Skip to content
GitLab
  • Menu
Projects Groups Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in / Register
  • sac2c sac2c
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 401
    • Issues 401
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 16
    • Merge requests 16
  • Deployments
    • Deployments
    • Releases
  • Wiki
    • Wiki
  • External wiki
    • External wiki
  • Activity
  • Graph
  • Create a new issue
  • Commits
  • Issue Boards
Collapse sidebar
  • sac-group
  • sac2csac2c
  • Issues
  • #2523
Closed
Open
Created Dec 02, 2025 by Daan Spijkers@daanrs

sac2c cuda does not support sm_52

I have a Nvidia 970, which has compute capabilities 5.2. However, there is no sm_52 target in setup/sac2crc.backend.cuda.in. I dont know if there is a specific reason for this, but I am unable to build sac2c with cuda support. I get the following error:

Error: You are trying to compile for the CUDA architecture 'sm_52`. The configuration files do not contain target 'arch_sm_52`. You may choose a different target from the compiler-specific configuration file '/home/daan/code/uni/sac2c-develop/build/sac2crc_p` or your local ones in '~/.sac2crc/`. You may also add a new target in a file in that folder, e.g. in '~/.sac2crc/sac2crc.mine`.
Compilation failed while initial, 1 error(s).
CMake Error at sac2c-variables.cmake:24 (MESSAGE):

  /home/daan/code/uni/sac2c-develop/build/sac2c_p;-target;cuda;-cti-message-length;0;-cti-single-line;-generic
  seems not to work, cannot determine SBI data, exiting...
Call Stack (most recent call first):
  /home/daan/code/uni/sac2c-develop/src/libsac/CMakeLists.txt:2 (INCLUDE)

If I add such a target then I can compile just fine, although I have not tested whether sac will actually work using cuda for me as a target.

To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information
Assignee
Assign to
Time tracking