SIMD 16-bit Add/Subtract Instructions

__STATIC_FORCEINLINE unsigned long __RV_ADD16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_CRAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_CRSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_KADD16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_KCRAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_KCRSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_KSTAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_KSTSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_KSUB16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_RADD16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_RCRAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_RCRSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_RSTAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_RSTSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_RSUB16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_STAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_STSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_SUB16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_UKADD16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_UKCRAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_UKCRSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_UKSTAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_UKSTSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_UKSUB16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_URADD16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_URCRAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_URCRSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_URSTAS16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_URSTSA16 (unsigned long a, unsigned long b)

__STATIC_FORCEINLINE unsigned long __RV_URSUB16 (unsigned long a, unsigned long b)

group SIMD 16-bit Add/Subtract Instructions

SIMD 16-bit Add/Subtract Instructions.

Based on the combination of the types of the two 16-bit arithmetic operations, the SIMD 16-bit add/subtract instructions can be classified into 6 main categories: Addition (two 16-bit addition), Subtraction (two 16-bit subtraction), Crossed Add & Sub (one addition and one subtraction), and Crossed Sub & Add (one subtraction and one addition), Straight Add & Sub (one addition and one subtraction), and Straight Sub & Add (one subtraction and one addition). Based on the way of how an overflow condition is handled, the SIMD 16-bit add/subtract instructions can be classified into 5 groups: Wrap-around (dropping overflow), Signed Halving (keeping overflow by dropping 1 LSB bit), Unsigned Halving, Signed Saturation (clipping overflow), and Unsigned Saturation. Together, there are 30 SIMD 16-bit add/subtract instructions.

Functions

__STATIC_FORCEINLINE unsigned long __RV_ADD16 (unsigned long a, unsigned long b)

ADD16 (SIMD 16-bit Addition)

Type: SIMD

Syntax:

ADD16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit integer element additions simultaneously.

Description

:

This instruction adds the 16-bit integer elements in Rs1 with the 16-bit integer elements in Rs2, and then writes the 16-bit element results to Rd.

Note

:

This instruction can be used for either signed or unsigned addition.

Operations:

Rd.H[x] = Rs1.H[x] + Rs2.H[x];
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_CRAS16 (unsigned long a, unsigned long b)

CRAS16 (SIMD 16-bit Cross Addition & Subtraction)

Type: SIMD

Syntax:

CRAS16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit integer element addition and 16-bit integer element subtraction in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks.

Description

:

This instruction adds the 16-bit integer element in [31:16] of 32-bit chunks in Rs1 with the 16-bit integer element in [15:0] of 32-bit chunks in Rs2, and writes the result to [31:16] of 32-bit chunks in Rd; at the same time, it subtracts the 16-bit integer element in [31:16] of 32-bit chunks in Rs2 from the 16-bit integer element in [15:0] of 32-bit chunks, and writes the result to [15:0] of 32- bit chunks in Rd.

Note

:

This instruction can be used for either signed or unsigned operations.

Operations:

Rd.W[x][31:16] = Rs1.W[x][31:16] + Rs2.W[x][15:0];
Rd.W[x][15:0] = Rs1.W[x][15:0] - Rs2.W[x][31:16];
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_CRSA16 (unsigned long a, unsigned long b)

CRSA16 (SIMD 16-bit Cross Subtraction & Addition)

Type: SIMD

Syntax:

CRSA16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit integer element subtraction and 16-bit integer element addition in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks.

Description

:

This instruction subtracts the 16-bit integer element in [15:0] of 32-bit chunks in Rs2 from the 16-bit integer element in [31:16] of 32-bit chunks in Rs1, and writes the result to [31:16] of 32-bit chunks in Rd; at the same time, it adds the 16-bit integer element in [31:16] of 32-bit chunks in Rs2 with the 16-bit integer element in [15:0] of 32-bit chunks in Rs1, and writes the result to [15:0] of 32-bit chunks in Rd.

Note

:

This instruction can be used for either signed or unsigned operations.

