FPR-to-GPR and GPR-to-FPR

Draft Status under development, for submission as an RFC

Links:

• https://bugs.libre-soc.org/show_bug.cgi?id=650
• https://bugs.libre-soc.org/show_bug.cgi?id=230#c71
• https://bugs.libre-soc.org/show_bug.cgi?id=230#c74
• https://bugs.libre-soc.org/show_bug.cgi?id=230#c76

Introduction:

High-performance CPU/GPU software needs to often convert between integers and floating-point, therefore fast conversion/data-movement instructions are needed. Also given that initialisation of floats tends to take up considerable space (even to just load 0.0) the inclusion of compact format float immediate is up for consideration using BF16

Libre-SOC will be compliant with the Scalar Floating-Point Subset (SFFS) i.e. is not implementing VMX/VSX, and with its focus on modern 3D GPU hybrid workloads represents an important new potential use-case for OpenPOWER.

Prior to the formation of the Compliancy Levels first introduced in v3.0C and v3.1 the progressive historic development of the Scalar parts of the Power ISA assumed that VSX would always be there to complement it. However With VMX/VSX not available in the newly-introduced SFFS Compliancy Level, the existing non-VSX conversion/data-movement instructions require load/store instructions (slow and expensive) to transfer data between the FPRs and the GPRs. For a 3D GPU this kills any modern competitive edge. Also, because SimpleV needs efficient scalar instructions in order to generate efficient vector instructions, adding new instructions for data-transfer/conversion between FPRs and GPRs multiplies the savings.

In addition, the vast majority of GPR <-> FPR data-transfers are as part of a FP <-> Integer conversion sequence, therefore reducing the number of instructions required to the minimum seems necessary.

Therefore, we are proposing adding:

• FPR load-immediate using BF16 as the constant
• FPR <-> GPR data-transfer instructions that just copy bits without conversion
• FPR <-> GPR combined data-transfer/conversion instructions that do Integer <-> FP conversions

If we're adding new Integer <-> FP conversion instructions, we may as well take this opportunity to modernise the instructions and make them well suited for common/important conversion sequences:

• standard Integer -> FP IEEE754 conversion (used by most languages and CPUs)
• standard OpenPower FP -> Integer conversion (saturation with NaN converted to minimum valid integer)
• Rust FP -> Integer conversion (saturation with NaN converted to 0)
• JavaScript FP -> Integer conversion (modular with Inf/NaN converted to 0)

The assembly listings in the appendix show how costly some of these language-specific conversions are: Javascript is 35 scalar instructions, including four branches.

FP -> Integer conversions

Different programming languages turn out to have completely different semantics for FP to Integer conversion. This section gives an overview of the different variants, listing the languages and hardware that implements each variant.

standard Integer -> FP conversion

This conversion is outlined in the IEEE754 specification. It is used by nearly all programming languages and CPUs. In the case of OpenPOWER, the rounding mode is read from FPSCR

standard OpenPower FP -> Integer conversion

This conversion, instead of exact IEEE754 Compliance, performs "saturation with NaN converted to minimum valid integer". This is also exactly the same as the x86 ISA conversion senantics. OpenPOWER however has instructions for both:

• rounding mode read from FPSCR
• rounding mode always set to truncate

Rust FP -> Integer conversion

For the sake of simplicity, the FP -> Integer conversion semantics generalized from those used by Rust's as operator will be referred to as Rust conversion semantics.

Those same semantics are used in some way by all of the following languages (not necessarily for the default conversion method):

• Rust's FP -> Integer conversion using the as operator
• Java's FP -> Integer conversion
• LLVM's llvm.fptosi.sat and llvm.fptoui.sat intrinsics
• SPIR-V's OpenCL dialect's OpConvertFToU and OpConvertFToS instructions when decorated with the SaturatedConversion decorator.

JavaScript FP -> Integer conversion

For the sake of simplicity, the FP -> Integer conversion semantics generalized from those used by JavaScripts's ToInt32 abstract operation will be referred to as JavaScript conversion semantics.

