AndyGrover
>On 03/17/2017 01:04 AM, Xiubo Li wrote:
>> [...]
>>> These days what I have gotten is that the unmap_mapping_range()
>could
>>> be used.
>>> At the same time I have deep into the mm code and fixed the double
>>> usage of
>>> the data blocks and possible page fault call trace bugs mentioned
>above.
>>>
>>> Following is the V3 patch. I have test this using 4 targets & fio
>for
>>> about 2 days, so
>>> far so good.
>>>
>>> I'm still testing this using more complex test case.
>>>
>> I have test it the whole day today:
>> - using 4 targets
>> - setting TCMU_GLOBAL_MAX_BLOCKS = [512 1K 1M 1G 2G]
>> - each target here needs more than 450 blocks when running
>> - fio: -iodepth [1 2 4 8 16] -thread -rw=[read write] -bs=[1K 2K 3K
>5K
>> 7K 16K 64K 1M] -size=20G -numjobs=10 -runtime=1000 ...
>
>Hi Xiubo,
>
>V3 is sounding very good. I look forward to reviewing it after it is
>posted.
>
Yes, I will post it later after more test and checking.
Thanks,
BRs
Xiubo
>Thanks -- Regards -- Andy
