[dpdk-dev] [PATCH 01/18] net/ixgbe: store SYN filter
Zhao1, Wei
wei.zhao1 at intel.com
Thu Dec 22 10:48:03 CET 2016
Hi, Yigit
> -----Original Message-----
> From: Yigit, Ferruh
> Sent: Wednesday, December 21, 2016 12:56 AM
> To: Zhao1, Wei <wei.zhao1 at intel.com>; dev at dpdk.org
> Cc: Lu, Wenzhuo <wenzhuo.lu at intel.com>
> Subject: Re: [dpdk-dev] [PATCH 01/18] net/ixgbe: store SYN filter
>
> On 12/2/2016 10:42 AM, Wei Zhao wrote:
> > From: wei zhao1 <wei.zhao1 at intel.com>
> >
> > Add support for storing SYN filter in SW.
>
> Do you think does it makes more clear to refer as TCP SYN filter? Or SYN filter
> is clear enough?
>
Ok, I will change to " TCP SYN filter " to make it more clear
> >
> > Signed-off-by: Wenzhuo Lu <wenzhuo.lu at intel.com>
> > Signed-off-by: wei zhao1 <wei.zhao1 at intel.com>
>
> Can you please update sign-off to your actual name?
>
Ok, I will change to " Signed-off-by: Wei Zhao <wei.zhao1 at intel.com>"
> > ---
> > drivers/net/ixgbe/ixgbe_ethdev.c | 12 ++++++++++--
> > drivers/net/ixgbe/ixgbe_ethdev.h | 2 ++
> > 2 files changed, 12 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c
> > b/drivers/net/ixgbe/ixgbe_ethdev.c
> > index edc9b22..7f10cca 100644
> > --- a/drivers/net/ixgbe/ixgbe_ethdev.c
> > +++ b/drivers/net/ixgbe/ixgbe_ethdev.c
> > @@ -1287,6 +1287,8 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev)
> > memset(filter_info->fivetuple_mask, 0,
> > sizeof(uint32_t) * IXGBE_5TUPLE_ARRAY_SIZE);
> >
> > + /* initialize SYN filter */
> > + filter_info->syn_info = 0;
>
> can it be an option to memset all filter_info? (and of course move list init
> after memset)
>
Maybe, change to the following code?
memset(filter_info, 0, sizeof(struct ixgbe_filter_info));
TAILQ_INIT(&filter_info->fivetuple_list);
But that wiil mix /* initialize ether type filter */ and /* initialize 5tuple filter list */ two process together,
Because struct ixgbe_filter_info store two type info of ether and 5tuple.
So, not to change ?
struct ixgbe_filter_info {
uint8_t ethertype_mask; /* Bit mask for every used ethertype filter */
/* store used ethertype filters*/
struct ixgbe_ethertype_filter ethertype_filters[IXGBE_MAX_ETQF_FILTERS];
/* Bit mask for every used 5tuple filter */
uint32_t fivetuple_mask[IXGBE_5TUPLE_ARRAY_SIZE];
struct ixgbe_5tuple_filter_list fivetuple_list;
/* store the SYN filter info */
uint32_t syn_info;
};
> > return 0;
> > }
> >
> > @@ -5509,15 +5511,19 @@ ixgbe_syn_filter_set(struct rte_eth_dev *dev,
> > bool add)
> > {
> > struct ixgbe_hw *hw =
> > IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> > + struct ixgbe_filter_info *filter_info =
> > + IXGBE_DEV_PRIVATE_TO_FILTER_INFO(dev->data-
> >dev_private);
> > + uint32_t syn_info;
> > uint32_t synqf;
> >
> > if (filter->queue >= IXGBE_MAX_RX_QUEUE_NUM)
> > return -EINVAL;
> >
> > + syn_info = filter_info->syn_info;
> > synqf = IXGBE_READ_REG(hw, IXGBE_SYNQF);
> >
> > if (add) {
> > - if (synqf & IXGBE_SYN_FILTER_ENABLE)
> > + if (syn_info & IXGBE_SYN_FILTER_ENABLE)
>
> If these checks will be done on syn_info, shouldn't syn_info be assigned to
> synqf before this. Specially for first usage, synqf may be different than hw
> register.
>
> Or perhaps can keep continue to use synqf. Since synqf assigned to
> filter_info->syn_info after updated.
>
Let me have a deeper think of this to reply you.
> > return -EINVAL;
> > synqf = (uint32_t)(((filter->queue <<
> IXGBE_SYN_FILTER_QUEUE_SHIFT) &
> > IXGBE_SYN_FILTER_QUEUE) |
> IXGBE_SYN_FILTER_ENABLE); @@ -5527,10
> > +5533,12 @@ ixgbe_syn_filter_set(struct rte_eth_dev *dev,
> > else
> > synqf &= ~IXGBE_SYN_FILTER_SYNQFP;
> > } else {
> > - if (!(synqf & IXGBE_SYN_FILTER_ENABLE))
> > + if (!(syn_info & IXGBE_SYN_FILTER_ENABLE))
> > return -ENOENT;
> > synqf &= ~(IXGBE_SYN_FILTER_QUEUE |
> IXGBE_SYN_FILTER_ENABLE);
> > }
> > +
> > + filter_info->syn_info = synqf;
> > IXGBE_WRITE_REG(hw, IXGBE_SYNQF, synqf);
> > IXGBE_WRITE_FLUSH(hw);
> > return 0;
> <...>
>
More information about the dev
mailing list