Operations:

Rd.W[x][31:16] = Rs1.W[x][31:16] - Rs2.W[x][15:0];
Rd.W[x][15:0] = Rs1.W[x][15:0] + Rs2.W[x][31:16];
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_KADD16 (unsigned long a, unsigned long b)

KADD16 (SIMD 16-bit Signed Saturating Addition)

Type: SIMD

Syntax:

KADD16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element saturating additions simultaneously.

Description

:

This instruction adds the 16-bit signed integer elements in Rs1 with the 16-bit signed integer elements in Rs2. If any of the results are beyond the Q15 number range (-2^15 <= Q15 <= 2^15-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to Rd.

Operations:

res[x] = Rs1.H[x] + Rs2.H[x];
if (res[x] > 32767) {
  res[x] = 32767;
  OV = 1;
} else if (res[x] < -32768) {
  res[x] = -32768;
  OV = 1;
}
Rd.H[x] = res[x];
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_KCRAS16 (unsigned long a, unsigned long b)

KCRAS16 (SIMD 16-bit Signed Saturating Cross Addition & Subtraction)

Type: SIMD

Syntax:

KCRAS16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element saturating addition and 16-bit signed integer element saturating subtraction in a 32-bit chunk simultaneously. Operands are from crossed positions in 32- bit chunks.

Description

:

This instruction adds the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs1 with the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs2; at the same time, it subtracts the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs2 from the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs1. If any of the results are beyond the Q15 number range (-2^15 <= Q15 <= 2^15-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to [31:16] of 32-bit chunks in Rd for addition and [15:0] of 32-bit chunks in Rd for subtraction.

Operations:

res1 = Rs1.W[x][31:16] + Rs2.W[x][15:0];
res2 = Rs1.W[x][15:0] - Rs2.W[x][31:16];
for (res in [res1, res2]) {
  if (res > (2^15)-1) {
    res = (2^15)-1;
    OV = 1;
  } else if (res < -2^15) {
    res = -2^15;
    OV = 1;
  }
}
Rd.W[x][31:16] = res1;
Rd.W[x][15:0] = res2;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_KCRSA16 (unsigned long a, unsigned long b)

KCRSA16 (SIMD 16-bit Signed Saturating Cross Subtraction & Addition)

Type: SIMD

Syntax:

KCRSA16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element saturating subtraction and 16-bit signed integer element saturating addition in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks.

Description

:

This instruction subtracts the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs2 from the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs1; at the same time, it adds the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs2 with the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs1. If any of the results are beyond the Q15 number range (-2^15 <= Q15 <= 2^15-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to [31:16] of 32-bit chunks in Rd for subtraction and [15:0] of 32-bit chunks in Rd for addition.

Operations:

res1 = Rs1.W[x][31:16] - Rs2.W[x][15:0];
res2 = Rs1.W[x][15:0] + Rs2.W[x][31:16];
for (res in [res1, res2]) {
  if (res > (2^15)-1) {
    res = (2^15)-1;
    OV = 1;
  } else if (res < -2^15) {
    res = -2^15;
    OV = 1;
  }
}
Rd.W[x][31:16] = res1;
Rd.W[x][15:0] = res2;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_KSTAS16 (unsigned long a, unsigned long b)

KSTAS16 (SIMD 16-bit Signed Saturating Straight Addition & Subtraction)

Type: SIMD

Syntax:

KSTAS16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element saturating addition and 16-bit signed integer element saturating subtraction in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks.

Description

:

This instruction adds the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs1 with the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs2; at the same time, it subtracts the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs2 from the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs1. If any of the results are beyond the Q15 number range (-2^15 <= Q15 <= 2^15-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to [31:16] of 32-bit chunks in Rd for addition and [15:0] of 32-bit chunks in Rd for subtraction.

