Re: [PATCH v5 5/7] iommu/amd: Tailored gather logic for AMD

[Nadav Amit]

> On Jul 13, 2021, at 11:40 AM, Robin Murphy <robin.murphy@xxxxxxx> wrote:
> 
> On 2021-07-13 10:41, Nadav Amit wrote:
>> From: Nadav Amit <namit@xxxxxxxxxx>
>> AMD's IOMMU can flush efficiently (i.e., in a single flush) any range.
>> This is in contrast, for instnace, to Intel IOMMUs that have a limit on
>> the number of pages that can be flushed in a single flush.  In addition,
>> AMD's IOMMU do not care about the page-size, so changes of the page size
>> do not need to trigger a TLB flush.
>> So in most cases, a TLB flush due to disjoint range is not needed for
>> AMD. Yet, vIOMMUs require the hypervisor to synchronize the virtualized
>> IOMMU's PTEs with the physical ones. This process induce overheads, so
>> it is better not to cause unnecessary flushes, i.e., flushes of PTEs
>> that were not modified.
>> Implement and use amd_iommu_iotlb_gather_add_page() and use it instead
>> of the generic iommu_iotlb_gather_add_page(). Ignore disjoint regions
>> unless "non-present cache" feature is reported by the IOMMU
>> capabilities, as this is an indication we are running on a physical
>> IOMMU. A similar indication is used by VT-d (see "caching mode"). The
>> new logic retains the same flushing behavior that we had before the
>> introduction of page-selective IOTLB flushes for AMD.
>> On virtualized environments, check if the newly flushed region and the
>> gathered one are disjoint and flush if it is.
>> Cc: Joerg Roedel <joro@xxxxxxxxxx>
>> Cc: Will Deacon <will@xxxxxxxxxx>
>> Cc: Jiajun Cao <caojiajun@xxxxxxxxxx>
>> Cc: Lu Baolu <baolu.lu@xxxxxxxxxxxxxxx>
>> Cc: iommu@xxxxxxxxxxxxxxxxxxxxxxxxxx
>> Cc: linux-kernel@xxxxxxxxxxxxxxx>
>> Cc: Robin Murphy <robin.murphy@xxxxxxx>
>> Signed-off-by: Nadav Amit <namit@xxxxxxxxxx>
>> ---
>>  drivers/iommu/amd/iommu.c | 23 ++++++++++++++++++++++-
>>  1 file changed, 22 insertions(+), 1 deletion(-)
>> diff --git a/drivers/iommu/amd/iommu.c b/drivers/iommu/amd/iommu.c
>> index bfae3928b98f..cc55c4c6a355 100644
>> --- a/drivers/iommu/amd/iommu.c
>> +++ b/drivers/iommu/amd/iommu.c
>> @@ -2048,6 +2048,27 @@ static int amd_iommu_map(struct iommu_domain *dom, unsigned long iova,
>>  	return ret;
>>  }
>>  +static void amd_iommu_iotlb_gather_add_page(struct iommu_domain *domain,
>> +					    struct iommu_iotlb_gather *gather,
>> +					    unsigned long iova, size_t size)
>> +{
>> +	/*
>> +	 * AMD's IOMMU can flush as many pages as necessary in a single flush.
>> +	 * Unless we run in a virtual machine, which can be inferred according
>> +	 * to whether "non-present cache" is on, it is probably best to prefer
>> +	 * (potentially) too extensive TLB flushing (i.e., more misses) over
>> +	 * mutliple TLB flushes (i.e., more flushes). For virtual machines the
>> +	 * hypervisor needs to synchronize the host IOMMU PTEs with those of
>> +	 * the guest, and the trade-off is different: unnecessary TLB flushes
>> +	 * should be avoided.
>> +	 */
>> +	if (amd_iommu_np_cache && gather->end != 0 &&
> 
> iommu_iotlb_gather_is_disjoint() is also checking "gather->end != 0", so I don't think we need both. Strictly it's only necessary here since the other call from iommu_iotlb_gather_add_page() equivalently asserts that the gather is already non-empty via its gather->pgsize check, but one could argue it either way and I don't have a hugely strong preference.

You are correct (even if the compiler would have eliminated the redundancy).

I will remove the redundant check.

> 
> Otherwise, I love how neat this has all ended up, thanks for persevering!

Thank you for the thorough review!

Regards,
Nadav