(original) (raw)
Great, thanks!
On Sat, Sep 3, 2016, 17:59 Guido van Rossum <gvanrossum@gmail.com> wrote:
Brett, I have not followed everything here but I have no problem with tweaks at this level as long as you are happy with it.
--Guido (mobile)
On Sep 3, 2016 5:39 PM, "Brett Cannon" <brett@python.org> wrote:On Sat, 3 Sep 2016 at 17:27 Yury Selivanov <yselivanov.ml@gmail.com> wrote:
On 2016-09-03 5:19 PM, Brett Cannon wrote:
\>
\>
\> On Sat, 3 Sep 2016 at 16:43 Yury Selivanov <yselivanov.ml@gmail.com
\> yselivanov.ml@gmail.com>> wrote:
\>
\>
\>
\> On 2016-09-03 4:15 PM, Christian Heimes wrote:
\> > On 2016-09-04 00:03, Yury Selivanov wrote:
\> >>
\> >> On 2016-09-03 12:27 PM, Brett Cannon wrote:
\> >>> Below is the \`co\_extra\` section of PEP 523 with the update
\> saying that
\> >>> users are expected to put a tuple in the field for easier
\> simultaneous
\> >>> use of the field.
\> >>>
\> >>> Since the \`co\_extra\` discussions do not affect CPython itself I'm
\> >>> planning on landing the changes stemming from the PEP probably
\> on Monday.
\> >> Tuples are immutable. If you have multiple co\_extra users then
\> they
\> >> will have to either mutate tuple (which isn't always possible, for
\> >> instance, you can't increase size), or to replace it with
\> another tuple.
\> >>
\> >> Creating lists is a bit more expensive, but item access speed
\> should be
\> >> in the same ballpark.
\> >>
\> >> Another question -- sorry if this was discussed before -- why
\> do we want
\> >> a PyObject\* there at all? I.e. why don't we create a dedicated
\> struct
\> >> CoExtraContainer to manage the stuff in co\_extra? My
\> understanding is
\> >> that the users of co\_extra are C-level python optimizers and
\> profilers,
\> >> which don't need the overhead of CPython API.
\>
\>
\> As Chris pointed out in another email, the overhead is only in the
\> allocation, not the iteration/access if you use the PyTuple macros to
\> get the size and index into the tuple the overhead is negligible.
Yes, my point was that it's as cheap to use a list as a tuple for
co\_extra. If we decide to store PyObject in co\_extra.
\> >>
\> >> This way my work to add an extra caching layer (which I'm very much
\> >> willing to continue to work on) wouldn't require another set of
\> extra
\> >> fields for code objects.
\> > Quick idea before I go to bed:
\> >
\> > You could adopt a similar API to OpenSSL's CRYPTO\_get\_ex\_new\_index()
\> > API,
\> >
\> https://www.openssl.org/docs/manmaster/crypto/CRYPTO\_get\_ex\_new\_index.html
\> >
\> >
\> > static int code\_index = 0;
\> >
\> > int PyCodeObject\_NewIndex() {
\> > return code\_index++;
\> > }
\> >
\> > A library like Pyjion has to acquire an index first. In further
\> calls it
\> > uses the index as offset into the new co\_extra field. Libraries
\> don't
\> > have to hard-code their offset and two libraries will never
\> conflict.
\> > PyCode\_New() can pre-populate co\_extra with a PyTuple of size
\> > code\_index. This avoids most resizes if you load Pyjion early. For
\> > code\_index == 0 leaf the field NULL.
\>
\> Sounds like a very good idea!
\>
\>
\> The problem with this is the pre-population. If you don't get your
\> index assigned before the very first code object is allocated then you
\> still have to manage the size of the tuple in co\_extra. So what this
\> would do is avoid the iteration but not the allocation overhead.
\>
\> If we open up the can of worms in terms of custom functions for this
\> (which I was trying to avoid), then you end up with Py\_ssize\_t
\> \_PyCode\_ExtraIndex(), PyObject \*
\> \_PyCode\_GetExtra(PyCodeObject \*code, Py\_ssize\_t index), and int
\> \_PyCode\_SetExtra(PyCodeObject \*code, Py\_ssize\_t index, PyObject \*data)
\> which does all the right things for creating or resizing the tuple as
\> necessary and which I think matches mostly what Nick had proposed
\> earlier. But the pseudo-code for \_PyCode\_GetExtra() would be::
\>
\> if co\_extra is None:
\> co\_extra = (None,) \* \_next\_extra\_index;
\> return None
\> elif len(co\_extra) < index - 1:
\> ... pad out tuple
\> return None
\> else:
\> return co\_extra\[index\]
\>
\> Is that going to save us enough to want to have a custom API for this?
But without that new API (basically what Christian proposed) you'd need
to iterate over the list in order to find the object that belongs to
Pyjion.Yes.If we manage to implement my opcode caching idea, we'll have at
least two known users of co\_extra. Without a way to claim a particular
index in co\_extra you will have some overhead to locate your objects.Two things. One, I would want any new API to start with an underscore so people know we can and will change its semantics as necessary. Two, Guido would have to re-accept the PEP as this is a shift in the use of the field if this is how people want to go.\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev