hi all:
i got an error when i try to build vp, following is the error information:
"/lib/libnvdla_cmod.so: undefined reference to sc_core::sc_vector_base::make_name(char const*, unsigned long)' "/lib/libnvdla_cmod.so: undefined reference to sc_dt::sc_int_base::to_string(sc_dt::sc_numrep, bool) const'
collect2: error: ld returned 1 exit status"

I both use gcc 4.8.5 in hw and vp.

./nvdla_runtime --loadable ./VPtestfile/basic.nvdla --image ./VPtestfile/000456.jpg

creating new runtime context...
[ 76.686715] random: crng init done
Emulator starting
surface format: 46
dlaimg height: 375 x 500 x 3: LS: 1504 SS: 0 Size: 564000
[ 3223.246633] nvdla_runtime[1273]: unhandled level 3 translation fault (11) at 0xffff7aec1000, esr 0x92000047, in libc-2.23.so[ffff91760000+12c000]
[ 3223.247392] CPU: 0 PID: 1273 Comm: nvdla_runtime Tainted: G O 4.13.3 #1
[ 3223.247657] Hardware name: linux,dummy-virt (DT)
[ 3223.247878] task: ffff80003daf6200 task.stack: ffff80003ca70000
[ 3223.248266] PC is at 0xffff917d8494
[ 3223.248403] LR is at 0x409768
[ 3223.248521] pc : [<0000ffff917d8494>] lr : [<0000000000409768>] pstate: 20000000
[ 3223.250152] sp : 0000ffffe5a1cee0
[ 3223.250308] x29: 0000ffffe5a1cee0 x28: 0000000000000000
[ 3223.250510] x27: 0000000000000000 x26: 0000000000000000
[ 3223.250705] x25: 0000000000000000 x24: 0000000000000000
[ 3223.250899] x23: 0000000000000000 x22: 0000000000000000
[ 3223.251081] x21: 0000000000403408 x20: 0000000000000000
[ 3223.251239] x19: 0000ffffe5a1df2d x18: 0000ffffe5a1ce6c
[ 3223.251394] x17: 0000ffff917d8380 x16: 0000000000440348
[ 3223.251551] x15: 00000000fffffffc x14: 0000000000000000
[ 3223.251700] x13: 0000000000000000 x12: 0000000000000000
[ 3223.251855] x11: 000031a632a72f88 x10: 0000000000000000
[ 3223.252037] x9 : 0000000000000000 x8 : 0000000000000000
[ 3223.252192] x7 : 0000310532062ec7 x6 : 0000ffff7aec0ff0
[ 3223.252344] x5 : 0000000000000040 x4 : 0000000000000000
[ 3223.252503] x3 : 0000ffff7ad2e000 x2 : 0000000000425d00
[ 3223.254206] x1 : 0000ffff7a87e040 x0 : 0000ffff7ad2e000
Segmentation fault

Hi,

I have been having this issue for two days! when I run ./nvdla_runtime with flatbuffs, i get it. If I run it with nvdla loadble, it runs fine. can anyone please please help me?

according to the readme of vp, I have built a linux image (http://nvdla.org/vp.html#building-linux-kernel-for-nvdla-virtual-simulator). But when I continue to run the ./build/bin/aarch64_toplevel -c conf/aarch64_nvdla.lua, it report :

             SystemC 2.3.0-ASI --- Jun 20 2018 20:29:36
        Copyright (c) 1996-2012 by all Contributors,
        ALL RIGHTS RESERVED

No sc_log specified, will use the default setting
verbosity_level = SC_MEDIUM
bridge: tlm2c_elaborate..
toplevel: -device virtio-9p-device,fsdev=r,mount_tag=r: 'virtio-9p-device' is not a valid device model name

how to fix this?

thanks.

I got a link error as below:

[ 87%] Linking CXX executable aarch64_toplevel
/usr/bin/ld: CMakeFiles/aarch64_toplevel.dir/src/aarch64_toplevel.cpp.o: undefined reference to symbol '_ZNKSs4findEPKcm@@GLIBCXX_3.4'
//usr/lib/x86_64-linux-gnu/libstdc++.so.6: error adding symbols: DSO missing from command line
collect2: error: ld returned 1 exit status
CMakeFiles/aarch64_toplevel.dir/build.make:106: recipe for target 'aarch64_toplevel' failed
make[2]: *** [aarch64_toplevel] Error 1
CMakeFiles/Makefile2:111: recipe for target 'CMakeFiles/aarch64_toplevel.dir/all' failed
make[1]: *** [CMakeFiles/aarch64_toplevel.dir/all] Error 2
Makefile:149: recipe for target 'all' failed
make: *** [all] Error 2

Who can tell me what happens? Where findEPK defines?
Thanks a lot!

/vp# ./build/bin/aarch64_toplevel -c conf/aarch64_nvdla.lua

         SystemC 2.3.0-ASI --- Jul 22 2018 23:30:28
    Copyright (c) 1996-2012 by all Contributors,
    ALL RIGHTS RESERVED

No sc_log specified, will use the default setting
verbosity_level = SC_MEDIUM
DMI mode enable
段错误 (核心已转储)
??????

/home/nicolast0604/vp/hw/outdir/nv_full/cmod/release/lib/libnvdla_cmod.so: undefined reference to `sc_core::sc_vector_base::make_name(char const*, unsigned long)'
collect2: error: ld returned 1 exit status
CMakeFiles/aarch64_toplevel.dir/build.make:106: recipe for target 'aarch64_toplevel' failed
make[2]: *** [aarch64_toplevel] Error 1
CMakeFiles/Makefile2:111: recipe for target 'CMakeFiles/aarch64_toplevel.dir/all' failed
make[1]: *** [CMakeFiles/aarch64_toplevel.dir/all] Error 2
Makefile:149: recipe for target 'all' failed
make: *** [all] Error 2

Hi,

I am trying to run virtual simulator with petalinux image and rootfs. I have given the path of petalinux image and rootfs in aarch64_nvdla.lua file. when running ./aarch64_toplevel -c conf/aarch64_nvdla.lua, I am getting following error,

         SystemC 2.3.0-ASI --- Oct 21 2019 12:36:09
    Copyright (c) 1996-2012 by all Contributors,
    ALL RIGHTS RESERVED

No sc_log specified, will use the default setting
verbosity_level = SC_MEDIUM
DMI mode enable
RAM base address: 0xc0000000
bridge: tlm2c_elaborate..
rom: requested regions overlap (rom dtb. free=0x000000004b80f200, addr=0x0000000048000000)
toplevel: rom check and register reset failed

Can anyone help if you know about this issue? thanks in advance.

Hello,
I have built the VP for nv_full and for nv_small project and successfully booted with prebuilt Linux from: https://github.com/nvdla/sw/tree/master/prebuilt/linux .
After that i have executed mount -t 9p -o trans=virtio r /mnt command and started the nvdla_bdma_mmio test. This test results with systemC crash:

./aarch64_nvdla_bdma_mmio

Start programming...
Write reg 0x00004004, value 0x00000000
Write reg 0x00004000, value 0xc0000000
Write reg 0x0000400c, value 0x00000000
Write reg 0x00004008, value 0xc00017c0
Write reg 0x0000402c, value 0x00000800
Write reg 0x00004020, value 0x00000100
Write reg 0x00004028, value 0x00000800
Write reg 0x0000401c, value 0x00000100
Write reg 0x00004024, value 0x00000000
Write reg 0x00004018, value 0x00000007
Write reg 0x00004010, value 0x00000007
Write reg 0x00004014, value 0x00000003
Write reg 0x00004030, value 0x00000001
Write reg 0x00004034, value 0x00000001
Finish programming...

Error: (E115) sc_signal cannot have more than one driver:
signal nvdla.nvdla_core.nvdla_core.bdma.signal_2' (sc_signal) first driver nvdla.nvdla_core.nvdla_core.bdma.bdma_core.ResetThread' (sc_method_process)
second driver `nvdla.nvdla_core.nvdla_core.bdma.bdma_core.ReadRequestSequenceGeneratorThread' (sc_thread_process)
In file: sc_signal.cpp:75
In process: nvdla.nvdla_core.nvdla_core.bdma.bdma_core.ReadRequestSequenceGeneratorThread @ 77500 ms

Can You help?

