Received-SPF: pass (google.com: domain of linux-kernel+bounces-177264-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Message-ID: <26d4ad30-c0fe-4286-9802-aa6afbd8074a@gmail.com>
Date: Mon, 13 May 2024 09:33:27 +0200
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
To: hch@lst.de
Cc: kbusch@kernel.org, kbusch@meta.com, linux-kernel@vger.kernel.org,
 linux-nvme@lists.infradead.org, ming.lei@redhat.com, tglx@linutronix.de
References: <20240510151047.GA10486@lst.de>
Subject: Re: [PATCH 2/2] nvme-pci: allow unmanaged interrupts
Content-Language: en-US
From: Benjamin Meier <benjamin.meier70@gmail.com>
In-Reply-To: <20240510151047.GA10486@lst.de>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit

 > From: Christoph Hellwig <hch@lst.de>
 >
 > So let them argue why.  I'd rather have a really, really, really
 > good argument for this crap, and I'd like to hear it from the horses
 > mouth.

I reached out to Keith to explore the possibility of manually defining
which cores handle NVMe interrupts.

The application which we develop and maintain (in the company I work)
has very high requirements regarding latency. We have some isolated cores
and we run our application on those.

Our system is using kernel 5.4 which unfortunately does not support
"isolcpus=managed_irq". Actually, we did not even know about that
option, because we are focussed on kernel 5.4. It solves part
of our problem, but being able to specify where exactly interrupts
are running is still superior in our opinion.

E.g. assume the number of house-keeping cores is small, because we
want to have full control over the system. In our case we have threads
of different priorities where some get an exclusive core. Some other threads
share a core (or a group of cores) with other threads. Now we are still
happy to assign some interrupts to some of the cores which we consider as
"medium-priority". Due to the small number of non-isolated cores, it can
be tricky to assign all interrupts to those without a performance-penalty.

Given these requirements, manually specifying interrupt/core assignments
would offer greater flexibility and control over system performance.
Moreover, the proposed code changes appear minimal and have no
impact on existing functionalities.