Operations:

res1 = Rs1.W[x][31:16] + Rs2.W[x][31:16];
res2 = Rs1.W[x][15:0] - Rs2.W[x][15:0];
for (res in [res1, res2]) {
  if (res > (2^15)-1) {
    res = (2^15)-1;
    OV = 1;
  } else if (res < -2^15) {
    res = -2^15;
    OV = 1;
  }
}
Rd.W[x][31:16] = res1;
Rd.W[x][15:0] = res2;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_KSTSA16 (unsigned long a, unsigned long b)

KSTSA16 (SIMD 16-bit Signed Saturating Straight Subtraction & Addition)

Type: SIMD

Syntax:

KSTSA16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element saturating subtraction and 16-bit signed integer element saturating addition in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks.

Description

:

This instruction subtracts the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs2 from the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs1; at the same time, it adds the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs2 with the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs1. If any of the results are beyond the Q15 number range (-2^15 <= Q15 <= 2^15-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to [31:16] of 32-bit chunks in Rd for subtraction and [15:0] of 32-bit chunks in Rd for addition.

Operations:

res1 = Rs1.W[x][31:16] - Rs2.W[x][31:16];
res2 = Rs1.W[x][15:0] + Rs2.W[x][15:0];
for (res in [res1, res2]) {
  if (res > (2^15)-1) {
    res = (2^15)-1;
    OV = 1;
  } else if (res < -2^15) {
    res = -2^15;
    OV = 1;
  }
}
Rd.W[x][31:16] = res1;
Rd.W[x][15:0] = res2;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_KSUB16 (unsigned long a, unsigned long b)

KSUB16 (SIMD 16-bit Signed Saturating Subtraction)

Type: SIMD

Syntax:

KSUB16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer elements saturating subtractions simultaneously.

Description

:

This instruction subtracts the 16-bit signed integer elements in Rs2 from the 16-bit signed integer elements in Rs1. If any of the results are beyond the Q15 number range (-2^15 <= Q15 <= 2^15-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to Rd.

Operations:

res[x] = Rs1.H[x] - Rs2.H[x];
if (res[x] > (2^15)-1) {
  res[x] = (2^15)-1;
  OV = 1;
} else if (res[x] < -2^15) {
  res[x] = -2^15;
  OV = 1;
}
Rd.H[x] = res[x];
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_RADD16 (unsigned long a, unsigned long b)

RADD16 (SIMD 16-bit Signed Halving Addition)

Type: SIMD

Syntax:

RADD16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element additions simultaneously. The results are halved to avoid overflow or saturation.

Description

:

This instruction adds the 16-bit signed integer elements in Rs1 with the 16-bit signed integer elements in Rs2. The results are first arithmetically right-shifted by 1 bit and then written to Rd.

Examples:

* Rs1 = 0x7FFF, Rs2 = 0x7FFF, Rd = 0x7FFF
* Rs1 = 0x8000, Rs2 = 0x8000, Rd = 0x8000
* Rs1 = 0x4000, Rs2 = 0x8000, Rd = 0xE000

Operations:

Rd.H[x] = (Rs1.H[x] + Rs2.H[x]) s>> 1; for RV32: x=1...0, for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_RCRAS16 (unsigned long a, unsigned long b)

RCRAS16 (SIMD 16-bit Signed Halving Cross Addition & Subtraction)

Type: SIMD

Syntax:

RCRAS16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element addition and 16-bit signed integer element subtraction in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks. The results are halved to avoid overflow or saturation.

Description

:

This instruction adds the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs1 with the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs2, and subtracts the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs2 from the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs1. The element results are first arithmetically right-shifted by 1 bit and then written to [31:16] of 32-bit chunks in Rd and [15:0] of 32-bit chunks in Rd.

Examples:

Please see `RADD16` and `RSUB16` instructions.

Operations:

Rd.W[x][31:16] = (Rs1.W[x][31:16] + Rs2.W[x][15:0]) s>> 1;
Rd.W[x][15:0] = (Rs1.W[x][15:0] - Rs2.W[x][31:16]) s>> 1;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_RCRSA16 (unsigned long a, unsigned long b)

RCRSA16 (SIMD 16-bit Signed Halving Cross Subtraction & Addition)

Type: SIMD

Syntax:

RCRSA16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element subtraction and 16-bit signed integer element addition in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks. The results are halved to avoid overflow or saturation.