Hi,

I'm using the docker image provided on DockerHub to run the virtual platform.
When I try to use nvdla_compiler to generate a loadable from my LeNet model, it crashes with this message:

./nvdla_compiler --prototxt lenet.prototxt --caffemodel lenet_iter_10000.caffe
model --configtarget nv_small --cprecision int8 --profile basic --calibtable cal
ib_table.json
./nvdla_compiler: line 2: syntax error: unexpected redirection
# ./nvdla_compiler: line 1: ELF�����: not found

If I compile my model on my machine using the nvdla_compiler from nvdla/sw/prebuilt/linux, it compiles into a loadable with no problem.
Unfortunately when I try to use that model in the nvdla_runtime in the docker vp, the runtime starts running but then crashes and exits the QEMU instance:

# ./nvdla_runtime --loadable nvdla/lenet_model_nosoftmax_nocalib_basic.nvdla --i
mage nvdla/digits/four_inv.pgm
creating new runtime context...
[  460.522980] random: crng init done
Emulator starting
pgm2dimg 1 28 28 1 224 6272 401408
submitting tasks...
[  465.391729] Enter:dla_read_network_config
[  465.392504] Exit:dla_read_network_config status=0
[  465.392763] Enter: dla_initiate_processors
[  465.393181] Enter: dla_submit_operation
[  465.393407] Prepare Convolution operation index 0 ROI 0 dep_count 1
[  465.393722] Enter: dla_prepare_operation
[  465.394232] processor:Convolution group:0, rdma_group:0 available
[  465.394657] Enter: dla_read_config
[  465.396656] Exit: dla_read_config
[  465.396903] Exit: dla_prepare_operation status=0
[  465.397186] Enter: dla_program_operation
[  465.397424] Program Convolution operation index 0 ROI 0 Group[0]
root@545d7e33f93c:/usr/local/nvdla# 

Any suggestions on getting the compiler and the runtime to work in docker?

Thank you

Hi,

I am new to using Github and need your help.

Under vp/libs/ directory, 3 of the subdirectories (greenlib/, qbox/ & tlm2c) seem to be greyed out and cannot navigate further into them. However I'm able to get into log/ directory. Even the downloaded image has empty directories for these 3. Due to this, cmake build fails with errors saying these directories doesn't contain CMakeLists.txt file.

Am I missing something here? Can you please let me know, how do I resolve this? Thanks much.

Regards
Kishore

hi,
I have build the virtual platform according to NVDLA document and run nvdla_bdma_mmio successfully. But when I debugged nvdla_bdma_mmio.c with gdb, there is something wrong.

According to NVDLA document, i build virtual platform again for GDB debug. 

"-DCMAKE_BUILD_TYPE=Debug" option was added to cmake command.

after this, I attached aarch64_toplevel process successfully, but when I run the program in GDB command-line, i got below errors:
starting program: .../tests/nvdla_bdma_mmio/aarch64_nvdla_bdma_mmio
/bin/bash: .../tests/nvdla_bdma_mmio/aarch64_nvdla_bdma_mmio: cannot execute binary file: Exec format error
...

Thank you very much and i sincerely look forward to you reply.

Is it possible to find out the inference time of the CNN model using vp at the momemt?

hi,

1. when i run VP simulation following (http://nvdla.org/vp.html#running-the-virtual-simulator-from-docker) , which nvdla hardware configuration is based on for this VP from docker ? (nv_small/nv_large/nv_full?)
2. when i run ./nvda_runtime, it can pass but with no register access information. according to (http://nvdla.org/vp.html#debugging-the-virtual-simulator) , we need enable debug feature when build the VP. does the VP version from docker support debug feature ? if yes, how can i enable it ?

Hello,

I am trying to install VP and I have been successful till step 2.4.2 of VP.

When I run "make" I get these errors:

make [ 0%] Building CXX object libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/__/src/greenreg.cpp.o In file included from /usr/local/systemc-2.3.0/include/sysc/packages/boost/mem_fn.hpp:25:0, from /usr/local/systemc-2.3.0/include/sysc/packages/boost/bind.hpp:26, from /usr/local/systemc-2.3.0/include/sysc/kernel/sc_boost.h:44, from /usr/local/systemc-2.3.0/include/sysc/kernel/sc_dynamic_processes.h:36, from /usr/local/systemc-2.3.0/include/systemc:124, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/sysc_ext/utils/gr_report.h:50, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:42, from /home/me/vp/libs/greenlib/greenreg/src/greenreg.cpp:35: /usr/local/systemc-2.3.0/include/sysc/packages/boost/get_pointer.hpp:21:40: warning: ‘template<class> class std::auto_ptr’ is deprecated [-Wdeprecated-declarations] template<class T> T * get_pointer(std::auto_ptr<T> const& p) ^~~~~~~~ In file included from /usr/include/c++/7/memory:80:0, from /usr/local/systemc-2.3.0/include/sysc/packages/boost/get_pointer.hpp:8, from /usr/local/systemc-2.3.0/include/sysc/packages/boost/mem_fn.hpp:25, from /usr/local/systemc-2.3.0/include/sysc/packages/boost/bind.hpp:26, from /usr/local/systemc-2.3.0/include/sysc/kernel/sc_boost.h:44, from /usr/local/systemc-2.3.0/include/sysc/kernel/sc_dynamic_processes.h:36, from /usr/local/systemc-2.3.0/include/systemc:124, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/sysc_ext/utils/gr_report.h:50, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:42, from /home/me/vp/libs/greenlib/greenreg/src/greenreg.cpp:35: /usr/include/c++/7/bits/unique_ptr.h:51:28: note: declared here template<typename> class auto_ptr; ^~~~~~~~ In file included from /home/me/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_extension_support_base.h:144:0, from /home/me/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_bind_checker.h:24, from /home/me/vp/libs/greenlib/greensocket/include/greensocket/initiator/green_initiator_socket_base.h:28, from /home/me/vp/libs/greenlib/greensocket/include/greensocket/initiator/single_socket.h:26, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/generic_transaction.h:29, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:32, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45, from /home/me/vp/libs/greenlib/greenreg/src/greenreg.cpp:35: /home/me/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_extension_support_base.tpp: In static member function ‘static T* gs::socket::extension_support_base<TRAITS>::get_extension(const gs::ext::gs_data_id&, gs::socket::extension_support_base<TRAITS>::payload_type&)’: /home/me/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_extension_support_base.tpp:127:29: error: expected primary-expression before ‘>’ token T* tmp=txn.get_extension<T>(); //get the pointer ^ /home/me/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_extension_support_base.tpp:127:31: error: expected primary-expression before ‘)’ token T* tmp=txn.get_extension<T>(); //get the pointer ^ In file included from /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/config.h:50:0, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:36, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45, from /home/me/vp/libs/greenlib/greenreg/src/greenreg.cpp:35: /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/cnf_api_if.h: At global scope: /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/cnf_api_if.h:237:65: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated] virtual sc_event& getUpdateEvent(const std::string &parname) throw (RegisterObserverFailedException) = 0; ^~~~~ /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/cnf_api_if.h:325:40: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated] virtual sc_event& getNewParamEvent() throw (RegisterObserverFailedException) = 0; ^~~~~ In file included from /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/config.h:56:0, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:36, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45, from /home/me/vp/libs/greenlib/greenreg/src/greenreg.cpp:35: /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/GCnf_Api_t.h:1013:57: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated] sc_event& getUpdateEvent(const std::string &parname) throw (RegisterObserverFailedException) { ^~~~~ /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/GCnf_Api_t.h:1134:32: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated] sc_event& getNewParamEvent() throw (RegisterObserverFailedException) { ^~~~~ In file included from /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/config.h:57:0, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:36, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45, from /home/me/vp/libs/greenlib/greenreg/src/greenreg.cpp:35: /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_private_Api/gcnf_private_api_t.h:486:59: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated] sc_event& getUpdateEvent(const std::string &parname) throw (RegisterObserverFailedException) { ^~~~~ /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_private_Api/gcnf_private_api_t.h:506:34: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated] sc_event& getNewParamEvent() throw (RegisterObserverFailedException) { ^~~~~ In file included from /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_class.h:33:0, from /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/config.h:71, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:36, from /home/me/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53, from /home/me/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45, from /home/me/vp/libs/greenlib/greenreg/src/greenreg.cpp:35: /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_native.hpp: In member function ‘gs::cnf::gs_param<bool>::my_type& gs::cnf::gs_param<bool>::operator++()’: /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_operator_macros.h:79:7: warning: use of an operand of type ‘bool’ in ‘operator++’ is deprecated [-Wdeprecated] ++tmp_val; \ ^ /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_native.hpp:261:3: note: in expansion of macro ‘GC_SPECIALISATIONS_INCREMENT_OPERATORS’ GC_SPECIALISATIONS_INCREMENT_OPERATORS; ^ /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_native.hpp: In member function ‘gs::cnf::gs_param<bool>::val_type gs::cnf::gs_param<bool>::operator++(int)’: /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_operator_macros.h:89:16: warning: use of an operand of type ‘bool’ in ‘operator++’ is deprecated [-Wdeprecated] tmp_set_val++; \ ^ /home/me/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_native.hpp:261:3: note: in expansion of macro ‘GC_SPECIALISATIONS_INCREMENT_OPERATORS’ GC_SPECIALISATIONS_INCREMENT_OPERATORS; ^ libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/build.make:62: recipe for target 'libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/__/src/greenreg.cpp.o' failed make[2]: *** [libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/__/src/greenreg.cpp.o] Error 1 CMakeFiles/Makefile2:1810: recipe for target 'libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/all' failed make[1]: *** [libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/all] Error 2 Makefile:151: recipe for target 'all' failed make: *** [all] Error 2

Any ideas how to resolve this? The error start like this:

green_socket_extension_support_base.tpp:127:29: error: expected primary-expression before ‘>’ token

Using "git submodule update --init --recursive" update libs.

