DPDK patches and discussions
 help / color / mirror / Atom feed
From: Jeremy Spewock <jspewock@iol.unh.edu>
To: Luca Vizzarro <Luca.Vizzarro@arm.com>
Cc: Dean Marx <dmarx@iol.unh.edu>,
	probb@iol.unh.edu, npratte@iol.unh.edu,
	 yoan.picchi@foss.arm.com, Honnappa.Nagarahalli@arm.com,
	 paul.szczepanek@arm.com, juraj.linkes@pantheon.tech,
	dev@dpdk.org
Subject: Re: [PATCH v2] dts: add flow rule dataclass to testpmd shell
Date: Fri, 2 Aug 2024 16:53:37 -0400	[thread overview]
Message-ID: <CAAA20UTEENawtpkuLouRKp9zBfVh5mqcnK-j=VKE4t-3UdYrBw@mail.gmail.com> (raw)
In-Reply-To: <15f1aeb5-3123-4c87-90b7-46b43323a6db@arm.com>

On Thu, Aug 1, 2024 at 5:24 AM Luca Vizzarro <Luca.Vizzarro@arm.com> wrote:
>
> Hi Dean, thank you for your work! Some minor comments.
>
> On 26/07/2024 15:15, Dean Marx wrote:
> > add dataclass for passing in flow rule creation arguments, as well as a
> > __str__ method for converting to a sendable testpmd command. Add
> > flow_create method to TestPmdShell class for initializing flow rules.
> >
> > Signed-off-by: Dean Marx <dmarx@iol.unh.edu>
> > ---
> >   dts/framework/remote_session/testpmd_shell.py | 58 ++++++++++++++++++-
> >   1 file changed, 57 insertions(+), 1 deletion(-)
> >
> > diff --git a/dts/framework/remote_session/testpmd_shell.py b/dts/framework/remote_session/testpmd_shell.py
> > index eda6eb320f..d6c111da0a 100644
> > --- a/dts/framework/remote_session/testpmd_shell.py
> > +++ b/dts/framework/remote_session/testpmd_shell.py
> > @@ -19,7 +19,7 @@
> >   from dataclasses import dataclass, field
> >   from enum import Flag, auto
> >   from pathlib import PurePath
> > -from typing import ClassVar
> > +from typing import ClassVar, Optional
> >
> >   from typing_extensions import Self, Unpack
> >
> > @@ -577,6 +577,43 @@ class TestPmdPortStats(TextParser):
> >       tx_bps: int = field(metadata=TextParser.find_int(r"Tx-bps:\s+(\d+)"))
> >
> >
> > +@dataclass
> > +class flow_func:
> Class names should be UpperCamelCase, also should be a suitable name for
> what it's describing. I believe FlowRule should work.
> > +    """Dataclass for setting flow rule parameters."""
> > +
> > +    #:
> > +    port_id: int
> > +    #:
> > +    ingress: bool
> > +    #:
> > +    pattern: str
> > +    #:
> > +    actions: str
> > +
> > +    #:
> > +    group_id: Optional[int] = None
> > +    #:
> > +    priority_level: Optional[int] = None
> > +    #:
> > +    user_id: Optional[int] = None
> Optional[..] is an outdated notation. `int | None` is preferred instead.
> See PEP 604[1].
> > +
> > +    def __str__(self) -> str:
> > +        """Returns the string representation of a flow_func instance.
> > +
> > +        In this case, a properly formatted flow create command that can be sent to testpmd.
> > +        """
> > +        ret = []
> > +        ret.append(f"flow create {self.port_id} ")
> > +        ret.append(f"group {self.group_id} " if self.group_id is not None else "")
> > +        ret.append(f"priority {self.priority_level} " if self.priority_level is not None else "")
> > +        ret.append("ingress " if self.ingress else "egress ")
> > +        ret.append(f"user_id {self.user_id} " if self.user_id is not None else "")
> > +        ret.append(f"pattern {self.pattern} ")
> > +        ret.append(" / end actions ")
> > +        ret.append(f"{self.actions} / end")
> > +        return "".join(ret)
>
> Using a list with inline conditional appending is not particularly
> readable. A regular string with conditional appending should do:
>
>     ret = f"flow create {self.port_id} "
>     if self.group_id is not None:
>          ret += f"group {self.group_id} "
>     ...
>
> Also the latest three append lines can all be in one, if you like the
> separation you can just do a multi-line string:
>
>     ret += (
>          f"pattern {self.pattern} / end "
>          f"actions {self.actions} / end"
>     )
>     # or actually this may be just fine:
>     ret += f"pattern {self.pattern} / end "
>     ret += f"actions {self.actions} / end"
>
> I guess the way it's split is more of a game changer.
>
> If you really want to use a list (in a way that is similar to what I've
> described here) then I'd take advantage of it... by omitting leading and
> trailing whitespaces and then use the join to add them in between: "
> ".join(ret)
>
> > +
> > +
> >   class TestPmdShell(DPDKShell):
> >       """Testpmd interactive shell.
> >
> > @@ -804,6 +841,25 @@ def show_port_stats(self, port_id: int) -> TestPmdPortStats:
> >
> >           return TestPmdPortStats.parse(output)
> >
> > +    def flow_create(self, cmd: str, verify: bool = True) -> None:
>
> Not a comment, but a discussion point. Normally we'd want a function to
> be read as an action such as:
>
>     create_flow
>
> But I understand this is basically mirroring the command format... I
> wonder which one would be the best. I am personally inclined in verb
> first. Maybe others can give their opinion.

