard Henderson
2024-07-31 12:38     ` Daniel Henrique Barboza
2024-07-17 15:30 ` [RFC 2/2] target/riscv: rvv: improve performance of RISC-V vector loads and stores on large amounts of data Paolo Savini
2024-07-26 12:27   ` Daniel Henrique Barboza
2024-07-27  7:15   ` Richard Henderson [this message]
2024-09-10 11:20     ` Paolo Savini
2024-09-10 18:18       ` Richard Henderson
2024-07-26 12:31 ` [RFC 0/2] Improve the performance of unit-stride RVV ld/st on Daniel Henrique Barboza

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=8e0c2afd-4c31-47f8-ade9-60a83ca20859@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=alistair.francis@wdc.com \
    --cc=bmeng.cn@gmail.com \
    --cc=dbarboza@ventanamicro.com \
    --cc=helene.chelin@embecosm.com \
    --cc=liwei1518@gmail.com \
    --cc=max.chou@sifive.com \
    --cc=palmer@dabbelt.com \
    --cc=paolo.savini@embecosm.com \
    --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 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 NNTP newsgroup(s).