linux-kernel - kernel in spinlock or deschduled : packet loss during 1 Mpps transfer

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 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]

Date:	Fri, 8 Aug 2014 06:52:14 +0200
From:	anshul makkar <anshulmakkar@...il.com>
To:	linux-kernel@...r.kernel.org
Subject: kernel in spinlock or deschduled : packet loss during 1 Mpps transfer

Hi,

I am transferring 1 Million packets from send process to receive
process. Both are running in separate cores. I am using Using DMA
transfer from usermode to the ethernet card.

After transfer I can see 700 - 800 packets loss per transaction.

I doubt that during the transfer kernel may be entering some spinlocks
or descheduling the receiver processor for long enough.

Please can you share how can I debug this case. Please share as to
what approach I can take to detect that. The time involved in the
tranfer is very less around 1 Million packets per second and in that
second packet loss is happening.

Thanks
Anshul Makkar
www.justkernel.com
http://www.linkedin.com/groups/Just-Kernel-3033180
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/