Hello!
The check for sse2 (lines below) is failing on android/aarch64 (specifically Termux on Android). It incorrectly detects that SSE2 is available (which is most certainly is not). I suspect the problem lies in using clang-9 vs gcc-9...
I tested on an RPi running linux (aarch64) with GCC-9 and it seemed to detect properly.
|
else() |
|
set(CMAKE_POSITION_INDEPENDENT_CODE ON) |
|
check_cxx_compiler_flag("-msse2" SUPPORT_MSSE2) |
|
if(SUPPORT_MSSE2) |
Likewise, the Makefile makes an implicit assumption that one is running on Intel.
|
|
|
ifndef USE_SSE |
|
USE_SSE = 1 |
|
endif |
|
|
|
ifeq ($(USE_SSE), 1) |
|
CFLAGS += -msse2 |
|
endif |
This is in stark contrast to rabit where at least in the Makefile it detects architectures (although, I would argue that the detection logic should really be the other way around: if the detected ARCH is Intel, then + -msse2; else: -march=native)
https://github.com/dmlc/rabit/blob/74bf00a5ab4594f1695a8ea960394ce89f4a44d0/Makefile#L20-L25
Hello!
The check for sse2 (lines below) is failing on android/aarch64 (specifically Termux on Android). It incorrectly detects that SSE2 is available (which is most certainly is not). I suspect the problem lies in using clang-9 vs gcc-9...
I tested on an RPi running linux (aarch64) with GCC-9 and it seemed to detect properly.
dmlc-core/CMakeLists.txt
Lines 187 to 190 in ebcaa42
Likewise, the Makefile makes an implicit assumption that one is running on Intel.
dmlc-core/Makefile
Lines 20 to 27 in ebcaa42
This is in stark contrast to rabit where at least in the Makefile it detects architectures (although, I would argue that the detection logic should really be the other way around: if the detected ARCH is Intel, then + -msse2; else: -march=native)
https://github.com/dmlc/rabit/blob/74bf00a5ab4594f1695a8ea960394ce89f4a44d0/Makefile#L20-L25