[Vm-dev] Cog Primitive Performance

Clément Bera bera.clement at gmail.com
Tue Apr 18 16:57:54 UTC 2017


On Tue, Apr 18, 2017 at 9:09 AM, Nicolas Cellier <
nicolas.cellier.aka.nice at gmail.com> wrote:

>
>
>
> 2017-04-18 16:03 GMT+02:00 David T. Lewis <lewis at mail.msen.com>:
>
>>
>> On Tue, Apr 18, 2017 at 10:27:41AM +0200, Nicolas Cellier wrote:
>> >
>> > 2017-04-18 4:02 GMT+02:00 Andres Valloud <avalloud at smalltalk.comcastbiz
>> .net>
>> > :
>> >
>> > >
>> > > On 4/17/17 16:45 , Eliot Miranda wrote:
>> > >
>> > >> FYI, hash multiply is
>> > >> hashMultiply
>> > >> | low |
>> > >> low := self bitAnd: 16383.
>> > >> ^(16r260D * low + ((16r260D * (self bitShift: -14) + (16r0065 * low)
>> > >> bitAnd: 16383) * 16384))
>> > >> bitAnd: 16r0FFFFFFF
>> > >>
>> > >> and when written as a primitive is
>> > >> hashMultiply
>> > >> | low |
>> > >> low := self bitAnd: 16383.
>> > >> ^(16r260D * low + (16r260D * (self bitShift: -14) + (16r0065 * low) *
>> > >> 16384))
>> > >> bitAnd: 16r0FFFFFFF
>> > >> because we don't care about the multiply by 16384 overflowing; when
>> > >> written as a primitive it won't overflow into a LargePositiveInteger.
>> > >>
>> > >
>> > > Hopefully this doesn't mean the primitive got implemented by actually
>> > > doing these operations verbatim.  As you have probably seen, that
>> > > convoluted arithmetic is done this way in Smalltalk only to simulate a
>> > > 32x32 multiplication bit-anded down to 28 bits without overflowing
>> into
>> > > large integers (the original code from August 2000 had my initials).
>> > >
>> > > At a sufficiently low level such as C, all that complexity is just an
>> > > unsigned multiplication by 1664525.  The image code should still have
>> a
>> > > comment to that effect, did it get lost?
>> > >
>> > > Andres.
>> > >
>> >
>> > More: if it's a 64 bit image, then we have 60 bit long unsigned small
>> int
>> > since 1664525 highBit = 21, and self is a hash result not exceeding 30
>> > bits, we can implement hashMultiply as
>> >     ^self * 1664525 bitAnd: 16r0FFFFFFF
>> >
>> > Maybe the JIT can be given a second chance too.
>>
>> AFAIK, primitiveStringHash has always been translated directly from
>> ByteArray class>>hashBytes:startingWith: and the hashMultiply logic that
>> we are
>> discussing here currently is repeated in four different methods in Squeak
>> trunk
>> (String class>>stringHash:initialHash:, TypeString
>> class>>stringHash:initialHash:,
>> SmallInteger hashMultiply, and ByteArray class hashBytes:startingWith:).
>>
>> At this point, and particularly given the differences in various VMs, it
>> would
>> make sense to reimplement primitiveStringHash directly in
>> MiscPrimitivesPlugin
>> with whatever optimizations are needed (maybe differently based on size
>> of a
>> small integers), rather than translating it from a default implementation
>> in
>> the image.
>>
>> Hi David,
> My opinion is rather that we should get rid of MiscPrimitivesPlugin.
> 1) If String hash is performance critical (and it is), it's not really
> Misc.
> 2) Eliot's post shows that regular primitive invocation is slow vs modern
> spur/jit, so it might not pay
> 3) other than that, we don't properly version control MiscPrimitivePlugin
> source code
>    it's not handled in VMMaker repository nor in one of the well
> identified external plugin repositories.
>    but it's spreaded over Squeak and/or Pharo packages, and thus subject
> to uncontrolled changes and divergences
>
> Maybe it's time to do it.
>

I agree. I am ready to collaborate on this and update the Pharo packages.

For each primitive, we need to evaluate the performance with and without
the primitive. In each case, we either move the primitive to numbered
primitive or we just use normal Smalltalk code. We looked into string
hashing with Eliot and it seems having only hashMultiply as a primitive is
enough.

>
> But I think we may be missing Eliot's main point. There is overhead
>> associated
>> with accessing the stack (at least on a stack interpreter) and in
>> returning
>> values to image, and this will be the case regardless of what
>> calculations are
>> being done in the primitive itself. So at least for some VMs we may be
>> reaching
>> the point where that overhead becomes a significant part of the
>> performance profile.
>>
>> Whether that overhead would justify creating a new primitive callout
>> mechanism
>> may be another question, but it is certainly worth understanding where the
>> overhead occurs, and how that performance profile changes as the Cog jit
>> and
>> Sista progress.
>
> Dave
>>
>>
> Yes, exactly, but the new callout mechanism has severe stack limitations
> preventing general purpose usage.
>

The idea there is that each primitive can be analysed to determine if it
can be called that way or not. I believe large integers primitives may be
able to be called that way for example.


> Concerning the specific case of hashMultiply, since it's just an imul and
> a bitAnd, the primitive should be inlined by the VM just like other basic
> arithmetic operations IMO.
> So the remarks of Andres makes sense, even if we can't generalize to other
> Misc primitives.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/vm-dev/attachments/20170418/edfbdf74/attachment-0001.html>


More information about the Vm-dev mailing list