pytorch: FTBFS on ppc64el

Source: pytorch
Version: 2.1.2+dfsg-4
Severity: serious
Tags: ftbfs

Hello,

pytorch FTBFS on ppc64el: Open bughttps://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1071655

*FAILED*:
caffe2/CMakeFiles/torch_cpu.dir/__/aten/src/ATen/native/Col2Im.cpp.o
/usr/bin/c++ -DAT_PER_OPERATOR_HEADERS -DCAFFE2_BUILD_MAIN_LIB
-DFMT_HEADER_ONLY=1 -DHAVE_MALLOC_USABLE_SIZE=1 -DHAVE_MMAP=1
-DHAVE_SHM_OPEN=1 -DHAVE_SHM_UNLINK=1
-DMINIZ_DISABLE_ZIP_READER_CRC32_CHECKS -DONNXIFI_ENABLE_EXT=1
-DONNX_ML=1 -DONNX_NAMESPACE=onnx -DTORCH_ENABLE_LLVM -DUSE_C10D_GLOO
-DUSE_DISTRIBUTED -DUSE_EXTERNAL_MZCRC -DUSE_RPC -DUSE_TENSORPIPE
-D_FILE_OFFSET_BITS=64 -Dtorch_cpu_EXPORTS
-I/home/merkys/pytorch-2.1.2+dfsg/build/aten/src
-I/home/merkys/pytorch-2.1.2+dfsg/aten/src
-I/home/merkys/pytorch-2.1.2+dfsg/build
-I/home/merkys/pytorch-2.1.2+dfsg
-I/home/merkys/pytorch-2.1.2+dfsg/cmake/../third_party/benchmark/include
-I/usr/lib/llvm-16/include -I/home/merkys/pytorch-2.1.2+dfsg/debian/foxi
-I/home/merkys/pytorch-2.1.2+dfsg/build/debian/foxi
-I/home/merkys/pytorch-2.1.2+dfsg/torch/csrc/api
-I/home/merkys/pytorch-2.1.2+dfsg/torch/csrc/api/include
-I/home/merkys/pytorch-2.1.2+dfsg/caffe2/aten/src/TH
-I/home/merkys/pytorch-2.1.2+dfsg/build/caffe2/aten/src/TH
-I/home/merkys/pytorch-2.1.2+dfsg/build/caffe2/aten/src
-I/home/merkys/pytorch-2.1.2+dfsg/build/caffe2/../aten/src
-I/home/merkys/pytorch-2.1.2+dfsg/torch/csrc
-I/home/merkys/pytorch-2.1.2+dfsg/third_party/miniz-2.1.0
-I/home/merkys/pytorch-2.1.2+dfsg/debian/kineto/libkineto/include
-I/home/merkys/pytorch-2.1.2+dfsg/debian/kineto/libkineto/src
-I/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/..
-I/home/merkys/pytorch-2.1.2+dfsg/c10/..
-I/home/merkys/pytorch-2.1.2+dfsg/third_party/flatbuffers/include
-isystem /home/merkys/pytorch-2.1.2+dfsg/build/third_party/gloo -isystem
/home/merkys/pytorch-2.1.2+dfsg/cmake/../third_party/gloo -isystem
/home/merkys/pytorch-2.1.2+dfsg/cmake/../third_party/googletest/googlemock/include
-isystem
/home/merkys/pytorch-2.1.2+dfsg/cmake/../third_party/googletest/googletest/include
-isystem /usr/include/opencv4 -isystem /usr/include/eigen3 -isystem
/home/merkys/pytorch-2.1.2+dfsg/caffe2 -Wdate-time -D_FORTIFY_SOURCE=2
-g -O2 -ffile-prefix-map=/home/merkys/pytorch-2.1.2+dfsg=.
-fstack-protector-strong -Wformat -Werror=format-security -gsplit-dwarf
-Wno-dangling-reference  -I/usr -D_GLIBCXX_USE_CXX11_ABI=1
-fvisibility-inlines-hidden -DUSE_KINETO -DLIBKINETO_NOCUPTI
-DLIBKINETO_NOROCTRACER -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC
-Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor
-Werror=range-loop-construct -Werror=bool-operation -Wnarrowing
-Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds
-Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function
-Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing
-Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic
-Wno-error=old-style-cast -Wno-invalid-partial-specialization
-Wno-unused-private-field -Wno-aligned-allocation-unavailable
-Wno-missing-braces -fdiagnostics-color=always -faligned-new
-Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno
-fno-trapping-math -Werror=format -Werror=cast-function-type
-Wno-stringop-overflow -O2 -g -DNDEBUG -std=gnu++17 -fPIC
-DCAFFE2_USE_GLOO -DTH_HAVE_THREAD -Wall -Wextra -Wno-unused-parameter
-Wno-unused-function -Wno-unused-result -Wno-missing-field-initializers
-Wno-unknown-pragmas -Wno-type-limits -Wno-array-bounds
-Wno-strict-overflow -Wno-strict-aliasing -Wno-missing-braces
-Wno-maybe-uninitialized -fvisibility=hidden -O2 -fopenmp -MD -MT
caffe2/CMakeFiles/torch_cpu.dir/__/aten/src/ATen/native/Col2Im.cpp.o -MF
caffe2/CMakeFiles/torch_cpu.dir/__/aten/src/ATen/native/Col2Im.cpp.o.d
-o caffe2/CMakeFiles/torch_cpu.dir/__/aten/src/ATen/native/Col2Im.cpp.o
-c /home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/native/Col2Im.cpp
In file included from
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/cpu/vec/vec256/vsx/vec256_common_vsx.h:8,
                  from
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/cpu/vec/vec256/vec256.h:19,
                  from
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/cpu/vec/vec.h:6,
                  from
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/native/cpu/utils.h:4,
                  from
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/native/im2col.h:7,
                  from
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/native/Col2Im.cpp:6:
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/cpu/vec/vec256/vsx/vec256_double_vsx.h:
In member function ‘at::vec::CPU_CAPABILITY::Vectorized<double>
at::vec::CPU_CAPABILITY::Vectorized<double>::acos() const’:
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/cpu/vec/vec256/vsx/vec256_double_vsx.h:220:14:
error: ‘Sleef_acosd2_u10’ was not declared in this scope; did you mean
‘Sleef_acoshf_u10’?
   220 |      return {Sleef_acosd2_u10(_vec0), Sleef_acosd2_u10(_vec1)};
       |              ^~~~~~~~~~~~~~~~
       |              Sleef_acoshf_u10
/home/merkys/pytorch-2.1.2+dfsg/aten/src/ATen/cpu/vec/vec256/vsx/vec256_double_vsx.h:220:62:
error: could not convert ‘{<expression error>, <expression error>}’ from
‘<brace-enclosed initializer list>’ to
‘at::vec::CPU_CAPABILITY::Vectorized<double>’
   220 |      return {Sleef_acosd2_u10(_vec0), Sleef_acosd2_u10(_vec1)};
       |                                                              ^
       |                                                              |
       |
<brace-enclosed initializer list>

This regression is still seeing more than a month. Can you provide any detail about this?

Thanks and regards,
Aditi