This instruction is present in ARM assembler as FJCVTZS https://developer.arm.com/documentation/dui0801/g/hko1477562192868

Other languages

TODO: review and investigate other language semantics

Proposed New Scalar Instructions

All of the following instructions use the standard OpenPower conversion to/from 64-bit float format when reading/writing a 32-bit float from/to a FPR. All integers however are sourced/stored in the GPR.

Integer operands and results being in the GPR is the key differentiator between the proposed instructions (the entire rationale) compated to existing Scalar Power ISA. In all existing Power ISA Scalar conversion instructions, all operands are FPRs, even if the format of the source or destination data is actually a scalar integer.

Note that source and destination widths can be overridden by SimpleV SVP64, and that SVP64 also has Saturation Modes in addition to those independently described here. SVP64 Overrides and Saturation work on both Fixed and Floating Point operands and results. The interactions with SVP64 are explained in the appendix

FPR to GPR moves

• fmvtg RT, FRA
• fmvtg. RT, FRA

move a 64-bit float from a FPR to a GPR, just copying bits directly. As a direct bitcopy, no exceptions occur and no status flags are set.

Rc=1 tests RT and sets CR0, exactly like all other Scalar Fixed-Point operations.

• fmvtgs RT, FRA
• fmvtgs. RT, FRA

move a 32-bit float from a FPR to a GPR, just copying bits. Converts the 64-bit float in FRA to a 32-bit float, then writes the 32-bit float to RT. Effectively, fmvtgs is a macro-fusion of frsp fmvtg and therefore has the exact same exception and flags behaviour of frsp

Unlike frsp however, with RT being a GPR, Rc=1 follows standard integer behaviour, i.e. tests RT and sets CR0.

GPR to FPR moves

fmvfg FRT, RA

move a 64-bit float from a GPR to a FPR, just copying bits. No exceptions are raised, no flags are altered of any kind.

Rc=1 tests FRT and sets CR1

fmvfgs FRT, RA

move a 32-bit float from a GPR to a FPR, just copying bits. Converts the 32-bit float in RA to a 64-bit float, then writes the 64-bit float to FRT. Effectively, fmvfgs is a macro-fusion of fmvfg frsp and therefore has the exact same exception and flags behaviour of frsp

Rc=1 tests FRT and sets CR1

TODO: clear statement on evaluation as to whether exceptions or flags raised as part of the FP conversion (not the int bitcopy part, the conversion part. the semantics should really be the same as frsp)

v3.0C section 4.6.7.1 states:

FPRF is set to the class and sign of the result, except for Invalid Operation Exceptions when VE=1.

Special Registers Altered:
  FPRF FR FI
  FX OX UX XX VXSNAN
  CR1 (if Rc=1)

Float load immediate (kinda a variant of fmvfg)

fmvis FRT, FI

Reinterprets FI << 16 as a 32-bit float, which is then converted to a 64-bit float and written to FRT. This is equivalent to reinterpreting FI as a BF16 and converting to 64-bit float.

There is no need for an Rc=1 variant because this is an immediate loading instruction. This frees up one extra bit in the X-Form format for packing a full BF16.

Example:

# clearing a FPR
fmvis f4, 0 # writes +0.0 to f4
# loading handy constants
fmvis f4, 0x8000 # writes -0.0 to f4
fmvis f4, 0x3F80 # writes +1.0 to f4
fmvis f4, 0xBF80 # writes -1.0 to f4
fmvis f4, 0xBFC0 # writes -1.5 to f4
fmvis f4, 0x7FC0 # writes +qNaN to f4
fmvis f4, 0x7F80 # writes +Infinity to f4
fmvis f4, 0xFF80 # writes -Infinity to f4
fmvis f4, 0x3FFF # writes +1.9921875 to f4

# clearing 128 FPRs with 2 SVP64 instructions
# by issuing 32 vec4 (subvector length 4) ops
setvli VL=MVL=32
sv.fmvis/vec4 f0, 0 # writes +0.0 to f0-f127

Important: If the float load immediate instruction(s) are left out, change all GPR to FPR conversion instructions to instead write +0.0 if RA is register 0, at least allowing clearing FPRs.

