Issue #12589 has been updated by vmakarov (Vladimir Makarov).


normalperson (Eric Wong) wrote:
> vmakarov / redhat.com wrote:
>  > https://github.com/vnmakarov/ruby/compare/rtl_mjit_branch_base...rtl_mjit_branch
>  
>  I've only taken a light look at it; but I think RTL seems
>  interesting promise.  I needed the following patch to remove
>  "restrict" to compile on Debian stable:
>  
>  	https://80x24.org/spew/20170408231647.8664-1-e / 80x24.org/raw
>  
>  I also noted some rubyspec failures around break/while loops which
>  might be RTL related (make update-rubyspec && make test-rubyspec):
>  
>  	https://80x24.org/spew/20170408231930.GA11999@starla/
>  
>  (The Random.urandom can be ignored since you're on an old version)
>  

Thank you for your feedback, Eric.  I'll work on issues you found.

So far I spent about 80% of my MRI efforts on RTL.  But probably it was because of the learning curve.  I did not try RTL on serious Ruby applications yet.  On small benchmarks, I got from 0% to 100% (for a simple while loop) improvement.  I'd say the average improvement could be 10%.  MRI has too many calls on which majority of time spent.  So savings on less insn dispatching and memory traffic have a small impact.  In some cases RTL can be even worse.  For example, `o.m(a1, a2, a3)` has the following stack insns and RTL insns:

```
  push <o index>
  push <a1 index>
  push <a2 index>
  push <a3 index>
  send <callinfo> <cache>
```  

```
  loc2temp -2, <a1 index>
  loc2temp -3, <a2 index>
  loc2temp -4, <a3 index>
  call_recv <call data>, <o index>, -1
```

RTL insns are 18% longer for this example.  I am going to investigate what the overall length of executed stack insns vs RTL insns when I resume my work on the project.

>  I haven't tried JIT, yet, as I'm already unhappy with current
>  Ruby memory usage; but if RTL alone can provide small speed
>  improvements without significant footprint I can deal with it.
>

I believe there would be no additional footprint for RTL insn or there would an insignificant increase (1-2%).

JIT is ready only for small benchmarks right now.  My big worry is in using exec wrapper when we go from JITed code execution to interpreted code execution to another JITed code and so on.  It might increase stack usage.  But I am going to work on removing exec wrapper usage in some cases.

If you are not happy with the current MRI memory footprint, you will be definitely unhappy with any JIT because their work will require much more peak memory (at least in order of magnitude) than the current MRI footprint.

But I think with my approach I can use much less memory and CPUs (JITs might require more CPU usage because of the compilations) than jruby or Graal.  My JIT will also have no startup delay which is huge for jruby and Graal. Still achieving a better performance (wall clock execution) should be the first priority of my JIT project.

By the way, I forgot to mention that my approach also opens a possibility in future to distribute gems in C code without binaries and it might help gems portability.
  
>  I'm currently running dtas-player with RTL to play music and it
>  seems fine <https://80x24.org/dtas/>

Great! Thank you for sharing this.


----------------------------------------
Feature #12589: VM performance improvement proposal
https://bugs.ruby-lang.org/issues/12589#change-64128

* Author: vmakarov (Vladimir Makarov)
* Status: Open
* Priority: Normal
* Assignee: 
* Target version: 
----------------------------------------
  Hello.  I'd like to start a big MRI project but I don't want to
disrupt somebody else plans.  Therefore I'd like to have MRI
developer's opinion on the proposed project or information if somebody
is already working on an analogous project.

  Basically I want to improve overall MRI VM performance:

  * First of all, I'd like to change VM insns and move from
    **stack-based** insns to **register transfer** ones.  The idea behind
    it is to decrease VM dispatch overhead as approximately 2 times
    less RTL insns are necessary than stack based insns for the same
    program (for Ruby it is probably even less as a typical Ruby program
    contains a lot of method calls and the arguments are passed through
    the stack).

    But *decreasing memory traffic* is even more important advantage
    of RTL insns as an RTL insn can address temporaries (stack) and
    local variables in any combination.  So there is no necessity to
    put an insn result on the stack and then move it to a local
    variable or put variable value on the stack and then use it as an
    insn operand.  Insns doing more also provide a bigger scope for C
    compiler optimizations.

    The biggest changes will be in files compile.c and insns.def (they
    will be basically rewritten).  **So the project is not a new VM
    machine.  MRI VM is much more than these 2 files.**

    The disadvantage of RTL insns is a bigger insn memory footprint
    (which can be upto 30% more) although as I wrote there are fewer
    number of RTL insns.

    Another disadvantage of RTL insns *specifically* for Ruby is that
    insns for call sequences will be basically the same stack based
    ones but only bigger as they address the stack explicitly.

  * Secondly, I'd like to **combine some frequent insn sequences** into
    bigger insns.  Again it decreases insn dispatch overhead and
    memory traffic even more.  Also it permits to remove some type
    checking.

