Fix Amber patch target specification (#26687)
Co-authored-by: Scott McMillan <smcmillan@nvidia.com>
This commit is contained in:
parent
31b0fe21a2
commit
b937aa205b
@ -101,10 +101,10 @@ class Amber(Package, CudaPackage):
|
||||
sha256=checksum, level=0, when='@{0}'.format(ver))
|
||||
|
||||
# Patch to add ppc64le in config.guess
|
||||
patch('ppc64le.patch', when='@18: target=ppc64le')
|
||||
patch('ppc64le.patch', when='@18: target=ppc64le:')
|
||||
|
||||
# Patch to add aarch64 in config.guess
|
||||
patch('aarch64.patch', when='@18: target=aarch64')
|
||||
patch('aarch64.patch', when='@18: target=aarch64:')
|
||||
|
||||
# Workaround to modify the AmberTools script when using the NVIDIA
|
||||
# compilers
|
||||
|
Loading…
Reference in New Issue
Block a user