[dpdk-stable] [PATCH 17.11] net/bnxt: replace memory barrier for doorbell response

[Luca Boccassi]

On Tue, 2020-01-07 at 11:25 +0000, Luca Boccassi wrote:
> On Thu, 2020-01-02 at 15:20 +0800, Gavin Hu wrote:
> > [ upstream commit 21ecbde99b2f0772d36eef8276a686f3135e5de9 ]
> > 
> > To read the doorbell response, which is held in the host CIO
> > memory,
> > rte_cio_rmb is sufficient.
> > 
> > Fixes: 804e746c7b73 ("net/bnxt: add hardware resource manager init
> > code")
> > 
> > Signed-off-by: Gavin Hu <gavin.hu at arm.com>
> > Acked-by: Ajit Khaparde <ajit.khaparde at broadcom.com>
> > ---
> >  drivers/net/bnxt/bnxt_hwrm.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> Thanks, applied

Actually, this doesn't compile as rte_cio_rmb is not available in
17.11:

drivers/net/bnxt/bnxt_hwrm.c: In function ‘bnxt_hwrm_send_message’:
drivers/net/bnxt/bnxt_hwrm.c:151:3: error: implicit declaration of
function ‘rte_cio_rmb’; did you mean ‘rte_io_rmb’? [-Werror=implicit-
function-declaration]
   rte_cio_rmb();
   ^~~~~~~~~~~
   rte_io_rmb

It was added by the following commits:

52bd60a5cf0b ("eal/arm64: define coherent I/O memory barriers")
bbc504dfbb5e ("eal/arm32: define coherent I/O memory barriers")
cbb3c4162899 ("eal/ppc64: define coherent I/O memory barriers")
0079e4fb6922 ("eal/x86: define coherent I/O memory barriers")
cd36cc49e59c ("eal: introduce coherent I/O memory barriers")

Should I drop this patch for this release?

-- 
Kind regards,
Luca Boccassi