diff mbox series

[07/38] target/riscv: SIMD 8-bit Shift Instructions

Message ID 20210212150256.885-8-zhiwei_liu@c-sky.com (mailing list archive)
State New, archived
Headers show
Series target/riscv: support packed extension v0.9.2 | expand

Commit Message

LIU Zhiwei Feb. 12, 2021, 3:02 p.m. UTC 
Signed-off-by: LIU Zhiwei <zhiwei_liu@c-sky.com>
---
 target/riscv/helper.h                   |   9 +++
 target/riscv/insn32.decode              |  17 ++++
 target/riscv/insn_trans/trans_rvp.c.inc |  16 ++++
 target/riscv/packed_helper.c            | 102 ++++++++++++++++++++++++
 4 files changed, 144 insertions(+)

Comments

Alistair Francis March 15, 2021, 9:27 p.m. UTC | #1 
On Fri, Feb 12, 2021 at 10:18 AM LIU Zhiwei <zhiwei_liu@c-sky.com> wrote:
>
> Signed-off-by: LIU Zhiwei <zhiwei_liu@c-sky.com>

Acked-by: Alistair Francis <alistair.francis@wdc.com>

Alistair

> ---
>  target/riscv/helper.h                   |   9 +++
>  target/riscv/insn32.decode              |  17 ++++
>  target/riscv/insn_trans/trans_rvp.c.inc |  16 ++++
>  target/riscv/packed_helper.c            | 102 ++++++++++++++++++++++++
>  4 files changed, 144 insertions(+)
>
> diff --git a/target/riscv/helper.h b/target/riscv/helper.h
> index 20bf400ac2..0ecd4d53f9 100644
> --- a/target/riscv/helper.h
> +++ b/target/riscv/helper.h
> @@ -1193,3 +1193,12 @@ DEF_HELPER_3(sll16, tl, env, tl, tl)
>  DEF_HELPER_3(ksll16, tl, env, tl, tl)
>  DEF_HELPER_3(kslra16, tl, env, tl, tl)
>  DEF_HELPER_3(kslra16_u, tl, env, tl, tl)
> +
> +DEF_HELPER_3(sra8, tl, env, tl, tl)
> +DEF_HELPER_3(sra8_u, tl, env, tl, tl)
> +DEF_HELPER_3(srl8, tl, env, tl, tl)
> +DEF_HELPER_3(srl8_u, tl, env, tl, tl)
> +DEF_HELPER_3(sll8, tl, env, tl, tl)
> +DEF_HELPER_3(ksll8, tl, env, tl, tl)
> +DEF_HELPER_3(kslra8, tl, env, tl, tl)
> +DEF_HELPER_3(kslra8_u, tl, env, tl, tl)
> diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
> index 6f053bfeb7..cc782fcde5 100644
> --- a/target/riscv/insn32.decode
> +++ b/target/riscv/insn32.decode
> @@ -24,6 +24,7 @@
>
>  %sh10    20:10
>  %sh4    20:4
> +%sh3    20:3
>  %csr    20:12
>  %rm     12:3
>  %nf     29:3                     !function=ex_plus_1
> @@ -61,6 +62,7 @@
>
>  @sh      ......  ...... .....  ... ..... ....... &shift  shamt=%sh10      %rs1 %rd
>  @sh4     ......  ...... .....  ... ..... ....... &shift  shamt=%sh4      %rs1 %rd
> +@sh3     ......  ...... .....  ... ..... ....... &shift  shamt=%sh3      %rs1 %rd
>  @csr     ............   .....  ... ..... .......               %csr     %rs1 %rd
>
>  @atom_ld ..... aq:1 rl:1 ..... ........ ..... ....... &atomic rs2=0     %rs1 %rd
