DOCA DPACC Compiler

Contents:

This document describes DOCA DPACC compiler and instructions about DPA toolchain setup and usage.

DPACC is a high-level compiler for the DPA processor which compiles code targeted for the data-path accelerator (DPA) processor into a device executable and generates a DPA program.

The DPA program is a host library with interfaces encapsulating the device executable. This DPA program is linked with the host application to generate a host executable. The host executable can invoke the DPA code through FlexIO runtime API.

DPACC uses DPA compiler (dpa-clang) to compile code targeted for DPA. dpa-clang is part of the DPA toolchain package which is an LLVM-based cross-compiling bare-metal toolchain. It provides Clang compiler, LLD linker targeting DPA architecture, and other utilities.

Glossary

Offloading Work on DPA

To invoke a DPA function from host, the following things are required:

• DPA device code – C programs, targeted to run on the DPA. DPA device code may contain one or more entry functions.

• Host application code – the corresponding host application. Refer to DPA Subsystem for more details.

• Runtime – FlexIO or DOCA DPA library provides the runtime

The generated DPA program, when linked with a host application results in a host executable which also contains the device executable. The host application oversees loading the device executable on the device.

DPACC Predefined Macros

DPACC predefines the following macros:

Writing DPA Applications

DPA device code is a C code with some restrictions and special definitions.

FlexIO or DOCA-DPA APIs provide interfaces to DPA.

Language Support

The DPA is programmed using a subset of the C11 language standard. The compiler documents any constructs that are not available. Language constructs, where available, retain their standard definitions.

Restrictions on DPA Code

• Use of C thread local storage is not allowed for any variables

• Identifiers with _dpacc/__dpacc prefix are reserved by the compiler. Use of such identifiers may result in an error or undefined behavior

• DPA processor does not have native floating-point support; use of floating point operations is disabled

DPA RPC Functions

A remote procedure call function is a synchronous call that triggers work in DPA and waits for its completion. These functions return a type uint64_t value. They are annotated with a __dpa_rpc__ attribute.

DPA Global Functions

A DPA global function is an event handler device function referenced from the host code. These functions do not return anything. They are annotated with a __dpa_global__ attribute.

For more information, refer to DPA Subsystem documentation.

Characteristics of Annotated Functions

• Global functions must have void return type and RPC functions must have uint64_t return type

• Annotated functions cannot accept C pointers and arrays as arguments (e.g., void my_global (int *ptr, int arr[]))

• Annotated functions cannot accept a variable number of arguments

• Inline specifier is not allowed on annotated functions

Handling User-defined Data Types

User-defined data types, when used as global function arguments, require special handling. They must be annotated with a __dpa_global__ attribute.

If the user-defined data type is typedef'd, the typedef statement must be annotated with a __dpa_global__ attribute along the data type itself.

Characteristics of Annotated Types

• They must have a copy of the definition in all translation units where they are used as global function arguments

• They cannot have pointers, variable length arrays, and flexible arrays as members

• Fixed-size arrays as C structure members are supported

• These characteristics apply recursively to any user-defined/typedef'd types that are members of an annotated type

DPACC processes all annotated functions along with annotated types and generates host and device interfaces to facilitate the function launch.

DPA Intrinsics

DPA features such as fences and processor-specific instructions are exposed via intrinsics by the DPA compiler. All intrinsics defined in the header file dpaintrin.h are guarded by the DPA_INTRIN_VERSION_USED macro. The current DPA_INTRIN_VERSION is 1.3.

Example:

            

            
#define DPA_INTRIN_VERSION_USED (DPA_INTRIN_VERSION(1, 3))
#include <dpaintrin.h>
…
__dpa_thread_writeback_window();   // Fence for write barrier
        

For more information, refer to DPA Subsystem documentation.

DPACC Inputs and Outputs

DPACC can produce DPA programs in a single command by accepting all source files as input. DPACC also offers the flexibility of producing DPA object files or libraries from input files.

DPA object files contain both host stub objects (DPACC-generated interfaces) and device objects. These DPA object files can later be given to DPACC as input to produce the DPA library.

