lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Message-ID: <39fe64b4-84bd-a40c-2638-c61652642302@gmail.com>
Date:   Wed, 10 Mar 2021 21:22:19 +0100
From:   Heiner Kallweit <hkallweit1@...il.com>
To:     Eric Dumazet <edumazet@...gle.com>
Cc:     "netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Enabling gro_flush_timeout/napi_defer_hard_irqs per default in a
 driver?

In r8169 standard irq coalescing is disabled per default, and setting
gro_flush_timeout = 20000 and napi_defer_hard_irqs = 1 reduces interrupt
number significantly, e.g. in iperf.
Therefore I wonder whether it would make sense to set reasonable defaults
for both parameters in the driver, similar to drivers enabling irq
coalescing per default. This way users could benefit who are not
experienced enough to tweak sysfs parameters.
Now to the actual question:
Are you aware of any use cases where setting these parameters could
have a negative impact, what would speak for leaving them disabled
per default?

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