From: Andrew Morton <akpm@linux-foundation.org>
To: akpm@linux-foundation.org, dchinner@redhat.com,
hannes@cmpxchg.org, hch@lst.de, hughd@google.com, jack@suse.cz,
kirill.shutemov@linux.intel.com, linux-mm@kvack.org,
mm-commits@vger.kernel.org, torvalds@linux-foundation.org,
william.kucharski@oracle.com, willy@infradead.org,
yang.shi@linux.alibaba.com
Subject: [patch 001/118] mm: make pagecache tagged lookups return only head pages
Date: Thu, 25 Feb 2021 17:15:25 -0800 [thread overview]
Message-ID: <20210226011525.jSx19jby_%akpm@linux-foundation.org> (raw)
In-Reply-To: <20210225171452.713967e96554bb6a53e44a19@linux-foundation.org>
From: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Subject: mm: make pagecache tagged lookups return only head pages
Patch series "Overhaul multi-page lookups for THP", v4.
This THP prep patchset changes several page cache iteration APIs to only
return head pages.
- It's only possible to tag head pages in the page cache, so only
return head pages, not all their subpages.
- Factor a lot of common code out of the various batch lookup routines
- Add mapping_seek_hole_data()
- Unify find_get_entries() and pagevec_lookup_entries()
- Make find_get_entries only return head pages, like find_get_entry().
These are only loosely connected, but they seem to make sense together as
a series.
This patch (of 14):
Pagecache tags are used for dirty page writeback. Since dirtiness is
tracked on a per-THP basis, we only want to return the head page rather
than each subpage of a tagged page. All the filesystems which use huge
pages today are in-memory, so there are no tagged huge pages today.
Link: https://lkml.kernel.org/r/20201112212641.27837-2-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: William Kucharski <william.kucharski@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Yang Shi <yang.shi@linux.alibaba.com>
Cc: Dave Chinner <dchinner@redhat.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
mm/filemap.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
--- a/mm/filemap.c~mm-make-pagecache-tagged-lookups-return-only-head-pages
+++ a/mm/filemap.c
@@ -2062,7 +2062,7 @@ retry:
EXPORT_SYMBOL(find_get_pages_contig);
/**
- * find_get_pages_range_tag - find and return pages in given range matching @tag
+ * find_get_pages_range_tag - Find and return head pages matching @tag.
* @mapping: the address_space to search
* @index: the starting page index
* @end: The final page index (inclusive)
@@ -2070,8 +2070,9 @@ EXPORT_SYMBOL(find_get_pages_contig);
* @nr_pages: the maximum number of pages
* @pages: where the resulting pages are placed
*
- * Like find_get_pages, except we only return pages which are tagged with
- * @tag. We update @index to index the next page for the traversal.
+ * Like find_get_pages(), except we only return head pages which are tagged
+ * with @tag. @index is updated to the index immediately after the last
+ * page we return, ready for the next iteration.
*
* Return: the number of pages which were found.
*/
@@ -2105,9 +2106,9 @@ unsigned find_get_pages_range_tag(struct
if (unlikely(page != xas_reload(&xas)))
goto put_page;
- pages[ret] = find_subpage(page, xas.xa_index);
+ pages[ret] = page;
if (++ret == nr_pages) {
- *index = xas.xa_index + 1;
+ *index = page->index + thp_nr_pages(page);
goto out;
}
continue;
_
next prev parent reply other threads:[~2021-02-26 1:15 UTC|newest]
Thread overview: 123+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-26 1:14 incoming Andrew Morton
2021-02-26 1:15 ` Andrew Morton [this message]
2021-02-26 1:15 ` [patch 002/118] mm/shmem: use pagevec_lookup in shmem_unlock_mapping Andrew Morton
2021-02-26 1:15 ` [patch 003/118] mm/swap: optimise get_shadow_from_swap_cache Andrew Morton
2021-02-26 1:15 ` [patch 004/118] mm: add FGP_ENTRY Andrew Morton