DPACC can accept the following file types as input:

Based on the mode of operations, DPACC can generate the following output files:

The following provides the commands to generate different kinds of supported output file types for each input file type:

dpacc -hostcc=<cc> -mcpu=<targets> in.c -o libprog.a

dpacc -hostcc=<cc> -mcpu=<targets> in.c -c

dpacc -hostcc=<cc> -mcpu=<targets> in.c -o lib<name> -gen-libs

dpacc -hostcc=<cc> -mcpu=<targets> in.dpa.o -o libprog.a

dpacc -hostcc=<cc> -mcpu=<targets> in.dpa.o -o lib<name> -gen-libs

dpacc -hostcc=<cc> -mcpu=<targets> in.a -o libprog.a

dpacc -hostcc=<cc> -mcpu=<targets> in.a -o lib<name> -gen-libs

DPA Program

When invoked in compile-and-link mode, dpacc produces a DPA program, which is a host library containing:

• DPACC-generated host stubs – Used to register the DPA application and facilitate invocation of the DPA entry-point from the host application.

• Device executable – Produced by compiling and linking input DPA device code.

The resulting DPA program library must be linked with a host application that uses the appropriate runtime APIs to load and execute the device code on the target hardware.

A DPA program may contain device executables for multiple hardware targets. These target-specific executables are packaged into a fatbinary container, which is embedded as a dedicated section within the host object that forms part of the host library.

DPA Object

When dpacc is run in compile-only mode, it produces a DPA object file—a host object with a structure similar to that of a DPA program. It includes:

• Host stubs – Generated by dpacc to enable future integration with host code.

• Device object – Compiled from the input DPA device code.

Like a DPA program, a DPA object may contain device objects for multiple hardware targets, encapsulated in a fatbinary container that is embedded within the host object file.

DPA Library

A DPA library is composed of two separate static archives:

• DPA host library – Contains host interface objects that correspond to the device objects found in the DPA device library.

• DPA device library – Contains device objects compiled from DPA device source code.

These libraries serve distinct roles:

• The DPA device library is consumed by dpacc during DPA program generation.

• The DPA host library can optionally be linked with other host-side code and redistributed.

The DPA device library is packaged as a specialized fatbinary container format, which itself contains archives of device objects for different targets.

The fatbinary container—illustrated by the dark-gray box in the associated diagram—is intended to be treated as an opaque object. It should not be manually extracted or modified by end users.

DPACC Trajectory

The following diagram illustrates DPACC compile-and-link mode trajectory.

Modes of Operation

In all modes described below, dpacc accepts one or more target names via the --mcpu option. The compiler then generates output that supports all specified targets, enabling multi-target deployment from a single build.

Compile-and-link Mode

This is a one-step mode that accepts C source files or DPA object files and produces the DPA program. Specifying the output library name is mandatory in this mode.

Example commands:

            

            
$ dpacc in1.c in2.c -o myLib1.a -hostcc=gcc -mcpu=nv-dpa-bf3                               # Takes C sources to produce myLib1.a library which supports a single target - nv-dpa-bf3
$ dpacc in3.dpa.o in4.dpa.o -o myLib2.a -hostcc=gcc -mcpu=nv-dpa-bf3,nv-dpa-cx8            # Takes DPA object files to produce myLib2.a library which supports multiple targets - nv-dpa-bf3 and nv-dpa-cx8
$ dpacc in1.c in3.dpa.o -o myLib3.a -hostcc=gcc -mcpu=nv-dpa-bf3,nv-dpa-cx7,nv-dpa-cx8     # Takes C source and DPA object to produce myLib3.a library which supports multiple targets - nv-dpa-bf3, nv-dpa-cx7 and nv-dpa-cx8
        

Compile-only Mode

This mode accepts C source code and produces .dpa.o object files. These files can be given to DPACC to produce the DPA program. The mode is invoked by the --compile or -c option.

The user can explicitly provide the output object file name using the --output-file or -o option.

