DPDK patches and discussions
 help / color / mirror / Atom feed
From: Olivier Matz <olivier.matz@6wind.com>
To: Qi Zhang <qi.z.zhang@intel.com>
Cc: konstantin.ananyev@intel.com, helin.zhang@intel.com,
	wenzhuo.lu@intel.com, ray.kinsella@intel.com, dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH v2] net/ixgbe: enable PTYPE offload for x86 vector PMD
Date: Tue, 6 Jun 2017 17:45:26 +0200	[thread overview]
Message-ID: <20170606174526.1205227d@platinum> (raw)
In-Reply-To: <1496273426-15176-1-git-send-email-qi.z.zhang@intel.com>

Hi Qi,

On Wed, 31 May 2017 19:30:26 -0400, Qi Zhang <qi.z.zhang@intel.com> wrote:
> Hardware PTYPE in Rx desc will be parsed to fill
> mbuf's packet_type.
> 
> Signed-off-by: Ray Kinsella <ray.kinsella@intel.com>
> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
> ---
> 
> v2:
> - replace large macro that parse packet type with inline function
> - fix couple check patch issues.
> 
>  drivers/net/ixgbe/ixgbe_ethdev.c       |   8 +
>  drivers/net/ixgbe/ixgbe_rxtx.c         | 623 ++++++++++++++-------------------
>  drivers/net/ixgbe/ixgbe_rxtx.h         |  92 +++++
>  drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c |  65 ++++
>  4 files changed, 434 insertions(+), 354 deletions(-)
> 

I tried to compile your patch with RTE_MACHINE=default, and I
have the following compilation error:

gcc -Wp,-MD,./.ixgbe_rxtx_vec_sse.o.d.tmp  -m64 -pthread  -march=core2 -DRTE_MACHINE_CPUFLAG_SSE -DRTE_MACHINE_CPUFLAG_SSE2 -DRTE_MACHINE_CPUFLAG_SSE3 -DRTE_MACHINE_CPUFLAG_SSSE3  -I/home/user/dpdk.org/build/include -include /home/user/dpdk.org/build/include/rte_config.h -O3 -W -Wall -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wold-style-definition -Wpointer-arith -Wcast-align -Wnested-externs -Wcast-qual -Wformat-nonliteral -Wformat-security -Wundef -Wwrite-strings -Werror -Wno-deprecated    -o ixgbe_rxtx_vec_sse.o -c /home/user/dpdk.org/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c 
In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/x86intrin.h:43:0,
                 from /home/user/dpdk.org/build/include/rte_vect.h:70,
                 from /home/user/dpdk.org/build/include/rte_memcpy.h:46,
                 from /home/user/dpdk.org/build/include/rte_ether.h:50,
                 from /home/user/dpdk.org/build/include/rte_ethdev.h:185,
                 from /home/user/dpdk.org/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c:35:
/home/user/dpdk.org/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c: In function ‘desc_to_ptype_v’:
/usr/lib/gcc/x86_64-linux-gnu/6/include/smmintrin.h:447:1: error: inlining failed in call to always_inline ‘_mm_extract_epi32’: target specific option mismatch
 _mm_extract_epi32 (__m128i __X, const int __N)
 ^~~~~~~~~~~~~~~~~
/home/user/dpdk.org/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c:300:13: note: called from here
  pkt_info = _mm_extract_epi32(ptype0, 3);
             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/x86intrin.h:43:0,
                 from /home/user/dpdk.org/build/include/rte_vect.h:70,
                 from /home/user/dpdk.org/build/include/rte_memcpy.h:46,
                 from /home/user/dpdk.org/build/include/rte_ether.h:50,
                 from /home/user/dpdk.org/build/include/rte_ethdev.h:185,
                 from /home/user/dpdk.org/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c:35:
[...]


To reproduce:
 make config T=x86_64-native-linuxapp-gcc
 sed -i 's,CONFIG_RTE_MACHINE="native",CONFIG_RTE_MACHINE="default",' build/.config
 make

Do we still want to support the core2 target?


Thanks,
Olivier

  parent reply	other threads:[~2017-06-06 15:45 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-31 23:30 Qi Zhang
2017-06-04 23:04 ` Ananyev, Konstantin
2017-06-06  4:25   ` Zhang, Qi Z
2017-06-06 15:45 ` Olivier Matz [this message]
2017-06-06 16:03   ` Zhang, Qi Z

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=20170606174526.1205227d@platinum \
    --to=olivier.matz@6wind.com \
    --cc=dev@dpdk.org \
    --cc=helin.zhang@intel.com \
    --cc=konstantin.ananyev@intel.com \
    --cc=qi.z.zhang@intel.com \
    --cc=ray.kinsella@intel.com \
    --cc=wenzhuo.lu@intel.com \
    /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).