Recent Content
System PLL of Agliex5 PCIE example design cannot be locked after configuration
Hi all, The device is Agilex 5 E series FPGA, development kit is plugged in main board via a x16 card edge. Both System PLL reference clock and PCIE AXI Stream Hard IP reference clock are driven from PCIE card edge. After power up, IO PLL locked but System PLL cannot be locked, PCIE hard IP remains in reset status, regardless the configuration method of FPGA (that is, via JTAG or QSPI flash). Here are my questions: 1、Is it valid for System PLL to have its reference clock driven from PCIE card edge？According to 4.1.1 of GTS AXI Streaming IP for PCI Express* User Guide: Agilex 5 and Agilex 3 FPGAs and SoCs (ug813754), reference clock for System PLL should from a independent and free-running local clock source. 2、If the answer of above question is positve, how should I debug to make the System PLL work? Best regards.
AshlingRISCFree IDE Build system: 'source directory does not appear to contain CMakeLists.txt"
Hello Altera Gurus, I am now having much trouble building my projects with the AshlingRISCFree IDE using a NIOSV/m processor on a MAX10 FPGA targeted at a MAX10 Development kit. I am using Quartus Standard edition 25.1 on a Windows 10 PC. The process i am following is this: I created an FPGA top level System Verilog file for a new MAX10 Project. created a Qsys Platform which has a NIOSV/m processor connected to a onchip RAM for program storage and a onchip RAM for the DMA data. Added one mSGDMA engine for transmit data operations Added a second mSGDMA engine for receivedata operations Added two RAM onchip memories for the DMA decsriptors and wired up everything created the address map and interrupt mapping saved and generated the Qsys platform In the top level SV file is loopback the Tx -> RX for the two mSGDMAs Used the niosv-bsp-editor in a niosv console to created a BSP using the Qsys sopcinfo file Generated the BSP, created a simple C main application to configure the mSGDMAs and NIOSV/ m processor etc. Imported both the HAL_BSP and HAL_APP folders using: 'Import NIOS-V CMake Project... In the AshlingRSICFree IDE i can successfully run a 'Build all' and compile the HAL_BSP. BUT When i select the hal_app folder and try to build the Project i get these error messages: 17:37:20 Buildscript generation: hal_app::Default in D:\VAREX_mSGDMA_Eval\software\hal_app\build\Default cmake -DCMAKE_EXPORT_COMPILE_COMMANDS:BOOL=ON -G "Unix Makefiles" "D:\\VAREX_mSGDMA_Eval\\software\\hal_app" CMake Warning: Ignoring extra path from command line: "D:\VAREX_mSGDMA_Eval\software\hal_app" CMake Error: The source directory "D:/VAREX_mSGDMA_Eval/software/hal_app" does not appear to contain CMakeLists.txt. Specify --help for usage, or press the help button on the CMake GUI. Problems : Description Resource Path Location Type CMake Error: The source directory "D:/VAREX_mSGDMA_Eval/software/hal_app" does not appear to contain CMakeLists.txt. hal_app de.marw.cdt.cmake.core.internal.CMakeErrorParser CMake Problem cmake exited with status 1. See CDT global build console for details. hal_app de.marw.cdt.cmake.core.internal.BuildscriptGenerator Buildscript Generation Problem Looking at the hal_bsl folder i can see the CMakeLists.txt is present, it is not present (automatically anyway) in the hal_app folder. I assume it would be if it was part of the BSP generate flow, but it ins't there so i assumed it shouldn't be there (in the hal_app folder i mean). Even if i add it manually then try to do a project build again i then see an error message saying the CmakeCache.txt file has not been created. This seems like a big tools flow mess to me. The Project is automatically setup to use a CMake Compile and CMake Build flow. But its not working. I am trying to use the AshlingRISC IDE GDB Debugger to load my ELF file to the NIOSV processor to allow me to debug my project, but of course because i can't even build it this is impossible. I have tried using niosv cli commands to build my ELF file ...and they seem to work, which means the AshlingRISC IDE is the culprit in the failed IDE build process: Here are my NIOSV cli commands : mSGDMA Test: $ niosv-app --bsp-dir=D:/VAREX_mSGDMA_Eval/software/hal_bsp --app-dir=D:/VAREX_mSGDMA_Eval/software/hal_app --srcs=D:/VAREX_mSGDMA_Eval/software/hal_app/msgdma_loopback.c $ cmake -S D:/VAREX_mSGDMA_Eval/software/hal_app -B D:/VAREX_mSGDMA_Eval/software/hal_app/build -G "Unix Makefiles" -DCMAKE_BUILD_TYPE=Debug $ make -j4 -C D:/VAREX_mSGDMA_Eval/software/hal_app/build After i try and fail to do a project build i can also no loner clean this project, it gets stuck in red with the same error. The only way i can get it back to the start state is : File -> Restart ...this is not great !! Does anybody know why i get these errors and how to fix them please ? : Here i have also linked to an older post here in the knowledge base Claiming that "This problem is fixed starting with the Intel® Quartus® Prime Pro Edition Software version 21.4. and later". This appears to be NOT the case though :) Why does CMake Error: The source directory "<project_directory>/intel_niosv_m_0_EXAMPLE_DESIGN" does not contain CMakeLists.txt. when compiling the Nios® V processor application in Command Line Interface? | Altera Community - 338917 NOTE: I have attached 3 screenshots, 2 show the output from the NIOSV CLI when i run the 3 compile commands. The third one shows what happens when i try to load the ELF file which was created after the 3 Compilation steps run to completion. It looks like the GDB debugger detects the NIOSV/m processor (the 1 hard message) and then promptly crashed during part of the boot process. Does anybody have any ideas about why that might be and what is going on please ? Thanks for any help, Dr Barry H
how to reduce clock skew between synchronous clock
I am working on Agilex 7 FPGA with quartus 25.3 software. In my project, there is 2 synchronous clocks that fast_clock is 416M and slow_clock is 208M. the clock scheme is ref_clk_100m -> IOPLL -> 416M -> clk_ctrl_div_ip -> 416M/208M. And there is data transfer between clk_fast and clk_slow. after compiling the whole design, I found there is large timing violation in the path that from clk_fast to clk_slow by timing report. from the clk_fast-clk_slow timing path, the clock skew has -700ps from clk_fast-clk_fast or clk_slow-clk_slow timing path, the clock skew is -5ps that is expected. so how to reduce clock skew between synchronous clock? and could you provide more suggestions about how to implement data transfer clk_2x to clk_1x?
