wiki:GcnInstrsVopc

Version 1 (modified by trac, 4 years ago) (diff)

--

Back to Table of content

GCN ISA VOPC/VOP3 instructions

VOPC instructions can be encoded in the VOPC encoding and the VOP3A/VOP3B encoding. List of fields for VOPC encoding:

Bits Name Description
0-8 SRC0 First (scalar or vector) source operand
9-16 VSRC1 Second (scalar or vector) source operand
17-24 OPCODE Operation code
25-31 ENCODING Encoding type. Must be 0b0111110

Syntax: INSTRUCTION VCC, SRC0, VSRC1

List of fields for VOP3A/VOP3B encoding (GCN 1.0/1.1):

Bits Name Description
0-7 SDST Scalar destination operand
8-10 ABS Absolute modifiers for source operands (VOP3A)
8-14 SDST Scalar destination operand (VOP3B)
11 CLAMP CLAMP modifier (VOP3A)
15 CLAMP CLAMP modifier (VOP3B)
17-25 OPCODE Operation code
26-31 ENCODING Encoding type. Must be 0b110100
32-40 SRC0 First (scalar or vector) source operand
41-49 SRC1 Second (scalar or vector) source operand
50-58 SRC2 Third (scalar or vector) source operand
59-60 OMOD OMOD modifier. Multiplication modifier
61-63 NEG Negation modifier for source operands

List of fields for VOP3A/VOP3B encoding (GCN 1.2):

Bits Name Description
0-7 SDST Scalar destination operand
8-10 ABS Absolute modifiers for source operands (VOP3A)
8-14 SDST Scalar destination operand (VOP3B)
15 CLAMP CLAMP modifier
16-25 OPCODE Operation code
26-31 ENCODING Encoding type. Must be 0b110100
32-40 SRC0 First (scalar or vector) source operand
41-49 SRC1 Second (scalar or vector) source operand
50-58 SRC2 Third (scalar or vector) source operand
59-60 OMOD OMOD modifier. Multiplication modifier
61-63 NEG Negation modifier for source operands

Syntax: INSTRUCTION SDST(2), SRC0, SRC1 [MODIFIERS]

Modifiers:

  • -SRC - negate floating point value from source operand. Applied after ABS modifier.
  • ABS(SRC) - apply absolute value to source operand

NOTE: ABS and negation is applied to source operand for any instruction.

Negation and absolute value can be combined: -ABS(V0).

Limitations for operands:

  • only one SGPR can be read by instruction. Multiple occurrences of this same SGPR is allowed
  • only one literal constant can be used, and only when a SGPR or M0 is not used in source operands
  • only SRC0 can holds LDS_DIRECT

VOPC opcodes (0-255) and VOP3 opcodes are same.

Tables of opcodes and their descriptions (GCN 1.0/1.1)

Table of floating point comparison instructions by opcode:

Opcode range Instruction Description
0-15 (0x00-0x0f) V_CMP_{OP16}_F32 Signal on sNAN input only. Single FP values.
16-31 (0x10-0x1f) V_CMPX_{OP16}_F32 Signal on sNAN input only. Write result to EXEC. Single FP values.
32-47 (0x20-0x2f) V_CMP_{OP16}_F64 Signal on sNAN input only. Double FP values.
48-63 (0x30-0x3f) V_CMPX_{OP16}_F64 Signal on sNAN input only. Write result to EXEC. Double FP values.
64-79 (0x40-0x4f) V_CMP_{OP16}_F32 Signal on any sNAN. Single FP values.
80-95 (0x50-0x5f) V_CMPX_{OP16}_F32 Signal on any sNAN. Write result to EXEC. Single FP values.
96-111 (0x60-0x6f) V_CMP_{OP16}_F64 Signal on any sNAN. Double FP values.
112-127 (0x70-0x7f) V_CMPX_{OP16}_F64 Signal on any sNAN. Write result to EXEC. Double FP values.

Table of OP16 (compare operations) for floating point values comparisons:

Opcode offset OP16 name Description
0 (0x0) F SDST(LANEID) = 0
1 (0x1) LT SDST(LANEID) = (TYPE)SRC0 < (TYPE)SRC1
2 (0x2) EQ SDST(LANEID) = (TYPE)SRC0 == (TYPE)SRC1
3 (0x3) LE SDST(LANEID) = (TYPE)SRC0 <= (TYPE)SRC1
4 (0x4) GT SDST(LANEID) = (TYPE)SRC0 > (TYPE)SRC1
5 (0x5) LG SDST(LANEID) = (TYPE)SRC0 != (TYPE)SRC1
6 (0x6) GE SDST(LANEID) = (TYPE)SRC0 >= (TYPE)SRC1
7 (0x7) O SDST(LANEID) = ((TYPE)SRC0!=NAN && (TYPE)SRC1!=NAN)
8 (0x8) U SDST(LANEID) = ((TYPE)SRC0!=NAN
9 (0x9) NGE SDST(LANEID) = !((TYPE)SRC0 >= (TYPE)SRC1)
10 (0xa) NLG SDST(LANEID) = !((TYPE)SRC0 != (TYPE)SRC1)
11 (0xb) NGT SDST(LANEID) = !((TYPE)SRC0 > (TYPE)SRC1)
12 (0xc) NLE SDST(LANEID) = !((TYPE)SRC0 <= (TYPE)SRC1)
13 (0xd) NEG SDST(LANEID) = !((TYPE)SRC0 == (TYPE)SRC1)
14 (0xe) NLT SDST(LANEID) = !((TYPE)SRC0 < (TYPE)SRC1)
15 (0xf) T SDST(LANEID) = 1

NOTE: Comparison operators (<,<=,!=,==) compares only non NaN values. If any operand is NaN then returns false. By contrast, negations of comparisons (NLT, NGT) returns true if any operand is NaN value. This feature distinguish for example NGE from LT.

LANEID in description is lane id. TYPE is type of compared values (FLOAT for _FP32, DOUBLE for _FP64).

Sample instructions:
V_CMPX_LT_F32 VCC, V0, V1 # V0<V1 V_CMPSX_EQ_F32 VCC, V0, V1 # V0==V1, store result to EXEC, signal for any sNaN V_CMPX_LT_F64 VCC, V[2:3], V[4:5] # V[2:3]<V[4:5]

Table of integer comparison instructions by opcode:

Opcode range Instruction Description
128-135 (0x80-0x87) V_CMP_{OP8}_I32 Signed 32-bit values.
144-151 (0x90-0x97) V_CMPX_{OP8}_I32 Write result to EXEC. Signed 32-bit values.
160-167 (0xa0-0xa7) V_CMP_{OP8}_I64 Signed 64-bit values.
176-183 (0xb0-0xb7) V_CMPX_{OP8}_I64 Write result to EXEC. Signed 64-bit values.
192-199 (0xc0-0xc7) V_CMP_{OP8}_U32 Unsigned 32-bit values.
208-215 (0xd0-0xd7) V_CMPX_{OP8}_U32 Write result to EXEC. Unsigned 32-bit values.
224-231 (0xe0-0xe7) V_CMP_{OP8}_U64 Unsigned 64-bit values.
240-247 (0xf0-0xf7) V_CMPX_{OP8}_U64 Write result to EXEC. Unsigned 64-bit values.

