Re: [PATCH v8 08/20] blockjob.h: introduce block_job _locked() APIs

From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
To: Stefan Hajnoczi <stefanha@redhat.com>
Cc: qemu-block@nongnu.org, Kevin Wolf <kwolf@redhat.com>,
	Hanna Reitz <hreitz@redhat.com>,
	Paolo Bonzini <pbonzini@redhat.com>, John Snow <jsnow@redhat.com>,
	Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>,
	Wen Congyang <wencongyang2@huawei.com>,
	Xie Changlong <xiechanglong.d@gmail.com>,
	Markus Armbruster <armbru@redhat.com>, Fam Zheng <fam@euphon.net>,
	qemu-devel@nongnu.org
Subject: Re: [PATCH v8 08/20] blockjob.h: introduce block_job _locked() APIs
Date: Tue, 5 Jul 2022 10:12:10 +0200	[thread overview]
Message-ID: <d5910e65-68e0-225c-c704-94a8f20fa9c6@redhat.com> (raw)
In-Reply-To: <YsPvLS+c3SNPO7Qw@stefanha-x1.localdomain>

Am 05/07/2022 um 09:58 schrieb Stefan Hajnoczi:
> On Wed, Jun 29, 2022 at 10:15:26AM -0400, Emanuele Giuseppe Esposito wrote:
>> +BlockJob *block_job_next(BlockJob *bjob)
>>  {
>> -    Job *job = job_get(id);
>> +    JOB_LOCK_GUARD();
>> +    return block_job_next_locked(bjob);
>> +}
> 
> This seems unsafe for the same reason as job_ref(). How can the caller
> be sure bjob is still valid if it doesn't hold the mutex and has no
> reference to it?
> 
> Maybe the assumption is that the next()/get()/unref() APIs are
> GLOBAL_STATE_CODE(), so there can be no race between them?
> 

Same answer as job_ref. Unfortunately if we want to keep this logic in
this serie that's the price to pay (even though it's just till patch 13).
No assumption I would say.

Emanuele