Cloning into 'libs/tlm2c'...
fatal: unable to access 'https://git.greensocs.com/tlm/tlm2c.git/': Failed to connect to git.greensocs.com port 443: Connection refused
fatal: clone of 'https://git.greensocs.com/tlm/tlm2c.git' into submodule path 'libs/tlm2c' failed

Not sure if somebody meets this issue also?

$ make
[ 5%] Built target tlm2c
[ 6%] No force-build step for 'qbox'
[ 6%] Performing build step for 'qbox'
[ 6%] Performing install step for 'qbox'
[ 8%] Completed 'qbox'
[ 13%] Built target qbox
[ 68%] Built target greenreg
[ 70%] Building CXX object fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/src/nvdla_cosim_sc_wrapper.cpp.o
In file included from /usr/local/systemc-2.3.0/include/sysc/packages/boost/mem_fn.hpp:25:0,
from /usr/local/systemc-2.3.0/include/sysc/packages/boost/bind.hpp:26,
from /usr/local/systemc-2.3.0/include/sysc/kernel/sc_boost.h:44,
from /usr/local/systemc-2.3.0/include/sysc/kernel/sc_dynamic_processes.h:36,
from /usr/local/systemc-2.3.0/include/systemc:124,
from /usr/local/systemc-2.3.0/include/systemc.h:246,
from /home/liyang/Github/NVDLA/vp/fpga/aws-fpga/cosim_sc_wrapper/include/cosim_sc_wrapper/nvdla_cosim_sc_wrapper.h:14,
from /home/liyang/Github/NVDLA/vp/fpga/aws-fpga/cosim_sc_wrapper/src/nvdla_cosim_sc_wrapper.cpp:11:
/usr/local/systemc-2.3.0/include/sysc/packages/boost/get_pointer.hpp:21:40: error: ‘template class std::auto_ptr’ is deprecated [-Werror=deprecated-declarations]
template T * get_pointer(std::auto_ptr const& p)
^
In file included from /usr/include/c++/5/memory:81:0,
from /usr/local/systemc-2.3.0/include/systemc.h:53,
from /home/liyang/Github/NVDLA/vp/fpga/aws-fpga/cosim_sc_wrapper/include/cosim_sc_wrapper/nvdla_cosim_sc_wrapper.h:14,
from /home/liyang/Github/NVDLA/vp/fpga/aws-fpga/cosim_sc_wrapper/src/nvdla_cosim_sc_wrapper.cpp:11:
/usr/include/c++/5/bits/unique_ptr.h:49:28: note: declared here
template class auto_ptr;
^
cc1plus: all warnings being treated as errors
fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/build.make:62: recipe for target 'fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/src/nvdla_cosim_sc_wrapper.cpp.o' failed
make[2]: *** [fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/src/nvdla_cosim_sc_wrapper.cpp.o] Error 1
CMakeFiles/Makefile2:3101: recipe for target 'fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/all' failed
make[1]: *** [fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/all] Error 2
Makefile:151: recipe for target 'all' failed
make: *** [all] Error 2

I have installed systemc-2.3.0. I'm did something wrong?

clone of neither git://git.qemu.org/QemuMacDrivers.git nor https://git.qemu.org/QemuMacDrivers.git failed.
I notice there is a thread talking about the problem. clone

They seemed to fix the bug however, I still could not fetch it.
May I know whether could I change the qbox repository to the github mirror?

revise the qbox/.gitsubmodule url to something like 'https://github.com/qemu/sgabios.git'

I got this output when i executed the command: tools/bin/tmake -build cmod_top

make: Entering directory '/root/vp/hw/spec/defs'
/home/utils/gcc-4.9.3/bin/cpp -undef -nostdinc -P -C nv_full.spec -o /root/vp/hw/outdir/nv_full/spec/defs/project.def
make: /home/utils/gcc-4.9.3/bin/cpp: Command not found
make: *** [/root/vp/hw/outdir/nv_full/spec/defs/project.def] Error 127
Makefile:22: recipe for target '/root/vp/hw/outdir/nv_full/spec/defs/project.def' failed
make: Leaving directory '/root/vp/hw/spec/defs'
make: Entering directory '/root/vp/hw/spec/manual'
/home/utils/java/jdk1.8.0_131/bin/java -jar Ordt.jar -parms test.parms -systemverilog /root/vp/hw/outdir/nv_full/spec/manual/sv/ -verilog /root/vp/hw/outdir/nv_full/spec/manual/regs_v.v -uvmregs /root/vp/hw/outdir/nv_full/spec/manual/regs_ral.sv -cppmod /root/vp/hw/outdir/nv_full/spec/manual/cmod -cppdrvmod /root/vp/hw/outdir/nv_full/spec/manual/dmod test.rdl
make: /home/utils/java/jdk1.8.0_131/bin/java: Command not found
make: *** [default] Error 127
Makefile:20: recipe for target 'default' failed
make: Leaving directory '/root/vp/hw/spec/manual'
make: Entering directory '/root/vp/hw/cmod'
mkdir -p ../outdir/nv_full/cmod/../cmod/nvdla_payload/
/home/utils/gcc-4.9.3/bin/g++ -I./include -I../cmod/include/nvdla_ness_header -I../cmod/include -I../cmod/nvdla_payload -I../cmod/bdma -I../cmod/cacc -I../cmod/cbuf -I../cmod/cdma -I../cmod/cdp -I../cmod/cmac -I../cmod/csb_master -I../cmod/csc -I../cmod/cvif -I../cmod/glb -I../cmod/hls_wrapper -I../cmod/mcif -I../cmod/nvdla_core -I../cmod/pdp -I../cmod/rubik -I../cmod/sdp -I../cmod/bdma/gen -I../cmod/cacc/gen -I../cmod/cbuf/gen -I../cmod/cdma/gen -I../cmod/cdp/gen -I../cmod/cmac/gen -I../cmod/csb_master/gen -I../cmod/csc/gen -I../cmod/cvif/gen -I../cmod/glb/gen -I../cmod/hls_wrapper/gen -I../cmod/mcif/gen -I../cmod/nvdla_core/gen -I../cmod/pdp/gen -I../cmod/rubik/gen -I../cmod/sdp/gen -I../cmod/hls/include -I../cmod/hls/cdma_libs -I../cmod/hls/cdp_libs -I../cmod/hls/csc_libs -I../cmod/hls/sdp -I../cmod/hls/vlibs -I/usr/local/systemc-2.3.0//include -MMD -MP -fPIC -Wall -Werror -DSC_INCLUDE_DYNAMIC_PROCESSES -Wp,-w -std=c++11 -c ../cmod/nvdla_payload/dla_b_transport_payload.cpp -o ../outdir/nv_full/cmod/../cmod/nvdla_payload/dla_b_transport_payload.cpp.o
make: /home/utils/gcc-4.9.3/bin/g++: Command not found
make: *** [../outdir/nv_full/cmod/../cmod/nvdla_payload/dla_b_transport_payload.cpp.o] Error 127
Makefile:165: recipe for target '../outdir/nv_full/cmod/../cmod/nvdla_payload/dla_b_transport_payload.cpp.o' failed
make: Leaving directory '/root/vp/hw/cmod'
logfile: outdir/build.log

Filehandle GEN1 opened only for input at /usr/share/perl5/IO/Tee.pm line 132.
==================BUILD PASS==================
Filehandle GEN1 opened only for input at /usr/share/perl5/IO/Tee.pm line 132.

Filehandle GEN1 opened only for input at /usr/share/perl5/IO/Tee.pm line 132.

These are the addresses included in my PATH variable:
/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/home/utils/java/jdk1.8.0_131/bin:/home/utils/gcc-4.9.3/bin

Please help me in identifying the problem.
Thanks

Hi,

After successful compilation of VP with nv_full configuration, I am using prebuilt linux image. Initially i was attempting to run some sanity tests under sw/regressions//, which were not successful. After going through posts in this forum, in one of the post it is suggested not to run unit tests. So I attempted to download bvlc_alexnet from Caffe zoo and successfully compiled with nvdla_compiler.

However with nvdla_runtime, I provided a 227x227x3 dog image as input. But I see the run doesn't complete even after 12hrs. What is the typical runtime on VP? I see the below message trail, especially that " 99 HWLs done, totally 290 layers". So is it normal for nvdla_compiler to flatten nn models to this huge size of 290 layers? Or please help my understanding, if I miss something basic here. Thanks.

