Re: [PATCH net-next 2/7] dma: avoid expensive redundant calls for sync operations

[Alexander Lobakin]

From: Robin Murphy <robin.murphy@xxxxxxx>
Date: Mon, 29 Jan 2024 14:29:43 +0000

> On 2024-01-29 2:07 pm, Alexander Lobakin wrote:
>> From: Alexander Lobakin <aleksander.lobakin@xxxxxxxxx>
>> Date: Fri, 26 Jan 2024 17:45:11 +0100
>>
>>> From: Robin Murphy <robin.murphy@xxxxxxx>
>>> Date: Fri, 26 Jan 2024 15:48:54 +0000
>>>
>>>> On 26/01/2024 1:54 pm, Alexander Lobakin wrote:
>>>>> From: Eric Dumazet <edumazet@xxxxxxxxxx>
>>>>>
>>>>> Quite often, NIC devices do not need dma_sync operations on x86_64
>>>>> at least.
>>>>> Indeed, when dev_is_dma_coherent(dev) is true and
>>>>> dev_use_swiotlb(dev) is false, iommu_dma_sync_single_for_cpu()
>>>>> and friends do nothing.
>>>>>
>>>>> However, indirectly calling them when CONFIG_RETPOLINE=y consumes
>>>>> about
>>>>> 10% of cycles on a cpu receiving packets from softirq at ~100Gbit
>>>>> rate.
>>>>> Even if/when CONFIG_RETPOLINE is not set, there is a cost of about 3%.
>>>>>
>>>>> Add dev->skip_dma_sync boolean which is set during the device
>>>>> initialization depending on the setup: dev_is_dma_coherent() for
>>>>> direct
>>>>> DMA, !(sync_single_for_device || sync_single_for_cpu) or positive
>>>>> result
>>>>> from the new callback, dma_map_ops::can_skip_sync for non-NULL DMA
>>>>> ops.
>>>>> Then later, if/when swiotlb is used for the first time, the flag
>>>>> is turned off, from swiotlb_tbl_map_single().
>>>>
>>>> I think you could probably just promote the dma_uses_io_tlb flag from
>>>> SWIOTLB_DYNAMIC to a general SWIOTLB thing to serve this purpose now.
>>>
>>> Nice catch!
>>
>> BTW, this implies such hotpath check:
>>
>>     if (dev->dma_skip_sync && !READ_ONCE(dev->dma_uses_io_tlb))
>>         // ...
>>
>> This seems less effective than just resetting dma_skip_sync on first
>> allocation.
> 
> Well, my point is not to have a dma_skip_sync at all; I'm suggesting the
> check would be:
> 
>     if (dev_is_dma_coherent(dev) && dev_uses_io_tlb(dev))
>         ...

Aaah, okay, but what about dma_map_ops?
It would then become:

	if ((!dev->dma_ops ||
	     (!dev->dma_ops->sync_single_for_device &&
	      !dev->dma_ops->sync_single_for_cpu)) ||
	     (dev->dma_ops->flags & DMA_F_SKIP_SYNC)) &&
	    dev_is_dma_coherent(dev) && !dev_uses_io_tlb(dev))
		dma_sync_ ...

Quite a lot and everything except dev_uses_io_tlb() is known at device
probing time, that's why I decided to cache the result into a new flag...

> 
> where on the platform which cares about this most, that first condition
> is a compile-time constant (and as implied, the second would want to be
> similarly wrapped for !SWIOTLB configs).
> 
> Thanks,
> Robin.

Thanks,
Olek