On 12/23/2015 07:57 AM, Joonsoo Kim wrote:
What are the cases where pageblock_pfn_to_page() is used for a subset of
the pageblock and the result would be problematic for compaction? I.e.,
do we actually care to use pageblocks that are not contiguous at all?
The problematic pageblocks are those
On Wed, Dec 23, 2015 at 07:14:21AM +0100, Vlastimil Babka wrote:
> On 22.12.2015 23:17, David Rientjes wrote:
> > On Mon, 21 Dec 2015, Joonsoo Kim wrote:
> >
> >> Before vs After
> >> Max: 1096 MB/s vs 1325 MB/s
> >> Min: 635 MB/s 1015 MB/s
> >> Avg: 899 MB/s 1194 MB/s
> >>
> >> Avg is improved by
On 22.12.2015 23:17, David Rientjes wrote:
> On Mon, 21 Dec 2015, Joonsoo Kim wrote:
>
>> Before vs After
>> Max: 1096 MB/s vs 1325 MB/s
>> Min: 635 MB/s 1015 MB/s
>> Avg: 899 MB/s 1194 MB/s
>>
>> Avg is improved by roughly 30% [2].
>>
>
> Wow, ok!
>
> I'm wondering if it would be better to main
On Mon, 21 Dec 2015, Joonsoo Kim wrote:
> There is a performance drop report due to hugepage allocation and in there
> half of cpu time are spent on pageblock_pfn_to_page() in compaction [1].
> In that workload, compaction is triggered to make hugepage but most of
> pageblocks are un-available for
2015-12-21 21:18 GMT+09:00 Joonsoo Kim :
> 2015-12-21 19:46 GMT+09:00 Vlastimil Babka :
>> On 12/21/2015 07:13 AM, Joonsoo Kim wrote:
>>>
>>> There is a performance drop report due to hugepage allocation and in there
>>> half of cpu time are spent on pageblock_pfn_to_page() in compaction [1].
>>> I
2015-12-21 19:46 GMT+09:00 Vlastimil Babka :
> On 12/21/2015 07:13 AM, Joonsoo Kim wrote:
>>
>> There is a performance drop report due to hugepage allocation and in there
>> half of cpu time are spent on pageblock_pfn_to_page() in compaction [1].
>> In that workload, compaction is triggered to make
On 12/21/2015 07:13 AM, Joonsoo Kim wrote:
There is a performance drop report due to hugepage allocation and in there
half of cpu time are spent on pageblock_pfn_to_page() in compaction [1].
In that workload, compaction is triggered to make hugepage but most of
pageblocks are un-available for com
There is a performance drop report due to hugepage allocation and in there
half of cpu time are spent on pageblock_pfn_to_page() in compaction [1].
In that workload, compaction is triggered to make hugepage but most of
pageblocks are un-available for compaction due to pageblock type and
skip bit so
On 12/15/2015 02:06 AM, Aaron Lu wrote:
On 12/14/2015 11:25 PM, Joonsoo Kim wrote:
2015-12-14 19:29 GMT+09:00 Vlastimil Babka :
Unless I'm mistaken, these results also include my RFC series (Aaron can you
clarify?). These patches should better be tested standalone on top of base,
as being simpl
On 12/14/2015 11:25 PM, Joonsoo Kim wrote:
> 2015-12-14 19:29 GMT+09:00 Vlastimil Babka :
>> On 12/14/2015 06:02 AM, Joonsoo Kim wrote:
>>> Before vs After
>>> Max: 1096 MB/s vs 1325 MB/s
>>> Min: 635 MB/s 1015 MB/s
>>> Avg: 899 MB/s 1194 MB/s
>>>
>>> Avg is improved by roughly 30% [2].
>>
>>
>> Un
2015-12-14 19:29 GMT+09:00 Vlastimil Babka :
> On 12/14/2015 06:02 AM, Joonsoo Kim wrote:
>>
>> There is a performance drop report due to hugepage allocation and in there
>> half of cpu time are spent on pageblock_pfn_to_page() in compaction [1].
>> In that workload, compaction is triggered to make
On 12/14/2015 06:02 AM, Joonsoo Kim wrote:
There is a performance drop report due to hugepage allocation and in there
half of cpu time are spent on pageblock_pfn_to_page() in compaction [1].
In that workload, compaction is triggered to make hugepage but most of
pageblocks are un-available for com
There is a performance drop report due to hugepage allocation and in there
half of cpu time are spent on pageblock_pfn_to_page() in compaction [1].
In that workload, compaction is triggered to make hugepage but most of
pageblocks are un-available for compaction due to pageblock type and
skip bit so
13 matches
Mail list logo