Skip to content

Openmvs 240 - #2074

Draft
DodgySpaniard wants to merge 8 commits into
masterfrom
openmvs-240
Draft

Openmvs 240#2074
DodgySpaniard wants to merge 8 commits into
masterfrom
openmvs-240

Conversation

@DodgySpaniard

@DodgySpaniard DodgySpaniard commented Aug 31, 2026

Copy link
Copy Markdown
Member

Bumping OpenMVS to v2.4.0 branch

  • Pinned a commit hash (v2.4.0) (instead of the 355 branch)
  • Updated pixi dependencies: cgal 5.6 -> 6, added nanoflann.
  • Added new arguments to ODM and forwarded them to preserve default behaviour from v2.2.0
  • Patching GPU flags for GPU compilation

AI Disclaimer: LLMs were used for research and testing

DodgySpaniard and others added 8 commits August 26, 2026 16:00
Pixi:
- Updating pixi dependencies.
- Runtime gperftools only, fixes from spwoodcock (commits 6c47d7f, 2038f0d)
- Added OpenCL (required for OpenSfM:v1.1.0+).
cmake:
- Removed aarch64 guard (nuked upstream).
- Linked to OpenSfM:osfm-v110
- Updated CMake install.
…uracy

opensfm no longer supports planar algo, and applies anisotropic GPS weighting
- Added flag to control tmalloc linking.
- Adding gperftools to all dependences (not linking it now).
 - Pined a commit hash (v2.4.0)
 - Updated pixi dependencies: cgal 5.6 -> 6, added nanoflann.
 - Added new arguments to preserver default behaviour from v2.2.0
We have the policy for empty variables:

  IF(POLICY CMP0104)
	# Initialize CMAKE_CUDA_ARCHITECTURES when CMAKE_CUDA_COMPILER_ID is NVIDIA. Raise an error if CUDA_ARCHITECTURES is empty.
	CMAKE_POLICY(SET CMP0104 NEW)
  ENDIF()

and then:

   UNSET(CMAKE_CUDA_ARCHITECTURES)

Setting:

   -DCMAKE_CUDA_ARCHITECTURES=OFF

Prevents the policy violation while preserving the intent.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants