[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20180117.164932.1269304606476934540.davem@davemloft.net>
Date: Wed, 17 Jan 2018 16:49:32 -0500 (EST)
From: David Miller <davem@...emloft.net>
To: torvalds@...ux-foundation.org
Cc: efault@....de, peterz@...radead.org, edumazet@...gle.com,
dima@...sta.com, frederic@...nel.org, linux-kernel@...r.kernel.org,
0x7f454c46@...il.com, akpm@...ux-foundation.org,
fweisbec@...il.com, hannes@...essinduktion.org, mingo@...nel.org,
alexander.levin@...izon.com, pabeni@...hat.com,
paulmck@...ux.vnet.ibm.com, rrendec@...sta.com, riel@...hat.com,
sgruszka@...hat.com, tglx@...utronix.de, wanpeng.li@...mail.com
Subject: Re: [RFC 1/2] softirq: Defer net rx/tx processing to ksoftirqd
context
From: Linus Torvalds <torvalds@...ux-foundation.org>
Date: Wed, 17 Jan 2018 13:06:58 -0800
> It was in some way always a "poor mans interrupt thread" (with no
> blocking like a real thread context, but at least not impacting actual
> interrupt latency).
Or in this loopback device case (and tunnel decapsulation) a poor
man's longjmp, releasing the current stack frame to keep the depth
in check.
Anyways...
> That said, this made me wonder a bit. I wonder how bounded the latency
> is for raising a softirq from process context. We only _check_ the
> softirq on the last hardirq exit, I think.
System call return checks it, otherwise this situation would be
completely bolixed.
> I wonder if we should run softirqs on return to user mode (and make
> softirq set a thread flag if not in interrupt context).
I'm pretty sure we already do.
Powered by blists - more mailing lists