Description

:

This instruction subtracts the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs2 from the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs1, and adds the 16-bit signed element integer in [15:0] of 32-bit chunks in Rs1 with the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs2. The two results are first arithmetically right-shifted by 1 bit and then written to [31:16] of 32-bit chunks in Rd and [15:0] of 32-bit chunks in Rd.

Examples:

Please see `RADD16` and `RSUB16` instructions.

Operations:

Rd.W[x][31:16] = (Rs1.W[x][31:16] - Rs2.W[x][15:0]) s>> 1;
Rd.W[x][15:0] = (Rs1.W[x][15:0] + Rs2.W[x][31:16]) s>> 1;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_RSTAS16 (unsigned long a, unsigned long b)

RSTAS16 (SIMD 16-bit Signed Halving Straight Addition & Subtraction)

Type: SIMD

Syntax:

RSTAS16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element addition and 16-bit signed integer element subtraction in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks. The results are halved to avoid overflow or saturation.

Description

:

This instruction adds the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs1 with the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs2, and subtracts the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs2 from the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs1. The element results are first arithmetically right-shifted by 1 bit and then written to [31:16] of 32-bit chunks in Rd and [15:0] of 32-bit chunks in Rd.

Examples:

Please see `RADD16` and `RSUB16` instructions.

Operations:

Rd.W[x][31:16] = (Rs1.W[x][31:16] + Rs2.W[x][31:16]) s>> 1;
Rd.W[x][15:0] = (Rs1.W[x][15:0] - Rs2.W[x][15:0]) s>> 1;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_RSTSA16 (unsigned long a, unsigned long b)

RSTSA16 (SIMD 16-bit Signed Halving Straight Subtraction & Addition)

Type: SIMD

Syntax:

RSTSA16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element subtraction and 16-bit signed integer element addition in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks. The results are halved to avoid overflow or saturation.

Description

:

This instruction subtracts the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs2 from the 16-bit signed integer element in [31:16] of 32-bit chunks in Rs1, and adds the 16-bit signed element integer in [15:0] of 32-bit chunks in Rs1 with the 16-bit signed integer element in [15:0] of 32-bit chunks in Rs2. The two results are first arithmetically right-shifted by 1 bit and then written to [31:16] of 32-bit chunks in Rd and [15:0] of 32-bit chunks in Rd.

Examples:

Please see `RADD16` and `RSUB16` instructions.

Operations:

Rd.W[x][31:16] = (Rs1.W[x][31:16] - Rs2.W[x][31:16]) s>> 1;
Rd.W[x][15:0] = (Rs1.W[x][15:0] + Rs2.W[x][15:0]) s>> 1;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_RSUB16 (unsigned long a, unsigned long b)

RSUB16 (SIMD 16-bit Signed Halving Subtraction)

Type: SIMD

Syntax:

RSUB16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit signed integer element subtractions simultaneously. The results are halved to avoid overflow or saturation.

Description

:

This instruction subtracts the 16-bit signed integer elements in Rs2 from the 16-bit signed integer elements in Rs1. The results are first arithmetically right-shifted by 1 bit and then written to Rd.

Examples:

* Ra = 0x7FFF, Rb = 0x8000, Rt = 0x7FFF
* Ra = 0x8000, Rb = 0x7FFF, Rt = 0x8000
* Ra = 0x8000, Rb = 0x4000, Rt = 0xA000

Operations:

Rd.H[x] = (Rs1.H[x] - Rs2.H[x]) s>> 1;
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_STAS16 (unsigned long a, unsigned long b)

STAS16 (SIMD 16-bit Straight Addition & Subtraction)

Type: SIMD

Syntax:

STAS16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit integer element addition and 16-bit integer element subtraction in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks.

Description

:

This instruction adds the 16-bit integer element in [31:16] of 32-bit chunks in Rs1 with the 16-bit integer element in [31:16] of 32-bit chunks in Rs2, and writes the result to [31:16] of 32-bit chunks in Rd; at the same time, it subtracts the 16-bit integer element in [15:0] of 32-bit chunks in Rs2 from the 16-bit integer element in [15:0] of 32-bit chunks, and writes the result to [15:0] of 32- bit chunks in Rd.