Example commands:

            

            
$ dpacc -c input1.c -hostcc=gcc -mcpu=nv-dpa-cx7                                     # Produces input1.dpa.o which supports a single target - nv-dpa-cx7
$ dpacc -c input2.c -o myObj.dpa.o -hostcc=gcc -mcpu=nv-dpa-cx8,nv-dpa-cx7           # Produces myObj.dpa.o which supports multiple targets - nv-dpa-cx7 and nv-dpa-cx8
$ dpacc -c input3.c input4.c -hostcc=gcc -mcpu=nv-dpa-bf3,nv-dpa-cx7,nv-dpa-cx8      # Produces input3.dpa.o and input4.dpa.o which support multiple targets - nv-dpa-bf3, nv-dpa-cx7 and nv-dpa-cx8
        

Library Generation Mode

This mode accepts C source files or DPA object files and produces the DPA program. Specifying the output DPA library name is mandatory in this mode.

Example commands:

            

            
$ dpacc in1.c in2.c -o libdummy1 -hostcc=gcc -mcpu=nv-dpa-cx8 -gen-libs                                    # Takes C sources to produce a DPA-Library (libdummy1_host.a and libdummy_device.a archives) which supports a single target - nv-dpa-cx8
$ dpacc in3.dpa.o in4.dpa.o -o libdummy2 -hostcc=gcc -mcpu=nv-dpa-cx8,nv-dpa-bf3 -gen-libs                 # Takes DPA object files to produce a DPA-Library (libdummy2_host.a and libdummy2_device.a archives) which supports multiple targets - nv-dpa-bf3 and nv-dpa-cx8
$ dpacc in1.c in3.dpa.o -o outdir/libdummy3 -hostcc=gcc -mcpu=nv-dpa-bf3,nv-dpa-cx7,nv-dpa-cx8 -gen-libs   # Takes C source and DPA object to produce a DPA-Library (outdir/libdummy3_host.a and outdir/libdummy3_device.a archives) which supports multiple targets - nv-dpa-bf3, nv-dpa-cx7 and nv-dpa-cx8
        

To execute DOCA DPACC compiler:

            

            
Usage: dpacc <list-of-input-files> -hostcc=<path> -mcpu=<targets> [other options]
Helper Flags:
  -h, --help                          Print help information about DPACC
  -V, --version                       Print DPACC version information
  -v, --verbose                       List the compilation commands generated by this invocation while also executing every command in verbose mode
  -dryrun, --dryrun                   Only list the compilation commands generated by DPACC, without executing them
  -keep, --keep                       Keep all intermediate files that are generated during internal compilation steps in the current directory
  -keep-dir, --keep-dir               Keep all intermediate files that are generated during internal compilation steps in the given directory
  -optf, --options-file <file>,...    Include command line options from the specified file
        

Mandatory Arguments

Commonly Used Arguments

DPA Hardware Architectures

The table below outlines the supported DPA hardware architectures, including the corresponding values used with the compiler’s --mcpu option, as well as the predefined macros the compiler uses to identify each architecture.

Link Compatibility

Only relocatable objects that are link-compatible can be linked together. If incompatible objects are detected during linking, the toolchain will emit an error.

To ensure successful linking, the linker's toolchain version must match the version of the compiler used to produce the input objects.

If two architectures, A and B, are link-compatible, and B is newer than A, then:

• Valid: Objects built for A can be linked to produce an application targeting B.

• Invalid: Objects built for B cannot be linked to build an application for A.

If two architectures 'A' and 'B' are link-compatible and 'B' is newer than 'A', objects built for target 'A' can be linked to build an app for target 'B'. However, the inverse i.e. linking objects built for target 'B' to build an app for target 'A' is not valid.

For example, BlueField-3/ConnectX-7 and ConnectX-8 are link-compatible. This means that objects built for BlueField-3/ConnectX-7 can be linked together to produce an application targeting ConnectX-8.

Architecture Macros

The compiler defines architecture identifier macros for each supported DPA hardware version, as listed in the DPA Hardware Architectures section. Each macro is assigned a unique integer value, where:

• Newer DPA hardware generations are assigned strictly greater values than older ones.

• Known aliases (e.g., BlueField-3 and ConnectX-7) share the same macro value.

