Date:   Fri, 3 Feb 2023 10:34:22 +0100 (CET)
From:   Christoph Lameter <cl@gentwo.de>
To:     Marcelo Tosatti <mtosatti@redhat.com>
cc:     Aaron Tomlin <atomlin@atomlin.com>,
        Frederic Weisbecker <frederic@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH 3/5] mm/vmstat: use cmpxchg loop in cpu_vm_stats_fold
In-Reply-To: <Y9vcpO5Dh2G3Y6ib@tpad>
Message-ID: <5615572-974e-74cc-6c34-1de618b777cf@gentwo.de>
References: <20230201195013.881721887@redhat.com> <20230201195104.460373427@redhat.com> <fb371ac-cf69-28a6-eca0-6caccda4e0b8@gentwo.de> <Y9vcpO5Dh2G3Y6ib@tpad>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Precedence: bulk

On Thu, 2 Feb 2023, Marcelo Tosatti wrote:

> > I thought you would only run this while the kernel is not active on the
> > remote cpu? Then you dont need any cmpxchg and you can leave the function
> > as is.
>
> The remote cpu can enter kernel mode while this function executes.

Isnt there some lock/serializtion to stall the kernel until you are done?

> There is no mode which indicates userspace cannot enter the kernel.

There are lot of thinngs that happen upon entry to the kernel. I would
hope that you can do something there. Scheduler?