Avalon-ST configuration with Agilex 3 fails
Hi, I have implemented a kind of passive serial programming from a CPU using SPI and a shift register. The signals nSTATUS, nCONFIG, CONF_DONE, READY, and VALID are directly controlled by GPIOs. This works well except after a power cycle. After the system has powered up, the programming fails — CONF_DONE does not go high. All retries afterward succeed. I already went through the debugging guidelines but couldn’t find an issue. However, I have observed two things: The nSTATUS pin follows exactly the timing of the nCONFIG pin during the first attempt. Normally, nSTATUS is delayed and goes high later. The CPU must finish the complete programming cycle before retrying; otherwise, the FPGA remains stuck in this erroneous state. I recorded some curves with a logic analyzer: full_timing.png: Power cycle First configuration cycle fails Retry works Another cycle also works 2_start.png: Beginning of cycle 2. Here, the nSTATUS pin follows exactly the timing of the nCONFIG pin. 2_3_restart.png: End of cycle 2 and beginning of cycle 3. 4_start.png: Another configuration cycle that works. Any idea what could cause this problem? Regards Samuel
Operating system kernel-level FPGA bridge communication
We are using a custom (Agilex 5) platform and need to access the FPGA bridges from the Linux kernel. We are unable to locate the corresponding device tree nodes or modifications required to access these memory-mapped regions. We're aware of the devmem2 package, but we want to access the FPGA from the kernel side without relying on it. Please guide us on how to configure the device tree and use kernel-level commands or interfaces to access the HPS-to-FPGA and Lightweight HPS-to-FPGA bridges. configuration from Linux. CONFIG_OF_RESOLVE y CONFIG_OF_OVERLAY y CONFIG_OF_CONFIGFS y CONFIG_FPGA_MGR_STRATIX10_SOC y CONFIG_FPGA_BRIDGE y CONFIG_FPGA_REGION y CONFIG_OF_FPGA_REGION y CONFIG_OVERLAY_FS yError (169008): Can't turn on open-drain option for differential I/O pin HPS_DDR3_DQS_N[1]
Hello. Please, suggest how to resolve this error. Quartus 25.1, Cyclone V, Start Fitter. inout [3:0] HPS_DDR3_DQS_N, inout [3:0] HPS_DDR3_DQS_P, .memory_mem_dqs (HPS_DDR3_DQS_P), // .mem_dqs .memory_mem_dqs_n (HPS_DDR3_DQS_N), // .mem_dqs_n
Recommended Quartus Prime Standard Edition for Nios V Development on MAX 10 FPGA (10M25DAF4817G)
Hi all, I am developing on a MAX 10 FPGA (specifically, the 10M25DAF4817G) using the Nios V processor. I need advice on the recommended Quartus Prime Standard Edition version for this workflow.Here is my situation and question: My Target FPGA: Intel MAX 10 (10M25DAF4817G). Reference Design: I started with the official: (Introduction • MAX® 10 FPGA - Helloworld on Nios® V/m Processor Design Example • Altera Documentation and Resources Center) . The documentation for this example states it is validated with Quartus Prime Standard Edition 23.1. My Experience: In Quartus Prime 23.1, I downloaded this example, made my modifications, and successfully got the design to work on my board.However, when I tried to migrate my project to Quartus Prime 25.1 and followed the same process (specifically, during the "Downloading the Software ELF File" step as per the 3. Hello World on MAX 10 FPGA 10M50 Evaluation Kit • AN 985: Nios V Processor Tutorial • Altera Documentation and Resources Center), I encountered some issues. [Quartus/Nios V] Nios V processor debug failure: "Could not halt the target: timeout occurred" with Quartus 25.1 generated SOF Given that the official design example is validated for 23.1, but a newer tool version (25.1) is available: What is the current community recommendation for the Quartus Prime Standard Edition version for stable Nios V development on MAX 10 FPGAs?Should I stick with 23.1 as the known stable version for my device family?Is 25.1 (or another version) now fully supported and recommended? If so, are there any known migration steps or workarounds for the ELF download issue? Any insights would be greatly appreciated. Thank you.
