From: Neil Horman <nhorman-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org>
To: Vijayakumar Muthuvel Manickam
<mmvijay-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: "dev-VfR2kkLFssw@public.gmane.org" <dev-VfR2kkLFssw@public.gmane.org>
Subject: Re: Running 32bit secondary with 64bit primary DPDK process
Date: Sat, 6 Dec 2014 08:48:01 -0500 [thread overview]
Message-ID: <20141206134801.GA13596@localhost.localdomain> (raw)
In-Reply-To: <CADJ2ZGPJxNXxd3HxTWgJqywfv9xxx+kaXjNxcePy5PiCwE+S9w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
On Fri, Dec 05, 2014 at 11:58:10PM -0800, Vijayakumar Muthuvel Manickam wrote:
> Hi,
>
> I have a patch that enables 32bit secondary DPDK processes to attach with
> 64bit primary DPDK process and share the mempool and ring datastructures. I
> developed this patch to enable our product(from Brocade) which uses a 32bit
> userspace network stack to be able to work with PMDs like ivshmem and some
> proprietary PMDs that are available only for 64bit version of the primary
> process.
>
> To achieve this,
> 1) I changed the mmap() of hugepages in primary process to use MAP_32BIT
> flag so that all hugepages are mapped within the 4GB virtual address limit
> and
> 2) Added a dummy pad_pointer member variable next to each pointer member
> within the shared datastructures like struct rte_ring, struct
> rte_mempool, struct rte_pktmbuf, TAILQ_HEAD, struct hugepage_file that both
> primary and secondary share.
>
> I have been able to successfully run a 64bit primary process with multiple
> 32bit secondary processes doing packet processing with the pipeline model.
>
> I would like to know if this compatibility support is of interest to the
> community and if so I can submit my patch.
>
> Thanks,
> Vijay
>
It never hurts to post something.
Neil
prev parent reply other threads:[~2014-12-06 13:48 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-12-06 7:58 Running 32bit secondary with 64bit primary DPDK process Vijayakumar Muthuvel Manickam
[not found] ` <CADJ2ZGPJxNXxd3HxTWgJqywfv9xxx+kaXjNxcePy5PiCwE+S9w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-12-06 13:48 ` Neil Horman [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=20141206134801.GA13596@localhost.localdomain \
--to=nhorman-2xusbdqka4r54taoqtywwq@public.gmane.org \
--cc=dev-VfR2kkLFssw@public.gmane.org \
--cc=mmvijay-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.