hughperkins

li sp, 280
jal x1, _Z10sum_floatsPfjS_
halt
	.text
	.attribute	4, 16
	.attribute	5, "rv32i2p0_m2p0_f2p0"
	.file	"sum_floats.cpp"
	.globl	_Z10sum_floatsPfjS_             # -- Begin function _Z10sum_floatsPfjS_
	.p2align	2
	.type	_Z10sum_floatsPfjS_,@function

hughperkins

// This is a GENERATED file. Do not modify by hand.
// Created by toy_proc/generation/dadda.py

// Multiply two 24-bit integers 'a' and 'b', and put result in 47-bit integer 'out'.

module dadda(
    input [23:0] a,
    input [23:0] b,
    output [46:0] out
);

hughperkins

# This is the CMakeCache file.
# For build in directory: /media/psf/OpenTimer/build_linux
# It was generated by CMake: /usr/local/bin/cmake
# You can edit this file to change values found and used by cmake.
# If you do not want to change any of the values, simply exit the editor.
# If you do want to change a value, simply edit, save, and exit the editor.
# The syntax for the file is as follows:
# KEY:TYPE=VALUE
# KEY is the name of a variable in the cache.
# TYPE is a hint to GUIs for the type of VALUE, DO NOT EDIT TYPE!.

hughperkins

""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
import numpy as np
import cPickle as pickle
import gym

# hyperparameters
H = 200 # number of hidden layer neurons
batch_size = 10 # every how many episodes to do a param update?
learning_rate = 1e-4
gamma = 0.99 # discount factor for reward

hughperkins

ERROR: /home/admin/srcs/tf-coriander/tensorflow/python/BUILD:1793:1: Linking of rule '//tensorflow/python:_pywrap_tensorflow.so' failed: link_dynamic_library.sh failed: error executing command 
  (cd /home/admin/.cache/bazel/_bazel_admin/5248d45850a159c0f72dd5c3cd93076b/execroot/tf-coriander && \
  exec env - \
  external/bazel_tools/tools/cpp/link_dynamic_library.sh no ignored ignored ignored tensorflow/tools/cocl/clang/bin/crosstool_wrapper_driver_is_not_gcc -shared -o bazel-out/local_linux-py3-fastbuild/bin/tensorflow/python/_pywrap_tensorflow.so -Lbazel-out/local_linux-py3-fastbuild/bin/_solib_local/_U_S_Sthird_Uparty_Ccocl___Uthird_Uparty_Scoriander_Sbuild '-Wl,-rpath,$ORIGIN/../../_solib_local/_U_S_Sthird_Uparty_Ccocl___Uthird_Uparty_Scoriander_Sbuild' -Wl,-whole-archive bazel-out/local_linux-py3-fastbuild/bin/tensorflow/python/_objs/_pywrap_tensorflow.so/tensorflow/python/pywrap_tensorflow.pic.o -Wl,-no-whole-archive -Wl,-whole-archive bazel-out/local_linux-py3-fastbuild/bin/tensorflow/python/libcpp_sh

hughperkins

lib/libLLVMCore.a(AsmWriter.cpp.o):
                 U __DefaultRuneLocale
0000000000000530 T isalnum(int)
0000000000018e30 T isalpha(int)
000000000000efd0 T isascii(int)
0000000000000510 T isdigit(int)
00000000000001f0 T isprint(int)
000000000000ef50 T __istype(int, unsigned long)
000000000000eff0 T __isctype(int, unsigned long)

