All of lore.kernel.org
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: LIU Zhiwei <zhiwei_liu@c-sky.com>,
	qemu-devel@nongnu.org, qemu-riscv@nongnu.org
Cc: palmer@dabbelt.com, bin.meng@windriver.com, Alistair.Francis@wdc.com
Subject: Re: [RFC PATCH 02/13] target/riscv: Support UXL32 for branch instructions
Date: Thu, 5 Aug 2021 09:06:16 -1000	[thread overview]
Message-ID: <840d76cc-fd1c-6324-19cc-a6ec0075d032@linaro.org> (raw)
In-Reply-To: <20210805025312.15720-3-zhiwei_liu@c-sky.com>

On 8/4/21 4:53 PM, LIU Zhiwei wrote:
> +static TCGv gpr_src_u(DisasContext *ctx, int reg_num)
> +{
> +    if (reg_num == 0) {
> +        return ctx->zero;
> +    }
> +    if (ctx->uxl32) {
> +        tcg_gen_ext32u_tl(cpu_gpr[reg_num], cpu_gpr[reg_num]);
> +    }
> +    return cpu_gpr[reg_num];
> +}
> +
> +static TCGv gpr_src_s(DisasContext *ctx, int reg_num)
> +{
> +    if (reg_num == 0) {
> +        return ctx->zero;
> +    }
> +    if (ctx->uxl32) {
> +        tcg_gen_ext32s_tl(cpu_gpr[reg_num], cpu_gpr[reg_num]);
> +    }
> +    return cpu_gpr[reg_num];
> +}

This is bad: you cannot modify the source registers like this.

These incorrect modifications will be visible to the kernel on transition back to S-mode.

> 
> +static bool gen_branch_u(DisasContext *ctx, arg_b *a, TCGCond cond)
> +{
> +    TCGv src1 = gpr_src_u(ctx, a->rs1);
> +    TCGv src2 = gpr_src_u(ctx, a->rs2);
> +
> +    return gen_branch_internal(ctx, a, cond, src1, src2);
> +}

This is unnecessary.  Unsigned comparisons work just fine with sign-extended values.  It 
will be simpler to keep all values sign-extended.


r~


WARNING: multiple messages have this Message-ID (diff)
From: Richard Henderson <richard.henderson@linaro.org>
To: LIU Zhiwei <zhiwei_liu@c-sky.com>,
	qemu-devel@nongnu.org, qemu-riscv@nongnu.org
Cc: Alistair.Francis@wdc.com, palmer@dabbelt.com, bin.meng@windriver.com
Subject: Re: [RFC PATCH 02/13] target/riscv: Support UXL32 for branch instructions
Date: Thu, 5 Aug 2021 09:06:16 -1000	[thread overview]
Message-ID: <840d76cc-fd1c-6324-19cc-a6ec0075d032@linaro.org> (raw)
In-Reply-To: <20210805025312.15720-3-zhiwei_liu@c-sky.com>

On 8/4/21 4:53 PM, LIU Zhiwei wrote:
> +static TCGv gpr_src_u(DisasContext *ctx, int reg_num)
> +{
> +    if (reg_num == 0) {
> +        return ctx->zero;
> +    }
> +    if (ctx->uxl32) {
> +        tcg_gen_ext32u_tl(cpu_gpr[reg_num], cpu_gpr[reg_num]);
> +    }
> +    return cpu_gpr[reg_num];
> +}
> +
> +static TCGv gpr_src_s(DisasContext *ctx, int reg_num)
> +{
> +    if (reg_num == 0) {
> +        return ctx->zero;
> +    }
> +    if (ctx->uxl32) {
> +        tcg_gen_ext32s_tl(cpu_gpr[reg_num], cpu_gpr[reg_num]);
> +    }
> +    return cpu_gpr[reg_num];
> +}

This is bad: you cannot modify the source registers like this.

These incorrect modifications will be visible to the kernel on transition back to S-mode.

