qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>, qemu-devel@nongnu.org
Cc: qemu-riscv@nongnu.org, palmer@dabbelt.com,
	alistair.francis@wdc.com, dbarboza@ventanamicro.com,
	liwei1518@gmail.com, bmeng.cn@gmail.com,
	TANG Tiancheng <tangtiancheng.ttc@alibaba-inc.com>
Subject: Re: [PATCH v3 08/14] tcg/riscv: Implement vector cmp ops
Date: Thu, 5 Sep 2024 00:12:28 -0700	[thread overview]
Message-ID: <9d40d3f4-731d-478f-81ba-1c380ef9d0ae@linaro.org> (raw)
In-Reply-To: <20240904142739.854-9-zhiwei_liu@linux.alibaba.com>

On 9/4/24 07:27, LIU Zhiwei wrote:
> @@ -2322,6 +2411,51 @@ static void tcg_out_vec_op(TCGContext *s, TCGOpcode opc,
>           riscv_set_vec_config_vl(s, type);
>           tcg_out_opc_vi(s, OPC_VXOR_VI, a0, a1, -1, true);
>           break;
> +    case INDEX_op_cmpsel_vec:
> +        TCGArg a3, a4;
> +        int c3, c4;
> +        TCGCond cond;

While I suppose this compiles, it's not great to have new variables added randomly within 
a switch.  At minimum, add { } around the block, but consider breaking out a separate 
tcg_out_cmpsel function, akin to tcg_out_movcond et al.

> @@ -2332,10 +2466,27 @@ static void tcg_out_vec_op(TCGContext *s, TCGOpcode opc,
>   void tcg_expand_vec_op(TCGOpcode opc, TCGType type, unsigned vece,
>                          TCGArg a0, ...)
>   {
> +    va_list va;
> +    TCGv_vec v0, v1;
> +    TCGArg a2, a3;
> +
> +    va_start(va, a0);
> +    v0 = temp_tcgv_vec(arg_temp(a0));
> +    v1 = temp_tcgv_vec(arg_temp(va_arg(va, TCGArg)));
> +    a2 = va_arg(va, TCGArg);
> +
>       switch (opc) {
> +    case INDEX_op_cmp_vec:
> +        a3 = va_arg(va, TCGArg);
> +        vec_gen_6(INDEX_op_cmpsel_vec, type, vece, tcgv_vec_arg(v0),
> +                    tcgv_vec_arg(v1), a2,
> +                    tcgv_i64_arg(tcg_constant_i64(-1)),
> +                    tcgv_i64_arg(tcg_constant_i64(0)), a3);
> +        break;
>       default:
>           g_assert_not_reached();
>       }
> +    va_end(va);
>   }

Better to use "TCGArg a0, a1".  Converting through arg_tmp + temp_tcgv_vec to v0/v1 and 
then undoing that with tcgv_vec_arg is confusing.


r~


  reply	other threads:[~2024-09-05  7:13 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-04 14:27 [PATCH v3 00/14] Add support for vector LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 01/14] tcg/op-gvec: Fix iteration step in 32-bit operation LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 02/14] util: Add RISC-V vector extension probe in cpuinfo LIU Zhiwei
2024-09-05  3:34   ` Richard Henderson
2024-09-09  7:18     ` LIU Zhiwei
2024-09-09 15:45       ` Richard Henderson
2024-09-10  2:47         ` LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 03/14] tcg/riscv: Add basic support for vector LIU Zhiwei
2024-09-05  4:05   ` Richard Henderson
2024-09-10  2:49     ` LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 04/14] tcg/riscv: Add riscv vset{i}vli support LIU Zhiwei
2024-09-05  6:03   ` Richard Henderson
2024-09-10  2:46     ` LIU Zhiwei
2024-09-10  4:34       ` Richard Henderson
2024-09-10  7:03         ` LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 05/14] tcg/riscv: Implement vector load/store LIU Zhiwei
2024-09-05  6:39   ` Richard Henderson
2024-09-10  3:04     ` LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 06/14] tcg/riscv: Implement vector mov/dup{m/i} LIU Zhiwei
2024-09-05  6:56   ` Richard Henderson
2024-09-10  1:13     ` LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 07/14] tcg/riscv: Add support for basic vector opcodes LIU Zhiwei
2024-09-05  6:57   ` Richard Henderson
2024-09-04 14:27 ` [PATCH v3 08/14] tcg/riscv: Implement vector cmp ops LIU Zhiwei
2024-09-05  7:12   ` Richard Henderson [this message]
2024-09-10  1:17     ` LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 09/14] tcg/riscv: Implement vector neg ops LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 10/14] tcg/riscv: Implement vector sat/mul ops LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 11/14] tcg/riscv: Implement vector min/max ops LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 12/14] tcg/riscv: Implement vector shs/v ops LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 13/14] tcg/riscv: Implement vector roti/v/x shi ops LIU Zhiwei
2024-09-04 14:27 ` [PATCH v3 14/14] tcg/riscv: Enable native vector support for TCG host LIU Zhiwei
2024-09-05 23:46 ` [PATCH v3 00/14] Add support for vector Alistair Francis
2024-09-10  3:08   ` LIU Zhiwei

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=9d40d3f4-731d-478f-81ba-1c380ef9d0ae@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=alistair.francis@wdc.com \
    --cc=bmeng.cn@gmail.com \
    --cc=dbarboza@ventanamicro.com \
    --cc=liwei1518@gmail.com \
    --cc=palmer@dabbelt.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-riscv@nongnu.org \
    --cc=tangtiancheng.ttc@alibaba-inc.com \
    --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).