Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

2016-02-15 19:06 GMT+09:00 Xishi Qiu : > On 2016/2/15 10:42, Joonsoo Kim wrote: > >>> >>> I have a question about the zone continuity. because hole exists at >>> arbitrary position in a page block. Therefore, only pageblock_pf_to_page() >>> is insufficiency, whether pageblock

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

2016-02-15 19:06 GMT+09:00 Xishi Qiu : > On 2016/2/15 10:42, Joonsoo Kim wrote: > >>> >>> I have a question about the zone continuity. because hole exists at >>> arbitrary position in a page block. Therefore, only pageblock_pf_to_page() >>> is insufficiency, whether pageblock aligned pfn or not ,

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On 2016/2/15 10:42, Joonsoo Kim wrote: >> >> I have a question about the zone continuity. because hole exists at >> arbitrary position in a page block. Therefore, only pageblock_pf_to_page() >> is insufficiency, whether pageblock aligned pfn or not , the >> pfn_valid_within() >> is necessary. >>

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On 2016/2/15 10:42, Joonsoo Kim wrote: >> >> I have a question about the zone continuity. because hole exists at >> arbitrary position in a page block. Therefore, only pageblock_pf_to_page() >> is insufficiency, whether pageblock aligned pfn or not , the >> pfn_valid_within() >> is necessary. >>

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On Sun, Feb 14, 2016 at 06:21:03PM +0800, zhong jiang wrote: > On 2016/2/6 0:11, Joonsoo Kim wrote: > > 2016-02-05 9:49 GMT+09:00 Andrew Morton : > >> On Thu, 4 Feb 2016 15:19:35 +0900 Joonsoo Kim wrote: > >> > >>> There is a performance drop report

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On Sun, Feb 14, 2016 at 06:21:03PM +0800, zhong jiang wrote: > On 2016/2/6 0:11, Joonsoo Kim wrote: > > 2016-02-05 9:49 GMT+09:00 Andrew Morton : > >> On Thu, 4 Feb 2016 15:19:35 +0900 Joonsoo Kim wrote: > >> > >>> There is a performance drop report due to hugepage allocation and in there > >>>

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On 2016/2/6 0:11, Joonsoo Kim wrote: > 2016-02-05 9:49 GMT+09:00 Andrew Morton : >> On Thu, 4 Feb 2016 15:19:35 +0900 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].

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On 2016/2/6 0:11, Joonsoo Kim wrote: > 2016-02-05 9:49 GMT+09:00 Andrew Morton : >> On Thu, 4 Feb 2016 15:19:35 +0900 Joonsoo Kim wrote: >> >>> There is a performance drop report due to hugepage allocation and in there >>> half of cpu time are spent

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

2016-02-11 3:58 GMT+09:00 Andrew Morton : > On Wed, 10 Feb 2016 14:42:57 +0100 Vlastimil Babka wrote: > >> > --- a/mm/memory_hotplug.c >> > +++ b/mm/memory_hotplug.c >> > @@ -509,6 +509,8 @@ int __ref __add_pages(int nid, struct zone *zone, >> > unsigned long phys_start_pfn, >> > int

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On Wed, 10 Feb 2016 14:42:57 +0100 Vlastimil Babka wrote: > > --- a/mm/memory_hotplug.c > > +++ b/mm/memory_hotplug.c > > @@ -509,6 +509,8 @@ int __ref __add_pages(int nid, struct zone *zone, > > unsigned long phys_start_pfn, > > int start_sec, end_sec; > > struct vmem_altmap *altmap; >

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On 02/09/2016 09:53 PM, Andrew Morton wrote: > On Tue, 9 Feb 2016 18:58:32 +0100 Vlastimil Babka wrote: > >> On 02/05/2016 05:11 PM, Joonsoo Kim wrote: >>> Yeah, it seems wrong to me. :) >>> Here goes fix. >> >> Doesn't apply for me, even after fixing the most obvious line wraps. >> Seems like

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On 02/09/2016 09:53 PM, Andrew Morton wrote: > On Tue, 9 Feb 2016 18:58:32 +0100 Vlastimil Babka wrote: > >> On 02/05/2016 05:11 PM, Joonsoo Kim wrote: >>> Yeah, it seems wrong to me. :) >>> Here goes fix. >> >> Doesn't apply for me, even after fixing the most obvious line wraps.

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On Wed, 10 Feb 2016 14:42:57 +0100 Vlastimil Babka wrote: > > --- a/mm/memory_hotplug.c > > +++ b/mm/memory_hotplug.c > > @@ -509,6 +509,8 @@ int __ref __add_pages(int nid, struct zone *zone, > > unsigned long phys_start_pfn, > > int start_sec, end_sec; > > struct

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

2016-02-11 3:58 GMT+09:00 Andrew Morton : > On Wed, 10 Feb 2016 14:42:57 +0100 Vlastimil Babka wrote: > >> > --- a/mm/memory_hotplug.c >> > +++ b/mm/memory_hotplug.c >> > @@ -509,6 +509,8 @@ int __ref __add_pages(int nid, struct zone *zone, >> >

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On Tue, 9 Feb 2016 18:58:32 +0100 Vlastimil Babka wrote: > On 02/05/2016 05:11 PM, Joonsoo Kim wrote: > > Yeah, it seems wrong to me. :) > > Here goes fix. > > Doesn't apply for me, even after fixing the most obvious line wraps. > Seems like the version in mmotm is still your original patch and

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On 02/05/2016 05:11 PM, Joonsoo Kim wrote: > Yeah, it seems wrong to me. :) > Here goes fix. Doesn't apply for me, even after fixing the most obvious line wraps. Seems like the version in mmotm is still your original patch and Andrew's hotfix?

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On Tue, 9 Feb 2016 18:58:32 +0100 Vlastimil Babka wrote: > On 02/05/2016 05:11 PM, Joonsoo Kim wrote: > > Yeah, it seems wrong to me. :) > > Here goes fix. > > Doesn't apply for me, even after fixing the most obvious line wraps. > Seems like the version in mmotm is still your

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On 02/05/2016 05:11 PM, Joonsoo Kim wrote: > Yeah, it seems wrong to me. :) > Here goes fix. Doesn't apply for me, even after fixing the most obvious line wraps. Seems like the version in mmotm is still your original patch and Andrew's hotfix?

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

2016-02-05 9:49 GMT+09:00 Andrew Morton : > On Thu, 4 Feb 2016 15:19:35 +0900 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

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

2016-02-05 9:49 GMT+09:00 Andrew Morton : > On Thu, 4 Feb 2016 15:19:35 +0900 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

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On Thu, 4 Feb 2016 15:19:35 +0900 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

Re: [PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

On Thu, 4 Feb 2016 15:19:35 +0900 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

[PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

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

[PATCH v2 3/3] mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous

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