[41797.438024] Update dependency operation index 98 ROI 0 DEP_COUNT=1
[41797.443687] enable SDP in dla_update_dependency as depdency are resolved
[41797.444089] Enter: dla_enable_operation
[41797.444333] exit dla_enable_operation without actual enable due to processor hasn't been programmed
[41797.444856] Exit: dla_enable_operation status=0
[41797.445134] Exit: dla_op_programmed
[41797.445359] Exit: dla_program_operation status=0
[41797.445637] Enter: dla_enable_operation
[41797.445882] Enable SDP operation index 97 ROI 0
[41797.446526] Enter: dla_op_enabled
[41797.446772] Update dependency operation index 87 ROI 0 DEP_COUNT=1
[41797.447127] enable Convolution in dla_update_dependency as depdency are resolved
[41797.447552] Enter: dla_enable_operation
[41797.447796] Enable Convolution operation index 87 ROI 0
[41797.452570] Enter: dla_op_enabled
[41797.468014] Exit: dla_op_enabled
[41797.468249] Exit: dla_enable_operation status=0
[41797.468532] Exit: dla_op_enabled
[41797.468762] Exit: dla_enable_operation status=0
[41797.469082] Exit: dla_submit_operation
[41797.469329] Exit: dla_dequeue_operation
[41797.469579] Enter: dla_free_op_desc op desc index 95 ROI 0
[41797.469941] Exit: dla_free_op_desc
[41797.470175] Exit:dla_op_completion processor SDP group1 status=0
[41797.470549] Handle op complete event, processor SDP group 0
[41797.470889] Enter:dla_op_completion processor SDP group0
[41797.471207] Completed SDP operation index 96 ROI 0
[41797.471504] Update dependency operation index 106 ROI 0 DEP_COUNT=2
[41797.474576] 99 HWLs done, totally 290 layers
[41797.475804] Enter: dla_free_op_desc op desc index 106 ROI 0
[41797.476178] Exit: dla_free_op_desc
[41797.476418] Enter: dla_free_op_desc op desc index 86 ROI 0
[41797.476793] Exit: dla_free_op_desc
[41797.477022] Enter: dla_dequeue_operation
[41797.477275] Dequeue op from SDP processor, index=98 ROI=0
[41797.477606] Enter: dla_submit_operation
[41797.477856] Prepare SDP operation index 98 ROI 0 dep_count 0
[41797.478186] Enter: dla_prepare_operation
[41797.582824] processor:SDP group:0, rdma_group:0 available
[41797.583366] Enter: dla_read_config

I meet a error which may not googled previously, Anybody can kindly give a help.

[ 86%] Built target nvdla
Scanning dependencies of target aarch64_toplevel
[ 87%] Building CXX object CMakeFiles/aarch64_toplevel.dir/src/aarch64_toplevel.cpp.o
[ 87%] Linking CXX executable aarch64_toplevel
//usr/lib/x86_64-linux-gnu/libsnappy.so.1: undefined reference to std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::_M_append(char const*, unsigned long)@GLIBCXX_3.4.21' //usr/lib/x86_64-linux-gnu/libsnappy.so.1: undefined reference to std::__cxx11::basic_string<char, std::char_traits, std::allocator >::resize(unsigned long, char)@GLIBCXX_3.4.21'
collect2: error: ld returned 1 exit status
CMakeFiles/aarch64_toplevel.dir/build.make:106: recipe for target 'aarch64_toplevel' failed
make[2]: *** [aarch64_toplevel] Error 1
CMakeFiles/Makefile2:111: recipe for target 'CMakeFiles/aarch64_toplevel.dir/all' failed
make[1]: *** [CMakeFiles/aarch64_toplevel.dir/all] Error 2
Makefile:149: recipe for target 'all' failed
make: *** [all] Error 2

I observed that nv_full INT8 inference on VP is taking more time than with FP16 inference.
NVDLA HW branch: nvdlav1, config: nv_full
NVDLA SW branch: Latest with INT8 option in nvdla_compiler

Please let me know whether this is expected? And whether this will happen when running on HW in say FPGA.

Hi,
I am learning to use fabulous NVDLA architecture.
I am using AWS FPGA AMI to execute the basic tests.
I am using the prebuilt Virtual Simulator on NVIDIA AMI as per http://nvdla.org/vp_fpga.html#using-the-virtual-simulator-on-nvidia-ami

I am trying to run inference test using the following command:
$ ./nvdla_runtime --loadable kmd/NN/NN_L0_0_fbuf --image /mnt/sw/regression/images/digits/one.pgm

The Virtual Simulator seems to work fine but it get hung i.e. does not return to # prompt of the simulator. Also the run does not generate output.dimg

Please let me know, if any steps are missing or I need to recompile the SW.

The detailed steps followed are as follows:
AWS FPGA AMI used has prebuilt NVDLA repository.
Ubuntu 14.04.5 LTS (GNU/Linux 3.13.0-143-generic x86_64)
Following steps are performed:
$ cd /home/ubuntu/nvdla
$ source /usr/local/nvdla/aws-fpga/sdk_setup.sh

$ ./setup_aws.sh
$ ./setup_fpga.sh
I am using NVIDIA Sample AFI, as I want to run inference test on AWS FPGA platform.

$ cd /usr/local/nvdla
$ sudo ./aarch64_toplevel -c aarch64_nvdla.lua --fpga
Login the kernel with account 'root' and password 'nvdla'

$ mount -t 9p -o trans=virtio r /mnt
$ cd /mnt/sw/prebuilt/linux/
$ insmod drm.ko
$ insmod opendla_small.ko
Now run the inference test:
$ ./nvdla_runtime --loadable kmd/NN/NN_L0_0_fbuf --image /mnt/sw/regression/imag
es/digits/one.pgm

creating new runtime context...
Emulator starting
ppgminfo 1 28 28
pgm2dimg 1 28 28 1 32 896 896
submitting tasks...
Work Found!
Work Done
[ 286.175452] Enter:dla_read_network_config
[ 286.177945] Exit:dla_read_network_config status=0
[ 286.178284] Enter: dla_initiate_processors
[ 286.179167] Enter: dla_submit_operation
[ 286.179426] Prepare Convolution operation index 0 ROI 0 dep_count 1
[ 286.179777] Enter: dla_prepare_operation
[ 286.180182] processor:Convolution group:0, rdma_group:0 available
[ 286.180585] Enter: dla_read_config
[ 286.191015] Exit: dla_read_config
[ 286.191313] Exit: dla_prepare_operation status=0
[ 286.191631] Enter: dla_program_operation
[ 286.191886] Program Convolution operation index 0 ROI 0 Group[0]
[ 286.197917] no desc get due to index==-1
[ 286.198930] no desc get due to index==-1
[ 286.199171] no desc get due to index==-1
[ 286.199400] no desc get due to index==-1
[ 286.199643] no desc get due to index==-1
[ 286.199896] Enter: dla_op_programmed
[ 286.200226] Update dependency operation index 3 ROI 0 DEP_COUNT=3
[ 286.200589] Update dependency operation index 1 ROI 0 DEP_COUNT=1
[ 286.202398] enable SDP in dla_update_dependency as depdency are resolved
[ 286.202833] Enter: dla_enable_operation
[ 286.203098] exit dla_enable_operation without actual enable due to processor hasn't been programmed
[ 286.203600] Exit: dla_enable_operation status=0
[ 286.203920] Exit: dla_op_programmed
[ 286.204153] Exit: dla_program_operation status=0
[ 286.204453] Exit: dla_submit_operation
[ 286.206215] Enter: dla_dequeue_operation
[ 286.206544] Dequeue op from Convolution processor, index=3 ROI=0
[ 286.206946] Enter: dla_submit_operation
[ 286.207188] Prepare Convolution operation index 3 ROI 0 dep_count 2
[ 286.207535] Enter: dla_prepare_operation
[ 286.207858] processor:Convolution group:1, rdma_group:0 available
[ 286.208213] Enter: dla_read_config
[ 286.218391] Exit: dla_read_config
[ 286.218660] Exit: dla_prepare_operation status=0
[ 286.218935] Enter: dla_program_operation
[ 286.219198] Program Convolution operation index 3 ROI 0 Group[1]
[ 286.222859] no desc get due to index==-1
[ 286.225318] no desc get due to index==-1
[ 286.225610] no desc get due to index==-1
[ 286.225848] no desc get due to index==-1
[ 286.226120] no desc get due to index==-1
[ 286.226386] Enter: dla_op_programmed
[ 286.226640] Update dependency operation index 6 ROI 0 DEP_COUNT=3
[ 286.227028] Update dependency operation index 4 ROI 0 DEP_COUNT=2
[ 286.227411] Exit: dla_op_programmed
[ 286.227647] Exit: dla_program_operation status=0
[ 286.227953] Exit: dla_submit_operation
[ 286.228237] Exit: dla_dequeue_operation
[ 286.228526] Enter: dla_submit_operation
[ 286.231000] Prepare SDP operation index 1 ROI 0 dep_count 0
[ 286.231355] Enter: dla_prepare_operation
[ 286.231725] processor:SDP group:0, rdma_group:0 available
[ 286.232037] Enter: dla_read_config
[ 286.241702] Exit: dla_read_config
[ 286.242057] Exit: dla_prepare_operation status=0
[ 286.242367] Enter: dla_program_operation
[ 286.242615] Program SDP operation index 1 ROI 0 Group[0]
[ 286.246296] no desc get due to index==-1
[ 286.246583] no desc get due to index==-1
[ 286.247224] no desc get due to index==-1
[ 286.247468] no desc get due to index==-1
[ 286.247727] Enter: dla_op_programmed
[ 286.247957] Update dependency operation index 4 ROI 0 DEP_COUNT=1
[ 286.248295] enable SDP in dla_update_dependency as depdency are resolved
[ 286.250030] Enter: dla_enable_operation
[ 286.250332] exit dla_enable_operation without actual enable due to processor hasn't been programmed
[ 286.250835] Exit: dla_enable_operation status=0
[ 286.251101] Exit: dla_op_programmed
[ 286.251321] Exit: dla_program_operation status=0
[ 286.251600] Enter: dla_enable_operation
[ 286.251880] Enable SDP operation index 1 ROI 0
[ 286.252405] Enter: dla_op_enabled
[ 286.252652] Update dependency operation index 0 ROI 0 DEP_COUNT=1
[ 286.254637] enable Convolution in dla_update_dependency as depdency are resolved
[ 286.255053] Enter: dla_enable_operation
[ 286.255291] Enable Convolution operation index 0 ROI 0
[ 286.256139] Enter: dla_op_enabled
[ 286.256369] Exit: dla_op_enabled
[ 286.256585] Exit: dla_enable_operation status=0
[ 286.258190] Exit: dla_op_enabled
[ 286.258443] Exit: dla_enable_operation status=0
[ 286.258730] Exit: dla_submit_operation
[ 286.258970] Enter: dla_dequeue_operation
[ 286.259213] Dequeue op from SDP processor, index=4 ROI=0
[ 286.259530] Enter: dla_submit_operation
[ 286.259786] Prepare SDP operation index 4 ROI 0 dep_count 0
[ 286.260122] Enter: dla_prepare_operation
[ 286.262535] processor:SDP group:1, rdma_group:1 available
[ 286.262890] Enter: dla_read_config
[ 286.271270] Exit: dla_read_config
[ 286.272520] Exit: dla_prepare_operation status=0
[ 286.273591] Enter: dla_program_operation
[ 286.273894] Program SDP operation index 4 ROI 0 Group[1]
[ 286.275691] no desc get due to index==-1
[ 286.275933] no desc get due to index==-1
[ 286.276962] no desc get due to index==-1
[ 286.278397] no desc get due to index==-1
[ 286.278695] Enter: dla_op_programmed
[ 286.278931] Update dependency operation index 7 ROI 0 DEP_COUNT=2
[ 286.279277] Exit: dla_op_programmed
[ 286.279507] Exit: dla_program_operation status=0
[ 286.279788] Enter: dla_enable_operation
[ 286.280071] Enable SDP operation index 4 ROI 0
[ 286.280524] Enter: dla_op_enabled
[ 286.282219] Update dependency operation index 3 ROI 0 DEP_COUNT=2
[ 286.282598] Exit: dla_op_enabled
[ 286.282808] Exit: dla_enable_operation status=0
[ 286.283080] Exit: dla_submit_operation
[ 286.283311] Exit: dla_dequeue_operation
[ 286.283547] Enter: dla_submit_operation
[ 286.283790] Prepare PDP operation index 2 ROI 0 dep_count 1
[ 286.284104] Enter: dla_prepare_operation
[ 286.284460] processor:PDP group:0, rdma_group:0 available
[ 286.286352] Enter: dla_read_config

