Re: [Qemu-devel] [PATCH v5 1/4] spapr: Implement VPA dispatch counter and prod bit on tcg

From: Nicholas Piggin <npiggin@gmail.com>
To: Greg Kurz <groug@kaod.org>
Cc: qemu-devel@nongnu.org, qemu-ppc@nongnu.org,
	"Cédric Le Goater" <clg@kaod.org>,
	"David Gibson" <david@gibson.dropbear.id.au>
Subject: Re: [Qemu-devel] [PATCH v5 1/4] spapr: Implement VPA dispatch counter and prod bit on tcg
Date: Thu, 18 Jul 2019 12:18:30 +1000	[thread overview]
Message-ID: <1563416246.khbnwcjhsc.astroid@bobo.none> (raw)
In-Reply-To: <20190717172910.426f3d22@bahia.lab.toulouse-stg.fr.ibm.com>

Greg Kurz's on July 18, 2019 1:29 am:
> On Wed, 17 Jul 2019 15:39:49 +1000
> Nicholas Piggin <npiggin@gmail.com> wrote:
> 
>> -    cpu->machine_data = g_new0(SpaprCpuState, 1);
>> +    spapr_cpu = g_new0(SpaprCpuState, 1);
>> +    cpu->machine_data = spapr_cpu;
> 
> What's the purpose of this change since there's no other
> user of spapr_cpu in this function ?

It is an orphan from when the previous patch kept a dispatch_counter
in the state data structure. Now we just use the VPA directly I can
remove this completely, good catch.

>> @@ -10624,6 +10646,9 @@ static void ppc_cpu_class_init(ObjectClass *oc, void *data)
>>      cc->tcg_initialize = ppc_translate_init;
>>      cc->tlb_fill = ppc_cpu_tlb_fill;
>>  #endif
>> +    cc->cpu_exec_enter = ppc_cpu_exec_enter;
>> +    cc->cpu_exec_exit = ppc_cpu_exec_exit;
>> +
> 
> This code only makes sense with system emulation. I guess it
> should be guarded with !defined(CONFIG_USER_ONLY).

I can do that.

Thanks,
Nick