Table of OP16 (compare operations) for integer values comparisons:

Opcode offset OP8 name Description
0 (0x0) F SDST(LANEID) = 0
1 (0x1) LT SDST(LANEID) = (TYPE)SRC0 < (TYPE)SRC1
2 (0x2) EQ SDST(LANEID) = (TYPE)SRC0 == (TYPE)SRC1
3 (0x3) LE SDST(LANEID) = (TYPE)SRC0 <= (TYPE)SRC1
4 (0x4) GT SDST(LANEID) = (TYPE)SRC0 > (TYPE)SRC1
5 (0x5) LG, NE SDST(LANEID) = (TYPE)SRC0 != (TYPE)SRC1
6 (0x6) GE SDST(LANEID) = (TYPE)SRC0 >= (TYPE)SRC1
7 (0x7) T SDST(LANEID) = 1

LANEID in description is lane id. TYPE is type of compared values (UINT32 for _U32, INT32 for _I32,...).

Sample instructions:
V_CMP_LT_U32 VCC, V0, V1 # V0<V1 V_CMPX_EQ_U32 VCC, V0, V1 # V0==V1, store result to EXEC, signal for any sNaN

Table of class instructions:

Tables of opcodes and their descriptions (GCN 1.2)

V_CMP_CLASS_F32

Opcode: 0x88
Syntax VOPC: V_CMP_CLASS_F32 VCC, SRC0, SRC1
Syntax VOP3: V_CMP_CLASS_F32 SDST, SRC0, SRC1
Operation: Check whether SSRC0 single floating point value belongs to one of specified class. Classes are specified as set bits in SRC1. If that condition is satisfied then store 1 to bit of SDST with number of current lane id, otherwise clear that bit. No flushing denormalized values for SRC0. List of classes:

Bit Description
0 Signaling NaN
1 quiet Nan
2 -INF
3 negative normalized value
4 negative dernormalized value
5 negative zero
6 positive zero
7 positive denormalized value
8 positive normalized value
9 +INF

V_CMPX_CLASS_F32

Opcode: 0x98
Syntax VOPC: V_CMPX_CLASS_F32 VCC, SRC0, SRC1
Syntax VOP3: V_CMPX_CLASS_F32 SDST, SRC0, SRC1
Operation: Check whether SSRC0 single floating point value belongs to one of specified class. Classes are specified as set bits in SRC1. If that condition is satisfied then store 1 to bit of SDST and EXEC with number of current lane id, otherwise clear that bit. No flushing denormalized values for SRC0. List of classes:

Bit Description
0 Signaling NaN
1 quiet Nan
2 -INF
3 negative normalized value
4 negative dernormalized value
5 negative zero
6 positive zero
7 positive denormalized value
8 positive normalized value
9 +INF

V_CMP_CLASS_F64

Opcode: 0xa8
Syntax VOPC: V_CMP_CLASS_F64 VCC, SRC0, SRC1(2)
Syntax VOP3: V_CMP_CLASS_F64 SDST, SRC0(2), SRC1(2)
Operation: Check whether SSRC0 double floating point value belongs to one of specified class. Classes are specified as set bits in SRC1. If that condition is satisfied then store 1 to bit of SDST with number of current lane id, otherwise clear that bit. No flushing denormalized values for SRC0. List of classes:

Bit Description
0 Signaling NaN
1 quiet Nan
2 -INF
3 negative normalized value
4 negative dernormalized value
5 negative zero
6 positive zero
7 positive denormalized value
8 positive normalized value
9 +INF

V_CMPX_CLASS_F64

Opcode: 0xb8
Syntax VOPC: V_CMPX_CLASS_F64 VCC, SRC0(2), SRC1(2)
Syntax VOP3: V_CMPX_CLASS_F64 SDST, SRC0(2), SRC1(2)
Operation: Check whether SSRC0 double floating point value belongs to one of specified class. Classes are specified as set bits in SRC1. If that condition is satisfied then store 1 to bit of SDST and EXEC with number of current lane id, otherwise clear that bit. No flushing denormalized values for SRC0. List of classes:

Bit Description
0 Signaling NaN
1 quiet Nan
2 -INF
3 negative normalized value
4 negative dernormalized value
5 negative zero
6 positive zero
7 positive denormalized value
8 positive normalized value
9 +INF