At this point, the VP simulator hung.

Please let me know, if any steps are missing or I need to recompile the SW.

Regards,
-Sushil Behere

can anyone explain the working of vp how the input from kernel goes to qemu and router and hw communicate with the router

Hi, friends,
I came across the following error when compile qbox. I guess I may miss some QEMU configuration. Can anybody give any help? I really appreciate that!
The compile environment:
Ubuntu: 18.04
gcc: 4.9.3
g++: 4.9.3
glibc: 2.27-3

Error Message:
`[ 70%] Performing build step for qbox

 BISON dtc-parser.tab.c
 DEP dtc-parser.tab.c
 DEP dtc-lexer.lex.c
    CC      util/memfd.o
    /home/test/nvdla/vp/libs/qbox/util/memfd.c:40:12: error: static declaration of memfd_create follows non-static declaration
   static int memfd_create(const char *name, unsigned int flags)

`

zhanglong@onap-2:~/nvdla/vp$ cmake -DCMAKE_INSTALL_PREFIX=build -DSYSTEMC_PREFIX=/usr/local/systemc-2.3.3 -DNVDLA_HW_PREFIX=/home/zhanglong/nvdla/hw -DNVDLA_HW_PROJECT=nv_full
-- Searching for SystemC
-- SystemC version = 2.3.3
-- SystemC library = /usr/local/systemc-2.3.3/lib-linux64/libsystemc.so
-- Searching for TLM
running ls /usr/local/systemc-2.3.3/include/tlm.h 2>&1
/usr/local/systemc-2.3.3/include/tlm.h
-- TLM library = /usr/local/systemc-2.3.3/include/tlm.h
-- Searching for NVDLA CMOD
-- NVDLA HW VERSION = nvdla_os_initial
-- NVDLA CMOD include directory = /home/zhanglong/nvdla/hw/outdir/nv_full/cmod/release/include
-- NVDLA CMOD library directory = /home/zhanglong/nvdla/hw/outdir/nv_full/cmod/release/lib
-- NVDLA CMOD library = /home/zhanglong/nvdla/hw/outdir/nv_full/cmod/release/lib/libnvdla_cmod.so
-- Searching for SystemC
-- SystemC version = 2.3.3
-- SystemC library = /usr/local/systemc-2.3.3/lib-linux64/libsystemc.so
-- Searching for TLM
running ls /usr/local/systemc-2.3.3/include/tlm.h 2>&1
/usr/local/systemc-2.3.3/include/tlm.h
-- TLM library = /usr/local/systemc-2.3.3/include/tlm.h
-- Boost version: 1.65.1
-- Searching for SystemC
-- SystemC version = 2.3.3
-- SystemC library = /usr/local/systemc-2.3.3/lib-linux64/libsystemc.so
-- Searching for SystemC
-- SystemC version = 2.3.3
-- SystemC library = /usr/local/systemc-2.3.3/lib-linux64/libsystemc.so
-- Searching for TLM
running ls /usr/local/systemc-2.3.3/include/tlm.h 2>&1
/usr/local/systemc-2.3.3/include/tlm.h
-- TLM library = /usr/local/systemc-2.3.3/include/tlm.h
-- Searching for GreenLib
-- Boost version: 1.65.1
-- GreenLib include directory = /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include
-- GreenLib library directory = /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/lib
-- Searching for TLM2C
-- TLM2C include directory = /home/zhanglong/nvdla/vp/libs/tlm2c/include
-- TLM2C library directory = /home/zhanglong/nvdla/vp/libs/tlm2c.build
-- Searching for SystemC
-- SystemC version = 2.3.3
-- SystemC library = /usr/local/systemc-2.3.3/lib-linux64/libsystemc.so
-- Searching for TLM
running ls /usr/local/systemc-2.3.3/include/tlm.h 2>&1
/usr/local/systemc-2.3.3/include/tlm.h
-- TLM library = /usr/local/systemc-2.3.3/include/tlm.h
-- Searching for GreenLib
-- Boost version: 1.65.1
-- GreenLib include directory = /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include
-- GreenLib library directory = /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/lib
-- Searching for SystemC
-- SystemC version = 2.3.3
-- SystemC library = /usr/local/systemc-2.3.3/lib-linux64/libsystemc.so
-- Searching for TLM
running ls /usr/local/systemc-2.3.3/include/tlm.h 2>&1
/usr/local/systemc-2.3.3/include/tlm.h
-- TLM library = /usr/local/systemc-2.3.3/include/tlm.h
-- Searching for GreenLib
-- Boost version: 1.65.1
-- GreenLib include directory = /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include
-- GreenLib library directory = /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/lib
-- Searching for NVDLA CMOD
-- NVDLA HW VERSION = nvdla_os_initial
-- NVDLA CMOD include directory = /home/zhanglong/nvdla/hw/outdir/nv_full/cmod/release/include
-- NVDLA CMOD library directory = /home/zhanglong/nvdla/hw/outdir/nv_full/cmod/release/lib
-- NVDLA CMOD library = /home/zhanglong/nvdla/hw/outdir/nv_full/cmod/release/lib/libnvdla_cmod.so
-- Searching for SystemC
-- SystemC version = 2.3.3
-- SystemC library = /usr/local/systemc-2.3.3/lib-linux64/libsystemc.so
-- Searching for TLM
running ls /usr/local/systemc-2.3.3/include/tlm.h 2>&1
/usr/local/systemc-2.3.3/include/tlm.h
-- TLM library = /usr/local/systemc-2.3.3/include/tlm.h
-- Searching for GreenLib
-- Boost version: 1.65.1
-- GreenLib include directory = /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include
-- GreenLib library directory = /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/lib
-- Searching for SimpleCPU model.
-- SimpleCPU include directory = /home/zhanglong/nvdla/vp/models/cpu/include
-- SimpleCPU library directory = /home/zhanglong/nvdla/vp/models/cpu.build
-- Configuring done
-- Generating done
-- Build files have been written to: /home/zhanglong/nvdla/vp

