1
67%
main: 67%

Ran 08 Aug 2025 10:51AM UTC

Files 35

Run time 206min

Badge

Embed ▾

Committed 08 Aug 2025 10:44AM UTC coverage: 27.219% (-0.3%) from 27.565%

Job # 16828462206.1

Build Type

push

github

Committed by

web-flow

Commit Message

Generic 2nd link for offloading (#196)

* Support for generic 2nd link pass that can be used to add new runtime libraries or entire object files that were only discovered during our thin-LTO phase (offloaded CUDA and TT kernels)

* Added CUDA kernel compilation support back into thinLTO stage with the other IR compilations

* Switched linker wrapper and new compile call to execvp to inherit env (needed for CPATH etc. with dev-docc builds)

 * silently accept unresolved symbols in first linking, we will always link a 2nd time to force the errors

* Moving secondary compilation (cuda kernels) to after generating the main region header

* Update sdfglib-auto and update thinLTO plugin-internal CUDA compilation to clang-19

* Add env-based option to stop double-linking for testing purposes

Run Details

1279 of 4699 relevant lines covered (27.22%)

7.08 hits per line

daisytuner / docc / 16828462206 / 1
67%
main: 67%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 16828462206.1

daisytuner / docc / 16828462206 / 1 67% main: 67%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 16828462206.1

daisytuner / docc / 16828462206 / 1
67%
main: 67%

README BADGES
x