You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As a user can include their own -ccbin / --compiler-bindir flags that would conflict with our flags added to NVCC_PREPEND_FLAGS (see below), it would be good to add a check to only add our flags if they are not -ccbin / --compiler-bindir in NVCC_PREPEND_FLAGS
In the nvcc wrapper script, this was handled on this line. Though --compiler-bindir is not checked. So maybe that logic can be adapted and tweaked here.
As a user can include their own
-ccbin
/--compiler-bindir
flags that would conflict with our flags added toNVCC_PREPEND_FLAGS
(see below), it would be good to add a check to only add our flags if they are not-ccbin
/--compiler-bindir
inNVCC_PREPEND_FLAGS
cuda-nvcc-feedstock/recipe/activate.sh
Line 38 in f602ed9
In the
nvcc
wrapper script, this was handled on this line. Though--compiler-bindir
is not checked. So maybe that logic can be adapted and tweaked here.Edit: Originally came up in PR ( conda-forge/gromacs-feedstock#17 )
The text was updated successfully, but these errors were encountered: