DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Dumitrescu, Cristian" <cristian.dumitrescu@intel.com>
To: Aws Ismail <aws.ismail@gmail.com>, "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] Explanation of the QoS offset values used in the QoS scheduler example app.
Date: Mon, 16 Feb 2015 11:34:35 +0000	[thread overview]
Message-ID: <3EB4FA525960D640B5BDFFD6A3D8912632315B1F@IRSMSX108.ger.corp.intel.com> (raw)
In-Reply-To: <CAMHMu0OVM9eWima6__dJ1ETrd1S4Y67p-1sn21C2oxQvwdVySA@mail.gmail.com>

Hi,

These are byte offsets used for reading these packet fields, considering that packet bytes are stored in memory in network order, while the CPU is little endian, so byte swapping takes place on read.

This is probably not the best way to write this code, and I agree this portion of the app code is a bit more cryptic than it should be. Using data structures to describe the header format for the input packet (Ethernet/SVLAN/CVLAN/IPv4) and using portable byte swapping macros is probably a better alternative.

This being said, the code implementation, code comments and Sample App Guide description seem to be consistent and correct.

Regards,
Cristian


-----Original Message-----
From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Aws Ismail
Sent: Saturday, February 14, 2015 7:35 PM
To: dev@dpdk.org
Subject: [dpdk-dev] Explanation of the QoS offset values used in the QoS scheduler example app.

Hi everyone,

I am looking at this portion of the code in the app_thread.c file of the
QoS scheduler example application:

/*
* QoS parameters are encoded as follows:
* Outer VLAN ID defines subport
* Inner VLAN ID defines pipe
* Destination IP 0.0.XXX.0 defines traffic class
* Destination IP host (0.0.0.XXX) defines queue
* Values below define offset to each field from start of frame
*/
#define SUBPORT_OFFSET 7
#define PIPE_OFFSET 9
#define TC_OFFSET 20
#define QUEUE_OFFSET 20
#define COLOR_OFFSET 19

static inline int get_pkt_sched(struct rte_mbuf *m, uint32_t *subport,
uint32_t *pipe, uint32_t *traffic_class, uint32_t *queue, uint32_t *color)
{
uint16_t *pdata = rte_pktmbuf_mtod(m, uint16_t *);

*subport = (rte_be_to_cpu_16(pdata[SUBPORT_OFFSET]) & 0x0FFF) &
(port_params.n_subports_per_port - 1); /* Outer VLAN ID*/

*pipe = (rte_be_to_cpu_16(pdata[PIPE_OFFSET]) & 0x0FFF) &
(port_params.n_pipes_per_subport - 1); /* Inner VLAN ID */

*traffic_class = (pdata[QUEUE_OFFSET] & 0x0F) &
(RTE_SCHED_TRAFFIC_CLASSES_PER_PIPE - 1); /* Destination IP */

*queue = ((pdata[QUEUE_OFFSET] >> 8) & 0x0F) &
(RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS - 1) ; /* Destination IP */

*color = pdata[COLOR_OFFSET] & 0x03; /* Destination IP */

return 0;
}

The offset values do not make sense to me. According to the programmer
guide, the queue selection is SVID/CVID/TC/QID based. And those offset seem
off in this case. Is this because it is assuming that the packet is being
altered before it gets to this stage ?

Can anyone provide a better explanation or at least the reason behind
choosing those offset values shown above.

Thanks.
--------------------------------------------------------------
Intel Shannon Limited
Registered in Ireland
Registered Office: Collinstown Industrial Park, Leixlip, County Kildare
Registered Number: 308263
Business address: Dromore House, East Park, Shannon, Co. Clare

This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies.


  reply	other threads:[~2015-02-16 11:34 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CAMHMu0MwKozt2A2Gp_s2yHT=dt2cZBE6Np_2guAWGjKfaMLD_Q@mail.gmail.com>
2015-02-14 19:34 ` Aws Ismail
2015-02-16 11:34   ` Dumitrescu, Cristian [this message]
2015-02-16 13:05     ` Aws Ismail

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=3EB4FA525960D640B5BDFFD6A3D8912632315B1F@IRSMSX108.ger.corp.intel.com \
    --to=cristian.dumitrescu@intel.com \
    --cc=aws.ismail@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).