From: John Fastabend <john.r.fastabend@intel.com>
To: Dmitry Kravkov <dkravkov@gmail.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>,
Dmitry Kravkov <dmitry@broadcom.com>,
"davem@davemloft.net" <davem@davemloft.net>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
Ariel Elior <ariele@broadcom.com>,
Eilon Greenstein <eilong@broadcom.com>
Subject: Re: [PATCH net-next 09/13] bnx2x: add pri_map module parameter
Date: Thu, 10 Nov 2011 12:34:50 -0800 [thread overview]
Message-ID: <4EBC356A.6020907@intel.com> (raw)
In-Reply-To: <CAM8tLiN=E2=a0TEYgw6mc70-vJU0dHjP4F4PpKUk8+3zvQPnAg@mail.gmail.com>
On 11/10/2011 12:21 PM, Dmitry Kravkov wrote:
> Resending for netdev
> On Thu, Nov 10, 2011 at 10:18 PM, Dmitry Kravkov <dkravkov@gmail.com> wrote:
>>
>> On Thu, Nov 10, 2011 at 7:39 PM, Ben Hutchings <bhutchings@solarflare.com> wrote:
>>>
>>> On Thu, 2011-11-10 at 17:14 +0200, Dmitry Kravkov wrote:
>>>> From: Ariel Elior <ariele@broadcom.com>
>>>>
>>>> The optional parameter pri_map is used to map the
>>>> skb->priority to a Class Of Service (CoS) in the HW.
>>>> This 32 bit parameter is evaluated by the driver as 8
>>>> values of 4 bits each. Each nibble sets the desired
>>>> HW queue number for that priority.
>>> [...]
>>>
>>> You should implement ndo_tc_setup instead of adding a driver-specific
>>> parameter (that should be per-device anyway).
>>
>> bnx2x has ndo_setup_tc, but the callback configures device for number of traffic
>> only, when actual mapping of priority to cos is done according to DCB negotiation.
>> The parameter allow user to force mapping if the link partner is not DCB capable.
>>
> --
Use 'tc' to configure the mapping explicitly then. Or allow DCBnl
set commands to be sent down either ieee_setets() or setpgtccfgtx()
are used to configure priority to traffic class mappings. Then you
could use a user space tool to set up all your DCB attributes without
DCBX running (e.g. dcbtool/lldptol but I'm partial).
For example 'tc' commands,
# tc qdisc add dev eth3 root mqprio map 0 1 2 3
# tc qdisc add dev eth3 root mqprio help
Usage: ... mqprio [num_tc NUMBER] [map P0 P1 ...]
[queues count1@offset1 count2@offset2 ...] [hw 1|0]
It might be nice to add the ability to change the mapping after the tc
has been created though. Maybe I'll get to this in the next week or
two.
John.
next prev parent reply other threads:[~2011-11-10 20:34 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-10 15:14 [PATCH net-next 00/13] bnx2x series Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 01/13] bnx2x: allow FCoE and DCB for 578xx Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 02/13] bnx2x: use rx_queue index for skb_record_rx_queue() Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 03/13] bnx2x: remove unused variable Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 04/13] bnx2x: separate FCoE and iSCSI license initialization Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 05/13] bnx2x: propagate DCBX negotiation Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 06/13] bnx2x: DCBX: use #define instead of magic Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 07/13] bnx2x: simplify definition of RX_SGE_MASK_LEN and use it Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 08/13] bnx2x: remove unused #define Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 09/13] bnx2x: add pri_map module parameter Dmitry Kravkov
2011-11-10 17:39 ` Ben Hutchings
[not found] ` <CAM8tLiMATfsH1mqmgrALX9o+5nWEahntCAc2ewm0RF_kbt=82A@mail.gmail.com>
2011-11-10 20:21 ` Dmitry Kravkov
2011-11-10 20:34 ` John Fastabend [this message]
2011-11-13 12:30 ` Dmitry Kravkov
2011-11-12 5:50 ` David Miller
2011-11-10 15:14 ` [PATCH net-next 10/13] bnx2x: add fan failure event handling Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 11/13] bnx2x: prevent race in statistics flow Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 12/13] bnx2x: Remove on-stack napi struct variable Dmitry Kravkov
2011-11-10 15:14 ` [PATCH net-next 13/13] bnx2x: update driver version to 1.70.35-0 Dmitry Kravkov
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=4EBC356A.6020907@intel.com \
--to=john.r.fastabend@intel.com \
--cc=ariele@broadcom.com \
--cc=bhutchings@solarflare.com \
--cc=davem@davemloft.net \
--cc=dkravkov@gmail.com \
--cc=dmitry@broadcom.com \
--cc=eilong@broadcom.com \
--cc=netdev@vger.kernel.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).