Re: [PATCH] tests/gem_userptr_blits: Expanded userptr test cases

[Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>]

On 01/22/2014 11:38 AM, Tvrtko Ursulin wrote:
>
> On 01/22/2014 11:21 AM, Chris Wilson wrote:
>> On Wed, Jan 22, 2014 at 10:41:38AM +0000, Tvrtko Ursulin wrote:
>>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>
>>> A set of userptr test cases to support the new feature.
>>
>> Oh, and another thing...
>>
>> I guess we will want some benchmarks for mmu_notifier scaling since
>> tracking the performance implications seems important.
>
> What would be a realistic benchmark? My first thoughts were something like:
>
> for each [unsync, sync]
>    for each N [1, 10, 100, 1000]

Also zero userptr objects of course!

>      create N userptr bos
>          benchmark for T seconds: malloc/realloc/free
>          benchmark for T seconds: mmap/unmap
>          ???

+ benchmark for T seconds: normal bo create/destroy
+ benchmark for T seconds: userptr bo create/destroy

Tvrtko