Re: [PATCH] IB/rxe: Don't clamp residual length to mtu

[Doug Ledford <dledford@redhat.com>]

On Tue, 2017-04-25 at 09:29 +0200, Johannes Thumshirn wrote:
> On Thu, Apr 06, 2017 at 02:49:44PM +0200, Johannes Thumshirn wrote:
> > 
> > When reading a RDMA WRITE FIRST packet we copy the DMA length from
> > the RDMA
> > header into the qp->resp.resid variable for later use. Later in
> > check_rkey()
> > we clamp it to the MTU if the packet is an  RDMA WRITE packet and
> > has a
> > residual length bigger than the MTU. Later in write_data_in() we
> > subtract the
> > payload of the packet from the residual length. If the packet
> > happens to have a
> > payload of exactly the MTU size we end up with a residual length of
> > 0 despite
> > the packet not being the last in the conversation. When the next
> > packet in the
> > conversation arrives, we don't have any residual length left and
> > thus set the QP
> > into an error state.
> > 
> > This broke NVMe over Fabrics functionality over rdma_rxe.ko
> > 
> > The patch was verified using the following test.
> > 
> >  # echo eth0 > /sys/module/rdma_rxe/parameters/add
> >  # nvme connect -t rdma -a 192.168.155.101 -s 1023 -n nvmf-test
> >  # mkfs.xfs -fK /dev/nvme0n1
> >  meta-data=/dev/nvme0n1           isize=256    agcount=4,
> > agsize=65536 blks
> >           =                       sectsz=4096  attr=2,
> > projid32bit=1
> >           =                       crc=0        finobt=0, sparse=0
> >  data     =                       bsize=4096   blocks=262144,
> > imaxpct=25
> >           =                       sunit=0      swidth=0 blks
> >  naming   =version 2              bsize=4096   ascii-ci=0 ftype=1
> >  log      =internal log           bsize=4096   blocks=2560,
> > version=2
> >           =                       sectsz=4096  sunit=1 blks, lazy-
> > count=1
> >  realtime =none                   extsz=4096   blocks=0,
> > rtextents=0
> >  # mount /dev/nvme0n1 /tmp/
> >  [  148.923263] XFS (nvme0n1): Mounting V4 Filesystem
> >  [  148.961196] XFS (nvme0n1): Ending clean mount
> >  # dd if=/dev/urandom of=test.bin bs=1M count=128
> >  128+0 records in
> >  128+0 records out
> >  134217728 bytes (134 MB, 128 MiB) copied, 0.437991 s, 306 MB/s
> >  # sha256sum test.bin
> >  cde42941f045efa8c4f0f157ab6f29741753cdd8d1cff93a6b03649d83c4129a  
> > test.bin
> >  # cp test.bin /tmp/
> >  sha256sum /tmp/test.bin
> >  cde42941f045efa8c4f0f157ab6f29741753cdd8d1cff93a6b03649d83c4129a  
> > /tmp/test.bin
> > 
> > Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
> > Cc: Hannes Reinecke <hare@suse.de>
> > Cc: Sagi Grimberg <sagi@grimberg.me>
> > Cc: Max Gurtovoy <maxg@mellanox.com>
> > ---
> 
> Doug anything left here? I already have an Ack from Moni. This patch
> is needed
> to get NVMe over Fabrics working on rxe so I'd like to see it in
> v4.12.

Nope, it's all good.  I applied it today.

-- 
Doug Ledford <dledford@redhat.com>
    GPG KeyID: B826A3330E572FDD
   
Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333 0E57 2FDD