Re: [RFC PATCH 1/4] net: implement mechanism for HW based QOS

[John Fastabend <john.r.fastabend@intel.com>]

On 12/9/2010 12:46 PM, Eric Dumazet wrote:
> Le jeudi 09 décembre 2010 à 11:59 -0800, John Fastabend a écrit :
>> This patch provides a mechanism for lower layer devices to
>> steer traffic using skb->priority to tx queues. This allows
>> for hardware based QOS schemes to use the default qdisc without
>> incurring the penalties related to global state and the qdisc
>> lock. While reliably receiving skbs on the correct tx ring
>> to avoid head of line blocking resulting from shuffling in
>> the LLD. Finally, all the goodness from txq caching and xps/rps
>> can still be leveraged.
>>
>> Many drivers and hardware exist with the ability to implement
>> QOS schemes in the hardware but currently these drivers tend
>> to rely on firmware to reroute specific traffic, a driver
>> specific select_queue or the queue_mapping action in the
>> qdisc.
>>
>> By using select_queue for this drivers need to be updated for
>> each and every traffic type and we lose the goodness of much
>> of the upstream work. Firmware solutions are inherently
>> inflexible. And finally if admins are expected to build a
>> qdisc and filter rules to steer traffic this requires knowledge
>> of how the hardware is currently configured. The number of tx
>> queues and the queue offsets may change depending on resources.
>> Also this approach incurs all the overhead of a qdisc with filters.
>>
>> With the mechanism in this patch users can set skb priority using
>> expected methods ie setsockopt() or the stack can set the priority
>> directly. Then the skb will be steered to the correct tx queues
>> aligned with hardware QOS traffic classes. In the normal case with
>> a single traffic class and all queues in this class everything
>> works as is until the LLD enables multiple tcs.
>>
>> To steer the skb we mask out the lower 4 bits of the priority
>> and allow the hardware to configure upto 15 distinct classes
>> of traffic. This is expected to be sufficient for most applications
>> at any rate it is more then the 8021Q spec designates and is
>> equal to the number of prio bands currently implemented in
>> the default qdisc.
>>
>> This in conjunction with a userspace application such as
>> lldpad can be used to implement 8021Q transmission selection
>> algorithms one of these algorithms being the extended transmission
>> selection algorithm currently being used for DCB.
>>
> 
> Very nice Changelog !
> 
>> Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
>> ---
>>
>>  include/linux/netdevice.h |   65 +++++++++++++++++++++++++++++++++++++++++++++
>>  net/core/dev.c            |   39 ++++++++++++++++++++++++++-
>>  2 files changed, 103 insertions(+), 1 deletions(-)
>>
>> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>> index a9ac5dc..c0d4fb1 100644
>> --- a/include/linux/netdevice.h
>> +++ b/include/linux/netdevice.h
>> @@ -646,6 +646,12 @@ struct xps_dev_maps {
>>      (nr_cpu_ids * sizeof(struct xps_map *)))
>>  #endif /* CONFIG_XPS */
>>  
>> +/* HW offloaded queuing disciplines txq count and offset maps */
>> +struct netdev_tc_txq {
>> +	u16 count;
>> +	u16 offset;
>> +};
>> +
>>  /*
>>   * This structure defines the management hooks for network devices.
>>   * The following hooks can be defined; unless noted otherwise, they are
>> @@ -1146,6 +1152,10 @@ struct net_device {
>>  	/* Data Center Bridging netlink ops */
>>  	const struct dcbnl_rtnl_ops *dcbnl_ops;
>>  #endif
>> +	u8 max_tc;
>> +	u8 num_tc;
>> +	struct netdev_tc_txq *_tc_to_txq;
> 
> Given that this is up to 16*4 bytes (64), shouldnt we embed this in