zhanglong@onap-2:~/nvdla/vp$ make
Scanning dependencies of target greenreg
[ 0%] Building CXX object libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/__ /src/greenreg.cpp.o
In file included from /home/zhanglong/nvdla/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_extension_support_base.h:144:0,
from /home/zhanglong/nvdla/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_bind_checker.h:24,
from /home/zhanglong/nvdla/vp/libs/greenlib/greensocket/include/greensocket/initiator/green_initiator_socket_base.h:28,
from /home/zhanglong/nvdla/vp/libs/greenlib/greensocket/include/greensocket/initiator/single_socket.h:26,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/generic_transaction.h:29,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:32,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/src/greenreg.cpp:35:
/home/zhanglong/nvdla/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_extension_support_base.tpp: In static member function ‘static T* gs::socket::extension_support_base::get_extension(const gs::ext::gs_data_id&, gs::socket::extension_support_base::payload_type&)’:
/home/zhanglong/nvdla/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_extension_support_base.tpp:127:29: error: expected primary-expression before ‘>’ token
T* tmp=txn.get_extension(); //get the pointer
^
/home/zhanglong/nvdla/vp/libs/greenlib/greensocket/include/greensocket/generic/green_socket_extension_support_base.tpp:127:31: error: expected primary-expression before ‘)’ token
T* tmp=txn.get_extension(); //get the pointer
^
In file included from /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/config.h:50:0,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:36,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/src/greenreg.cpp:35:
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/cnf_api_if.h: At global scope:
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/cnf_api_if.h:237:65: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated]
virtual sc_event& getUpdateEvent(const std::string &parname) throw (RegisterObserverFailedException) = 0;
^~~~~
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/cnf_api_if.h:325:40: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated]
virtual sc_event& getNewParamEvent() throw (RegisterObserverFailedException) = 0;
^~~~~
In file included from /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/config.h:56:0,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:36,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/src/greenreg.cpp:35:
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/GCnf_Api_t.h:1013:57: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated]
sc_event& getUpdateEvent(const std::string &parname) throw (RegisterObserverFailedException) {
^~~~~
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_Api/GCnf_Api_t.h:1134:32: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated]
sc_event& getNewParamEvent() throw (RegisterObserverFailedException) {
^~~~~
In file included from /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/config.h:57:0,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:36,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/src/greenreg.cpp:35:
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_private_Api/gcnf_private_api_t.h:486:59: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated]
sc_event& getUpdateEvent(const std::string &parname) throw (RegisterObserverFailedException) {
^~~~~
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/GCnf_private_Api/gcnf_private_api_t.h:506:34: warning: dynamic exception specifications are deprecated in C++11 [-Wdeprecated]
sc_event& getNewParamEvent() throw (RegisterObserverFailedException) {
^~~~~
In file included from /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_class.h:33:0,
from /home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/config.h:71,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSocket.h:36,
from /home/zhanglong/nvdla/vp/libs/greenlib/gsgpsocket/include/gsgpsocket/transport/GSGPSlaveSocket.h:24,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/framework/core/gr_common.h:53,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/include/greenreg/greenreg.h:45,
from /home/zhanglong/nvdla/vp/libs/greenlib/greenreg/src/greenreg.cpp:35:
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_native.hpp: In member function ‘gs::cnf::gs_param::my_type& gs::cnf::gs_param::operator++()’:
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_operator_macros.h:79:7: warning: use of an operand of type ‘bool’ in ‘operator++’ is deprecated [-Wdeprecated]
++tmp_val;
^
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_native.hpp:261:3: note: in expansion of macro ‘GC_SPECIALISATIONS_INCREMENT_OPERATORS’
GC_SPECIALISATIONS_INCREMENT_OPERATORS;
^
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_native.hpp: In member function ‘gs::cnf::gs_param::val_type gs::cnf::gs_param::operator++(int)’:
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_operator_macros.h:89:16: warning: use of an operand of type ‘bool’ in ‘operator++’ is deprecated [-Wdeprecated]
tmp_set_val++;
^
/home/zhanglong/nvdla/vp/libs/greenlib/greencontrol/include/greencontrol/gcnf/apis/gs_param/gs_param_native.hpp:261:3: note: in expansion of macro ‘GC_SPECIALISATIONS_INCREMENT_OPERATORS’
GC_SPECIALISATIONS_INCREMENT_OPERATORS;
^
libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/build.make:62: recipe for target 'libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir//src/greenreg.cpp.o' failed
make[2]: *** [libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir//src/greenreg.cpp.o] Error 1
CMakeFiles/Makefile2:1810: recipe for target 'libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/all' failed
make[1]: *** [libs/greenlib/greenreg/lib/CMakeFiles/greenreg.dir/all] Error 2
Makefile:151: recipe for target 'all' failed
make: *** [all] Error 2

In what format can we expect the object detection inference output (output.dimg) to be?
Are both the coordinates and the labels supposed to be in the same output file?

And what is the significance on negative numbers in the output.
In my result, I got all negative numbers. Does it mean that there was no object found?

followed the instruction from http://nvdla.org/vp.html#prepare-kernel-image
after this command: ./build/bin/aarch64_toplevel -c conf/aarch64_nvdla.lua

         SystemC 2.3.0-ASI --- Jun 19 2019 10:14:15
    Copyright (c) 1996-2012 by all Contributors,
    ALL RIGHTS RESERVED

No sc_log specified, will use the default setting
verbosity_level = SC_MEDIUM
DMI mode enable
Segmentation fault (core dumped)

besides the location for Image nd rootfs.ext4, is there anything else i need to change in the aarch64_nvdla.lua file?
extra_arguments = "-machine virt -cpu cortex-a57 -machine type=virt -nographic -smp 1 -m 1024 -kernel /home/jingchi/vp/buildroot-2017.11-rc1/output/images/Image --append "root=/dev/vda" -drive file=/home/jingchi/vp/buildroot-2017.11-rc1/output/images/rootfs.ext4,if=none,format=raw,id=hd0 -device virtio-blk-device,drive=hd0 -fsdev local,id=r,path=.,security_model=none -device virtio-9p-device,fsdev=r,mount_tag=r -netdev user,id=user0,hostfwd=tcp::6666-:6666,hostfwd=tcp::6667-:22 -device virtio-net-device,netdev=user0"

http://nvdla.org/vp.html#running-kernel-image-in-the-virtual-simulator
I'm trying this step on the web,but i got the error below.

chi@chi:~/tools_for_cmod/vp$ ./build/bin/aarch64_toplevel -c conf/aarch64_nvdla.lua

         SystemC 2.3.0-ASI --- Jan 13 2018 00:30:21
    Copyright (c) 1996-2012 by all Contributors,
    ALL RIGHTS RESERVED

No sc_log specified, will use the default setting
verbosity_level = SC_MEDIUM
bridge: tlm2c_elaborate..
toplevel: -device virtio-9p-device,fsdev=r,mount_tag=r: 'virtio-9p-device' is not a valid device model name

I'm on ubuntu14.04.
How i shoud do to solve this problem?
Thanks

There will be an error using python3.5m. "error: ‘PyString_FromString’ was not declared in this scope".
Change to python2.7, this error disappear.

I still confuse two problems as follows, please help to solve:
1. After I complete VP installation, and I also run the command about "./nvdla_runtime --loadable XXXX --image YYYYY" without any errors. But I do not know what next operations will do?
2. After running "./nvdla_runtime" , How do I find or see the generated configurable information for XXXX file, such as NVDLA HW paramemters (register value, etc)? how to write data into memory? how does virtual CPU communicate with nvdla_core in the NVDLA Virtual Platform.

As the title. For example, when I run the regression case of SDP_small, I can see the CMOD send the WDMA requests to the address 0xc0005000. Though there will be logging when the $SC_LOG Environment Variable is set, is there any way to dump the memory without logging?

Thanks.

