[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v2 12/29] tcg/aarch64: Implement tcg_out_dupm_ve
From: |
Alex Bennée |
Subject: |
Re: [Qemu-devel] [PATCH v2 12/29] tcg/aarch64: Implement tcg_out_dupm_vec |
Date: |
Thu, 02 May 2019 14:26:26 +0100 |
User-agent: |
mu4e 1.3.1; emacs 26.1 |
Richard Henderson <address@hidden> writes:
> Signed-off-by: Richard Henderson <address@hidden>
> ---
> tcg/aarch64/tcg-target.inc.c | 38 ++++++++++++++++++++++++++++++++++--
> 1 file changed, 36 insertions(+), 2 deletions(-)
>
> diff --git a/tcg/aarch64/tcg-target.inc.c b/tcg/aarch64/tcg-target.inc.c
> index 4a3cfa778a..411fb463ac 100644
> --- a/tcg/aarch64/tcg-target.inc.c
> +++ b/tcg/aarch64/tcg-target.inc.c
> @@ -381,6 +381,9 @@ typedef enum {
> I3207_BLR = 0xd63f0000,
> I3207_RET = 0xd65f0000,
>
> + /* AdvSIMD load/store single structure. */
> + I3303_LD1R = 0x0d40c000,
> +
I can't recall where these magic numbers come from again? The (moving)
section numbers of the ARM ARM?
I was hoping the XML had a bit more guidance on the encoding names but
we get:
<iclass name="No offset" oneof="2" id="LD1R_asisdlso_R1" no_encodings="1"
isa="A64">
and
<iclass name="Post-index" oneof="2" id="as_post_index" no_encodings="2"
isa="A64">
Although the instruction does have:
<instructionsection id="LD1R_advsimd" title="LD1R -- A64" type="instruction">
> /* Load literal for loading the address at pc-relative offset */
> I3305_LDR = 0x58000000,
> I3305_LDR_v64 = 0x5c000000,
> @@ -414,6 +417,8 @@ typedef enum {
> I3312_LDRVQ = 0x3c000000 | 3 << 22 | 0 << 30,
> I3312_STRVQ = 0x3c000000 | 2 << 22 | 0 << 30,
>
> +
> +
> I3312_TO_I3310 = 0x00200800,
> I3312_TO_I3313 = 0x01000000,
>
> @@ -566,7 +571,14 @@ static inline uint32_t tcg_in32(TCGContext *s)
> #define tcg_out_insn(S, FMT, OP, ...) \
> glue(tcg_out_insn_,FMT)(S, glue(glue(glue(I,FMT),_),OP), ##
> __VA_ARGS__)
The above is basically a winge as to what do we really get out of this
"type checking"?
>
> -static void tcg_out_insn_3305(TCGContext *s, AArch64Insn insn, int imm19,
> TCGReg rt)
> +static void tcg_out_insn_3303(TCGContext *s, AArch64Insn insn, bool q,
> + TCGReg rt, TCGReg rn, unsigned size)
> +{
> + tcg_out32(s, insn | (rt & 0x1f) | (rn << 5) | (size << 10) | (q << 30));
> +}
> +
> +static void tcg_out_insn_3305(TCGContext *s, AArch64Insn insn,
> + int imm19, TCGReg rt)
> {
> tcg_out32(s, insn | (imm19 & 0x7ffff) << 5 | rt);
> }
> @@ -825,7 +837,29 @@ static bool tcg_out_dup_vec(TCGContext *s, TCGType type,
> unsigned vece,
> static bool tcg_out_dupm_vec(TCGContext *s, TCGType type, unsigned vece,
> TCGReg r, TCGReg base, intptr_t offset)
> {
> - return false;
> + if (offset != 0) {
> + AArch64Insn add_insn = I3401_ADDI;
> + TCGReg temp = TCG_REG_TMP;
> +
> + if (offset < 0) {
> + add_insn = I3401_SUBI;
> + offset = -offset;
> + }
> + if (offset <= 0xfff) {
> + tcg_out_insn_3401(s, add_insn, 1, temp, base, offset);
> + } else if (offset <= 0xffffff) {
> + tcg_out_insn_3401(s, add_insn, 1, temp, base, offset & 0xfff000);
> + if (offset & 0xfff) {
> + tcg_out_insn_3401(s, add_insn, 1, temp, base, offset &
> 0xfff);
> + }
> + } else {
> + tcg_out_movi(s, TCG_TYPE_PTR, temp, offset);
> + tcg_out_insn(s, 3502, ADD, 1, temp, temp, base);
> + }
> + base = temp;
> + }
> + tcg_out_insn(s, 3303, LD1R, type == TCG_TYPE_V128, r, base, vece);
> + return true;
> }
>
> static void tcg_out_movi(TCGContext *s, TCGType type, TCGReg rd,
--
Alex Bennée
- Re: [Qemu-devel] [PATCH v2 09/29] tcg: Manually expand INDEX_op_dup_vec, (continued)
- [Qemu-devel] [PATCH v2 10/29] tcg: Add tcg_out_dupm_vec to the backend interface, Richard Henderson, 2019/05/01
- [Qemu-devel] [PATCH v2 11/29] tcg/i386: Implement tcg_out_dupm_vec, Richard Henderson, 2019/05/01
- [Qemu-devel] [PATCH v2 13/29] tcg: Add INDEX_op_dup_mem_vec, Richard Henderson, 2019/05/01
- [Qemu-devel] [PATCH v2 16/29] tcg/aarch64: Support vector variable shift opcodes, Richard Henderson, 2019/05/01
- [Qemu-devel] [PATCH v2 12/29] tcg/aarch64: Implement tcg_out_dupm_vec, Richard Henderson, 2019/05/01
- Re: [Qemu-devel] [PATCH v2 12/29] tcg/aarch64: Implement tcg_out_dupm_vec,
Alex Bennée <=
- [Qemu-devel] [PATCH v2 15/29] tcg/i386: Support vector variable shift opcodes, Richard Henderson, 2019/05/01
- [Qemu-devel] [PATCH v2 14/29] tcg: Add gvec expanders for variable shift, Richard Henderson, 2019/05/01
- [Qemu-devel] [PATCH v2 17/29] tcg: Add gvec expanders for vector shift by scalar, Richard Henderson, 2019/05/01
- [Qemu-devel] [PATCH v2 18/29] tcg/i386: Support vector scalar shift opcodes, Richard Henderson, 2019/05/01
- [Qemu-devel] [PATCH v2 19/29] tcg: Add support for integer absolute value, Richard Henderson, 2019/05/01