Received-SPF: pass (google.com: domain of linux-kernel+bounces-57868-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Precedence: bulk
MIME-Version: 1.0
References: <CABXGCsNnUfCCYVSb_-j-a-cAdONu1r6Fe8p2OtQ5op_wskOfpw@mail.gmail.com>
 <Zb6D/5R8nNrxveAP@cae.in-ulm.de> <Zb/30qOGYAH4j6Mn@cae.in-ulm.de>
 <CABXGCsPu73D+JS9dpvzX78RktK2VOv_xT8vvuVaQ=B6zs2dMNQ@mail.gmail.com>
 <e7b96819-edf7-1f9f-7b01-e2e805c99b33@linux.intel.com> <CABXGCsPjW_Gr4fGBzYSkr_4tsn0fvuT72G-YJYXcb1a4kX=CQw@mail.gmail.com>
 <2d87509a-1515-520c-4b9e-bba4cd4fa2c6@linux.intel.com> <CABXGCsPdXqRG6v97KDGy+o59xc3ayaq3rLj267veC7YcKVp8ww@mail.gmail.com>
 <1126ed0a-bfc1-a752-1b5e-f1339d7a8aa5@linux.intel.com>
In-Reply-To: <1126ed0a-bfc1-a752-1b5e-f1339d7a8aa5@linux.intel.com>
From: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Date: Thu, 8 Feb 2024 15:32:05 +0500
Message-ID: <CABXGCsN5_O3iKDOyYxtsGTGDA6fw4962CjzXLSnOK3rscELq+Q@mail.gmail.com>
Subject: =?UTF-8?Q?Re=3A_This_is_the_fourth_time_I=E2=80=99ve_tried_to_find_wha?=
	=?UTF-8?Q?t_led_to_the_regression_of_outgoing_network_speed_and_each_t?=
	=?UTF-8?Q?ime_I_find_the_merge_commit_8c94ccc7cd691472461448f98e2372c7?=
	=?UTF-8?Q?5849406c?=
To: Mathias Nyman <mathias.nyman@linux.intel.com>
Cc: "Christian A. Ehrhardt" <lk@c--e.de>, niklas.neronin@linux.intel.com, 
	Linux List Kernel Mailing <linux-kernel@vger.kernel.org>, Greg KH <gregkh@linuxfoundation.org>, 
	linux-usb@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Feb 8, 2024 at 2:23=E2=80=AFPM Mathias Nyman
<mathias.nyman@linux.intel.com> wrote:
>
> My guess is that CPU0 spends more time with interrupts disabled than othe=
r CPUs.
> Either because it's handling interrupts from some other hardware, or runn=
ing
> code that disables interrupts (for example kernel code inside spin_lock_i=
rq),
> and thus not able to handle network adapter interrupts at the same rate a=
s CPU23
>

Can this be fixed?
Can I help you here with anything else?

--=20
Best Regards,
Mike Gavrilov.