During compilation, the macro __NV_DPA is defined to reflect the current target architecture. This allows conditional compilation of device code based on the target hardware. For example:

            

            
#if __NV_DPA == __NV_DPA_BF3
// Code for Bluefield-3 here
#elif __NV_DPA > __NV_DPA_BF3
// Code for devices after Bluefield-3 here
#endif
        

LTO Usage Guidelines

Restrictions

• Only the default linker script is supported with LTO

• Using options -fPIC/-fpic/-shared/-mcmodel=large through -devicecc-options is not supported when LTO is enabled

• Fat bitcode objects containing both LLVM bitcode and ELF representation are not supported

• Thin LTO is not supported

Compatibility

During compilation, LLVM generates the object as bitcode IR (intermediate representation) when LTO is enabled instead of ELF representation. The bitcode IR generated by the DPA compiler is only guaranteed to be compatible within the same version. The toolchain version of the compiler which builds the objects involved in link-time optimization (enabled with -flto) and the toolchain version of the linker which performs LTO must be the same.

Deprecated Features

• The '-ldpa' option which links with DOCA-DPA libraries is deprecated and will be removed in future releases. The option '-ldoca_dpa' is to be used instead of '-ldpa'.

Examples

This section provides some common use cases of DPACC and showcases the dpacc command.

Building Libraries

This example shows how to build DPA libraries using DPACC. Libraries for DPA typically contain two archives, one for the host and one for the device.

            

            
dpacc input.c -hostcc=gcc -mcpu=nv-dpa-bf3 -o lib<name> -gen-libs -hostcc-options="-fPIC"
        

This command generates the output files lib<name>_host.a and lib<name>_device.a.

The host stub archive can be linked with other host code to generate a shared/static host library.

• Generating a static host library:

  Copy

  Copied!

              
  
              
  ar x lib<name>_host.a                   # Extract objects to generate *.o
  ar cr lib<name>.a <*src.host.o> *.o     # Generate final static archive with all objects
          

• Generating a shared host library:

  Copy

  Copied!

              
  
              
  gcc -shared -o lib<name>.so <*src.host.o> -Wl,-whole-archive -l<name>_host -Wl,-no-whole-archive       # Link the generated archive to build a shared library
          

Linking with DPA Device Library

The DPA device library generated by DPACC using -gen-libs as part of a DPA library can be consumed by DPACC using the -device-libs option.

            

            
dpacc input.c -hostcc=gcc -mcpu=nv-dpa-bf3 -o libInput.a -device-libs="-L <path-to-library> -l<libName>"
        

Enabling Link-time Optimizations

Link-time optimizations can be enabled using -flto along with an optimization level specified for device compilation.

            

            
dpacc input1.c -hostcc=gcc -mcpu=nv-dpa-bf3 -c -flto -devicecc-options="-O2" 
dpacc input2.c -hostcc=gcc -mcpu=nv-dpa-bf3 -c -flto -devicecc-options="-O2"   
dpacc -mcpu=nv-dpa-bf3 input1.dpa.o input2.dpa.o -hostcc=gcc -o libInput.a
        

Including Headers

This example includes headers for device compilation using devicecc-options and host compilation using hostcc-options. You may also specify headers for any compilation on both the host and device side using the -I option.

            

            
dpacc input.c -hostcc=gcc -mcpu=nv-dpa-bf3 -o libInput.a -I <common-headers-path> -devicecc-options="-I <device-headers-path>" -hostcc-options="-I <host-headers-path>"
        

Dumping Targets Supported by a Fatbinary File

The dpa-fatbin tool can be used to list the target architectures supported by a fatbinary file. This is especially useful for inspecting device archives in a DPA library.

Examples:

            

            
dpa-fatbin --list libfoo_device.a
dpa-fatbin --list device_exec.fatbin
        

Dumping Target of Device ELF File

To identify the target architecture for which a device ELF file was built, use the dpa-objdump tool:

Example:

            

            
dpa-objdump --file-headers foo.o
        

Generating Output as Source Code

dpacc supports a --src-output option that generates the output as host-side C source code. This can be compiled using a standard host compiler (e.g., gcc) to produce the same output that dpacc would normally emit directly.