Note

:

This instruction can be used for either signed or unsigned operations.

Operations:

Rd.W[x][31:16] = Rs1.W[x][31:16] + Rs2.W[x][31:16];
Rd.W[x][15:0] = Rs1.W[x][15:0] - Rs2.W[x][15:0];
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_STSA16 (unsigned long a, unsigned long b)

STSA16 (SIMD 16-bit Straight Subtraction & Addition)

Type: SIMD

Syntax:

STSA16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit integer element subtraction and 16-bit integer element addition in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks.

Description

:

This instruction subtracts the 16-bit integer element in [31:16] of 32-bit chunks in Rs2 from the 16-bit integer element in [31:16] of 32-bit chunks in Rs1, and writes the result to [31:16] of 32-bit chunks in Rd; at the same time, it adds the 16-bit integer element in [15:0] of 32-bit chunks in Rs2 with the 16-bit integer element in [15:0] of 32-bit chunks in Rs1, and writes the result to [15:0] of 32-bit chunks in Rd.

Note

:

This instruction can be used for either signed or unsigned operations.

Operations:

Rd.W[x][31:16] = Rs1.W[x][31:16] - Rs2.W[x][31:16];
Rd.W[x][15:0] = Rs1.W[x][15:0] + Rs2.W[x][15:0];
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_SUB16 (unsigned long a, unsigned long b)

SUB16 (SIMD 16-bit Subtraction)

Type: SIMD

Syntax:

SUB16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit integer element subtractions simultaneously.

Description

:

This instruction subtracts the 16-bit integer elements in Rs2 from the 16-bit integer elements in Rs1, and then writes the result to Rd.

Note

:

This instruction can be used for either signed or unsigned subtraction.

Operations:

Rd.H[x] = Rs1.H[x] - Rs2.H[x];
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_UKADD16 (unsigned long a, unsigned long b)

UKADD16 (SIMD 16-bit Unsigned Saturating Addition)

Type: SIMD

Syntax:

UKADD16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit unsigned integer element saturating additions simultaneously.

Description

:

This instruction adds the 16-bit unsigned integer elements in Rs1 with the 16-bit unsigned integer elements in Rs2. If any of the results are beyond the 16-bit unsigned number range (0 <= RES <= 2^16-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to Rd.

Operations:

res[x] = Rs1.H[x] + Rs2.H[x];
if (res[x] > (2^16)-1) {
  res[x] = (2^16)-1;
  OV = 1;
}
Rd.H[x] = res[x];
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_UKCRAS16 (unsigned long a, unsigned long b)

UKCRAS16 (SIMD 16-bit Unsigned Saturating Cross Addition & Subtraction)

Type: SIMD

Syntax:

UKCRAS16 Rd, Rs1, Rs2

Purpose

:

Do one 16-bit unsigned integer element saturating addition and one 16-bit unsigned integer element saturating subtraction in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks.

Description

:

This instruction adds the 16-bit unsigned integer element in [31:16] of 32-bit chunks in Rs1 with the 16-bit unsigned integer element in [15:0] of 32-bit chunks in Rs2; at the same time, it subtracts the 16-bit unsigned integer element in [31:16] of 32-bit chunks in Rs2 from the 16-bit unsigned integer element in [15:0] of 32-bit chunks in Rs1. If any of the results are beyond the 16-bit unsigned number range (0 <= RES <= 2^16-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to [31:16] of 32-bit chunks in Rd for addition and [15:0] of 32-bit chunks in Rd for subtraction.

