DPDK usage discussions
 help / color / mirror / Atom feed
From: Suraj R Gupta <surajrgupta@iith.ac.in>
To: dev@dpdk.org, users@dpdk.org
Cc: "Trahe, Fiona" <fiona.trahe@intel.com>,
	tomasz.kantecki@intel.com,  marcel.d.cornu@intel.com,
	pablo.de.lara.guarch@intel.com
Subject: Re: [dpdk-users] Segmentation fault in aes_cmac_subkey_gen_avx512 and avx512 usage
Date: Sat, 7 Dec 2019 03:08:35 +0530	[thread overview]
Message-ID: <CAMfGv3ndCJ2=Dvp830dVd=wDzS9Jb=JyDZNy3SF=67T_gO1FsQ@mail.gmail.com> (raw)
In-Reply-To: <CAMfGv3kyxKpyM4Y0F3ibb1zNoRLV_k2t57wa2eFWkxJQUwDTCA@mail.gmail.com>

Some more Info.
Basically the CPU does not support AVX512.
So, I think I need to disable avx512 instructions when compiling
multibuffer library, DPDK and the application itself.
I tried using some -mno-avx512x flags but still error persisted.
Is the observation valid? I need help in disabling avx512 functions then.
would it affect performances nd usage?

On Sat, Dec 7, 2019 at 2:10 AM Suraj R Gupta <surajrgupta@iith.ac.in> wrote:

> Hi all,
> I am encountering some issues in using intel multi buffer library with dpdk
> crypto library.
>
>
>
>
> *DPDK version used - 18.05.1Intel Multibuffer library version - 0.52
> ( https://github.com/intel/intel-ipsec-mb/tree/v0.52
> <https://github.com/intel/intel-ipsec-mb/tree/v0.52> )CPU - InfoIntel(R)
> Core(TM) i7-8700 CPU @ 3.20GHz*
>
> 1. When I am using aes cmac with intel multi buffer library from DPDK. I am
> receiving an segmentation fault in aes_cmac_subkey_gen_avx512().
> This segmentation fault appears at times.
> Thread 5 "lcore-slave-1" received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 0x7feff43fe700 (LWP 24301)]
> 0x00007ffff730ab04 in aes_cmac_subkey_gen_avx512 () from
> /usr/lib/libIPSec_MB.so.0
>
> 2. On compiling the above intel multi buffer library and running LibTestApp
> from multi buffer library. It is showing Illegal Instruction.
> AES-CFB128 standard test vectors:
> ........................
> ...Pass
> SHA standard test vectors (N jobs = 1):
> ....................
> ...Pass
> Testing AVX512 interface
> Illegal instruction (core dumped)
>
> I am not able to understand the issue. Request you to help me in this
> regard.
> Please revert for any clarification.
> --
> Thanks and Regards
> Suraj R Gupta
>


-- 
Thanks and Regards
Suraj R Gupta

      reply	other threads:[~2019-12-06 21:38 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-06 20:40 Suraj R Gupta
2019-12-06 21:38 ` Suraj R Gupta [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAMfGv3ndCJ2=Dvp830dVd=wDzS9Jb=JyDZNy3SF=67T_gO1FsQ@mail.gmail.com' \
    --to=surajrgupta@iith.ac.in \
    --cc=dev@dpdk.org \
    --cc=fiona.trahe@intel.com \
    --cc=marcel.d.cornu@intel.com \
    --cc=pablo.de.lara.guarch@intel.com \
    --cc=tomasz.kantecki@intel.com \
    --cc=users@dpdk.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).