From: "John Daley (johndale)" <johndale@cisco.com>
To: Bruce Richardson <bruce.richardson@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
"bruce.richarsdon@intel.com" <bruce.richarsdon@intel.com>
Subject: Re: [dpdk-dev] [PATCH v3 07/13] enic: use Tx completion messages instead of descriptors
Date: Fri, 10 Jun 2016 22:28:01 +0000 [thread overview]
Message-ID: <6fdf2e090a4142a5ba4e3b78086267a5@XCH-RCD-007.cisco.com> (raw)
In-Reply-To: <20160610211813.GA15176@bricha3-MOBL3>
> -----Original Message-----
> From: Bruce Richardson [mailto:bruce.richardson@intel.com]
> Sent: Friday, June 10, 2016 2:18 PM
> To: John Daley (johndale) <johndale@cisco.com>
> Cc: dev@dpdk.org; bruce.richarsdon@intel.com
> Subject: Re: [dpdk-dev] [PATCH v3 07/13] enic: use Tx completion messages
> instead of descriptors
>
> On Thu, Jun 02, 2016 at 05:22:51PM -0700, John Daley wrote:
> > The NIC can either DMA a separate completion message for each
> > completed send or periodically just DMA an index of the last completed
> > send. Switch to the second method which improves cache locality and
> > performance.
> >
> > Signed-off-by: John Daley <johndale@cisco.com>
>
> Can you perhaps send me an updated wording for this commit message as
> the title and commit message conflict. The title says to use completion
> messages not descriptors, while the body talks about moving away from a
> completion message way of working.
> Is the former method a descriptor writeback method, while the latter a head
> pointer writeback? If so, I think the title could be:
>
> "enic: use Tx head pointer not descriptor writeback"
>
> or something similar.
>
> Again, if you send on the updated commit text, I'll just update it on apply. I'd
> ideally like to get this patchset pushed to next-net first thing Monday.
Ok, I agree that it is confusing.
We moved from having the hardware send a completion descriptor for every packet to having it send the index of the last completed packet every once in a while. We can use the word 'index' and 'message' to describe the 2 methods and drop the word 'descriptor'. Here is a suggestion:
enic: use Tx completion index instead of completion messages
The NIC can either DMA a separate completion message for each completed send or periodically just DMA an index of the last completed send. Switch to the latter method which improves cache locality and performance.
Thank you,
John
>
> /Bruce
next prev parent reply other threads:[~2016-06-10 22:28 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-24 6:32 [dpdk-dev] [PATCH v2 00/11] enic counter fixes and Tx optimization John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 01/11] enic: fix Rx drop counters John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 02/11] enic: drop bad packets and remove unused Rx error flag John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 03/11] enic: count truncated packets John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 04/11] enic: put Tx and Rx functions into same file John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 05/11] enic: remove some unused functions in Tx path John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 06/11] enic: streamline mbuf handling " John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 07/11] enic: use Tx completion messages instead of descriptors John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 08/11] enic: refactor Tx mbuf recycling John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 09/11] enic: optimize the Tx function John Daley
2016-05-30 10:05 ` Azarewicz, PiotrX T
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 10/11] enic: remove unused files and functions and variables John Daley
2016-05-24 6:32 ` [dpdk-dev] [PATCH v2 11/11] enic: add an enic assert macro John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 00/13] enic counter fixes and Tx optimization John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 01/13] enic: fix Rx drop counters John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 02/13] enic: drop bad packets and remove unused Rx error flag John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 03/13] enic: count truncated packets John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 04/13] enic: put Tx and Rx functions into same file John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 05/13] enic: remove some unused functions in Tx path John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 06/13] enic: streamline mbuf handling " John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 07/13] enic: use Tx completion messages instead of descriptors John Daley
2016-06-10 21:18 ` Bruce Richardson
2016-06-10 22:28 ` John Daley (johndale) [this message]
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 08/13] enic: refactor Tx mbuf recycling John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 09/13] enic: optimize the Tx function John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 10/13] enic: remove unused files and functions and variables John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 11/13] enic: add an enic assert macro John Daley
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 12/13] enic: expand local Tx mbuf flags variable to 64-bits John Daley
2016-06-03 8:05 ` Azarewicz, PiotrX T
2016-06-03 0:22 ` [dpdk-dev] [PATCH v3 13/13] enic: fix Tx IP and UDP/TCP checksum offload John Daley
2016-06-10 22:38 ` [dpdk-dev] [PATCH v3 00/13] enic counter fixes and Tx optimization Bruce Richardson
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=6fdf2e090a4142a5ba4e3b78086267a5@XCH-RCD-007.cisco.com \
--to=johndale@cisco.com \
--cc=bruce.richardson@intel.com \
--cc=bruce.richarsdon@intel.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).