[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <7c2af941-42a9-a59b-6a20-b331a4934a3@gentwo.de>
Date: Tue, 10 Jan 2023 14:39:08 +0100 (CET)
From: Christoph Lameter <cl@...two.de>
To: Frederic Weisbecker <frederic@...nel.org>
cc: Marcelo Tosatti <mtosatti@...hat.com>, atomlin@...mlin.com,
tglx@...utronix.de, mingo@...nel.org, peterz@...radead.org,
pauld@...hat.com, neelx@...hat.com, oleksandr@...alenko.name,
linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH v13 2/6] mm/vmstat: Use vmstat_dirty to track CPU-specific
vmstat discrepancies
On Tue, 10 Jan 2023, Frederic Weisbecker wrote:
> Note I'm absolutely clueless with vmstat. But I was wondering about it as well
> while reviewing Marcelo's series, so git blame pointed me to:
>
> 7c83912062c801738d7d19acaf8f7fec25ea663c ("vmstat: User per cpu atomics to avoid
> interrupt disable / enable")
>
> And this seem to mention that this can race with IRQs as well, hence the local
> cmpxchg operation.
The race with irq could be an issue but I thought we avoided that and were
content with disabling preemption.
But this issue illustrates the central problem of the patchset: It makes
the lightweight counters not so lightweight anymore. The basic primitives
add a lot of weight. And the pre cpu atomic updates operations require
the modification of multiple values. The operation cannot be "atomic" in
that sense anymore and we need some other form of synchronization that can
span multiple instructions.
Powered by blists - more mailing lists