Operations:

res1 = Rs1.W[x][31:16] + Rs2.W[x][15:0];
res2 = Rs1.W[x][15:0] - Rs2.W[x][31:16];
if (res1 > (2^16)-1) {
  res1 = (2^16)-1;
  OV = 1;
}
if (res2 < 0) {
  res2 = 0;
  OV = 1;
}
Rd.W[x][31:16] = res1;
Rd.W[x][15:0] = res2;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_UKCRSA16 (unsigned long a, unsigned long b)

UKCRSA16 (SIMD 16-bit Unsigned Saturating Cross Subtraction & Addition)

Type: SIMD

Syntax:

UKCRSA16 Rd, Rs1, Rs2

Purpose

:

Do one 16-bit unsigned integer element saturating subtraction and one 16-bit unsigned integer element saturating addition in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks.

Description

:

This instruction subtracts the 16-bit unsigned integer element in [15:0] of 32-bit chunks in Rs2 from the 16-bit unsigned integer element in [31:16] of 32-bit chunks in Rs1; at the same time, it adds the 16-bit unsigned integer element in [31:16] of 32-bit chunks in Rs2 with the 16- bit unsigned integer element in [15:0] of 32-bit chunks in Rs1. If any of the results are beyond the 16-bit unsigned number range (0 <= RES <= 2^16-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to [31:16] of 32-bit chunks in Rd for subtraction and [15:0] of 32-bit chunks in Rd for addition.

Operations:

res1 = Rs1.W[x][31:16] - Rs2.W[x][15:0];
res2 = Rs1.W[x][15:0] + Rs2.W[x][31:16];
if (res1 < 0) {
  res1 = 0;
  OV = 1;
} else if (res2 > (2^16)-1) {
  res2 = (2^16)-1;
  OV = 1;
}
Rd.W[x][31:16] = res1;
Rd.W[x][15:0] = res2;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_UKSTAS16 (unsigned long a, unsigned long b)

UKSTAS16 (SIMD 16-bit Unsigned Saturating Straight Addition & Subtraction)

Type: SIMD

Syntax:

UKSTAS16 Rd, Rs1, Rs2

Purpose

:

Do one 16-bit unsigned integer element saturating addition and one 16-bit unsigned integer element saturating subtraction in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks.

Description

:

This instruction adds the 16-bit unsigned integer element in [31:16] of 32-bit chunks in Rs1 with the 16-bit unsigned integer element in [31:16] of 32-bit chunks in Rs2; at the same time, it subtracts the 16-bit unsigned integer element in [15:0] of 32-bit chunks in Rs2 from the 16-bit unsigned integer element in [15:0] of 32-bit chunks in Rs1. If any of the results are beyond the 16-bit unsigned number range (0 <= RES <= 2^16-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to [31:16] of 32-bit chunks in Rd for addition and [15:0] of 32-bit chunks in Rd for subtraction.

Operations:

res1 = Rs1.W[x][31:16] + Rs2.W[x][31:16];
res2 = Rs1.W[x][15:0] - Rs2.W[x][15:0];
if (res1 > (2^16)-1) {
  res1 = (2^16)-1;
  OV = 1;
}
if (res2 < 0) {
  res2 = 0;
  OV = 1;
}
Rd.W[x][31:16] = res1;
Rd.W[x][15:0] = res2;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_UKSTSA16 (unsigned long a, unsigned long b)

UKSTSA16 (SIMD 16-bit Unsigned Saturating Straight Subtraction & Addition)

Type: SIMD

Syntax:

UKSTSA16 Rd, Rs1, Rs2

Purpose

:

Do one 16-bit unsigned integer element saturating subtraction and one 16-bit unsigned integer element saturating addition in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks.

Description

:

This instruction subtracts the 16-bit unsigned integer element in [31:16] of 32-bit chunks in Rs2 from the 16-bit unsigned integer element in [31:16] of 32-bit chunks in Rs1; at the same time, it adds the 16-bit unsigned integer element in [15:0] of 32-bit chunks in Rs2 with the 16- bit unsigned integer element in [15:0] of 32-bit chunks in Rs1. If any of the results are beyond the 16-bit unsigned number range (0 <= RES <= 2^16-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to [31:16] of 32-bit chunks in Rd for subtraction and [15:0] of 32-bit chunks in Rd for addition.

