> -----Original Message-----
> From: David Marchand <[email protected]>
> Sent: Wednesday, April 7, 2021 19:54
> To: Wang, Haiyue <[email protected]>; Zhang, Qi Z <[email protected]>
> Cc: dev <[email protected]>; Paolo Valerio <[email protected]>; Aaron Conole 
> <[email protected]>; Rong,
> Leyi <[email protected]>; Tu, Lijuan <[email protected]>; Ruifeng Wang 
> (Arm Technology China)
> <[email protected]>; Feifei Wang <[email protected]>; dpdk stable 
> <[email protected]>; Guo, Jia
> <[email protected]>; Richardson, Bruce <[email protected]>; Ananyev, 
> Konstantin
> <[email protected]>; Van Haaren, Harry 
> <[email protected]>; Thomas Monjalon
> <[email protected]>; Yigit, Ferruh <[email protected]>
> Subject: Re: [dpdk-stable] [PATCH v3] net/ixgbe: fix UDP zero checksum error
> 
> On Thu, Feb 25, 2021 at 11:54 AM David Marchand
> <[email protected]> wrote:
> >
> > Hello Haiyue,
> >
> > On Thu, Feb 4, 2021 at 3:56 PM Haiyue Wang <[email protected]> wrote:
> > >
> > > There is an 82599 errata that UDP frames with a zero checksum are
> > > incorrectly marked as checksum invalid by the hardware.  This was
> > > leading to misleading PKT_RX_L4_CKSUM_BAD flag.
> > >
> > > This patch changes the bad UDP checksum to PKT_RX_L4_CKSUM_UNKNOWN,
> > > so the software application will then have to recompute the checksum
> > > itself if needed.
> > >
> > > Bugzilla ID: 629
> > > Fixes: af75078fece3 ("first public release")
> > > Cc: [email protected]
> > >
> > > Reported-by: Paolo Valerio <[email protected]>
> > > Signed-off-by: Haiyue Wang <[email protected]>
> >
> > There was a previous mention of this issue in the past and it resulted
> > in dropping part of the hw statistics.
> > https://git.dpdk.org/dpdk/commit/?id=256ff05a9cae7484e2197cde4401dfa1f21d5a6f
> >
> > Does it make sense to restrict this "fix" to 82599 only?

Make sense.

> 
> ping.
> 
> 
> --
> David Marchand

Reply via email to