From: Claudio Fontana <cfontana@suse.de>
To: Eduardo Habkost <ehabkost@redhat.com>
Cc: "Laurent Vivier" <lvivier@redhat.com>,
"Thomas Huth" <thuth@redhat.com>,
"Stefano Stabellini" <sstabellini@kernel.org>,
"Paul Durrant" <paul@xen.org>, "Olaf Hering" <ohering@suse.de>,
"Jason Wang" <jasowang@redhat.com>,
"Marcelo Tosatti" <mtosatti@redhat.com>,
"Richard Henderson" <richard.henderson@linaro.org>,
qemu-devel@nongnu.org, "Peter Xu" <peterx@redhat.com>,
"Dario Faggioli" <dfaggioli@suse.com>,
"Roman Bolshakov" <r.bolshakov@yadro.com>,
"Cameron Esfahani" <dirty@apple.com>,
haxm-team@intel.com, "Wenchao Wang" <wenchao.wang@intel.com>,
"Anthony Perard" <anthony.perard@citrix.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Sunil Muthuswamy" <sunilmut@microsoft.com>,
"Bruce Rogers" <brogers@suse.com>,
"Philippe Mathieu-Daudé" <philmd@redhat.com>,
"Colin Xu" <colin.xu@intel.com>
Subject: Re: [RFC v4 9/9] i386: split cpu accelerators from cpu.c
Date: Fri, 20 Nov 2020 19:47:11 +0100 [thread overview]
Message-ID: <5ac27efa-0766-c5e4-be6c-7ba031997cd3@suse.de> (raw)
In-Reply-To: <20201120174447.GC2271382@habkost.net>
On 11/20/20 6:44 PM, Eduardo Habkost wrote:
> On Fri, Nov 20, 2020 at 03:49:09PM +0100, Claudio Fontana wrote:
>> split cpu.c into:
>>
>> cpu.c cpuid and common x86 cpu functionality
>> host-cpu.c host x86 cpu functions and "host" cpu type
>> kvm/cpu.c KVM x86 cpu type
>> hvf/cpu.c HVF x86 cpu type
>> tcg/cpu.c TCG x86 cpu type
>>
>> The link to the accel class is set in the X86CPUClass classes
>> at MODULE_INIT_ACCEL_CPU time, when the accelerator is known.
>>
>> Signed-off-by: Claudio Fontana <cfontana@suse.de>
> [...]
>> +static void hvf_cpu_accel_class_init(ObjectClass *oc, void *data)
>> +{
>> + X86CPUAccelClass *acc = X86_CPU_ACCEL_CLASS(oc);
>> +
>> + acc->cpu_realizefn = host_cpu_realizefn;
>> + acc->cpu_common_class_init = hvf_cpu_common_class_init;
>> + acc->cpu_instance_init = hvf_cpu_instance_init;
>> +};
>> +static const TypeInfo hvf_cpu_accel_type_info = {
>> + .name = X86_CPU_ACCEL_TYPE_NAME("hvf"),
>> +
>> + .parent = TYPE_X86_CPU_ACCEL,
>> + .class_init = hvf_cpu_accel_class_init,
>> +};
>> +static void hvf_cpu_accel_register_types(void)
>> +{
>> + type_register_static(&hvf_cpu_accel_type_info);
>> +}
>> +type_init(hvf_cpu_accel_register_types);
>> +
>> +static void hvf_cpu_accel_init(void)
>> +{
>> + if (hvf_enabled()) {
>> + x86_cpu_accel_init(X86_CPU_ACCEL_TYPE_NAME("hvf"));
>> + }
>> +}
>> +
>> +accel_cpu_init(hvf_cpu_accel_init);
>
> The point of my suggestion of using QOM is to not require
> separate accel_cpu_init() functions and (hvf|tcg|kvm)_enabled()
> checks.
>
> If we still have separate accel_cpu_init() functions for calling
> x86_cpu_accel_init() with the right argument, using a pointer to
> static variables like &hvf_cpu_accel (like you did before) was
> simpler and required less boilerplate code.
Yes I agree.
>
> However, the difference is that with the X86_CPU_ACCEL_TYPE_NAME
> macro + object_class_by_name(), you don't need the separate
> accel_cpu_init() functions for each accelerator.
>
> All you need is a single:
>
> x86_cpu_accel_init(X86_CPU_ACCEL_TYPE_NAME(chosen_accel_name));
>
> call somewhere in the initialization path.
Makes sense. The problem is just determining chosen_accel_name.
>
> A good place for the x86_cpu_accel_init() call would be
> do_configure_accelerator(), but the function is arch-specific.
> That's why I suggested a cpu_accel_arch_init() function at
> https://lore.kernel.org/qemu-devel/20201118220750.GP1509407@habkost.net
>
Fine by me. I'd use a specific init step for this, but that also works.
Ciao,
Clauidio
next prev parent reply other threads:[~2020-11-20 18:48 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-20 14:49 [RFC v4 0/9] i386 cleanup Claudio Fontana
2020-11-20 14:49 ` [RFC v4 1/9] i386: move kvm accel files into kvm/ Claudio Fontana
2020-11-20 14:49 ` [RFC v4 2/9] i386: move whpx accel files into whpx/ Claudio Fontana
2020-11-20 14:49 ` [RFC v4 3/9] i386: move hax accel files into hax/ Claudio Fontana
2020-11-20 14:49 ` [RFC v4 4/9] i386: hvf: remove stale MAINTAINERS entry for old hvf stubs Claudio Fontana
2020-11-20 14:49 ` [RFC v4 5/9] i386: move TCG accel files into tcg/ Claudio Fontana
2020-11-20 14:49 ` [RFC v4 6/9] i386: move cpu dump out of helper.c into cpu-dump.c Claudio Fontana
2020-11-20 14:49 ` [RFC v4 7/9] i386: move TCG cpu class initialization out of helper.c Claudio Fontana
2020-11-20 14:49 ` [RFC v4 8/9] module: introduce MODULE_INIT_ACCEL_CPU Claudio Fontana
2020-11-20 14:49 ` [RFC v4 9/9] i386: split cpu accelerators from cpu.c Claudio Fontana
2020-11-20 15:34 ` Claudio Fontana
2020-11-20 17:21 ` Eduardo Habkost
2020-11-20 17:44 ` Eduardo Habkost
2020-11-20 18:47 ` Claudio Fontana [this message]
2020-11-20 19:00 ` Eduardo Habkost
2020-11-23 9:17 ` Claudio Fontana
2020-11-20 15:12 ` [RFC v4 0/9] i386 cleanup no-reply
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=5ac27efa-0766-c5e4-be6c-7ba031997cd3@suse.de \
--to=cfontana@suse.de \
--cc=anthony.perard@citrix.com \
--cc=brogers@suse.com \
--cc=colin.xu@intel.com \
--cc=dfaggioli@suse.com \
--cc=dirty@apple.com \
--cc=ehabkost@redhat.com \
--cc=haxm-team@intel.com \
--cc=jasowang@redhat.com \
--cc=lvivier@redhat.com \
--cc=mtosatti@redhat.com \
--cc=ohering@suse.de \
--cc=paul@xen.org \
--cc=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=philmd@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=r.bolshakov@yadro.com \
--cc=richard.henderson@linaro.org \
--cc=sstabellini@kernel.org \
--cc=sunilmut@microsoft.com \
--cc=thuth@redhat.com \
--cc=wenchao.wang@intel.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).