hughperkins

lib/libLLVMCore.a:AsmWriter.cpp.o: 0000000000019560 T llvm::isa_impl_cl<llvm::LocalAsMetadata, llvm::ValueAsMetadata const*>::doit(llvm::ValueAsMetadata const*)
lib/libLLVMCore.a:AsmWriter.cpp.o: 00000000000193f0 T llvm::isa_impl_cl<llvm::ValueAsMetadata, llvm::Metadata const*>::doit(llvm::Metadata const*)
lib/libLLVMCore.a:AsmWriter.cpp.o: 00000000000194e0 T llvm::isa_impl_wrap<llvm::LocalAsMetadata, llvm::ValueAsMetadata const* const, llvm::ValueAsMetadata const*>::doit(llvm::ValueAsMetadata const* const&)
lib/libLLVMCore.a:AsmWriter.cpp.o: 0000000000019510 T llvm::isa_impl_wrap<llvm::LocalAsMetadata, llvm::ValueAsMetadata const*, llvm::ValueAsMetadata const*>::doit(llvm::ValueAsMetadata const* const&)
lib/libLLVMCore.a:AsmWriter.cpp.o: 0000000000019390 T llvm::isa_impl_wrap<llvm::ValueAsMetadata, llvm::Metadata const* const, llvm::Metadata const*>::doit(llvm::Metadata const* const&)
lib/libLLVMCore.a:AsmWriter.cpp.o: 00000000000193c0 T llvm::isa_impl_wrap<llvm::ValueAsMetadata, llvm::Metadata const*, llvm:

hughperkins

lib/libLLVMCore.a:AsmWriter.cpp.o: 000000000001fae0 unsigned short orderValue(llvm::Value const*, (anonymous namespace)::OrderMap&)
lib/libLLVMCore.a:AsmWriter.cpp.o: 0000000000019640 unsigned short PrintLLVMName(llvm::raw_ostream&, llvm::Value const*)
lib/libLLVMCore.a:AsmWriter.cpp.o: 00000000000086b0 unsigned short getModuleFromVal(llvm::Value const*)
lib/libLLVMCore.a:AsmWriter.cpp.o: 000000000000e860 unsigned short printWithoutType(llvm::Value const&, llvm::raw_ostream&, llvm::SlotTracker*, llvm::Module const*)
lib/libLLVMCore.a:AsmWriter.cpp.o: 00000000000197b0 unsigned short createSlotTracker(llvm::Value const*)
lib/libLLVMCore.a:AsmWriter.cpp.o: 000000000000e8f0 unsigned short printAsOperandImpl(llvm::Value const&, llvm::raw_ostream&, bool, llvm::ModuleSlotTracker&)
lib/libLLVMCore.a:AsmWriter.cpp.o: 0000000000016c30 unsigned short WriteAsOperandInternal(llvm::raw_ostream&, llvm::Value const*, (anonymous namespace)::TypePrinting*, llvm::SlotTracker*, llvm::Module const*)
lib/libLLVMCore.a:AsmWriter.cp

hughperkins

; ModuleID = './fill_copy_sequence-device-noopt.ll'
source_filename = "./fill_copy_sequence.cu"
target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
target triple = "nvptx64-nvidia-cuda"

%"class.thrust::system::cuda::detail::bulk_::uninitialized" = type { %"union.thrust::system::cuda::detail::bulk_::detail::aligned_storage<24, 8>::type" }
%"union.thrust::system::cuda::detail::bulk_::detail::aligned_storage<24, 8>::type" = type { [24 x i8] }
%"class.thrust::system::cuda::detail::bulk_::detail::cuda_task" = type <{ %"class.thrust::system::cuda::detail::bulk_::detail::task_base", i32, [4 x i8] }>
%"class.thrust::system::cuda::detail::bulk_::detail::task_base" = type { %"class.thrust::system::cuda::detail::bulk_::detail::closure", %"class.thrust::system::cuda::detail::bulk_::parallel_group" }
%"class.thrust::system::cuda::detail::bulk_::detail::closure" = type { %"struct.thrust::system::cuda::detail::for_each_n_detail::for_each_kernel", %"class.thrust::tuple" }

hughperkins

// __vmem__ is just a marker, so we can see which bits are vmems
// It doesnt actually do anything; compiler ignores it
#define __vmem__

// vmem2 is a pointer to a pointer (so we have to unwrap twice)
#define __vmem2__

struct GlobalVars {
    local int *scratch;
    global char *clmem0;