That's funny because Patrick also raised this point about whether to
use a more testpmd-oriented naming scheme or a more human-readable
one. I forget which patch it was on exactly, but Patrick did raise a
good point that if our goal is to have DPDK developers be able to
easily pick up and work with this API, they'll probably be more
familiar with the testpmd methods as they are now. I also lean more
towards the verb first method just because it makes it more readable I
think, but I can't speak for DPDK developers. Even if it were
`create_flow` though I'm sure people can just type `testpmd.flow` and
their IDE will be able to help with the auto-complete for the flow
rule options.

Maybe it is something we should discuss more however since it is come
up twice now.

> > +        """Creates a flow rule in the testpmd session.
> > +
> > +        Args:
> > +            cmd: String from flow_func instance to send as a flow rule.
> > +            verify: If :data:`True`, the output of the command is scanned
> > +            to ensure the flow rule was created successfully.
> > +
> > +        Raises:
> > +            InteractiveCommandExecutionError: If flow rule is invalid.
> > +        """
> > +        flow_output = self.send_command(cmd)
> > +        if verify:
> > +            if "created" not in flow_output:
> > +                self._logger.debug(f"Failed to create flow rule:\n{flow_output}")
> > +                raise InteractiveCommandExecutionError(
> > +                    f"Failed to create flow rule:\n{flow_output}"
> > +                )
> > +
> >       def _close(self) -> None:
> >           """Overrides :meth:`~.interactive_shell.close`."""
> >           self.stop()
>
> [1] https://peps.python.org/pep-0604/
>

  reply	other threads:[~2024-08-02 20:53 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-26 14:15 Dean Marx
2024-08-01  9:24 ` Luca Vizzarro
2024-08-02 20:53   ` Jeremy Spewock [this message]
2024-07-26 14:22 [PATCH v1] " Dean Marx
2024-08-06 16:42 ` [PATCH v2] " Dean Marx

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAAA20UTEENawtpkuLouRKp9zBfVh5mqcnK-j=VKE4t-3UdYrBw@mail.gmail.com' \
    --to=jspewock@iol.unh.edu \
    --cc=Honnappa.Nagarahalli@arm.com \
    --cc=Luca.Vizzarro@arm.com \
    --cc=dev@dpdk.org \
    --cc=dmarx@iol.unh.edu \
    --cc=juraj.linkes@pantheon.tech \
    --cc=npratte@iol.unh.edu \
    --cc=paul.szczepanek@arm.com \
    --cc=probb@iol.unh.edu \
    --cc=yoan.picchi@foss.arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).