    The first thing on my mind is a sequence of a compare insn and a
    branch and using immediate operands besides temporary (stack) and
    local variables.  Also it is not a trivial task for Ruby as the
    compare can be implemented as a method.

  I already did some experiments.  RTL insns & combining insns permits
to speed the following micro-benchmark in more 2 times:

```
i = 0
while i<30_000_000 # benchmark loop 1
  i += 1
end
```

The generated RTL insns for the benchmark are

```
== disasm: #<ISeq:<main>@while.rb>======================================
== catch table
| catch type: break  st: 0007 ed: 0020 sp: 0000 cont: 0020
| catch type: next   st: 0007 ed: 0020 sp: 0000 cont: 0005
| catch type: redo   st: 0007 ed: 0020 sp: 0000 cont: 0007
|------------------------------------------------------------------------
local table (size: 2, temp: 1, argc: 0 [opts: 0, rest: -1, post: 0, block: -1, kw: -1@-1, kwrest: -1])
[ 2] i
0000 set_local_val    2, 0                                            (   1)
0003 jump             13                                              (   2)
0005 jump             13
0007 plusi            <callcache>, 2, 2, 1, -1                        (   3)
0013 btlti            7, <callcache>, -1, 2, 30000000, -1             (   2)
0020 local_ret        2, 0                                            (   3)
```

In this experiment I ignored trace insns (that is another story) and a
complication that a integer compare insn can be re-implemented as a
Ruby method.  Insn bflti is combination of LT immediate compare and
branch true.

A modification of fib benchmark is sped up in 1.35 times:

```
def fib_m n
  if n < 1
    1
  else
    fib_m(n-1) * fib_m(n-2)
  end
end

fib_m(40)
```

The RTL code of fib_m looks like

```
== disasm: #<ISeq:fib_m / fm.rb>==========================================
local table (size: 2, temp: 3, argc: 1 [opts: 0, rest: -1, post: 0, block: -1, kw: -1@-1, kwrest: -1])
[ 2] n<Arg>
0000 bflti            10, <callcache>, -1, 2, 1, -1                   (   2)
0007 val_ret          1, 16
0010 minusi           <callcache>, -2, 2, 1, -2                       (   5)
0016 simple_call_self <callinfo!mid:fib_m, argc:1, FCALL|ARGS_SIMPLE>, <callcache>, -1
0020 minusi           <callcache>, -3, 2, 2, -3
0026 simple_call_self <callinfo!mid:fib_m, argc:1, FCALL|ARGS_SIMPLE>, <callcache>, -2
0030 mult             <callcache>, -1, -1, -2, -1
0036 temp_ret         -1, 16
```

In reality, the improvement of most programs probably will be about
10%.  That is because of very dynamic nature of Ruby (a lot of calls,
checks for redefinition of basic type operations, checking overflows
to switch to GMP numbers).  For example, integer addition can not be
less than about x86-64 17 insns out of the current 50 insns on the
fast path.  So even if you make the rest (33) insns 2 times faster,
the improvement will be only 30%.

A very important part of MRI performance improvement is to make calls
fast because there are a lot of them in Ruby but as I read in some
Koichi Sasada's presentations he pays a lot of attention to it.  So I
don't want to touch it.

  * Thirdly.  I want to implement the insns as small inline functions
    for future AOT compiler, of course, if the projects described
    above are successful.  It will permit easy AOT generation of C code
    which will be basically calls of the functions.

    I'd like to implement AOT compiler which will generate a Ruby
    method code, call a C compiler to generate a binary shared code
    and load it into MRI for subsequent calls.  The key is to minimize
    the compilation time.  There are many approaches to do it but I
    don't want to discuss it right now.

    C generation is easy and most portable implementation of AOT but
    in future it is possible to use GCC JIT plugin or LLVM IR to
    decrease overhead of C scanner/parser.

    C compiler will see a bigger scope (all method insns) to do
    optimizations.  I think using AOT can give another 10%
    improvement.  It is not that big again because of dynamic nature
    of Ruby and any C compiler is not smart enough to figure out
    aliasing for typical generated C program.

    The life with the performance point of view would be easy if Ruby
    did not permit to redefine basic operations for basic types,
    e.g. plus for integer.  In this case we could evaluate types of
    operands and results using some data flow analysis and generate
    faster specialized insns.  Still a gradual typing if it is
    introduced in future versions of Ruby would help to generate such
    faster insns.

  Again I wrote this proposal for discussion as I don't want to be in
a position to compete with somebody else ongoing big project.  It
might be counterproductive for MRI development.  Especially I don't
want it because the project is big and long and probably will have a
lot of tehcnical obstacles and have a possibilty to be a failure.




-- 
https://bugs.ruby-lang.org/

Unsubscribe: <mailto:ruby-core-request / ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>