Target Zynq and NPU

[sampsyo]

We need to add the Zynq (ARM + FPGA) platform as a target for the ACCEPT toolchain.

This involves:

• Extend app.mk to cross-compile binaries for the bare-metal ARM.
• Also extend it to execute binaries using the Xilinx debugger (xmd), including sending the appropriate bitstream and possibly resetting the board with the Wemo switch.
• Collect output and performance numbers through the serial log rather than the filesystem.
• Make the experiments command aware of the platform it's running on so we can run x86 benchmarks on the cluster (in parallel) and Zynq experiments on pinga (in serial), and ne'er the twain shall meet.

---

Instructions from Thierry for running benchmarks on the Zynq:

1. ssh into sampa-gw.cs.washington.edu
2. ssh sampa-gw.cs.washington.edu
3. from there, ssh into pinga
4. ssh pinga
5. source the xilinx settings script
6. . /sampa/share/Xilinx/14.6/14.6/ISE_DS/settings64.sh
7. go to the following directory
8. cd /sampa/share/thierry/benchmarks/boot_files
9. to run a benchmark, type the following (replace inversek2j with the benchmark of your choice)
10.  make inversek2j
11. this will compile the benchmark source code in BenchNN/, and will copy the elf file into the elf/ directory, program the NPU and run the binary on the NPU
12. if you need to run your own elf file on the NPU, change the path of the elf file in the Makefile to your own (e.g. change elf/inversek2j.elf to /sampa/share/andreolb/inversek2j.elf).
13. to read the output, go to the mini com_logs directory. 
14. cd /sampa/share/thierry/minicom_logs
15. each output is timestamped with the current time the benchmark was running
16. vim minicom_14_06_09_13_42_44_inversek2j.txt

---

Andre's cross-compiler setup is in the npu branch's app.mk. Clang is invoked to target ARM and produce assembly files. Then gcc is used to produce the final binary:

arm-xilinx-eabi-gcc -Wl,-T -Wl,../lib/lscript.ld -L ../lib/bsp/lib inversek2j.s -o inversek2j.elf -Wl,--start-group,-lxil,-lgcc,-lc,-lm,--end-group

much like the msp430 workflow we have now.

[sampsyo]

Next step: some sort of post-processor script that will split the main output from the performance numbers. For ease of wrangling, this could use the same filename for the performance data as for the default acceptrt. It could even do the conversion to seconds from FPGA clocks.

[sampsyo]

All ready. Need to merge in the NPU branch so we can actually run the experiments.