All of lore.kernel.org
 help / color / mirror / Atom feed
From: Markus Armbruster <armbru@redhat.com>
To: John Snow <jsnow@redhat.com>
Cc: "Kevin Wolf" <kwolf@redhat.com>,
	pkrempa@redhat.com, "Daniel Berrange" <berrange@redhat.com>,
	"Eduardo Habkost" <ehabkost@redhat.com>,
	qemu-block@nongnu.org, "Juan Quintela" <quintela@redhat.com>,
	libvir-list@redhat.com, qemu-devel <qemu-devel@nongnu.org>,
	mdroth@linux.vnet.ibm.com,
	"Dr. David Alan Gilbert" <dgilbert@redhat.com>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Marc-André Lureau" <marcandre.lureau@redhat.com>,
	"Eric Blake" <eblake@redhat.com>,
	libguestfs@redhat.com
Subject: Re: [PATCH 5/9] qapi: Generalize struct member policy checking
Date: Tue, 26 Oct 2021 11:14:49 +0200	[thread overview]
Message-ID: <87mtmw5era.fsf@dusky.pond.sub.org> (raw)
In-Reply-To: <CAFn=p-a9w7Qn8kVK4AADKgj-c7ewhDRhvAHKKN-omwjiSstjSw@mail.gmail.com> (John Snow's message of "Mon, 25 Oct 2021 15:28:15 -0400")

John Snow <jsnow@redhat.com> writes:

> On Mon, Oct 25, 2021 at 1:25 AM Markus Armbruster <armbru@redhat.com> wrote:
>
>> The generated visitor functions call visit_deprecated_accept() and
>> visit_deprecated() when visiting a struct member with special feature
>> flag 'deprecated'.  This makes the feature flag visible to the actual
>> visitors.  I want to make feature flag 'unstable' visible there as
>> well, so I can add policy for it.
>>
>> To let me make it visible, replace these functions by
>> visit_policy_reject() and visit_policy_skip(), which take the member's
>> special features as an argument.  Note that the new functions have the
>> opposite sense, i.e. the return value flips.
>>
>> Signed-off-by: Markus Armbruster <armbru@redhat.com>

[...]

>> diff --git a/scripts/qapi/visit.py b/scripts/qapi/visit.py
>> index 9d9196a143..e13bbe4292 100644
>> --- a/scripts/qapi/visit.py
>> +++ b/scripts/qapi/visit.py
>> @@ -21,7 +21,7 @@
>>      indent,
>>      mcgen,
>>  )
>> -from .gen import QAPISchemaModularCVisitor, ifcontext
>> +from .gen import QAPISchemaModularCVisitor, gen_special_features, ifcontext
>>  from .schema import (
>>      QAPISchema,
>>      QAPISchemaEnumMember,
>> @@ -76,7 +76,6 @@ def gen_visit_object_members(name: str,
>>                       c_type=base.c_name())
>>
>>      for memb in members:
>> -        deprecated = 'deprecated' in [f.name for f in memb.features]
>>          ret += memb.ifcond.gen_if()
>>          if memb.optional:
>>              ret += mcgen('''
>> @@ -84,14 +83,15 @@ def gen_visit_object_members(name: str,
>>  ''',
>>                           name=memb.name, c_name=c_name(memb.name))
>>              indent.increase()
>> -        if deprecated:
>> +        special_features = gen_special_features(memb.features)
>> +        if special_features != '0':
>>
>
> Would it be possible for gen_special_features to return something false-y
> instead of '0'? Do we actually *use* the '0' return anywhere other than to
> test it to see if we should include additional code?
>
> If you actually use the '0' anywhere: Go ahead and treat this as an ack. If
> you don't, can we clean this up?

gen_special_features() returns a string holding C code for a special
features bitset.  The empty bitset is 0.

The "natural" use is interpolating this value into C code.  Two
instances are visible below.

The use right here is for testing "have we got special features", so we
can skip generating code that is of no use when we don't have any.  It's
admittedly slightly unclean.

> (Sorry, I find the mcgen stuff hard to read in patch form and I am trying
> to give you a quick review instead of NO review.)

Any kind of review is appreciated :)

> --js
>
>
>>              ret += mcgen('''
>> -    if (!visit_deprecated_accept(v, "%(name)s", errp)) {
>> +    if (visit_policy_reject(v, "%(name)s", %(special_features)s, errp)) {
>>          return false;
>>      }
>> -    if (visit_deprecated(v, "%(name)s")) {
>> +    if (!visit_policy_skip(v, "%(name)s", %(special_features)s)) {
>>  ''',
>> -                         name=memb.name)
>> +                         name=memb.name, special_features=special_features)

These are the "natural" uses I mentioned.

If gen_special_features() returned something other than '0', we'd have
to replace it by '0' here.