The above fits reasonably well with Minor 19 and follows the pattern shown by addpcis, which uses an entire column of Minor 19 XO. 15 bits of FI fit into bits 11 to 25, the top bit FI0 (MSB0 numbered 0) makes 16.

bf16 = FI0 || FI
fp32 = bf16 || [0]*16
FRT = Single_to_Double(fp32)

FPR to GPR conversions

X-Form:

Mode values:

fcvttgw RT, FRA, Mode

Convert from 64-bit float to 32-bit signed integer, writing the result to the GPR RT. Converts using mode Mode

fcvttguw RT, FRA, Mode

Convert from 64-bit float to 32-bit unsigned integer, writing the result to the GPR RT. Converts using mode Mode

fcvttgd RT, FRA, Mode

Convert from 64-bit float to 64-bit signed integer, writing the result to the GPR RT. Converts using mode Mode

fcvttgud RT, FRA, Mode

Convert from 64-bit float to 64-bit unsigned integer, writing the result to the GPR RT. Converts using mode Mode

fcvtstgw RT, FRA, Mode

Convert from 32-bit float to 32-bit signed integer, writing the result to the GPR RT. Converts using mode Mode

fcvtstguw RT, FRA, Mode

Convert from 32-bit float to 32-bit unsigned integer, writing the result to the GPR RT. Converts using mode Mode

fcvtstgd RT, FRA, Mode

Convert from 32-bit float to 64-bit signed integer, writing the result to the GPR RT. Converts using mode Mode

fcvtstgud RT, FRA, Mode

Convert from 32-bit float to 64-bit unsigned integer, writing the result to the GPR RT. Converts using mode Mode

GPR to FPR conversions

All of the following GPR to FPR conversions use the rounding mode from FPSCR.

fcvtfgw FRT, RA

Convert from 32-bit signed integer in the GPR RA to 64-bit float in FRT.

fcvtfgws FRT, RA

Convert from 32-bit signed integer in the GPR RA to 32-bit float in FRT.

fcvtfguw FRT, RA

Convert from 32-bit unsigned integer in the GPR RA to 64-bit float in FRT.

fcvtfguws FRT, RA

Convert from 32-bit unsigned integer in the GPR RA to 32-bit float in FRT.

fcvtfgd FRT, RA

Convert from 64-bit signed integer in the GPR RA to 64-bit float in FRT.

fcvtfgds FRT, RA

Convert from 64-bit signed integer in the GPR RA to 32-bit float in FRT.

fcvtfgud FRT, RA

Convert from 64-bit unsigned integer in the GPR RA to 64-bit float in FRT.

fcvtfguds FRT, RA

Convert from 64-bit unsigned integer in the GPR RA to 32-bit float in FRT.

FP to Integer Conversion Pseudo-code

Key for pseudo-code:

OpenPower conversion semantics (section A.2 page 999 (page 1023) of OpenPower ISA v3.1):

def fp_to_int_open_power<fp, int>(v: fp) -> int:
    if v is NaN:
        return int::MIN_VALUE
    if v >= int::MAX_VALUE:
        return int::MAX_VALUE
    if v <= int::MIN_VALUE:
        return int::MIN_VALUE
    return (int)rint(v, rounding_mode)

Rust conversion semantics (with adjustment to add non-truncate rounding modes):

def fp_to_int_rust<fp, int>(v: fp) -> int:
    if v is NaN:
        return 0
    if v >= int::MAX_VALUE:
        return int::MAX_VALUE
    if v <= int::MIN_VALUE:
        return int::MIN_VALUE
    return (int)rint(v, rounding_mode)

Section 7.1 of the ECMAScript / JavaScript conversion semantics (with adjustment to add non-truncate rounding modes):

def fp_to_int_java_script<fp, int>(v: fp) -> int:
    if v is NaN or infinite:
        return 0
    v = rint(v, rounding_mode)
    v = v mod int::VALUE_COUNT  # 2^32 for i32, 2^64 for i64, result is non-negative
    bits = (uint)v
    return (int)bits

Equivalent OpenPower ISA v3.0 Assembly Language for FP -> Integer Conversion Modes

Moved to appendix