Add evp_rand perftool

[andrewkdinh]

Summary

Adds a CLI tool evp_rand that generates random data with HASH-DRGB.
Runs for 5 seconds and prints the average execution time per computation.

openssl/project#1840

Features

• Two modes of operation:
  • evp_shared (default): Use EVP API and allow shared data between computations
  • evp_isolated: Use EVP API and don't allow shared data between computations
• Configurable number of times to update
• Terse output for easier CI automation (-t)
• Configurable thread count

TODO: Since freeze functionality hasn't been properly added yet (see openssl/project#1835 & openssl/openssl#29433), support for freezing the context store has not been added yet.

Usage

$ ./evp_rand -h
Usage: evp_rand [-h] [-t] [-o operation] [-V] thread-count
-h - print this help output
-t - terse output
-o operation - mode of operation. One of [evp_isolated, evp_shared] (default: evp_shared)
-V - print version information and exit
thread-count - number of thread

$ ./evp_rand -o evp_shared 10 # evp_shared operation mode, 10 threads
Average time per computation: 2.888452us

$ ./evp_rand -o evp_isolated 10 # now using evp_isolated mode
Average time per computation: 9.663231us

Findings

Initial results showed that evp_isolated takes about 3x time as evp_shared

[nhorman]

should this test completely replace the randbytes test? i.e. should that one be deleted in this PR?

[andrewkdinh]

@nhorman I don’t think the randbytes test should be removed. The randbytes test calls RAND_BYTES while this perftool focuses on EVP_RAND.

[Sashan]

looks good to me.

[Sashan]

should this test completely replace the randbytes test? i.e. should that one be deleted in this PR?

I don't feel strongly about that. however if we are going for this, than it should be seperate PR.