>>              indent.increase()
>>          ret += mcgen('''
>>      if (!visit_type_%(c_type)s(v, "%(name)s", &obj->%(c_name)s, errp)) {
>> @@ -100,7 +100,7 @@ def gen_visit_object_members(name: str,
>>  ''',
>>                       c_type=memb.type.c_name(), name=memb.name,
>>                       c_name=c_name(memb.name))
>> -        if deprecated:
>> +        if special_features != '0':
>>              indent.decrease()
>>              ret += mcgen('''
>>      }
>> --
>> 2.31.1
>>
>>



  reply	other threads:[~2021-10-26  9:17 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-25  5:25 [PATCH 0/9] Configurable policy for handling unstable interfaces Markus Armbruster
2021-10-25  5:25 ` [PATCH 1/9] qapi: New special feature flag "unstable" Markus Armbruster
2021-10-25  7:15   ` Juan Quintela
2021-10-25 12:05   ` Kashyap Chamarthy
2021-10-25 12:18     ` Philippe Mathieu-Daudé
2021-10-26  7:15     ` Markus Armbruster
2021-10-26  8:14       ` Kashyap Chamarthy
2021-10-25 19:01   ` John Snow
2021-10-26  5:35     ` Markus Armbruster
2021-10-26  7:37   ` Kevin Wolf
2021-10-26  8:36     ` [Libguestfs] " Kashyap Chamarthy
2021-10-26  9:22     ` Dr. David Alan Gilbert
2021-10-26  9:28       ` Daniel P. Berrangé
2021-10-26  9:37     ` Markus Armbruster
2021-10-26 11:34       ` Kevin Wolf
2021-10-28  8:36         ` Markus Armbruster
2021-10-26 14:56       ` Daniel P. Berrangé
2021-10-26 15:15         ` Markus Armbruster
2021-10-26 15:22           ` Daniel P. Berrangé
2021-10-27  5:29             ` Markus Armbruster
2021-10-25  5:25 ` [PATCH 2/9] qapi: Mark unstable QMP parts with feature 'unstable' Markus Armbruster
2021-10-25  7:16   ` Juan Quintela
2021-10-25 19:05   ` John Snow
2021-10-26  7:56     ` Markus Armbruster
2021-10-26 16:34       ` John Snow
2021-10-27  5:44         ` Markus Armbruster
2021-10-29 13:07   ` Eric Blake
2021-11-02  9:15     ` Kevin Wolf
2021-10-25  5:25 ` [PATCH 3/9] qapi: Eliminate QCO_NO_OPTIONS for a slight simplification Markus Armbruster
2021-10-25  7:17   ` Juan Quintela
2021-10-25 19:08   ` John Snow
2021-10-26  7:58     ` Markus Armbruster
2021-10-25  5:25 ` [PATCH 4/9] qapi: Tools for sets of special feature flags in generated code Markus Armbruster
2021-10-25 19:21   ` John Snow
2021-10-26  8:48     ` Markus Armbruster
2021-10-25  5:25 ` [PATCH 5/9] qapi: Generalize struct member policy checking Markus Armbruster
2021-10-25 19:28   ` John Snow
2021-10-26  9:14     ` Markus Armbruster [this message]
2021-10-26 15:43   ` Philippe Mathieu-Daudé
2021-10-27  5:56     ` Markus Armbruster
2021-10-25  5:25 ` [PATCH 6/9] qapi: Generalize command " Markus Armbruster
2021-10-25 12:04   ` Philippe Mathieu-Daudé
2021-10-26  9:40     ` Markus Armbruster
2021-10-25 19:30   ` John Snow
2021-10-25  5:25 ` [PATCH 7/9] qapi: Generalize enum member " Markus Armbruster
2021-10-25 19:36   ` John Snow
2021-10-26  9:43     ` Markus Armbruster
2021-10-26 16:14       ` John Snow
2021-10-25  5:25 ` [PATCH 8/9] qapi: Factor out compat_policy_input_ok() Markus Armbruster
2021-10-25 12:13   ` Philippe Mathieu-Daudé
2021-10-26  9:46     ` Markus Armbruster
2021-10-26 10:39       ` Philippe Mathieu-Daudé
2021-10-29 14:08         ` Markus Armbruster
2021-10-25 19:38   ` John Snow
2021-10-25  5:25 ` [PATCH 9/9] qapi: Extend -compat to set policy for unstable interfaces Markus Armbruster
2021-10-25 19:40   ` John Snow
2021-10-26 18:45   ` Dr. David Alan Gilbert
2021-10-27  5:59     ` Markus Armbruster

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=87mtmw5era.fsf@dusky.pond.sub.org \
    --to=armbru@redhat.com \
    --cc=berrange@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=eblake@redhat.com \
    --cc=ehabkost@redhat.com \
    --cc=jsnow@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=libguestfs@redhat.com \
    --cc=libvir-list@redhat.com \
    --cc=marcandre.lureau@redhat.com \
    --cc=mdroth@linux.vnet.ibm.com \
    --cc=pbonzini@redhat.com \
    --cc=pkrempa@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@redhat.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.