Hi, so I downloaded the VP for NVDLA. And ran the following commands

• sudo docker run -it -v /home:/home nvdla/vp
• cd /usr/local/nvdla
• aarch64_toplevel -c aarch64_nvdla.lua
  Then enter credentials (root,nvdla).
• # mount -t 9p -o trans=virtio r /mnt
• # cd /mnt/
• # insmod drm.ko
• # insmod opendla_2.ko
  Using opendla_2.ko as I am currently working on nv_small
• # ./nvdla_runtime --loadable ./sw-master/regression/flatbufs/kmd/NN/NN_L0_0_fbuf --image ./sw-master/regression/images/digits/seven.pgm --rawdump

Then the terminal shows this output

creating new runtime context...
Emulator starting
pgm2dimg 1 28 28 1 896 25088 25088
submitting tasks...
Work Found!
Work Done
[ 197.712727] Enter:dla_read_network_config
[ 197.713316] Exit:dla_read_network_config status=0
[ 197.713506] Enter: dla_initiate_processors
[ 197.713819] Enter: dla_submit_operation
[ 197.713984] Prepare Convolution operation index 0 ROI 0 dep_count 1
[ 197.714223] Enter: dla_prepare_operation
[ 197.714570] processor:Convolution group:0, rdma_group:0 available
[ 197.714801] Enter: dla_read_config
[ 197.715134] Exit: dla_read_config
[ 197.715274] Exit: dla_prepare_operation status=0
[ 197.715429] Enter: dla_program_operation
[ 197.715553] Program Convolution operation index 0 ROI 0 Group[0]
root@6f0c649fa2f5:/usr/local/nvdla#

As you can tell I'm back at where I started. This is consistently going on with every loadable file I have whether using the prebuilt flatbufs or the compiling my own .nvdla files.

Has anyone else faced this issue? Any help would be appreciated. Thanks a lot.

Hi,
I would like to know how to debug inside the virtual platform with gdb.
Is there a way to install gdb inside the image that gets loaded with the VP? Can we do it after the VP boots up with QEMU?
I am using the prebuilt image provided for the ARM architecture. How should I install gcc/gdb tools inside it, to compile the run time and debug it.

[ 74%] Linking CXX shared library libcosim_sc_wrapper.so
[ 74%] Built target cosim_sc_wrapper
Scanning dependencies of target greenscript
[ 74%] Building CXX object libs/greenlib/greenscript/lib/CMakeFiles/greenscript.dir//src/greenscript.cpp.o
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp: In constructor ‘gs::script::GreenScriptModule::GreenScriptModule(sc_core::sc_module_name, const char*)’:
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp:178:39: error: ‘PyString_FromString’ was not declared in this scope
name_py = PyString_FromString(name()); // new ref
^
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp: In member function ‘void gs::script::GreenScriptModule::load(const char*)’:
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp:221:46: error: ‘PyString_AsString’ was not declared in this scope
std::string s(PyString_AsString(path_py));
^
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp: In member function ‘void gs::script::GreenScriptModule::add_to_pythonpath(const char*)’:
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp:262:47: error: ‘PyString_FromString’ was not declared in this scope
PyObject path_py = PyString_FromString(path); // new ref
^
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp: In member function ‘bool gs::script::GreenScriptModule::private_load(const char)’:
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp:278:68: error: ‘PyFile_FromString’ was not declared in this scope
PyObject* PyFileObject = PyFile_FromString((char *)fullname, mode);
^
In file included from /usr/local/include/python3.7m/Python.h:129:0,
from /root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp:38:
/root/nvdla/vp/libs/greenlib/greenscript/src/greenscript.cpp:279:56: error: ‘PyFile_AsFile’ was not declared in this scope
PyObject *ret = PyRun_File(PyFile_AsFile(PyFileObject),
^
/usr/local/include/python3.7m/pythonrun.h:146:23: note: in definition of macro ‘PyRun_File’
PyRun_FileExFlags(fp, p, s, g, l, 0, NULL)
^
libs/greenlib/greenscript/lib/CMakeFiles/greenscript.dir/build.make:62: recipe for target 'libs/greenlib/greenscript/lib/CMakeFiles/greenscript.dir//src/greenscript.cpp.o' failed
make[2]: *** [libs/greenlib/greenscript/lib/CMakeFiles/greenscript.dir/__/src/greenscript.cpp.o] Error 1
CMakeFiles/Makefile2:2079: recipe for target 'libs/greenlib/greenscript/lib/CMakeFiles/greenscript.dir/all' failed
make[1]: *** [libs/greenlib/greenscript/lib/CMakeFiles/greenscript.dir/all] Error 2
Makefile:151: recipe for target 'all' failed
make: *** [all] Error 2

hi all:
i got an error when i try to build vp by the method in readme, behind is the error information,
"/lib/libnvdla_cmod.so: undefined reference to `sc_core::sc_vector_base::make_name(char const*, unsigned long)'
collect2: error: ld returned 1 exit status"

systemc is 2.3.0,
hw is on branch nvdlav1
it is looks like some wrong with hw ?

Hi,
I'm able to load the virtual platform from aarch64_toplevel, but when I try to mount using the command:
mount -t 9p -o trans=virtio r /mnt

I get an error:
mount: mounting r on /mnt failed: No such device

Please let me know how to solve this.

1. git clone the vp and set up environment.
2. git clone greenlib and install, then copy findgreenlib.cmake to vp directory.
3. change python3 to python2.7
4. cmake -DCMAKE_INSTALL_PREFIX=build -DSYSTEMC_PREFIX=/usr/local/systemc-2.3.0/ -DNVDLA_HW_PREFIX=/usr/local/nvdla/hw -DNVDLA_HW_PROJECT=nv_small
5. make.
   Then this step I get error information as follows:

/NVDLA/vp/libs/qbox/util/memfd.c:40:12: error: static declaration of ‘memfd_create’ follows non-static declaration static int memfd_create(const char *name, unsigned int flags) ^ In file included from /usr/include/x86_64-linux-gnu/bits/mman-linux.h:115:0, from /usr/include/x86_64-linux-gnu/bits/mman.h:45, from /usr/include/x86_64-linux-gnu/sys/mman.h:41, from /media/simon/360A152B0A14E99F/WorkSpace/NVDLA/vp/libs/qbox/include/sysemu/os-posix.h:29, from /NVDLA/vp/libs/qbox/include/qemu/osdep.h:104, from /NVDLA/vp/libs/qbox/util/memfd.c:28: /usr/include/x86_64-linux-gnu/bits/mman-shared.h:46:5: note: previous declaration of ‘memfd_create’ was here int memfd_create (const char *__name, unsigned int __flags) __THROW; ^ /NVDLA/vp/libs/qbox/rules.mak:66: recipe for target 'util/memfd.o' failed make[3]: *** [util/memfd.o] Error 1 CMakeFiles/qbox.dir/build.make:112: recipe for target 'libs/qbox.build/stampdir/qbox-build' failed make[2]: *** [libs/qbox.build/stampdir/qbox-build] Error 2 CMakeFiles/Makefile2:111: recipe for target 'CMakeFiles/qbox.dir/all' failed make[1]: *** [CMakeFiles/qbox.dir/all] Error 2 Makefile:151: recipe for target 'all' failed make: *** [all] Error 2

So anyone can give a suggestion about that?

Thanks.

two problems:

1. cmake flags does not take effect.
   when typing in

cmake -DCMAKE_INSTALL_PREFIX=build -DSYSTEMC_PREFIX=/home/nvdla/systemc-2.3.0/ -DNVDLA_HW_PREFIX=/home/nvdla/NVDLA/git-nvdla/nvdla -DNVDLA_HW_PROJECT=nv_small

finding package result is for nv_full

-- NVDLA HW VERSION = nvdla_os_initial -- NVDLA CMOD include directory = /home/nvdla/NVDLA/git-nvdlav1/outdir/nv_full/cmod/release/include -- NVDLA CMOD library directory = /home/nvdla/NVDLA/git-nvdlav1/outdir/nv_full/cmod/release/lib -- NVDLA CMOD library = /home/nvdla/NVDLA/git-nvdlav1/outdir/nv_full/cmod/release/lib/libnvdla_cmod.so

2. when I change the env like this

set(NVDLA_CMOD_INCLUDE_DIR "/home/nvdla/NVDLA/git-nvdla/nvdla/outdir/nv_small/cmod/release/include") set(NVDLA_CMOD_LIBRARY_DIR "/home/nvdla/NVDLA/git-nvdla/nvdla/outdir/nv_small/cmod/release/lib") set(NVDLA_CMOD_LIBRARY "/home/nvdla/NVDLA/git-nvdla/nvdla/outdir/nv_small/cmod/release/lib/libnvdla_cmod.so")

there are some errors related to the built libnvdla_cmod.so

Linking CXX shared library libnvdla.so [ 88%] Built target nvdla [ 89%] Building CXX object CMakeFiles/aarch64_toplevel.dir/src/aarch64_toplevel.cpp.o Linking CXX executable aarch64_toplevel /home/nvdla/NVDLA/git-nvdla/nvdla/outdir/nv_small/cmod/release/lib/libnvdla_cmod.so: undefined reference tosc_core::sc_signal<bool, (sc_core::sc_writer_policy)0>::do_update()'
/home/nvdla/NVDLA/git-nvdla/nvdla/outdir/nv_small/cmod/release/lib/libnvdla_cmod.so: undefined reference to sc_core::sc_api_version_2_3_1<&sc_core::SC_DISABLE_VIRTUAL_BIND_UNDEFINED_>::sc_api_version_2_3_1(sc_core::sc_writer_policy)' /home/nvdla/NVDLA/git-nvdla/nvdla/outdir/nv_small/cmod/release/lib/libnvdla_cmod.so: undefined reference to sc_core::sc_object::simulation_phase_callback()'
collect2: error: ld returned 1 exit status
make[2]: *** [aarch64_toplevel] Error 1
make[1]: *** [CMakeFiles/aarch64_toplevel.dir/all] Error 2
make: *** [all] Error 2
`

@prasshantg i am facing the same issue when running with NN_L0_0_fbuf.

