From: fwefew 4t4tg <7532yahoo@gmail.com>
To: Stephen Hemminger <stephen@networkplumber.org>
Cc: users@dpdk.org
Subject: Re: need definitions / references
Date: Thu, 17 Mar 2022 12:59:18 -0400 [thread overview]
Message-ID: <CA+Tq66VpEogPH4VY5XEgX-YsEUFzZLAMNV3ChpDYXkC8V670pg@mail.gmail.com> (raw)
In-Reply-To: <20220317094824.423d2a83@hermes.local>
[-- Attachment #1: Type: text/plain, Size: 2075 bytes --]
Mr. Hemminger,
Thank you once again for your help.
Now, I did see the one-liner DPDK puts in code which you reference:
struct rte_eth_thresh {
uint8_t pthresh; /**< Ring prefetch threshold. */
uint8_t hthresh; /**< Ring host threshold. */
uint8_t wthresh; /**< Ring writeback threshold. */
};
The problem for me is I don't know what "ring host threshold" or "ring
writeback" means.
"Ring prefetch threshold" I think I understand. This refers to how mbufs
are pending in the ring before it starts writing them onto the wire. This
seems to have some connection to tx_rs_thresh.
On Thu, Mar 17, 2022 at 12:48 PM Stephen Hemminger <
stephen@networkplumber.org> wrote:
> On Thu, 17 Mar 2022 12:05:33 -0400
> fwefew 4t4tg <7532yahoo@gmail.com> wrote:
>
> > DPDK provides but does not define (that I can see at least) these TXQ
> > config fields. What are these fields what do they do?
> >
> > https://doc.dpdk.org/api/structrte__eth__thresh.html
> > uint8_t pthresh
> > uint8_t hthresh
> > uint8_t wthresh
>
> These are low level details inherited from original Intel NIC's (ie ixgbe)
> they only apply to some NICs. They are used to control the ring threshold
> values relating to PCI.
>
> In rte_ethdev.h
> /**
> * A structure used to configure the ring threshold registers of an Rx/Tx
> * queue for an Ethernet port.
> */
> struct rte_eth_thresh {
> uint8_t pthresh; /**< Ring prefetch threshold. */
> uint8_t hthresh; /**< Ring host threshold. */
> uint8_t wthresh; /**< Ring writeback threshold. */
> };
>
>
> > And here: What is the RS bit?
> > https://doc.dpdk.org/api/structrte__eth__txconf.html
> > uint16_t tx_rs_thresh.
>
> Ready to Send?
>
> >
> > And some what unrelated: What is a ``doorbell"? this is something I
> > continue to run into esp. in Mellanox NIC papers? Doorbell counts are
> > reported for AWS NICs at least with xstats api:
> >
> > tx_q0_doorbells: 18812
> > tx_q1_doorbells: 18861
>
> These are how many times driver needed to poke hardware to tell it
> that new packets are ready to send.
>
[-- Attachment #2: Type: text/html, Size: 3039 bytes --]
prev parent reply other threads:[~2022-03-17 16:59 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-17 16:05 fwefew 4t4tg
2022-03-17 16:48 ` Stephen Hemminger
2022-03-17 16:59 ` fwefew 4t4tg [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=CA+Tq66VpEogPH4VY5XEgX-YsEUFzZLAMNV3ChpDYXkC8V670pg@mail.gmail.com \
--to=7532yahoo@gmail.com \
--cc=stephen@networkplumber.org \
--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).