[dpdk-dev] [PATCH v2] sched: fix port time rounding error

[Kevin Traynor]

On 21/08/2020 16:28, Kinsella, Ray wrote:
> 
> 
> On 20/08/2020 15:32, Kevin Traynor wrote:
>> Hi,
>>
>> On 25/06/2020 10:59, alangordondewar at gmail.com wrote:
>>> From: Alan Dewar <alan.dewar at att.com>
>>>
>>> The QoS scheduler works off port time that is computed from the number
>>> of CPU cycles that have elapsed since the last time the port was
>>> polled.   It divides the number of elapsed cycles to calculate how
>>> many bytes can be sent, however this division can generate rounding
>>> errors, where some fraction of a byte sent may be lost.
>>>
>>> Lose enough of these fractional bytes and the QoS scheduler
>>> underperforms.  The problem is worse with low bandwidths.
>>>
>>> To compensate for this rounding error this fix doesn't advance the
>>> port's time_cpu_cycles by the number of cycles that have elapsed,
>>> but by multiplying the computed number of bytes that can be sent
>>> (which has been rounded down) by number of cycles per byte.
>>> This will mean that port's time_cpu_cycles will lag behind the CPU
>>> cycles momentarily.  At the next poll, the lag will be taken into
>>> account.
>>>
>>> v2:
>>> If the cycles value wraps (100 year+) reset the port's cpu cycle back
>>> to zero.
>>>
>>> Fixes: de3cfa2c98 ("sched: initial import")
>>>
>>> Signed-off-by: Alan Dewar <alan.dewar at att.com>
>>> ---
>>>  lib/librte_sched/rte_sched.c | 11 +++++++++--
>>>  1 file changed, 9 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/lib/librte_sched/rte_sched.c b/lib/librte_sched/rte_sched.c
>>> index c0983ddda..7c022cd61 100644
>>> --- a/lib/librte_sched/rte_sched.c
>>> +++ b/lib/librte_sched/rte_sched.c
>>> @@ -222,6 +222,7 @@ struct rte_sched_port {
>>>  	uint64_t time_cpu_bytes;      /* Current CPU time measured in bytes */
>>>  	uint64_t time;                /* Current NIC TX time measured in bytes */
>>>  	struct rte_reciprocal inv_cycles_per_byte; /* CPU cycles per byte */
>>> +	uint64_t cycles_per_byte;
>>>  
>>
>> I was backporting this patch to 18.11. The older ABI checker complains
>> about this structure change.
>>
>> "cycles_per_byte has been added at the middle position of this
>> structural type."
>>
>> Isn't this an ABI break? Dropping from 18.11 for time being.
> 
> So it looks like rte_sched_port * is an opaque pointers as it's contents are only
> known to rte_sched.c and not outside. To everyone else it is an opaque data structure, 
> so structural changes to it would not be an ABI breakage.
> 

Thanks Ray, makes sense. I've included the fix in 18.11.10-rc1.

Kevin.

>>
>>>  	/* Grinders */
>>>  	struct rte_mbuf **pkts_out;
>>> @@ -852,6 +853,7 @@ rte_sched_port_config(struct rte_sched_port_params *params)
>>>  	cycles_per_byte = (rte_get_tsc_hz() << RTE_SCHED_TIME_SHIFT)
>>>  		/ params->rate;
>>>  	port->inv_cycles_per_byte = rte_reciprocal_value(cycles_per_byte);
>>> +	port->cycles_per_byte = cycles_per_byte;
>>>  
>>>  	/* Grinders */
>>>  	port->pkts_out = NULL;
>>> @@ -2673,16 +2675,21 @@ static inline void
>>>  rte_sched_port_time_resync(struct rte_sched_port *port)
>>>  {
>>>  	uint64_t cycles = rte_get_tsc_cycles();
>>> -	uint64_t cycles_diff = cycles - port->time_cpu_cycles;
>>> +	uint64_t cycles_diff;
>>>  	uint64_t bytes_diff;
>>>  	uint32_t i;
>>>  
>>> +	if (cycles < port->time_cpu_cycles)
>>> +		port->time_cpu_cycles = 0;
>>> +
>>> +	cycles_diff = cycles - port->time_cpu_cycles;
>>>  	/* Compute elapsed time in bytes */
>>>  	bytes_diff = rte_reciprocal_divide(cycles_diff << RTE_SCHED_TIME_SHIFT,
>>>  					   port->inv_cycles_per_byte);
>>>  
>>>  	/* Advance port time */
>>> -	port->time_cpu_cycles = cycles;
>>> +	port->time_cpu_cycles +=
>>> +		(bytes_diff * port->cycles_per_byte) >> RTE_SCHED_TIME_SHIFT;
>>>  	port->time_cpu_bytes += bytes_diff;
>>>  	if (port->time < port->time_cpu_bytes)
>>>  		port->time = port->time_cpu_bytes;
>>>
>>
>