DPDK patches and discussions
 help / color / mirror / Atom feed
From: Pablo de Lara <pablo.de.lara.guarch@intel.com>
To: bombermag@gmail.com
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [dpdk-users] Broken hash table functions with CFLAGS -O3
Date: Mon, 18 Jan 2016 13:47:42 +0000	[thread overview]
Message-ID: <1453124862-35513-1-git-send-email-pablo.de.lara.guarch@intel.com> (raw)
In-Reply-To: <CALq5hH1wwF+KyTgqNfYNM-Tpj6CuxffUfMBcHZS0B7Md4E4Lew@mail.gmail.com>

> -----Original Message-----
> From: users [mailto:users-bounces@dpdk.org] On Behalf Of Dmitriy Yakovlev
> Sent: Friday, January 15, 2016 1:20 AM
> To: users@dpdk.org
> Subject: [dpdk-users] Broken hash table functions with CFLAGS -O3
> 
> Hello.
> I trying to use hash tables for adding struct ip:port as key.
> When trying to add duplicated key it's successfuly added second time, not
> always, but sometimes.
> Please see this youtube video, where I'm demonstrating this bug:
> https://www.youtube.com/watch?v=vsXVxISpmnU
> Code on github: https://github.com/BombermaG/bug-app
> 
> When using CFLAGS -O2 this problem cannot be repeated.
> It's problem with -O3 flag and I must use O2 or how can I solve it?
> Thank you for helping.
> (Sorry if this message double posting).
> 
> Spec info. Tell me, if you need more info.
> dpdk 2.2.0

Hello Dmitriy,

Thanks for the detailed report. I am able to reproduce the issue with
gcc 4.7 and 4.8. The problem is that the signature sometimes changes,
due to a different key and therefore the bucket where it should be stored
is the wrong one.

Actually, the hash table is fine. The issue is that your key is a structure
of a 4-byte and 2-byte integers, so your key size should be 6, but instead,
it is 8, because sizeof(struct...) includes padding to have it multiple of 4.
Most times, that padding is 0, so it is fine, but when that padding is not 0,
the key is different and then it looks like you are adding a duplicate
(but actually you are adding a different key!).
So, either add padding manually in the key or manually set the key size to 6.

Regards,
Pablo
> 
> Linux 3.10.0-229.el7.x86_64 #1 SMP Fri Mar 6 11:36:42 UTC 2015 x86_64
> x86_64 x86_64 GNU/Linux
> 
> gcc -v
> Using built-in specs.
> COLLECT_GCC=gcc
> COLLECT_LTO_WRAPPER=/usr/libexec/gcc/x86_64-redhat-linux/4.8.5/lto-wrapper
> Target: x86_64-redhat-linux
> Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --
> infodir=/usr/share/info --with-bugurl=http://bugzilla.redhat.com/bugzilla
> --enable-bootstrap --enable-shared --enable-threads=posix --enable-
> checking=release --with-system-zlib --enable-__cxa_atexit --disable-
> libunwind-exceptions --enable-gnu-unique-object --enable-linker-build-id -
> -with-linker-hash-style=gnu --enable-languages=c,c++,objc,obj-
> c++,java,fortran,ada,go,lto
> --enable-plugin --enable-initfini-array --disable-libgcj --with-
> isl=/builddir/build/BUILD/gcc-4.8.5-20150702/obj-x86_64-redhat-linux/isl-
> install
> --with-cloog=/builddir/build/BUILD/gcc-4.8.5-20150702/obj-x86_64-redhat-
> linux/cloog-install
> --enable-gnu-indirect-function --with-tune=generic --with-arch_32=x86-64 -
> -build=x86_64-redhat-linux Thread model: posix gcc version 4.8.5 20150623
> (Red Hat 4.8.5-4) (GCC)
> 
> Intel(R) Xeon(R) CPU E5-2643 v2 @ 3.50GHz

       reply	other threads:[~2016-01-18 13:48 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CALq5hH1wwF+KyTgqNfYNM-Tpj6CuxffUfMBcHZS0B7Md4E4Lew@mail.gmail.com>
2016-01-18 13:47 ` Pablo de Lara [this message]
2016-01-18 14:18   ` De Lara Guarch, Pablo

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=1453124862-35513-1-git-send-email-pablo.de.lara.guarch@intel.com \
    --to=pablo.de.lara.guarch@intel.com \
    --cc=bombermag@gmail.com \
    --cc=dev@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).