Feature #9577: [PATCH] benchmark/driver.rb: align columns in text output

• Status: Closed
• Assignee: normalperson (Eric Wong)
• Target version: -

[normalperson (Eric Wong)]

It is easier to read output when columns are aligned properly.
I have an 80 column terminal, that's HUGE! Use it :D

pulL: git://80x24.org/ruby.git bmdriver-align
patch: http://bogomips.org/ruby.git/patch?id=32eed05c80

samples
before: http://80x24.org/bmlog-20140126-003136.7320.gz
after: http://80x24.org/bmlog-20140228-071544.1805.gz

I'll commit soon unless this breaks some tools (or this makes life hard
for the 40-column people :P)

[ko1 (Koichi Sasada)]

(2014/02/28 16:44), [e-mail] wrote:

It is easier to read output when columns are aligned properly.
I have an 80 column terminal, that's HUGE! Use it :D

Sorry, I can't read .gz file because of my environment.

What happen when many executables are specified (over 80 chars)?

--
// SASADA Koichi at atdot dot net

[normalperson (Eric Wong)]

SASADA Koichi [e-mail] wrote:

(2014/02/28 16:44), [e-mail] wrote:

It is easier to read output when columns are aligned properly.
I have an 80 column terminal, that's HUGE! Use it :D

Sorry, I can't read .gz file because of my environment.

unzipped versions:
before: http://80x24.org/bmlog-20140126-003136.7320.txt
after: http://80x24.org/bmlog-20140228-071544.1805.txt

What happen when many executables are specified (over 80 chars)?

This the timing lines after the text source is just aligned to the
longest line, so it looks like this:

ruby 2.2.0dev (2014-02-28 trunk 45211) [x86_64-linux]   0.071003417
built-ruby                                              0.0817603

Execution time summary header isn't changed, but the times themselves
look much better:

Execution time (sec)
name	ruby 2.2.0dev (2014-02-28 trunk 45211) [x86_64-linux]	built-ruby
app_answer                         0.071    0.082
app_aobench                       84.672   84.014
app_erb                            1.511    1.524

I didn't check the speedup ratio header, but maybe the "name\tbuilt-ruby"
looks a little odd. I can fix that. The first line
("compare with the result of") is unchanged (and too long, but I don't
read that).

Speedup ratio: compare with the result of `ruby 2.2.0dev (2014-02-28 trunk 45211) [x86_64-linux]' (greater is better)
name	built-ruby
app_answer                         0.868
app_aobench                        1.008
app_erb                            0.992

[naruse (Yui NARUSE)]

• Target version deleted (2.2.0)

[k0kubun (Takashi Kokubun)]

• Status changed from Open to Closed

I think this was already fixed by introducing benchmark-driver.

$ make benchmark OPTS="-r time -o simple" ITEM=app_
../revision.h unchanged
/home/k0kubun/.rbenv/shims/ruby --disable=gems -rrubygems -I../benchmark/lib ../benchmark/benchmark-driver/exe/benchmark-driver \
            --executables="compare-ruby::/home/k0kubun/.rbenv/shims/ruby --disable=gems -I.ext/common --disable-gem" \
            --executables="built-ruby::./miniruby -I../lib -I. -I.ext/common  -r../prelude --disable-gem" \
            $(find ../benchmark -maxdepth 1 -name '*app_*.yml' -o -name '*app_*.rb' | sort) -r time -o simple
Execution time (s):
                 compare-ruby  built-ruby
app_answer             0.022        0.019
app_aobench           31.575       30.098
app_erb                0.632        1.311