The following are examples for each output type.

DPA Program Source

Generate DPA Program source:

            

            
dpacc input.c -hostcc=gcc -mcpu=nv-dpa-bf3 -o libfoo.c --src-output
        

Compile the generated source into an object file and archive it:

            

            
gcc libfoo.c -c -I /opt/mellanox/flexio/include \
  -Wno-attributes -Wno-pedantic -Wno-unused-parameter -Wno-return-type -Wno-implicit-function-declaration \
  -D__DPACC_SRC_TARGET__
 
ar cr libfoo.a libfoo.o
        

DPA Library Source

Generate DPA Library source:

            

            
dpacc input.c -hostcc=gcc -mcpu=nv-dpa-bf3 -o libfoo --gen-libs --src-output
        

This produces:

• libfoo_device.a – device archive

• libfoo.lib.c, input.dpa.c – host source files

Compile and archive the host library:

            

            
gcc libfoo.lib.c input.dpa.c -c -I /opt/mellanox/flexio/include \
  -Wno-attributes -Wno-pedantic -Wno-unused-parameter -Wno-return-type -Wno-implicit-function-declaration \
  -D__DPACC_SRC_TARGET__
 
ar cr libfoo_host.a libfoo.lib.o input.dpa.o
        

DPA Object Source

Generate DPA Object source:

            

            
dpacc input.c -hostcc=gcc -mcpu=nv-dpa-bf3 -c --src-output
        

This creates a single host source file input.dpa.c. Compile it using:

            

            
gcc input.dpa.c -c -I /opt/mellanox/flexio/include \
  -Wno-attributes -Wno-pedantic -Wno-unused-parameter -Wno-return-type -Wno-implicit-function-declaration
        

DPA Compiler Usage

The DPA Compiler is an LLVM-based backend used by dpacc to compile and link DPA device code. Users can pass additional options to the underlying compiler and linker using the following dpacc options:

• --devicecc-options – pass options to the device compiler

• --devicelink-options – pass options to the device linker

To find valid flags that can be passed through these options, refer to:

• Clang Compiler User's Manual

• Clang command line argument reference

• Target-dependent compilation options

• LLD command line reference

dpacc-extract Command Line Options

dpacc-extract is a tool for extracting a device executable out of a DPA program or a host executable containing DPA program(s).

To execute dpacc-extract:

            

            
Usage: dpacc-extract <input-file> -o=<output-file> [other options]
Helper Flags:
 
  -o, --output-file                   Specify name of the output file
  -app-name, --app-name <name>        Specify name of the DPA application to extract
  -mcpu, --mcpu <target>              Specify name of the device for which the application is to be extracted
  -h, --help                          Print help information about dpacc-extract
  -V, --version                       Print dpacc-extract version information
  -optf, --options-file <file>,...    Include command line options from the specified file
        

Mandatory arguments:

Objdump Command Line Options

The dpa-objdump utility prints the contents of object files and final linked images named on the command line.

For more information, please refer to the Objdump command line reference.

Archiver Command Line Options

dpa-ar is a Unix ar-compatible archiver.

For more information, please refer to the Archiver command line reference.

NM Tool Command Line Options

The dpa-nm utility lists the names of symbols from object files and archives.

For more information, please refer to the NM tool command line reference.

Miscellaneous Notes

• Object files produced by LLD are not compatible with those generated by other linkers (e.g., GNU ld). Mixing them may result in linker errors or runtime issues.

• Ensure that your host application includes at least one reference to the device entry-point function defined in the DPA Program. Otherwise, the host linker may silently discard the DPA Program during linking, treating it as unused code.

Changes in DPACC 1.10.0

New Features

• Support for fatbinaries where DPACC accepts multiple targets through mcpu option

• Enforced linking policy where only compatible objects can be linked together

• dpa-objdump infers the target automatically without the need to explicitly specify mcpu option

• Set FLEXIO_VER_USED in generated host stubs

• Support for new builtin __dpa_thread_l1_flush on nv-dpa-cx8 target

Limitations

• DPACC generates a warning about unknown target when building a DPA Library from DPA Objects produced by v1.9.0 or older