[dpdk-dev] [PATCH] ixgbe: do not include CRC in Tx byte count
stephen at networkplumber.org
Fri Jul 10 01:03:37 CEST 2015
On Wed, 01 Apr 2015 09:45:02 +0200
Thomas Monjalon <thomas.monjalon at 6wind.com> wrote:
> 2015-03-24 23:54, Ananyev, Konstantin:
> > From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> > > On Mon, 23 Mar 2015 16:45:44 +0000
> > > "Ananyev, Konstantin" <konstantin.ananyev at intel.com> wrote:
> > > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of stephen at networkplumber.org
> > > > > From: Stephen Hemminger <shemming at brocade.com>
> > > > >
> > > > > The ixgbe driver was including CRC in the transmit packet byte
> > > > > count, but not for packets received.
> > > > > This was notice when forwarding and
> > > > > the number of bytes received was greater than the number of bytes transmitted
> > > > > for the same number of packets. Make the driver behave like other
> > > > > virtual devices and not include CRC in byte count. Use the same queue
> > > > > counters already computed and used for Rx.
> > > >
> > > > About RX side stats - as I remember it depends to what value hw_stip_crc is set at configure().
> > > > If hw_stip_crc==1, then, yes CRC bytes are not included into QBRC value.
> > > > I If hw_stip_crc==0, then CRC bytes are included into QBRC.
> > >
> > > That is an additional bug!
> > > * CRC should never be included in the byte count.
> > > This is not how Linux or other drivers report byte count.
> > I don't have any strong opinion here...
> > For me any method (with or without CRC) of counting bytes are ok, as long as this method is identical across all PMDs we support.
> > Which makes we wonder, what approach other PMDs use?
> > >
> > > * the byte count must be symmetrical Rx == Tx
> > >
> > > The Brocade router always set strip_crc to 1. So did not see the additional bug
> > > of CRC being included in byte count.
> > Are you going to submit v2, to make Rx==Tx for both cases?
> Any conclusion?
> Without update, this patch is going to be dropped.
My conclusion was that CRC should never be in byte count regardless of
configuration because that is what everyone (except CISCO) expects.
My patch did that.
More information about the dev