NvDlaSubmit: Error IOCTL failed (No such process)
(DLA_RUNTIME) Error 0x0003000f: (propagating from Runtime.cpp, function submitInternal(), line 666)
(DLA_TEST) Error 0x00000004: runtime->submit() failed (in RuntimeTest.cpp, function runTest(), line 387)
(DLA_TEST) Error 0x00000004: (propagating from RuntimeTest.cpp, function run(), line 440)

But when i am running my alexnet i got fast-math.nvdla and when i am running my input cat.pgm it is giving error like
./nvdla_runtime --loadable fast-math.nvdla --image cat.pgm

creating new runtime context...
Emulator starting
pgm2dimg 1 360 480 1 7264 1648928 16489280
(DLA_TEST) Error 0x00000004: Mismatched width: 480 != 227 (in TestUtils.cpp, function createImageCopy(), line 156)
(DLA_TEST) Error 0x00000004: (propagating from RuntimeTest.cpp, function copyImageToInputTensor(), line 100)
(DLA_TEST) Error 0x00000004: (propagating from RuntimeTest.cpp, function setupInputBuffer(), line 162)
(DLA_TEST) Error 0x00000004: (propagating from RuntimeTest.cpp, function runTest(), line 382)
(DLA_TEST) Error 0x00000004: (propagating from RuntimeTest.cpp, function run(), line 440)
Shutdown signal received, exiting
(DLA_TEST) Error 0x00000004: (propagating from main.cpp, function launchTest(), line 87)

Please @prasshantg help me out.Thanks

hi,
In above issue, i debug nvdla_bdma_mmio.c successfully with GDB in VP.
But how can i succeed to debug nvdla model with GDB in VP ?

Thank you very much and i am looking forward to you replay!

there is no CmakeLists.txt file in the directory vp/models/memory and vp/models/cpu

I trained and compiled the LeNet from Caffe github example, and Run the Test Inference using the test image from NVDLA_SW digits.
From the Inference log, I suppose the DLA running flow is OK, however there is an Error in the end.

./nvdla_runtime --softmax --loadable lenet.nvdla --image two.pgm

creating new runtime context...
ppgminfo 1 28 28
pgm2dimg 1 28 28 1 32 896 896
submitting tasks...
[ 727.191545] Enter:dla_read_network_config
[ 727.193579] Exit:dla_read_network_config status=0
[ 727.195396] Enter: dla_initiate_processors
[ 727.198834] Enter: dla_submit_operation
[ 727.206431] Prepare Convolution operation index 0 ROI 0 dep_count 1
[ 727.212430] Enter: dla_prepare_operation

.......
.......
.......

[ 785.230416] 11 HWLs done, totally 11 layers
[ 785.230782] Enter: dla_free_op_desc op desc index 9 ROI 0
[ 785.232123] Exit: dla_free_op_desc
[ 785.232799] Enter: dla_free_op_desc op desc index 10 ROI 0
[ 785.233598] Exit: dla_free_op_desc
[ 785.233964] Exit:dla_op_completion processor SDP group1 status=0
[ 785.234791] Exit:dla_handle_events, ret:0
[ 785.235707] Enter:dla_handle_events, processor:PDP
[ 785.236012] Exit:dla_handle_events, ret:0
[ 785.236168] Enter:dla_handle_events, processor:CDP
[ 785.236590] Exit:dla_handle_events, ret:0
[ 785.237620] Enter:dla_handle_events, processor:RUBIK
[ 785.238801] Exit:dla_handle_events, ret:0
[ 785.241088] reset engine done
(DLA_RUNTIME) Error 0x00000004: unrecognized interface 0 (in Runtime.cpp, function submitInternal(), line 487)
(DLA_TEST) Error 0x00000004: runtime->submit() failed (in RuntimeTest.cpp, function runTest(), line 355)
(DLA_TEST) Error 0x00000004: (propagating from RuntimeTest.cpp, function run(), line 379)
(DLA_TEST) Error 0x00000004: (propagating from main.cpp, function launchTest(), line 92)

Hello all,
I am trying to debug NVDLA virtual platform using gdb.
I have no idea how to debug this running process.
It is even harder because i have to debug QEMU process (aarch64_toplevel) to see NVDLA working..

Is there any clever way to see how NVDLA working in a step-by-step fashion?
Please give me some advice then i'll do the work and post the result in here.

Thanks!

Hi,

I'm facing build issue with virtual simulator at linking stage.
The build of 'hw' and SystemC passed without any issue.

The error message on building 'vp' is pasted below. Can you suggest what is the solution. I also tried with '-pthread' option instead of '-lpthread' in the file 'vp/CMakeFiles/aarch64_toplevel.dir/link.txt' that didn't solve the issue.

The command and error messages is given below.

uname -a
*** 4.15.0-43-generic #46~16.04.1-Ubuntu SMP Fri Dec 7 13:31:08 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

The configuration Command:
cmake -DCMAKE_CXX_COMPILER=/home/syed/utils/gcc-4.9.3/bin/g++ -DCMAKE_INSTALL_PREFIX=build -DSYSTEMC_PREFIX=/usr/local/systemc-2.3.0/ -DNVDLA_HW_PREFIX=/home/syed/workdla/nvdla/hw -DNVDLA_HW_PROJECT=nv_full

The build:

make -j6
.....
[ 87%] Built target greenreg
Scanning dependencies of target nvdla
Scanning dependencies of target cosim_sc_wrapper
[ 87%] Building CXX object models/nvdla.build/CMakeFiles/nvdla.dir/src/adaptors/gp2str.cpp.o
[ 89%] Building CXX object models/nvdla.build/CMakeFiles/nvdla.dir/src/nvdla_top.cpp.o
[ 91%] Building CXX object models/nvdla.build/CMakeFiles/nvdla.dir/src/adaptors/gsgp2tlm.cpp.o
[ 91%] Building CXX object models/nvdla.build/CMakeFiles/nvdla.dir/src/adaptors/tlm2gsgp.cpp.o
[ 93%] Building CXX object models/nvdla.build/CMakeFiles/nvdla.dir/src/adaptors/irq_adaptor.cpp.o
[ 94%] Building CXX object fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/src/nvdla_cosim_sc_wrapper.cpp.o
[ 94%] Building CXX object fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/src/extmem_cosim_sc_wrapper.cpp.o
[ 96%] Building CXX object fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/src/transactors/gsgp2cosim.cpp.o
[ 96%] Building CXX object fpga/aws-fpga/cosim_sc_wrapper.build/CMakeFiles/cosim_sc_wrapper.dir/src/transactors/irq2gss.cpp.o
[ 96%] Linking CXX shared library libnvdla.so
[ 96%] Built target nvdla
[ 98%] Linking CXX shared library libcosim_sc_wrapper.so
[ 98%] Built target cosim_sc_wrapper
Scanning dependencies of target aarch64_toplevel
[100%] Building CXX object CMakeFiles/aarch64_toplevel.dir/src/aarch64_toplevel.cpp.o
[100%] Linking CXX executable aarch64_toplevel
/home/syed/workdla/nvdla/vp/libs/qbox.build/aarch64-softmmu/libqbox-nvdla.so: undefined reference to `pthread_atfork'
collect2: error: ld returned 1 exit status
CMakeFiles/aarch64_toplevel.dir/build.make:95: recipe for target 'aarch64_toplevel' failed
make[2]: *** [aarch64_toplevel] Error 1
CMakeFiles/Makefile2:79: recipe for target 'CMakeFiles/aarch64_toplevel.dir/all' failed
make[1]: *** [CMakeFiles/aarch64_toplevel.dir/all] Error 2
Makefile:151: recipe for target 'all' failed
make: *** [all] Error 2

Hi,

I have run docker VP version on my Ubuntu16.04 compluter, and enter VP already.
I can run runtime inside VP.

But I don't know how to copy my loadable and test data from my local disk into VP.
Do you have any recomand on how to do this.

Thanks!

Hi.

When I run below command. Program seems stucked..

./nvdla_runtime --loadable ./VPtestfile/basic.nvdla -image ./VPtestfile/000456.jpg

I can only see one line after this command :
creating new runtime context...

I want to do some check. but don't know how to open another terminor to check the kernel log.

Thanks!

from the sw#9
it seems that currently, we can't get the execution time of a network.
However, this is a very important profiling data to estimate the performance of the DLA core.

May I know how to dump the execution time or cycle number of a CNN network by c model (or system c)?
Is there any plan from nvidia or guide if I want to dig the problem myself?

ps. Another issue troubles me is the compiler seems not support layers of prelu, deconv and so on. I might be better to open another page to talk about it.