2021-02-26 1:15 ` [patch 005/118] mm/filemap: rename find_get_entry to mapping_get_entry Andrew Morton
2021-02-26 1:15 ` [patch 006/118] mm/filemap: add helper for finding pages Andrew Morton
2021-02-26 1:15 ` [patch 007/118] mm/filemap: add mapping_seek_hole_data Andrew Morton
2021-02-26 1:15 ` [patch 008/118] iomap: use mapping_seek_hole_data Andrew Morton
2021-02-26 1:15 ` [patch 009/118] mm: add and use find_lock_entries Andrew Morton
2021-02-26 1:16 ` [patch 010/118] mm: add an 'end' parameter to find_get_entries Andrew Morton
2021-02-26 1:16 ` [patch 011/118] mm: add an 'end' parameter to pagevec_lookup_entries Andrew Morton
2021-02-26 1:16 ` [patch 012/118] mm: remove nr_entries parameter from pagevec_lookup_entries Andrew Morton
2021-02-26 1:16 ` [patch 013/118] mm: pass pvec directly to find_get_entries Andrew Morton
2021-02-26 1:16 ` [patch 014/118] mm: remove pagevec_lookup_entries Andrew Morton
2021-02-26 1:16 ` [patch 015/118] mm,thp,shmem: limit shmem THP alloc gfp_mask Andrew Morton
2021-02-26 1:16 ` [patch 016/118] mm,thp,shm: limit gfp mask to no more than specified Andrew Morton
2021-02-26 1:16 ` [patch 017/118] mm,thp,shmem: make khugepaged obey tmpfs mount flags Andrew Morton
2021-02-26 1:16 ` [patch 018/118] mm,shmem,thp: limit shmem THP allocations to requested zones Andrew Morton
2021-02-26 1:16 ` [patch 019/118] mm: cma: allocate cma areas bottom-up Andrew Morton
2021-02-26 1:16 ` [patch 020/118] mm/cma: expose all pages to the buddy if activation of an area fails Andrew Morton
2021-02-26 1:16 ` [patch 021/118] mm/page_alloc: count CMA pages per zone and print them in /proc/zoneinfo Andrew Morton
2021-02-26 1:16 ` [patch 022/118] mm: cma: print region name on failure Andrew Morton
2021-02-26 1:16 ` [patch 023/118] mm: vmstat: fix NOHZ wakeups for node stat changes Andrew Morton
2021-02-26 1:16 ` [patch 024/118] mm: vmstat: add some comments on internal storage of byte items Andrew Morton
2021-02-26 1:16 ` [patch 025/118] mm/vmstat.c: erase latency in vmstat_shepherd Andrew Morton
2021-02-26 1:16 ` [patch 026/118] mm: move pfn_to_online_page() out of line Andrew Morton
2021-02-26 1:17 ` [patch 027/118] mm: teach pfn_to_online_page() to consider subsection validity Andrew Morton
2021-02-26 1:17 ` [patch 028/118] mm: teach pfn_to_online_page() about ZONE_DEVICE section collisions Andrew Morton
2021-02-26 1:17 ` [patch 029/118] mm: fix memory_failure() handling of dax-namespace metadata Andrew Morton
2021-02-26 1:17 ` [patch 030/118] mm/memory_hotplug: rename all existing 'memhp' into 'mhp' Andrew Morton
2021-02-26 1:17 ` [patch 031/118] mm/memory_hotplug: MEMHP_MERGE_RESOURCE -> MHP_MERGE_RESOURCE Andrew Morton
2021-02-26 1:17 ` [patch 032/118] mm/memory_hotplug: use helper function zone_end_pfn() to get end_pfn Andrew Morton
2021-02-26 1:17 ` [patch 033/118] drivers/base/memory: don't store phys_device in memory blocks Andrew Morton
2021-02-26 1:17 ` [patch 034/118] Documentation: sysfs/memory: clarify some memory block device properties Andrew Morton
2021-02-26 1:17 ` [patch 035/118] mm/memory_hotplug: prevalidate the address range being added with platform Andrew Morton
2021-02-26 1:17 ` [patch 036/118] arm64/mm: define arch_get_mappable_range() Andrew Morton
2021-02-26 1:17 ` [patch 037/118] s390/mm: " Andrew Morton
2021-02-26 1:17 ` [patch 038/118] virtio-mem: check against mhp_get_pluggable_range() which memory we can hotplug Andrew Morton
2021-02-26 1:17 ` [patch 039/118] mm/mlock: stop counting mlocked pages when none vma is found Andrew Morton
2021-02-26 1:17 ` [patch 040/118] mm/rmap: correct some obsolete comments of anon_vma Andrew Morton
2021-02-26 1:17 ` [patch 041/118] mm/rmap: remove unneeded semicolon in page_not_mapped() Andrew Morton
2021-02-26 1:17 ` [patch 042/118] mm/rmap: fix obsolete comment in __page_check_anon_rmap() Andrew Morton