> @@ -652,3 +654,18 @@ ksll16     0110010  ..... ..... 000 ..... 1111111 @r
>  kslli16    0111010  1.... ..... 000 ..... 1111111 @sh4
>  kslra16    0101011  ..... ..... 000 ..... 1111111 @r
>  kslra16_u  0110011  ..... ..... 000 ..... 1111111 @r
> +
> +sra8       0101100  ..... ..... 000 ..... 1111111 @r
> +sra8_u     0110100  ..... ..... 000 ..... 1111111 @r
> +srai8      0111100  00... ..... 000 ..... 1111111 @sh3
> +srai8_u    0111100  01... ..... 000 ..... 1111111 @sh3
> +srl8       0101101  ..... ..... 000 ..... 1111111 @r
> +srl8_u     0110101  ..... ..... 000 ..... 1111111 @r
> +srli8      0111101  00... ..... 000 ..... 1111111 @sh3
> +srli8_u    0111101  01... ..... 000 ..... 1111111 @sh3
> +sll8       0101110  ..... ..... 000 ..... 1111111 @r
> +slli8      0111110  00... ..... 000 ..... 1111111 @sh3
> +ksll8      0110110  ..... ..... 000 ..... 1111111 @r
> +kslli8     0111110  01... ..... 000 ..... 1111111 @sh3
> +kslra8     0101111  ..... ..... 000 ..... 1111111 @r
> +kslra8_u   0110111  ..... ..... 000 ..... 1111111 @r
> diff --git a/target/riscv/insn_trans/trans_rvp.c.inc b/target/riscv/insn_trans/trans_rvp.c.inc
> index 848edab7e5..12a64849eb 100644
> --- a/target/riscv/insn_trans/trans_rvp.c.inc
> +++ b/target/riscv/insn_trans/trans_rvp.c.inc
> @@ -353,3 +353,19 @@ GEN_RVP_SHIFTI(slli16, sll16, tcg_gen_vec_shl16i_i64);
>  GEN_RVP_SHIFTI(srai16_u, sra16_u, NULL);
>  GEN_RVP_SHIFTI(srli16_u, srl16_u, NULL);
>  GEN_RVP_SHIFTI(kslli16, ksll16, NULL);
> +
> +/* SIMD 8-bit Shift Instructions */
> +GEN_RVP_SHIFT(sra8, tcg_gen_gvec_sars, 0);
> +GEN_RVP_SHIFT(srl8, tcg_gen_gvec_shrs, 0);
> +GEN_RVP_SHIFT(sll8, tcg_gen_gvec_shls, 0);
> +GEN_RVP_R_OOL(sra8_u);
> +GEN_RVP_R_OOL(srl8_u);
> +GEN_RVP_R_OOL(ksll8);
> +GEN_RVP_R_OOL(kslra8);
> +GEN_RVP_R_OOL(kslra8_u);
> +GEN_RVP_SHIFTI(srai8, sra8, tcg_gen_vec_sar8i_i64);
> +GEN_RVP_SHIFTI(srli8, srl8, tcg_gen_vec_shr8i_i64);
> +GEN_RVP_SHIFTI(slli8, sll8, tcg_gen_vec_shl8i_i64);
> +GEN_RVP_SHIFTI(srai8_u, sra8_u, NULL);
> +GEN_RVP_SHIFTI(srli8_u, srl8_u, NULL);
> +GEN_RVP_SHIFTI(kslli8, ksll8, NULL);
> diff --git a/target/riscv/packed_helper.c b/target/riscv/packed_helper.c
> index 7e31c2fe46..ab9ebc472b 100644
> --- a/target/riscv/packed_helper.c
> +++ b/target/riscv/packed_helper.c
> @@ -529,3 +529,105 @@ static inline void do_kslra16_u(CPURISCVState *env, void *vd, void *va,
>  }
>
>  RVPR(kslra16_u, 1, 2);
> +
> +/* SIMD 8-bit Shift Instructions */
> +static inline void do_sra8(CPURISCVState *env, void *vd, void *va,
> +                           void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] = a[i] >> shift;
> +}
> +
> +RVPR(sra8, 1, 1);
> +
> +static inline void do_srl8(CPURISCVState *env, void *vd, void *va,
> +                           void *vb, uint8_t i)
> +{
> +    uint8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] = a[i] >> shift;
> +}
> +
> +RVPR(srl8, 1, 1);
> +
> +static inline void do_sll8(CPURISCVState *env, void *vd, void *va,
> +                           void *vb, uint8_t i)
> +{
> +    uint8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] = a[i] << shift;
> +}
> +
> +RVPR(sll8, 1, 1);
> +
> +static inline void do_sra8_u(CPURISCVState *env, void *vd, void *va,
> +                             void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] =  vssra8(env, 0, a[i], shift);
> +}
> +
> +RVPR(sra8_u, 1, 1);
> +
> +static inline void do_srl8_u(CPURISCVState *env, void *vd, void *va,
> +                             void *vb, uint8_t i)
> +{
> +    uint8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] =  vssrl8(env, 0, a[i], shift);
> +}
> +
> +RVPR(srl8_u, 1, 1);
> +
> +static inline void do_ksll8(CPURISCVState *env, void *vd, void *va,
> +                            void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va, result;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +
> +    result = a[i] << shift;
> +    if (shift > (clrsb32(a[i]) - 24)) {
> +        env->vxsat = 0x1;
> +        d[i] = (a[i] & INT8_MIN) ? INT8_MIN : INT8_MAX;
> +    } else {
> +        d[i] = result;
> +    }
> +}
> +
> +RVPR(ksll8, 1, 1);
> +
> +static inline void do_kslra8(CPURISCVState *env, void *vd, void *va,
> +                             void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va;
> +    int32_t shift = sextract32((*(uint32_t *)vb), 0, 4);
> +
> +    if (shift >= 0) {
> +        do_ksll8(env, vd, va, vb, i);
> +    } else {
> +        shift = -shift;
> +        shift = (shift == 8) ? 7 : shift;
> +        d[i] = a[i] >> shift;
> +    }
> +}
> +
> +RVPR(kslra8, 1, 1);
> +
> +static inline void do_kslra8_u(CPURISCVState *env, void *vd, void *va,
> +                               void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va;
> +    int32_t shift = sextract32((*(uint32_t *)vb), 0, 4);
> +
> +    if (shift >= 0) {
> +        do_ksll8(env, vd, va, vb, i);
> +    } else {
> +        shift = -shift;
> +        shift = (shift == 8) ? 7 : shift;
> +        d[i] =  vssra8(env, 0, a[i], shift);
> +    }
> +}
> +
> +RVPR(kslra8_u, 1, 1);
> --
> 2.17.1
>
Palmer Dabbelt May 24, 2021, 4:46 a.m. UTC | #2 
On Fri, 12 Feb 2021 07:02:25 PST (-0800), zhiwei_liu@c-sky.com wrote:
> Signed-off-by: LIU Zhiwei <zhiwei_liu@c-sky.com>

