Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Tue, 26 Apr 2022 07:00:22 -0700
From:   Christoph Hellwig <hch@infradead.org>
To:     Donghai Qiao <dqiao@redhat.com>
Cc:     akpm@linux-foundation.org, sfr@canb.auug.org.au, arnd@arndb.de,
        peterz@infradead.org, heying24@huawei.com,
        andriy.shevchenko@linux.intel.com, axboe@kernel.dk,
        rdunlap@infradead.org, tglx@linutronix.de, gor@linux.ibm.com,
        donghai.w.qiao@gmail.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 00/11] smp: cross CPU call interface
Message-ID: <Ymf69iLnB9sJaej0@infradead.org>
References: <20220422200040.93813-1-dqiao@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220422200040.93813-1-dqiao@redhat.com>
Precedence: bulk

On Fri, Apr 22, 2022 at 04:00:29PM -0400, Donghai Qiao wrote:
> The motivation of submitting this patch set is intended to make the
> existing cross CPU call mechanism become a bit more formal interface
> and more friendly to the kernel developers.

As far as I can tell it mostly does the reverse.  The new interfaces
are a lot more verbose for no good reason.  E.g. this is a common
pattern:


-	on_each_cpu(common_shutdown_1, &args, 0);
+	smp_call(SMP_CALL_ALL, common_shutdown_1, &args, SMP_CALL_TYPE_ASYNC);

or this:

-		smp_call_function_single(boot_cpuid, do_remote_read, &x, 1);
+		smp_call(boot_cpuid, do_remote_read, &x, SMP_CALL_TYPE_SYNC);


The old interface more or less made some sense.  The new one is just a mess.

> Patch 1: The smp cross call related structures and definitions are
> consolidated from smp.c smp_types.h to smp.h. As a result, smp_types.h
> is deleted from the source tree.

And a lot of definitions that used to be private are not public.