On 7/10/2023 2:11 PM, Huang, Ying wrote:
Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx> writes:On my machine with below memory layout, and I can see it will take more time to skip the larger memory hole (range: 0x100000000 - 0x1800000000) when isolating free pages. So adding a new helper to skip the memory hole rapidly, which can reduce the time consumed from about 70us to less than 1us. [ 0.000000] Zone ranges: [ 0.000000] DMA [mem 0x0000000040000000-0x00000000ffffffff] [ 0.000000] DMA32 empty [ 0.000000] Normal [mem 0x0000000100000000-0x0000001fa7ffffff]The memory hole is at the beginning of zone NORMAL? If so, should zone
No, the memory hole range is 0x1000000000 - 0x1800000000, and the normal zone is start from 0x100000000.
I'm sorry I made a typo in the commit message, which confuses you. The memory hole range should be: 0x1000000000 - 0x1800000000. I updated the commit message to the following and addressed David's comment:
"
Just like commit 9721fd82351d ("mm: compaction: skip memory hole rapidly
when isolating migratable pages"), I can see it will also take more
time to skip the larger memory hole (range: 0x1000000000 - 0x1800000000)
when isolating free pages on my machine with below memory layout. So
like commit 9721fd82351d, adding a new helper to skip the memory hole
rapidly, which can reduce the time consumed from about 70us to less than
1us.
[ 0.000000] Zone ranges: [ 0.000000] DMA [mem 0x0000000040000000-0x00000000ffffffff] [ 0.000000] DMA32 empty [ 0.000000] Normal [mem 0x0000000100000000-0x0000001fa7ffffff] [ 0.000000] Movable zone start for each node [ 0.000000] Early memory node ranges [ 0.000000] node 0: [mem 0x0000000040000000-0x0000000fffffffff] [ 0.000000] node 0: [mem 0x0000001800000000-0x0000001fa3c7ffff] [ 0.000000] node 0: [mem 0x0000001fa3c80000-0x0000001fa3ffffff] [ 0.000000] node 0: [mem 0x0000001fa4000000-0x0000001fa402ffff] [ 0.000000] node 0: [mem 0x0000001fa4030000-0x0000001fa40effff] [ 0.000000] node 0: [mem 0x0000001fa40f0000-0x0000001fa73cffff] [ 0.000000] node 0: [mem 0x0000001fa73d0000-0x0000001fa745ffff] [ 0.000000] node 0: [mem 0x0000001fa7460000-0x0000001fa746ffff] [ 0.000000] node 0: [mem 0x0000001fa7470000-0x0000001fa758ffff] [ 0.000000] node 0: [mem 0x0000001fa7590000-0x0000001fa7ffffff] "
NORMAL start at 0x1800000000? And, the free pages will not be scanned there? Or my understanding were wrong. >[ 0.000000] Movable zone start for each node [ 0.000000] Early memory node ranges [ 0.000000] node 0: [mem 0x0000000040000000-0x0000000fffffffff] [ 0.000000] node 0: [mem 0x0000001800000000-0x0000001fa3c7ffff] [ 0.000000] node 0: [mem 0x0000001fa3c80000-0x0000001fa3ffffff] [ 0.000000] node 0: [mem 0x0000001fa4000000-0x0000001fa402ffff] [ 0.000000] node 0: [mem 0x0000001fa4030000-0x0000001fa40effff] [ 0.000000] node 0: [mem 0x0000001fa40f0000-0x0000001fa73cffff] [ 0.000000] node 0: [mem 0x0000001fa73d0000-0x0000001fa745ffff] [ 0.000000] node 0: [mem 0x0000001fa7460000-0x0000001fa746ffff] [ 0.000000] node 0: [mem 0x0000001fa7470000-0x0000001fa758ffff] [ 0.000000] node 0: [mem 0x0000001fa7590000-0x0000001fa7ffffff] Signed-off-by: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx> --- mm/compaction.c | 30 +++++++++++++++++++++++++++++- 1 file changed, 29 insertions(+), 1 deletion(-) diff --git a/mm/compaction.c b/mm/compaction.c index 43358efdbdc2..9641e2131901 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -249,11 +249,31 @@ static unsigned long skip_offline_sections(unsigned long start_pfn)return 0;} + +static unsigned long skip_offline_sections_reverse(unsigned long start_pfn) +{ + unsigned long start_nr = pfn_to_section_nr(start_pfn); + + if (!start_nr || online_section_nr(start_nr)) + return 0; + + while (start_nr-- > 0) { + if (online_section_nr(start_nr)) + return section_nr_to_pfn(start_nr) + PAGES_PER_SECTION - 1; + } + + return 0; +} #else static unsigned long skip_offline_sections(unsigned long start_pfn) { return 0; } + +static unsigned long skip_offline_sections_reverse(unsigned long start_pfn) +{ + return 0; +} #endif/*@@ -1668,8 +1688,16 @@ static void isolate_freepages(struct compact_control *cc)page = pageblock_pfn_to_page(block_start_pfn, block_end_pfn,zone); - if (!page) + if (!page) { + unsigned long next_pfn; + + next_pfn = skip_offline_sections_reverse(block_start_pfn); + if (next_pfn) + block_start_pfn = max(pageblock_start_pfn(next_pfn), + low_pfn); + continue; + }/* Check the block is suitable for migration */if (!suitable_migration_target(cc, page))
