Jetson Xavier NX ccminer

Jetson nanoでNiceHash Minerを試すHPをみてJetson Xavier NXでも動かせないか試そうとしていますが
$ sudo apt install git
$ cd
$ git clone -b linux GitHub - tpruvot/ccminer: CUDA Open Source miner project, for most nvidia cards
$ cd ccminer
$ ./build.sh
ここまで入力後に
In file included from ./crt/host_config.h:60:0,
from ./crt/host_config.h:60,
from ./cuda_runtime.h:100,
from ccminer.cpp:49:
./crt/host_defines.h:60:30: error: #include nested too deeply
#include “crt/host_defines.h”
^
util.cpp: In function ‘bool stratum_benchdata(json_t*, json_t*, int)’:
util.cpp:1697:21: error: ‘gpu_power’ was not declared in this scope
cgpu->gpu_power = gpu_power(cgpu); // mWatts
^~~~~~~~~
util.cpp:1700:2: error: ‘gpu_info’ was not declared in this scope
gpu_info(cgpu); // vid/pid
^~~~~~~~
util.cpp:1700:2: note: suggested alternative: ‘cgpu_info’
gpu_info(cgpu); // vid/pid
^~~~~~~~
cgpu_info
Makefile:1894: recipe for target ‘ccminer-util.o’ failed
make[2]: *** [ccminer-util.o] Error 1
make[2]: *** 未完了のジョブを待っています…
In file included from ./builtin_types.h:56:0,
from ./cuda_runtime.h:91,
from ccminer.cpp:49:
./device_types.h:66:6: error: use of enum ‘device_builtin’ without previous declaration
enum device_builtin cudaRoundMode
^~~~~~~~~~~~~~~~~~
./device_types.h:68:5: error: ‘cudaRoundNearest’ was not declared in this scope
cudaRoundNearest,
^~~~~~~~~~~~~~~~
./device_types.h:68:5: note: suggested alternative: ‘cudaRoundMode’
cudaRoundNearest,
^~~~~~~~~~~~~~~~
cudaRoundMode
./device_types.h:69:5: error: ‘cudaRoundZero’ was not declared in this scope
cudaRoundZero,
^~~~~~~~~~~~~
./device_types.h:69:5: note: suggested alternative: ‘cudaRoundMode’
cudaRoundZero,
^~~~~~~~~~~~~
cudaRoundMode
./device_types.h:70:5: error: ‘cudaRoundPosInf’ was not declared in this scope
cudaRoundPosInf,
^~~~~~~~~~~~~~~
./device_types.h:70:5: note: suggested alternative: ‘cudaRoundMode’
cudaRoundPosInf,
^~~~~~~~~~~~~~~
cudaRoundMode
./device_types.h:71:5: error: ‘cudaRoundMinInf’ was not declared in this scope
cudaRoundMinInf
^~~~~~~~~~~~~~~
./device_types.h:71:5: note: suggested alternative: ‘cudaRoundMode’
cudaRoundMinInf
^~~~~~~~~~~~~~~
cudaRoundMode
./device_types.h:66:25: error: scalar object ‘cudaRoundMode’ requires one element in initializer
enum device_builtin cudaRoundMode
^~~~~~~~~~~~~
In file included from ./driver_types.h:59:0,
from ./builtin_types.h:59,
from ./cuda_runtime.h:91,
from ccminer.cpp:49:
./vector_types.h:98:27: error: variable ‘device_builtin char1’ has initializer but incomplete type
struct device_builtin char1
^~~~~
./vector_types.h:100:5: error: expected primary-expression before ‘signed’
signed char x;
^~~~~~
./vector_types.h:100:5: error: expected ‘}’ before ‘signed’
./vector_types.h:100:5: error: expected ‘,’ or ‘;’ before ‘signed’
./vector_types.h:101:1: error: expected declaration before ‘}’ token
};
^
Makefile:1866: recipe for target ‘ccminer-ccminer.o’ failed
make[2]: *** [ccminer-ccminer.o] Error 1
mv -f .deps/ccminer-hefty1.Tpo .deps/ccminer-hefty1.Po
mv -f .deps/ccminer-pools.Tpo .deps/ccminer-pools.Po
make[2]: ディレクトリ ‘/home/yukio/ccminer’ から出ます
Makefile:2292: recipe for target ‘all-recursive’ failed
make[1]: *** [all-recursive] Error 1
make[1]: ディレクトリ ‘/home/yukio/ccminer’ から出ます
Makefile:693: recipe for target ‘all’ failed
make: *** [all] Error 2
となってしまいます。…同じエラーが続くので同じものは省かせてもらいました。

個別にファイルの配置が関係するのかと思いますが解決策などありますでしょうか?

We never tried that, may developers help to share experiences if they done something similar.

@8046yukio
It seems that ccminer is targeted to Discrete GPU devices [DGPU], while Jetson has Integrated GPU [iGPU]; They have different memory etc, that should be the cause of the failure.