Operations:

res1 = Rs1.W[x][31:16] - Rs2.W[x][31:16];
res2 = Rs1.W[x][15:0] + Rs2.W[x][15:0];
if (res1 < 0) {
  res1 = 0;
  OV = 1;
} else if (res2 > (2^16)-1) {
  res2 = (2^16)-1;
  OV = 1;
}
Rd.W[x][31:16] = res1;
Rd.W[x][15:0] = res2;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_UKSUB16 (unsigned long a, unsigned long b)

UKSUB16 (SIMD 16-bit Unsigned Saturating Subtraction)

Type: SIMD

Syntax:

UKSUB16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit unsigned integer elements saturating subtractions simultaneously.

Description

:

This instruction subtracts the 16-bit unsigned integer elements in Rs2 from the 16-bit unsigned integer elements in Rs1. If any of the results are beyond the 16-bit unsigned number range (0 <= RES <= 2^16-1), they are saturated to the range and the OV bit is set to 1. The saturated results are written to Rd.

Operations:

res[x] = Rs1.H[x] - Rs2.H[x];
if (res[x] < 0) {
  res[x] = 0;
  OV = 1;
}
Rd.H[x] = res[x];
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_URADD16 (unsigned long a, unsigned long b)

URADD16 (SIMD 16-bit Unsigned Halving Addition)

Type: SIMD

Syntax:

URADD16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit unsigned integer element additions simultaneously. The results are halved to avoid overflow or saturation.

Description

:

This instruction adds the 16-bit unsigned integer elements in Rs1 with the 16-bit unsigned integer elements in Rs2. The results are first logically right-shifted by 1 bit and then written to Rd.

Examples:

* Ra = 0x7FFF, Rb = 0x7FFF Rt = 0x7FFF
* Ra = 0x8000, Rb = 0x8000 Rt = 0x8000
* Ra = 0x4000, Rb = 0x8000 Rt = 0x6000

Operations:

Rd.H[x] = (Rs1.H[x] + Rs2.H[x]) u>> 1;
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_URCRAS16 (unsigned long a, unsigned long b)

URCRAS16 (SIMD 16-bit Unsigned Halving Cross Addition & Subtraction)

Type: SIMD

Syntax:

URCRAS16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit unsigned integer element addition and 16-bit unsigned integer element subtraction in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks. The results are halved to avoid overflow or saturation.

Description

:

This instruction adds the 16-bit unsigned integer in [31:16] of 32-bit chunks in Rs1 with the 16-bit unsigned integer in [15:0] of 32-bit chunks in Rs2, and subtracts the 16-bit unsigned integer in [31:16] of 32-bit chunks in Rs2 from the 16-bit unsigned integer in [15:0] of 32-bit chunks in Rs1. The element results are first logically right-shifted by 1 bit and then written to [31:16] of 32- bit chunks in Rd and [15:0] of 32-bit chunks in Rd.

Examples:

Please see `URADD16` and `URSUB16` instructions.

Operations:

Rd.W[x][31:16] = (Rs1.W[x][31:16] + Rs2.W[x][15:0]) u>> 1;
Rd.W[x][15:0] = (Rs1.W[x][15:0] - Rs2.W[x][31:16]) u>> 1;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_URCRSA16 (unsigned long a, unsigned long b)

URCRSA16 (SIMD 16-bit Unsigned Halving Cross Subtraction & Addition)

Type: SIMD

Syntax:

URCRSA16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit unsigned integer element subtraction and 16-bit unsigned integer element addition in a 32-bit chunk simultaneously. Operands are from crossed positions in 32-bit chunks. The results are halved to avoid overflow or saturation.

Description

:

This instruction subtracts the 16-bit unsigned integer in [15:0] of 32-bit chunks in Rs2 from the 16-bit unsigned integer in [31:16] of 32-bit chunks in Rs1, and adds the 16-bit unsigned integer in [15:0] of 32-bit chunks in Rs1 with the 16-bit unsigned integer in [31:16] of 32-bit chunks in Rs2. The two results are first logically right-shifted by 1 bit and then written to [31:16] of 32-bit chunks in Rd and [15:0] of 32-bit chunks in Rd.