I know it's always kind of akward for this type of patches, but IIUC 
they're all supposed to have some sort of description.

> ---
>  target/riscv/helper.h                   |   9 +++
>  target/riscv/insn32.decode              |  17 ++++
>  target/riscv/insn_trans/trans_rvp.c.inc |  16 ++++
>  target/riscv/packed_helper.c            | 102 ++++++++++++++++++++++++
>  4 files changed, 144 insertions(+)
>
> diff --git a/target/riscv/helper.h b/target/riscv/helper.h
> index 20bf400ac2..0ecd4d53f9 100644
> --- a/target/riscv/helper.h
> +++ b/target/riscv/helper.h
> @@ -1193,3 +1193,12 @@ DEF_HELPER_3(sll16, tl, env, tl, tl)
>  DEF_HELPER_3(ksll16, tl, env, tl, tl)
>  DEF_HELPER_3(kslra16, tl, env, tl, tl)
>  DEF_HELPER_3(kslra16_u, tl, env, tl, tl)
> +
> +DEF_HELPER_3(sra8, tl, env, tl, tl)
> +DEF_HELPER_3(sra8_u, tl, env, tl, tl)
> +DEF_HELPER_3(srl8, tl, env, tl, tl)
> +DEF_HELPER_3(srl8_u, tl, env, tl, tl)
> +DEF_HELPER_3(sll8, tl, env, tl, tl)
> +DEF_HELPER_3(ksll8, tl, env, tl, tl)
> +DEF_HELPER_3(kslra8, tl, env, tl, tl)
> +DEF_HELPER_3(kslra8_u, tl, env, tl, tl)
> diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
> index 6f053bfeb7..cc782fcde5 100644
> --- a/target/riscv/insn32.decode
> +++ b/target/riscv/insn32.decode
> @@ -24,6 +24,7 @@
>
>  %sh10    20:10
>  %sh4    20:4
> +%sh3    20:3
>  %csr    20:12
>  %rm     12:3
>  %nf     29:3                     !function=ex_plus_1
> @@ -61,6 +62,7 @@
>
>  @sh      ......  ...... .....  ... ..... ....... &shift  shamt=%sh10      %rs1 %rd
>  @sh4     ......  ...... .....  ... ..... ....... &shift  shamt=%sh4      %rs1 %rd
> +@sh3     ......  ...... .....  ... ..... ....... &shift  shamt=%sh3      %rs1 %rd
>  @csr     ............   .....  ... ..... .......               %csr     %rs1 %rd
>
>  @atom_ld ..... aq:1 rl:1 ..... ........ ..... ....... &atomic rs2=0     %rs1 %rd
> @@ -652,3 +654,18 @@ ksll16     0110010  ..... ..... 000 ..... 1111111 @r
>  kslli16    0111010  1.... ..... 000 ..... 1111111 @sh4
>  kslra16    0101011  ..... ..... 000 ..... 1111111 @r
>  kslra16_u  0110011  ..... ..... 000 ..... 1111111 @r
> +
> +sra8       0101100  ..... ..... 000 ..... 1111111 @r
> +sra8_u     0110100  ..... ..... 000 ..... 1111111 @r
> +srai8      0111100  00... ..... 000 ..... 1111111 @sh3
> +srai8_u    0111100  01... ..... 000 ..... 1111111 @sh3
> +srl8       0101101  ..... ..... 000 ..... 1111111 @r
> +srl8_u     0110101  ..... ..... 000 ..... 1111111 @r
> +srli8      0111101  00... ..... 000 ..... 1111111 @sh3
> +srli8_u    0111101  01... ..... 000 ..... 1111111 @sh3
> +sll8       0101110  ..... ..... 000 ..... 1111111 @r
> +slli8      0111110  00... ..... 000 ..... 1111111 @sh3
> +ksll8      0110110  ..... ..... 000 ..... 1111111 @r
> +kslli8     0111110  01... ..... 000 ..... 1111111 @sh3
> +kslra8     0101111  ..... ..... 000 ..... 1111111 @r
> +kslra8_u   0110111  ..... ..... 000 ..... 1111111 @r
> diff --git a/target/riscv/insn_trans/trans_rvp.c.inc b/target/riscv/insn_trans/trans_rvp.c.inc
> index 848edab7e5..12a64849eb 100644
> --- a/target/riscv/insn_trans/trans_rvp.c.inc
> +++ b/target/riscv/insn_trans/trans_rvp.c.inc
> @@ -353,3 +353,19 @@ GEN_RVP_SHIFTI(slli16, sll16, tcg_gen_vec_shl16i_i64);
>  GEN_RVP_SHIFTI(srai16_u, sra16_u, NULL);
>  GEN_RVP_SHIFTI(srli16_u, srl16_u, NULL);
>  GEN_RVP_SHIFTI(kslli16, ksll16, NULL);
> +
> +/* SIMD 8-bit Shift Instructions */
> +GEN_RVP_SHIFT(sra8, tcg_gen_gvec_sars, 0);
> +GEN_RVP_SHIFT(srl8, tcg_gen_gvec_shrs, 0);
