Files
openairinterface5g/openair1
Robert Schmidt f8d9c54eff Merge remote-tracking branch 'origin/multipath-testing-ci' into integration_2026_w07 (!3886)
Integrate GPU-Accelerated Channel Simulation + CI Test Pipeline

This MR builds on top of MR !3588 (feat: GPU Accelerated Channel
Simulation Pipeline).

The main portion of the changes is directly taken from MR !3588, and
focuses on integrating the GPU-accelerated channel simulation. On top of
that, this MR adds CI-related configurations to introduce a new channel
simulation testing pipeline into the CI --> RAN-Channel-Simulation

This MR introduces a CUDA-based pipeline to offload the
multipath_channel and add_noise channel simulation functions to the GPU.
This enables large-scale, real-time physical layer simulations not
possible on the CPU.

For full documentation, please see the file at openair1/SIMULATION/TOOLS/DOC/gpu_acceleration.md.

System & Compiler Requirements: An NVIDIA GPU with a compatible driver
and the CUDA Toolkit installed.

This feature was developed and tested on the GH200 server using the GCC
12 toolchain. To ensure reproducibility, building with GCC 12 is highly
recommended.  Build Instructions

The primary flag to enable this feature is -DCUDA_ENABLE=ON. The
following command shows a sample configuration used during development
on the GH200 server:

    ./build_oai --phy_simulators --cmake-opt "\
      -DCUDA_ENABLE=ON \
      -DUSE_UNIFIED_MEMORY=ON \
      -DUSE_ATS_MEMORY=OFF \
      -DCMAKE_C_COMPILER=/usr/bin/gcc-12 \
      -DCMAKE_CXX_COMPILER=/usr/bin/g++-12 \
      -DENABLE_TESTS=ON"

Memory Model Configuration

The pipeline supports multiple CUDA memory management modes, which can
be configured at compile time. ATS Memory (-DUSE_ATS_MEMORY=ON) is
enabled by default. To use another memory model, you must first disable
ATS.

For a detailed explanation of these models, please refer to Section 3 of
the gpu_acceleration.md report.
2026-02-11 10:21:44 +01:00
..
2026-02-05 13:33:03 +00:00

This folder contains the software for layer 1 of the OpenAirInterface eNB and UE as well as the unitary simulators

|-- PHY			   // contains all signal processing related to physical layer (used in real-time AND simulation)

General remarks: every directory contains at least defs.h (for declaration of structure and functions), vars.h (for definitions of variables), and extern.h (variables are defined only once in xxx-softmodem.c (for real-time HW operation) or a top-level simulation (in SIMULATION/xxx), if variables are needed in another file, use extern.h)

|   |-- CODING		   // turbo and convolutional (de-)coders, rate-matching, crc generation (implementation of 36-212)
|   |   `-- TESTBENCH      // This contains unitary testbenches for the different channel coders/decoders
|   |-- INIT		   // variables defined in vars.h are initialzed here (memory allocation)
|   |-- LTE_ESTIMATION     // estimation for LTE
|   |-- LTE_REFSIG 	   // reference signals for LTE (sync and pilot sequences) reference signals from 36-211 
|   |-- NR_REFSIG 	   // reference signals for NR (sync and pilot sequences) reference signals from 38-211
|   |-- LTE_TRANSPORT      // these are the top level routines for different transport and physical channels (for example DL-SCH, PSS) implements a subset of 36-211,36-212
|   |-- LTE_UE_TRANSPORT      // these are the top level routines for different transport and physical channels (for example DL-SCH, PSS) implements a subset of 36-211,36-212
|   |-- NR_TRANSPORT      // these are the top level routines for different transport and physical channels (for example DL-SCH, PSS) implements a subset of 38-211,38-212
|   |-- NR_UE_TRANSPORT      // these are the top level routines for different transport and physical channels (for example DL-SCH, PSS) implements a subset of 38-211,38-212
|   |-- NBIoT_TRANSPORT      // these are the top level routines for different transport and physical channels for NB-IoT
|   |-- MODULATION	   // modulation and demodulation includes FFT/SC-FDMA front-end processing from 36-211
|   |-- TOOLS		   // tools like (I)FFT, vector operations, matrix multiplication, etc.
|   |   `-- FFTTEST        // unitary testbench for FFT routines
|   |-- defs.h		  // top level defs.h (includes spec_defs, impl_defs and all defs.h of subdirectories)
|   |-- extern.h
|   |-- impl_defs.h       // non-LTE openair implementation
|   |-- impl_defs_top.h   // general implementaion defs
|   |-- spec_defs.h
|   |-- spec_defs_top.h
|   |-- types.h
|   `-- vars.h		  
|-- SCHED		  // schedules the different LTE eNB functions
|   |-- defs.h
|   |-- extern.h
|   |-- phy_procedures_lte_eNb.c // LTE PHY procedures for eNB (from 36-213) 
|   |-- phy_procedures_lte_common.c  // LTE PHY procedures common for UE and eNB (from 36-213)
|   |-- prach_procedures.c  //LTE PRACH  procedures (from 36-213)
|   |-- ru_procedures.c  //RU  procedures
|   |-- fapi_l1.c  //L1 side of the FAPI interface
|   |-- phy_mac_stub.c  //MAC stub that generates channels when used in phy-test-mode
|   |-- rt_compat.h
|   `-- vars.h
|-- SCHED_UE		  // schedules the different LTE UE functions
|   |-- phy_procedures_lte_ue.c  // LTE PHY procedures for UE (from 36-213)
|   |-- pucch_pc.c // power control for PUCCH
|   |-- pusch_pc.c // power control for PUSCH
|   |-- srs_pc.c // power control for SRS
|-- SCHED_NB_IOT		  // same as above for NB-IOT
|-- SCHED_NR		  // same as above for NR gNB
|-- SCHED_NR_UE		  // same as above for NR UE
|-- SIMULATION		 // contains simulation routines to test PHY
|   |-- LTE_PHY          // LTE simulation testbenches for unitary simulation of transport/physical channels
|   |   |-- pbchsim.c       // PBCH simulation testbench
|   |   |-- pdcchsim.c      // PCFICH/PHICH/PDCCH (DCI) simulation testbench
|   |   |-- dlsim.c         // PDSCH simulation testbench
|   |   |-- ulsim.c         // PUSCH simulation testbench
|   |   |-- pucchsim.c      // PUCCH simulation testbench
|   |-- RF  		    // RF simulation tools
|   |-- ETH_TRANSPORT
|   `-- TOOLS