Re: [dpdk-dev] [PATCH v4 30/32] net/ice: support basic RX/TX

[Thomas Monjalon <thomas@monjalon.net>]

14/12/2018 14:00, Ferruh Yigit:
> On 12/14/2018 8:35 AM, Wenzhuo Lu wrote:
> > Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
> > Signed-off-by: Qiming Yang <qiming.yang@intel.com>
> > Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
> > Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
> 
> <...>
> 
> > +
> > +	/* Check to make sure the last descriptor to clean is done */
> > +	desc_to_clean_to = sw_ring[desc_to_clean_to].last_id;
> > +	if (!(txd[desc_to_clean_to].cmd_type_offset_bsz &
> > +	    rte_cpu_to_le_64(ICE_TX_DESC_DTYPE_DESC_DONE))) {
> > +		PMD_TX_FREE_LOG(DEBUG, "TX descriptor %4u is not done "
> > +				"(port=%d queue=%d) value=0x%lx\n",
> > +				desc_to_clean_to,
> > +				txq->port_id, txq->queue_id,
> > +				txd[desc_to_clean_to].cmd_type_offset_bsz);
> 
> Causing build error for i686 [1], should use PRIx64 for 64bit variables.
> 
> Perhaps we should create a rule in checkpatch to check and warn %lx %lu formats
> `git grep -n '%l[xud]' drivers/net/ice/` shows only this occurrence in 'ice' but
> there are more in other drivers...

If it's clear to everybody that checkpatch can return some false positive,
yes I am for checking '%l[xud]'. It is most of the time a mistake.