> +GEN_RVP_SHIFT(sll8, tcg_gen_gvec_shls, 0);
> +GEN_RVP_R_OOL(sra8_u);
> +GEN_RVP_R_OOL(srl8_u);
> +GEN_RVP_R_OOL(ksll8);
> +GEN_RVP_R_OOL(kslra8);
> +GEN_RVP_R_OOL(kslra8_u);
> +GEN_RVP_SHIFTI(srai8, sra8, tcg_gen_vec_sar8i_i64);
> +GEN_RVP_SHIFTI(srli8, srl8, tcg_gen_vec_shr8i_i64);
> +GEN_RVP_SHIFTI(slli8, sll8, tcg_gen_vec_shl8i_i64);
> +GEN_RVP_SHIFTI(srai8_u, sra8_u, NULL);
> +GEN_RVP_SHIFTI(srli8_u, srl8_u, NULL);
> +GEN_RVP_SHIFTI(kslli8, ksll8, NULL);
> diff --git a/target/riscv/packed_helper.c b/target/riscv/packed_helper.c
> index 7e31c2fe46..ab9ebc472b 100644
> --- a/target/riscv/packed_helper.c
> +++ b/target/riscv/packed_helper.c
> @@ -529,3 +529,105 @@ static inline void do_kslra16_u(CPURISCVState *env, void *vd, void *va,
>  }
>
>  RVPR(kslra16_u, 1, 2);
> +
> +/* SIMD 8-bit Shift Instructions */
> +static inline void do_sra8(CPURISCVState *env, void *vd, void *va,
> +                           void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] = a[i] >> shift;
> +}
> +
> +RVPR(sra8, 1, 1);
> +
> +static inline void do_srl8(CPURISCVState *env, void *vd, void *va,
> +                           void *vb, uint8_t i)
> +{
> +    uint8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] = a[i] >> shift;
> +}
> +
> +RVPR(srl8, 1, 1);
> +
> +static inline void do_sll8(CPURISCVState *env, void *vd, void *va,
> +                           void *vb, uint8_t i)
> +{
> +    uint8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] = a[i] << shift;
> +}
> +
> +RVPR(sll8, 1, 1);
> +
> +static inline void do_sra8_u(CPURISCVState *env, void *vd, void *va,
> +                             void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] =  vssra8(env, 0, a[i], shift);
> +}
> +
> +RVPR(sra8_u, 1, 1);
> +
> +static inline void do_srl8_u(CPURISCVState *env, void *vd, void *va,
> +                             void *vb, uint8_t i)
> +{
> +    uint8_t *d = vd, *a = va;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +    d[i] =  vssrl8(env, 0, a[i], shift);
> +}
> +
> +RVPR(srl8_u, 1, 1);
> +
> +static inline void do_ksll8(CPURISCVState *env, void *vd, void *va,
> +                            void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va, result;
> +    uint8_t shift = *(uint8_t *)vb & 0x7;
> +
> +    result = a[i] << shift;
> +    if (shift > (clrsb32(a[i]) - 24)) {
> +        env->vxsat = 0x1;
> +        d[i] = (a[i] & INT8_MIN) ? INT8_MIN : INT8_MAX;
> +    } else {
> +        d[i] = result;
> +    }
> +}
> +
> +RVPR(ksll8, 1, 1);
> +
> +static inline void do_kslra8(CPURISCVState *env, void *vd, void *va,
> +                             void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va;
> +    int32_t shift = sextract32((*(uint32_t *)vb), 0, 4);
> +
> +    if (shift >= 0) {
> +        do_ksll8(env, vd, va, vb, i);
> +    } else {
> +        shift = -shift;
> +        shift = (shift == 8) ? 7 : shift;
> +        d[i] = a[i] >> shift;
> +    }
> +}
> +
> +RVPR(kslra8, 1, 1);
> +
> +static inline void do_kslra8_u(CPURISCVState *env, void *vd, void *va,
> +                               void *vb, uint8_t i)
> +{
> +    int8_t *d = vd, *a = va;
> +    int32_t shift = sextract32((*(uint32_t *)vb), 0, 4);
> +
> +    if (shift >= 0) {
> +        do_ksll8(env, vd, va, vb, i);
> +    } else {
> +        shift = -shift;
> +        shift = (shift == 8) ? 7 : shift;
> +        d[i] =  vssra8(env, 0, a[i], shift);
> +    }
> +}
> +
> +RVPR(kslra8_u, 1, 1);

Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
diff mbox series

Patch

diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index 20bf400ac2..0ecd4d53f9 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -1193,3 +1193,12 @@  DEF_HELPER_3(sll16, tl, env, tl, tl)
 DEF_HELPER_3(ksll16, tl, env, tl, tl)
 DEF_HELPER_3(kslra16, tl, env, tl, tl)
 DEF_HELPER_3(kslra16_u, tl, env, tl, tl)
+
+DEF_HELPER_3(sra8, tl, env, tl, tl)
+DEF_HELPER_3(sra8_u, tl, env, tl, tl)
+DEF_HELPER_3(srl8, tl, env, tl, tl)
+DEF_HELPER_3(srl8_u, tl, env, tl, tl)
+DEF_HELPER_3(sll8, tl, env, tl, tl)
+DEF_HELPER_3(ksll8, tl, env, tl, tl)
+DEF_HELPER_3(kslra8, tl, env, tl, tl)
+DEF_HELPER_3(kslra8_u, tl, env, tl, tl)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index 6f053bfeb7..cc782fcde5 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -24,6 +24,7 @@ 
 
 %sh10    20:10
 %sh4    20:4
+%sh3    20:3
 %csr    20:12
 %rm     12:3
 %nf     29:3                     !function=ex_plus_1
@@ -61,6 +62,7 @@ 
 
 @sh      ......  ...... .....  ... ..... ....... &shift  shamt=%sh10      %rs1 %rd
 @sh4     ......  ...... .....  ... ..... ....... &shift  shamt=%sh4      %rs1 %rd
+@sh3     ......  ...... .....  ... ..... ....... &shift  shamt=%sh3      %rs1 %rd
 @csr     ............   .....  ... ..... .......               %csr     %rs1 %rd
 
 @atom_ld ..... aq:1 rl:1 ..... ........ ..... ....... &atomic rs2=0     %rs1 %rd
@@ -652,3 +654,18 @@  ksll16     0110010  ..... ..... 000 ..... 1111111 @r
 kslli16    0111010  1.... ..... 000 ..... 1111111 @sh4
 kslra16    0101011  ..... ..... 000 ..... 1111111 @r
 kslra16_u  0110011  ..... ..... 000 ..... 1111111 @r
+
+sra8       0101100  ..... ..... 000 ..... 1111111 @r
+sra8_u     0110100  ..... ..... 000 ..... 1111111 @r
+srai8      0111100  00... ..... 000 ..... 1111111 @sh3
+srai8_u    0111100  01... ..... 000 ..... 1111111 @sh3
+srl8       0101101  ..... ..... 000 ..... 1111111 @r
+srl8_u     0110101  ..... ..... 000 ..... 1111111 @r
+srli8      0111101  00... ..... 000 ..... 1111111 @sh3
+srli8_u    0111101  01... ..... 000 ..... 1111111 @sh3
+sll8       0101110  ..... ..... 000 ..... 1111111 @r
+slli8      0111110  00... ..... 000 ..... 1111111 @sh3
+ksll8      0110110  ..... ..... 000 ..... 1111111 @r
+kslli8     0111110  01... ..... 000 ..... 1111111 @sh3
+kslra8     0101111  ..... ..... 000 ..... 1111111 @r
+kslra8_u   0110111  ..... ..... 000 ..... 1111111 @r
diff --git a/target/riscv/insn_trans/trans_rvp.c.inc b/target/riscv/insn_trans/trans_rvp.c.inc
index 848edab7e5..12a64849eb 100644
--- a/target/riscv/insn_trans/trans_rvp.c.inc
+++ b/target/riscv/insn_trans/trans_rvp.c.inc
@@ -353,3 +353,19 @@  GEN_RVP_SHIFTI(slli16, sll16, tcg_gen_vec_shl16i_i64);
 GEN_RVP_SHIFTI(srai16_u, sra16_u, NULL);
 GEN_RVP_SHIFTI(srli16_u, srl16_u, NULL);
 GEN_RVP_SHIFTI(kslli16, ksll16, NULL);
+
+/* SIMD 8-bit Shift Instructions */
+GEN_RVP_SHIFT(sra8, tcg_gen_gvec_sars, 0);
+GEN_RVP_SHIFT(srl8, tcg_gen_gvec_shrs, 0);
+GEN_RVP_SHIFT(sll8, tcg_gen_gvec_shls, 0);
+GEN_RVP_R_OOL(sra8_u);
+GEN_RVP_R_OOL(srl8_u);
+GEN_RVP_R_OOL(ksll8);
+GEN_RVP_R_OOL(kslra8);
+GEN_RVP_R_OOL(kslra8_u);
+GEN_RVP_SHIFTI(srai8, sra8, tcg_gen_vec_sar8i_i64);
+GEN_RVP_SHIFTI(srli8, srl8, tcg_gen_vec_shr8i_i64);
+GEN_RVP_SHIFTI(slli8, sll8, tcg_gen_vec_shl8i_i64);
+GEN_RVP_SHIFTI(srai8_u, sra8_u, NULL);
+GEN_RVP_SHIFTI(srli8_u, srl8_u, NULL);
+GEN_RVP_SHIFTI(kslli8, ksll8, NULL);
diff --git a/target/riscv/packed_helper.c b/target/riscv/packed_helper.c
index 7e31c2fe46..ab9ebc472b 100644
--- a/target/riscv/packed_helper.c
+++ b/target/riscv/packed_helper.c
@@ -529,3 +529,105 @@  static inline void do_kslra16_u(CPURISCVState *env, void *vd, void *va,
 }
 
 RVPR(kslra16_u, 1, 2);
+
+/* SIMD 8-bit Shift Instructions */
+static inline void do_sra8(CPURISCVState *env, void *vd, void *va,
+                           void *vb, uint8_t i)
+{
+    int8_t *d = vd, *a = va;
+    uint8_t shift = *(uint8_t *)vb & 0x7;
+    d[i] = a[i] >> shift;
+}
+
+RVPR(sra8, 1, 1);
+
+static inline void do_srl8(CPURISCVState *env, void *vd, void *va,
+                           void *vb, uint8_t i)
+{
+    uint8_t *d = vd, *a = va;
+    uint8_t shift = *(uint8_t *)vb & 0x7;
+    d[i] = a[i] >> shift;
+}
+
+RVPR(srl8, 1, 1);
+
+static inline void do_sll8(CPURISCVState *env, void *vd, void *va,
+                           void *vb, uint8_t i)
+{
+    uint8_t *d = vd, *a = va;
+    uint8_t shift = *(uint8_t *)vb & 0x7;
+    d[i] = a[i] << shift;
+}
+
+RVPR(sll8, 1, 1);
+
+static inline void do_sra8_u(CPURISCVState *env, void *vd, void *va,
+                             void *vb, uint8_t i)
+{
+    int8_t *d = vd, *a = va;
+    uint8_t shift = *(uint8_t *)vb & 0x7;
+    d[i] =  vssra8(env, 0, a[i], shift);
+}
+
+RVPR(sra8_u, 1, 1);
+
+static inline void do_srl8_u(CPURISCVState *env, void *vd, void *va,
+                             void *vb, uint8_t i)
+{
+    uint8_t *d = vd, *a = va;
+    uint8_t shift = *(uint8_t *)vb & 0x7;
+    d[i] =  vssrl8(env, 0, a[i], shift);
+}
+
+RVPR(srl8_u, 1, 1);
+
+static inline void do_ksll8(CPURISCVState *env, void *vd, void *va,
+                            void *vb, uint8_t i)
+{
+    int8_t *d = vd, *a = va, result;
+    uint8_t shift = *(uint8_t *)vb & 0x7;
+
+    result = a[i] << shift;
+    if (shift > (clrsb32(a[i]) - 24)) {
+        env->vxsat = 0x1;
+        d[i] = (a[i] & INT8_MIN) ? INT8_MIN : INT8_MAX;
+    } else {
+        d[i] = result;
+    }
+}
+
+RVPR(ksll8, 1, 1);
+
+static inline void do_kslra8(CPURISCVState *env, void *vd, void *va,
+                             void *vb, uint8_t i)
+{
+    int8_t *d = vd, *a = va;
+    int32_t shift = sextract32((*(uint32_t *)vb), 0, 4);
+
+    if (shift >= 0) {
+        do_ksll8(env, vd, va, vb, i);
+    } else {
+        shift = -shift;
+        shift = (shift == 8) ? 7 : shift;
+        d[i] = a[i] >> shift;
+    }
+}
+
+RVPR(kslra8, 1, 1);
+
+static inline void do_kslra8_u(CPURISCVState *env, void *vd, void *va,
+                               void *vb, uint8_t i)
+{
+    int8_t *d = vd, *a = va;
+    int32_t shift = sextract32((*(uint32_t *)vb), 0, 4);
+
+    if (shift >= 0) {
+        do_ksll8(env, vd, va, vb, i);
+    } else {
+        shift = -shift;
+        shift = (shift == 8) ? 7 : shift;
+        d[i] =  vssra8(env, 0, a[i], shift);
+    }
+}
+
+RVPR(kslra8_u, 1, 1);