Log tests

[abejgonzalez]

No description provided.

[abejgonzalez]

Similar to #237 but addresses tests.

[colinschmidt]

I'm concerned this is losing some functionality here but perhaps you are using some fancy bash trick that I don't understand.
Have you run this with a program that prints something to the console? I believe the current behavior is to separate that from the .out file so you should make sure it matches the current behavior.

[abejgonzalez]

I have tested this with both the bmark and asm tests. I get the following in the *.out files:

testing $random 86561c0c seed          1
  C0:          0 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          1 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          2 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          3 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          4 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          5 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          6 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          7 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          8 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:          9 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R[r31=7daa29fb7daa29fb] R[r 7
  C0:         10 [0] pc=[ffffffdbed90b8db] W[r 0=0000000000000000][0] R

and in the *.log files I get the following output (which matches what comes from the console):

Chronologic VCS simulator copyright 1991-2019
Contains Synopsys proprietary information.
 Compiler version P-2019.06-SP1_Full64; Runtime version P-2019.06-SP1_Full64;  Oct 14 21:07 2019
 $finish called from file "/scratch/abejgonza/chipyard-work/chipyard-fixes/sims/vcs/generated-src/ex
 $finish at simulation time              4781850
            V C S   S i m u l a t i o n   R e p o r t
 Time: 47818500 ps
 CPU Time:     11.410 seconds;       Data structure size: 514.6Mb
 Mon Oct 14 21:08:06 2019

or when getting console output from something like mm.riscv:

C0: reg block 4x5x6, cache block 24x25x24
C0: 24688 instructions
C0: 16331 cycles
C0: 28800 flops
C0: 1763 Mflops @ 1 GHz

Before stdout and stderr were swapped and the new stdout was piped to spike-dasm then to an *.out file. What I am doing now is piping stderr to spike-dasm then to *.out then stdout to the console and a *.log file. So this is the same as before but adds an extra *.log file.

[colinschmidt]

LGTM. Thanks for the response.

[abejgonzalez]

Hmm. For some reason, after running a test, the console output can't be seen unless I reset. Im looking into it for now.

[jerryz123]

Hmm. For some reason, after running a test, the console output can't be seen unless I reset. Im looking into it for now.

This is only a issue with the newer VCS version, if that helps

[abejgonzalez]

That seems to be the case, I get rid of all the output manipulation commands and it still causes this.