Hello together,
I just noticed, there is a difference in the default value of `/proc/sys/net/core/optmem_max` on armv7l:
On all arches it is 20480, but on armv7l it is 10240.
Is there any specific reason for limiting the maximum ancillary buffer size allowed per socket on this arch?
Cheers, Björn
I just noticed, there is a difference in the default value of `/proc/sys/net/core/optmem_max` on armv7l:
On all arches it is 20480, but on armv7l it is 10240.
Is there any specific reason for limiting the maximum ancillary buffer size allowed per socket on this arch?
No specific reason, I think that would be an upstream kernel default, I'm not aware we tweak that setting on any of the architectures so it should be the arch default.
On 01/11/2018 06:17 PM, Peter Robinson wrote:
I just noticed, there is a difference in the default value of `/proc/sys/net/core/optmem_max` on armv7l:
On all arches it is 20480, but on armv7l it is 10240.
Is there any specific reason for limiting the maximum ancillary buffer size allowed per socket on this arch?
No specific reason, I think that would be an upstream kernel default, I'm not aware we tweak that setting on any of the architectures so it should be the arch default.
The actual root cause is that the default is based off of sizeof(unsigned long) which is different on 32-bit targets.
Thanks, Laura
Am Montag, den 15.01.2018, 08:42 -0800 schrieb Laura Abbott:
On 01/11/2018 06:17 PM, Peter Robinson wrote:
I just noticed, there is a difference in the default value of `/proc/sys/net/core/optmem_max` on armv7l:
On all arches it is 20480, but on armv7l it is 10240.
Is there any specific reason for limiting the maximum ancillary buffer size allowed per socket on this arch?
No specific reason, I think that would be an upstream kernel default, I'm not aware we tweak that setting on any of the architectures so it should be the arch default.
The actual root cause is that the default is based off of sizeof(unsigned long) which is different on 32-bit targets.
I just checked and found `unsigned long` to be 32 bits on %{arm}, only. All other arches (even %{ix86}) have it set to 64 bits.
unsigned long long seems to be at least 64-bit on all arches.
This has been the setting for a very long time, is there a reason changing it via sysctl is not an option?
Well, yes… One cannot change it using sysctl (or using other ways) without being root (and thus isn't an option for Koji builds).
The preset can be changed to be the same on all arches (based on changing that in systemd) by adding `net.core.optmem_max = 20480` to `/usr/lib/sysctl.d/50-default.conf`, tho.
Any ideas towards that?
Let's keep the further discussion in [1].
Cheers, Björn
kernel@lists.fedoraproject.org
-
Björn 'besser82' Esser -
Florian Weimer -
Laura Abbott -
Peter Robinson