Transparent Hugepage support #12
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest35 patches 0 repliesNew
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/35 | define MADV_HUGEPAGEarch/ arch/ arch/ +2 more | +9 | 0 replies | |
| 2/35 | compound_lockinclude/ include/ | +26 | 0 replies | |
| 3/35 | alter compound get_page/put_pagearch/ arch/ include/ +1 more | +106 | 0 replies | |
| 4/35 | update futex compound knowledgekernel/ | +58 | 0 replies | |
| 5/35 | fix bad_page to show the real reason the page is badmm/ | +1 | 0 replies | |
| 6/35 | clear compound mappingmm/ | +2 | 0 replies | |
| 7/35 | add native_set_pmd_atarch/ | +6 | 0 replies | |
| 8/35 | add pmd paravirt opsarch/ arch/ arch/ | +32 | 0 replies | |
| 9/35 | no paravirt version of pmd opsarch/ | +3 | 0 replies | |
| 10/35 | export maybe_mkwriteinclude/ mm/ | +13 | 0 replies | |
| 11/35 | comment reminder in destroy_compound_pagemm/ | +1 | 0 replies | |
| 12/35 | config_transparent_hugepagemm/ | +14 | 0 replies | |
| 13/35 | special pmd_trans_* functionsarch/ arch/ include/ | +20 | 0 replies | |
| 14/35 | add pmd mangling generic functionsinclude/ | +125 | 0 replies | |
| 15/35 | add pmd mangling functions to x86arch/ arch/ arch/ | +175 | 0 replies | |
| 16/35 | bail out gup_fast on splitting pmdarch/ | +12 | 0 replies | |
| 17/35 | pte alloc trans splittinginclude/ mm/ mm/ | +25 | 0 replies | |
| 18/35 | add pmd mmu_notifier helpersinclude/ | +40 | 0 replies | |
| 19/35 | clear page compoundinclude/ | +16 | 0 replies | |
| 20/35 | add pmd_huge_pte to mm_structinclude/ kernel/ | +10 | 0 replies | |
| 21/35 | split_huge_page_mm/vmaarch/ mm/ mm/ +3 more | +6 | 0 replies | |
| 22/35 | split_huge_page pagingmm/ mm/ mm/ +1 more | +11 | 0 replies | |
| 23/35 | clear_copy_huge_pageinclude/ mm/ mm/ | +82 | 0 replies | |
| 24/35 | kvm mmu transparent hugepage supportarch/ | +9 | 0 replies | |
| 25/35 | _GFP_NO_KSWAPDinclude/ mm/ | +5 | 0 replies | |
| 26/35 | dont alloc harder for gfp nomemalloc even if nowaitmm/ | +5 | 0 replies | |
| 27/35 | transparent hugepage coreinclude/ include/ include/ +7 more | +1144 | 0 replies | |
| 28/35 | adapt to mm_counter in -mmmm/ | +3 | 0 replies | |
| 29/35 | verify pmd_trans_huge isnt leakingmm/ | +6 | 0 replies | |
| 30/35 | madvise(MADV_HUGEPAGE)include/ mm/ mm/ | +30 | 0 replies | |
| 31/35 | pmd_trans_huge migrate bugcheckinclude/ mm/ | +10 | 0 replies | |
| 32/35 | memcg compoundDocumentation/ mm/ | +52 | 0 replies | |
| 33/35 | memcg huge memorymm/ | +25 | 0 replies | |
| 34/35 | transparent hugepage vmstatfs/ include/ mm/ +2 more | +30 | 0 replies | |
| 35/35 | khugepagedinclude/ include/ include/ +2 more | +1049 | 0 replies |
Reviews trailers per patch
All 35 patches have a Reviewed-by or Acked-by.
| Patch | Rik van Riel | Arnd Bergmann | Christoph Lameter | Mel Gorman |
|---|---|---|---|---|
| 1/35 | A-b | A-b | ||
| 2/35 | A-b | |||
| 3/35 | A-b | |||
| 4/35 | A-b | A-b | ||
| 5/35 | A-b | A-b | ||
| 6/35 | A-b | A-b | ||
| 7/35 | A-b | A-b | ||
| 8/35 | A-b | A-b | ||
| 9/35 | A-b | A-b | ||
| 10/35 | A-b | A-b | ||
| 11/35 | A-b | A-b | ||
| 12/35 | A-b | A-b | ||
| 13/35 | A-b | |||
| 14/35 | A-b | |||
| 15/35 | A-b | |||
| 16/35 | A-b | A-b | ||
| 17/35 | A-b | A-b | ||
| 18/35 | A-b | |||
| 19/35 | A-b | R-b | ||
| 20/35 | A-b | |||
| 21/35 | A-b | A-b | ||
| 22/35 | A-b | A-b | ||
| 23/35 | A-b | A-b | ||
| 24/35 | A-b | |||
| 25/35 | A-b | |||
| 26/35 | A-b | |||
| 27/35 | A-b | |||
| 28/35 | A-b | |||
| 29/35 | A-b | |||
| 30/35 | A-b | |||
| 31/35 | A-b | |||
| 32/35 | A-b | |||
| 33/35 | A-b | |||
| 34/35 | A-b | |||
| 35/35 | A-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 4 reviewers.
Cover letter [patch 00/35] Transparent Hugepage support #12
This adds more commentary to khugepaged and it adds two defrag sysfs knobs:
find /sys/ -name defrag /sys/kernel/mm/transparent_hugepage/defrag /sys/kernel/mm/transparent_hugepage/khugepaged/defrag
find /sys/ -name defrag -exec cat {} \;
always [madvise] never
[yes] no
This should also fix memcg khugepaged accounting (thanks Kame).
I folded the page_anon_vma patch into transparent_hugepage. It was however good idea to keep it separated first time around IMHO to make reviewing of the changes feasible (it's still there to review on the list in the #11 submit if others wants to review).
Let me know if something else is needed to merge into -mm.
After that we can start to call memory compaction in alloc_hugepage(int defrag) if defrag == 1.
Thanks!
Andrea
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>