> net_device struct to avoid one dereference ?
> 
> 
>> +	u8 prio_tc_map[16];
>>  
>>  #if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
>>  	/* max exchange id for FCoE LRO by ddp */
>> @@ -1162,6 +1172,58 @@ struct net_device {
>>  #define	NETDEV_ALIGN		32
>>  
>>  static inline
>> +int netdev_get_prio_tc_map(const struct net_device *dev, u32 prio)
>> +{
>> +	return dev->prio_tc_map[prio & 15];
>> +}
>> +
>> +static inline
>> +int netdev_set_prio_tc_map(struct net_device *dev, u8 prio, u8 tc)
>> +{
>> +	if (tc >= dev->num_tc)
>> +		return -EINVAL;
>> +
>> +	dev->prio_tc_map[prio & 15] = tc & 15;
>> +	return 0;
>> +}
>> +
>> +static inline
>> +int netdev_set_tc_queue(struct net_device *dev, u8 tc, u16 count, u16 offset)
>> +{
>> +	struct netdev_tc_txq *tcp;
>> +
>> +	if (tc >= dev->num_tc)
>> +		return -EINVAL;
>> +
>> +	tcp = &dev->_tc_to_txq[tc];
>> +	tcp->count = count;
>> +	tcp->offset = offset;
>> +	return 0;
>> +}
>> +
>> +static inline
>> +struct netdev_tc_txq *netdev_get_tc_queue(const struct net_device *dev, u8 tc)
>> +{
>> +	return &dev->_tc_to_txq[tc];
>> +}
>> +
>> +static inline
>> +int netdev_set_num_tc(struct net_device *dev, u8 num_tc)
>> +{
>> +	if (num_tc > dev->max_tc)
>> +		return -EINVAL;
>> +
>> +	dev->num_tc = num_tc;
>> +	return 0;
>> +}
>> +
>> +static inline
>> +u8 netdev_get_num_tc(const struct net_device *dev)
>> +{
>> +	return dev->num_tc;
>> +}
>> +
>> +static inline
>>  struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
>>  					 unsigned int index)
>>  {
>> @@ -1386,6 +1448,9 @@ static inline void unregister_netdevice(struct net_device *dev)
>>  	unregister_netdevice_queue(dev, NULL);
>>  }
>>  
>> +extern int		netdev_alloc_max_tc(struct net_device *dev, u8 tc);
>> +extern void		netdev_free_tc(struct net_device *dev);
>> +
>>  extern int 		netdev_refcnt_read(const struct net_device *dev);
>>  extern void		free_netdev(struct net_device *dev);
>>  extern void		synchronize_net(void);
>> diff --git a/net/core/dev.c b/net/core/dev.c
>> index 55ff66f..cc00e66 100644
>> --- a/net/core/dev.c
>> +++ b/net/core/dev.c
>> @@ -2118,6 +2118,8 @@ static u32 hashrnd __read_mostly;
>>  u16 skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb)
>>  {
>>  	u32 hash;
>> +	u16 qoffset = 0;
>> +	u16 qcount = dev->real_num_tx_queues;
>>  
>>  	if (skb_rx_queue_recorded(skb)) {
>>  		hash = skb_get_rx_queue(skb);
>> @@ -2126,13 +2128,20 @@ u16 skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb)
>>  		return hash;
>>  	}
>>  
>> +	if (dev->num_tc) {
>> +		u8 tc = netdev_get_prio_tc_map(dev, skb->priority);
>> +		struct netdev_tc_txq *tcp = netdev_get_tc_queue(dev, tc);
>> +		qoffset = tcp->offset;
>> +		qcount = tcp->count;
>> +	}
>> +
>>  	if (skb->sk && skb->sk->sk_hash)
>>  		hash = skb->sk->sk_hash;
>>  	else
>>  		hash = (__force u16) skb->protocol ^ skb->rxhash;
>>  	hash = jhash_1word(hash, hashrnd);
>>  
>> -	return (u16) (((u64) hash * dev->real_num_tx_queues) >> 32);
>> +	return (u16) ((((u64) hash * qcount)) >> 32) + qoffset;
>>  }
>>  EXPORT_SYMBOL(skb_tx_hash);
>>  
>> @@ -5091,6 +5100,33 @@ void netif_stacked_transfer_operstate(const struct net_device *rootdev,
>>  }
>>  EXPORT_SYMBOL(netif_stacked_transfer_operstate);
>>  
>> +int netdev_alloc_max_tc(struct net_device *dev, u8 tcs)
>> +{
>> +	struct netdev_tc_txq *tcp;
>> +
>> +	if (tcs > 16)
>> +		return -EINVAL;
>> +
>> +	tcp = kcalloc(tcs, sizeof(*tcp), GFP_KERNEL);
> 
> common risk : allocating less than one cache line, and this possibly can
> have false sharing.
> 
> I would just embed the thing.
> 

Yes, I think you are right plus this simplifies the code a bit. I'll go ahead and do this. Thanks!