hipSOLVER precision support

hipSOLVER precision support#

This section provides an overview of the numerical precision types supported by the hipSOLVER library. hipSOLVER provides a consistent interface to linear algebra solvers that can run on AMD hardware.

This page lists the data types supported by the library itself and does not indicate hardware support. A type listed here is only usable if the GPU architecture also supports it; otherwise it is unsupported. For data type support across the other ROCm libraries and by GPU architecture, see the Data types and precision support page.

Supported data types overview#

The following table summarizes the input and output data types supported by hipSOLVER. For the precision-prefix naming convention and how precision appears in function signatures, see the sections that follow.

Icon

Definition

✅

Fully supported as both an input and output type.

⚠️

Partially supported as an input or output type.

Data types not listed in the table below are not supported.

Data type

Support

float32

✅

float64

✅

cfloat32

✅

cfloat64

✅

Supported precision types#

hipSOLVER supports four primary precision types across its functions:

Type prefix

C++ type

Description

s

float

Single-precision real (32-bit)

d

double

Double-precision real (64-bit)

c

hipFloatComplex

Single-precision complex (32-bit real, 32-bit imaginary)

z

hipDoubleComplex

Double-precision complex (64-bit real, 64-bit imaginary)

Function naming convention#

hipSOLVER follows the LAPACK naming convention where the first letter of the function name indicates the precision type:

  • Functions beginning with hipsolverS operate on single-precision real data.

  • Functions beginning with hipsolverD operate on double-precision real data.

  • Functions beginning with hipsolverC operate on single-precision complex data.

  • Functions beginning with hipsolverZ operate on double-precision complex data.

For example, the Cholesky factorization function potrf is implemented as:

  • hipsolverSpotrf - For single-precision real matrices

  • hipsolverDpotrf - For double-precision real matrices

  • hipsolverCpotrf - For single-precision complex matrices

  • hipsolverZpotrf - For double-precision complex matrices

In the documentation, these are sometimes represented generically as hipsolver<T>potrf(), where <T> is a placeholder for the precision type prefix.

Understanding precision in function signatures#

In the function signatures throughout the documentation, precision information is indicated directly in the parameter types. For example:

hipsolver_status hipsolverSpotrf(hipsolverHandle_t handle,
                                 hipsolverFillMode_t uplo,
                                 int n,
                                 float *A,
                                 int lda,
                                 float *work,
                                 int lwork,
                                 int *devInfo)

The parameter types (float, double, hipFloatComplex, or hipDoubleComplex) correspond to the function prefix and indicate the precision used by that specific function variant.

Real versus complex precision#

Some LAPACK-based functions in hipSOLVER have different behaviors or names when operating on real versus complex data:

  • Functions for symmetric matrices (prefix sy) use the same name for both real precision types.

  • Functions for Hermitian matrices (prefix he) are used for complex precision types.

  • Some auxiliary routines might be specific to real or complex precision types.

For example, hipsolverSsyevj and hipsolverDsyevj handle real symmetric matrices, while hipsolverCheevj and hipsolverZheevj handle complex Hermitian matrices.

Contents