[ovs-dev] SIGILL ovs branch-2.12/arm64/DPDK

[Yanqin Wei (Arm Technology China)]

Hi James,

CRC32 is not a mandatory feature for Armv8.0. What is the arm64 CPU used in your platform?
I'm not sure if it's necessary to use mandatory feature (neon) to optimize hash libraries, because I thought most arm64 CPUs supported CRC32.

Best Regards,
Wei Yanqin

-----Original Message-----
From: ovs-dev-bounces at openvswitch.org <ovs-dev-bounces at openvswitch.org> On Behalf Of James Page
Sent: Wednesday, September 4, 2019 4:50 PM
To: Ilya Maximets <i.maximets at samsung.com>
Cc: ovs-dev at openvswitch.org
Subject: Re: [ovs-dev] SIGILL ovs branch-2.12/arm64/DPDK

On Tue, Sep 3, 2019 at 5:48 PM Ilya Maximets <i.maximets at samsung.com> wrote:

> > Hi
> >
> > I've been testing non-x86 architecture builds for the upcoming 2.12
> release
> > of OVS and I'm hitting an issue with DPDK enabled builds on the
> > arm64 architecture.
> >
> > branch-2.12 includes improvements for native hashing under arm64;
> > these appear to work fine when DPDK is not in the mix, but with DPDK
> > enabled, I get a SIGILL in lib/hash.c:
>
> Hi.
>
> What is your target platform?
> One explanation I could imagine is that DPDK blindly forces
> -march=armv8-a+crc defining __ARM_FEATURE_CRC32 while your cpu doesn't
> support crc32 extension.
>
> Do you have crc32 in the list of cpu features in /proc/cpuinfo ?
>

It would appear that I do not:

Features : fp asimd evtstrm cpuid

Thanks for the pointer.
_______________________________________________
dev mailing list
dev at openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.