Zha0q1 commented on a change in pull request #19174:
URL: https://github.com/apache/incubator-mxnet/pull/19174#discussion_r523295383
##########
File path: CMakeLists.txt
##########
@@ -60,6 +60,7 @@ cmake_dependent_option(USE_SSE "Build with x86 SSE
instruction support" ON
"CMAKE_SYSTEM_PROCESSOR STREQUAL x86_64 OR CMAKE_SYSTEM_PROCESSOR STREQUAL
amd64" OFF)
option(USE_F16C "Build with x86 F16C instruction support" ON) # autodetects
support if ON
option(USE_LAPACK "Build with lapack support" ON)
+option(USE_LAPACKE_INTERFACE "USE LAPACKE interface for lapack support" OFF)
Review comment:
Actually there might be a reason to keep this flag. So the initial
motivation to use the LAPACKE apis is that we want to support ilp64
openblas(cblas+lapacke). As I dig into the ci I think even after enabling large
tensor by default we can just use ilp64 openblas on some pipelines (ubuntu cpu
gpu) rather than all, because there are many different platform and build
combinations to tweak if we choose otherwise. So we can use the flag to keep
things how they are there.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