Examples:

Please see `URADD16` and `URSUB16` instructions.

Operations:

Rd.W[x][31:16] = (Rs1.W[x][31:16] - Rs2.W[x][15:0]) u>> 1;
Rd.W[x][15:0] = (Rs1.W[x][15:0] + Rs2.W[x][31:16]) u>> 1;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_URSTAS16 (unsigned long a, unsigned long b)

URSTAS16 (SIMD 16-bit Unsigned Halving Straight Addition & Subtraction)

Type: SIMD

Syntax:

URSTAS16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit unsigned integer element addition and 16-bit unsigned integer element subtraction in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks. The results are halved to avoid overflow or saturation.

Description

:

This instruction adds the 16-bit unsigned integer in [31:16] of 32-bit chunks in Rs1 with the 16-bit unsigned integer in [31:16] of 32-bit chunks in Rs2, and subtracts the 16-bit unsigned integer in [15:0] of 32-bit chunks in Rs2 from the 16-bit unsigned integer in [15:0] of 32-bit chunks in Rs1. The element results are first logically right-shifted by 1 bit and then written to [31:16] of 32- bit chunks in Rd and [15:0] of 32-bit chunks in Rd.

Examples:

Please see `URADD16` and `URSUB16` instructions.

Operations:

Rd.W[x][31:16] = (Rs1.W[x][31:16] + Rs2.W[x][31:16]) u>> 1;
Rd.W[x][15:0] = (Rs1.W[x][15:0] - Rs2.W[x][15:0]) u>> 1;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_URSTSA16 (unsigned long a, unsigned long b)

URSTSA16 (SIMD 16-bit Unsigned Halving Straight Subtraction & Addition)

Type: SIMD

Syntax:

URCRSA16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit unsigned integer element subtraction and 16-bit unsigned integer element addition in a 32-bit chunk simultaneously. Operands are from corresponding positions in 32-bit chunks. The results are halved to avoid overflow or saturation.

Description

:

This instruction subtracts the 16-bit unsigned integer in [31:16] of 32-bit chunks in Rs2 from the 16-bit unsigned integer in [31:16] of 32-bit chunks in Rs1, and adds the 16-bit unsigned integer in [15:0] of 32-bit chunks in Rs1 with the 16-bit unsigned integer in [15:0] of 32-bit chunks in Rs2. The two results are first logically right-shifted by 1 bit and then written to [31:16] of 32-bit chunks in Rd and [15:0] of 32-bit chunks in Rd.

Examples:

Please see `URADD16` and `URSUB16` instructions.

Operations:

Rd.W[x][31:16] = (Rs1.W[x][31:16] - Rs2.W[x][31:16]) u>> 1;
Rd.W[x][15:0] = (Rs1.W[x][15:0] + Rs2.W[x][15:0]) u>> 1;
for RV32, x=0
for RV64, x=1...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type

__STATIC_FORCEINLINE unsigned long __RV_URSUB16 (unsigned long a, unsigned long b)

URSUB16 (SIMD 16-bit Unsigned Halving Subtraction)

Type: SIMD

Syntax:

URSUB16 Rd, Rs1, Rs2

Purpose

:

Do 16-bit unsigned integer element subtractions simultaneously. The results are halved to avoid overflow or saturation.

Description

:

This instruction subtracts the 16-bit unsigned integer elements in Rs2 from the 16-bit unsigned integer elements in Rs1. The results are first logically right-shifted by 1 bit and then written to Rd.

Examples:

* Ra = 0x7FFF, Rb = 0x8000 Rt = 0xFFFF
* Ra = 0x8000, Rb = 0x7FFF Rt = 0x0000
* Ra = 0x8000, Rb = 0x4000 Rt = 0x2000

Operations:

Rd.H[x] = (Rs1.H[x] - Rs2.H[x]) u>> 1;
for RV32: x=1...0,
for RV64: x=3...0

Parameters:

• a – [in] unsigned long type of value stored in a

• b – [in] unsigned long type of value stored in b

Returns:

value stored in unsigned long type