> 
> +static bool gen_branch_u(DisasContext *ctx, arg_b *a, TCGCond cond)
> +{
> +    TCGv src1 = gpr_src_u(ctx, a->rs1);
> +    TCGv src2 = gpr_src_u(ctx, a->rs2);
> +
> +    return gen_branch_internal(ctx, a, cond, src1, src2);
> +}

This is unnecessary.  Unsigned comparisons work just fine with sign-extended values.  It 
will be simpler to keep all values sign-extended.


r~


  reply	other threads:[~2021-08-05 19:07 UTC|newest]

Thread overview: 82+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-05  2:52 [RFC PATCH 00/13] Support UXL field in mstatus LIU Zhiwei
2021-08-05  2:52 ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 01/13] target/riscv: Add UXL to tb flags LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05  6:00   ` Alistair Francis
2021-08-05  6:00     ` Alistair Francis
2021-08-05 19:01   ` Richard Henderson
2021-08-05 19:01     ` Richard Henderson
2021-08-06  2:49     ` LIU Zhiwei
2021-08-06  2:49       ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 02/13] target/riscv: Support UXL32 for branch instructions LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05 19:06   ` Richard Henderson [this message]
2021-08-05 19:06     ` Richard Henderson
2021-08-09  1:45     ` LIU Zhiwei
2021-08-09  1:45       ` LIU Zhiwei
2021-08-09 19:34       ` Richard Henderson
2021-08-09 19:34         ` Richard Henderson
2021-08-11 14:57         ` LIU Zhiwei
2021-08-11 14:57           ` LIU Zhiwei
2021-08-11 17:56           ` Richard Henderson
2021-08-11 17:56             ` Richard Henderson
2021-08-11 22:40             ` LIU Zhiwei
2021-08-11 22:40               ` LIU Zhiwei
2021-08-12  4:42               ` Richard Henderson
2021-08-12  4:42                 ` Richard Henderson
2021-08-12  5:03                 ` LIU Zhiwei
2021-08-12  5:03                   ` LIU Zhiwei
2021-08-12  6:12                   ` Richard Henderson
2021-08-12  6:12                     ` Richard Henderson
2021-08-12  7:20                     ` LIU Zhiwei
2021-08-12  7:20                       ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 03/13] target/riscv: Support UXL32 on 64-bit cpu for load/store LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05 19:08   ` Richard Henderson
2021-08-05 19:08     ` Richard Henderson
2021-08-09  1:50     ` LIU Zhiwei
2021-08-09  1:50       ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 04/13] target/riscv: Support UXL32 for slit/sltiu LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05 19:09   ` Richard Henderson
2021-08-05 19:09     ` Richard Henderson
2021-08-09  7:28     ` LIU Zhiwei
2021-08-09  7:28       ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 05/13] target/riscv: Support UXL32 for shift instruction LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05 22:17   ` Richard Henderson
2021-08-05 22:17     ` Richard Henderson
2021-08-09  7:51     ` LIU Zhiwei
2021-08-09  7:51       ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 06/13] target/riscv: Fix div instructions LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05 22:18   ` Richard Henderson
2021-08-05 22:18     ` Richard Henderson
2021-08-09  7:53     ` LIU Zhiwei
2021-08-09  7:53       ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 07/13] target/riscv: Support UXL32 for RVM LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 08/13] target/riscv: Support UXL32 for vector instructions LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 09/13] target/riscv: Support UXL32 for atomic instructions LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 10/13] target/riscv: Support UXL32 for float instructions LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 11/13] target/riscv: Fix srow LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 12/13] target/riscv: Support UXL32 for RVB LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05  2:53 ` [RFC PATCH 13/13] target/riscv: Changing the width of U-mode CSR LIU Zhiwei
2021-08-05  2:53   ` LIU Zhiwei
2021-08-05  6:01 ` [RFC PATCH 00/13] Support UXL field in mstatus Alistair Francis
2021-08-05  6:01   ` Alistair Francis
2021-08-05  7:14   ` LIU Zhiwei
2021-08-05  7:14     ` LIU Zhiwei
2021-08-05  7:20     ` Bin Meng
2021-08-05  7:20       ` Bin Meng
2021-08-05  8:10       ` LIU Zhiwei
2021-08-05  8:10         ` LIU Zhiwei
2021-08-06 10:05     ` Alistair Francis
2021-08-06 10:05       ` Alistair Francis
2021-08-09  1:25       ` LIU Zhiwei
2021-08-09  1:25         ` 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=840d76cc-fd1c-6324-19cc-a6ec0075d032@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=Alistair.Francis@wdc.com \
    --cc=bin.meng@windriver.com \
    --cc=palmer@dabbelt.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-riscv@nongnu.org \
    --cc=zhiwei_liu@c-sky.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.