On 1/19/2018 7:54 PM, Kirill A. Shutemov wrote: > On Thu, Jan 18, 2018 at 06:26:50PM +0530, Vinayak Menon wrote: >> Based on Kirill's patch [1]. >> >> Currently, faultaround code produces young pte. This can screw up >> vmscan behaviour[2], as it makes vmscan think that these pages are hot >> and not push them out on first round. >> >> During sparse file access faultaround gets more pages mapped and all of >> them are young. Under memory pressure, this makes vmscan swap out anon >> pages instead, or to drop other page cache pages which otherwise stay >> resident. >> >> Modify faultaround to produce old ptes if sysctl 'want_old_faultaround_pte' >> is set, so they can easily be reclaimed under memory pressure. >> >> This can to some extend defeat the purpose of faultaround on machines >> without hardware accessed bit as it will not help us with reducing the >> number of minor page faults. >> >> Making the faultaround ptes old results in a unixbench regression for some >> architectures [3][4]. But on some architectures like arm64 it is not found >> to cause any regression. >> >> unixbench shell8 scores on arm64 v8.2 hardware with CONFIG_ARM64_HW_AFDBM >> enabled (5 runs min, max, avg): >> Base: (741,748,744) >> With this patch: (739,748,743) >> >> So by default produce young ptes and provide a sysctl option to make the >> ptes old. >> >> [1] http://lkml.kernel.org/r/1463488366-47723-1-git-send-email-kirill.shutemov@xxxxxxxxxxxxxxx >> [2] https://lkml.kernel.org/r/1460992636-711-1-git-send-email-vinmenon@xxxxxxxxxxxxxx >> [3] https://marc.info/?l=linux-kernel&m=146582237922378&w=2 >> [4] https://marc.info/?l=linux-mm&m=146589376909424&w=2 >> >> Signed-off-by: Vinayak Menon <vinmenon@xxxxxxxxxxxxxx> >> --- >> >> V2: >> 1. Removed the arch hook and want_old_faultaround_pte is made a sysctl >> 2. Renamed FAULT_FLAG_MKOLD to FAULT_FLAG_PREFAULT_OLD (suggested by Jan Kara) >> 3. Removed the saved fault address from vmf (suggested by Jan Kara) >> >> Documentation/sysctl/vm.txt | 22 ++++++++++++++++++++++ >> include/linux/mm.h | 3 +++ >> kernel/sysctl.c | 9 +++++++++ >> mm/filemap.c | 10 ++++++++++ >> mm/memory.c | 4 ++++ >> 5 files changed, 48 insertions(+) >> >> diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt >> index 17256f2..e015940 100644 >> --- a/Documentation/sysctl/vm.txt >> +++ b/Documentation/sysctl/vm.txt >> @@ -63,6 +63,7 @@ Currently, these files are in /proc/sys/vm: >> - vfs_cache_pressure >> - watermark_scale_factor >> - zone_reclaim_mode >> +- want_old_faultaround_pte >> >> ============================================================== >> >> @@ -887,4 +888,25 @@ Allowing regular swap effectively restricts allocations to the local >> node unless explicitly overridden by memory policies or cpuset >> configurations. >> >> +============================================================= >> + >> +want_old_faultaround_pte: >> + >> +By default faultaround code produces young pte. When want_old_faultaround_pte is >> +set to 1, faultaround produces old ptes. >> + >> +During sparse file access faultaround gets more pages mapped and when all of >> +them are young (default), under memory pressure, this makes vmscan swap out anon >> +pages instead, or to drop other page cache pages which otherwise stay resident. >> +Setting want_old_faultaround_pte to 1 avoids this. >> + >> +Making the faultaround ptes old can result in performance regression on some >> +architectures. This is due to cycles spent in micro-fault for TLB lookup of old >> +entry. > It's not for TLB lookup. Micro-fault would take page walk to set young bit in > the pte. Fixed it in v3. Thanks Kirill. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