2021-02-26 1:18 ` [patch 043/118] mm/rmap: use page_not_mapped in try_to_unmap() Andrew Morton
2021-02-26 1:18 ` [patch 044/118] mm/rmap: correct obsolete comment of page_get_anon_vma() Andrew Morton
2021-02-26 1:18 ` [patch 045/118] mm/rmap: fix potential pte_unmap on an not mapped pte Andrew Morton
2021-02-26 1:18 ` [patch 046/118] mm: zswap: clean up confusing comment Andrew Morton
2021-02-26 1:18 ` [patch 047/118] mm/zswap: add the flag can_sleep_mapped Andrew Morton
2021-02-26 1:18 ` [patch 048/118] mm: set the sleep_mapped to true for zbud and z3fold Andrew Morton
2021-02-26 1:18 ` [patch 049/118] mm/zsmalloc.c: convert to use kmem_cache_zalloc in cache_alloc_zspage() Andrew Morton
2021-02-26 1:18 ` [patch 050/118] zsmalloc: account the number of compacted pages correctly Andrew Morton
2021-02-26 1:18 ` [patch 051/118] mm/zsmalloc.c: use page_private() to access page->private Andrew Morton
2021-02-26 1:18 ` [patch 052/118] mm: page-flags.h: Typo fix (It -> If) Andrew Morton
2021-02-26 1:18 ` [patch 053/118] mm/dmapool: use might_alloc() Andrew Morton
2021-02-26 1:18 ` [patch 054/118] mm/backing-dev.c: " Andrew Morton
2021-02-26 1:18 ` [patch 055/118] mm/early_ioremap.c: use __func__ instead of function name Andrew Morton
2021-02-26 1:18 ` [patch 056/118] mm: add Kernel Electric-Fence infrastructure Andrew Morton
2021-02-26 1:18 ` [patch 057/118] x86, kfence: enable KFENCE for x86 Andrew Morton
2021-02-26 1:19 ` [patch 058/118] arm64, kfence: enable KFENCE for ARM64 Andrew Morton
2021-02-26 1:19 ` [patch 059/118] kfence: use pt_regs to generate stack trace on faults Andrew Morton
2021-02-26 1:19 ` [patch 060/118] mm, kfence: insert KFENCE hooks for SLAB Andrew Morton
2021-02-26 1:19 ` [patch 061/118] mm, kfence: insert KFENCE hooks for SLUB Andrew Morton
2021-02-26 1:19 ` [patch 062/118] kfence, kasan: make KFENCE compatible with KASAN Andrew Morton
2021-02-26 1:19 ` [patch 063/118] kfence, Documentation: add KFENCE documentation Andrew Morton
2021-02-26 1:19 ` [patch 064/118] kfence: add test suite Andrew Morton
2021-02-26 1:19 ` [patch 065/118] MAINTAINERS: add entry for KFENCE Andrew Morton
2021-02-26 1:19 ` [patch 066/118] kfence: report sensitive information based on no_hash_pointers Andrew Morton
2021-02-26 1:19 ` [patch 067/118] tracing: add error_report_end trace point Andrew Morton
2021-02-26 14:03 ` Steven Rostedt
2021-02-26 1:19 ` [patch 068/118] kfence: use error_report_end tracepoint Andrew Morton
2021-02-26 1:19 ` [patch 069/118] kasan: " Andrew Morton
2021-02-26 1:19 ` [patch 070/118] kasan, mm: don't save alloc stacks twice Andrew Morton
2021-02-26 1:19 ` [patch 071/118] kasan, mm: optimize kmalloc poisoning Andrew Morton
2021-02-26 1:20 ` [patch 072/118] kasan: optimize large " Andrew Morton
2021-02-26 1:20 ` [patch 073/118] kasan: clean up setting free info in kasan_slab_free Andrew Morton
2021-02-26 1:20 ` [patch 074/118] kasan: unify large kfree checks Andrew Morton
2021-02-26 1:20 ` [patch 075/118] kasan: rework krealloc tests Andrew Morton
2021-02-26 1:20 ` [patch 076/118] kasan, mm: fail krealloc on freed objects Andrew Morton
2021-02-26 1:20 ` [patch 077/118] kasan, mm: optimize krealloc poisoning Andrew Morton
2021-02-26 1:20 ` [patch 078/118] kasan: ensure poisoning size alignment Andrew Morton
2021-02-26 1:20 ` [patch 079/118] arm64: kasan: simplify and inline MTE functions Andrew Morton
2021-02-26 1:20 ` [patch 080/118] kasan: inline HW_TAGS helper functions Andrew Morton
2021-02-26 1:20 ` [patch 081/118] kasan: clarify that only first bug is reported in HW_TAGS Andrew Morton
2021-02-26 1:20 ` [patch 082/118] alpha: remove CONFIG_EXPERIMENTAL from defconfigs Andrew Morton
2021-02-26 1:20 ` [patch 083/118] proc/wchan: use printk format instead of lookup_symbol_name() Andrew Morton
2021-02-26 1:20 ` [patch 084/118] proc: use kvzalloc for our kernel buffer Andrew Morton
2021-02-26 1:20 ` [patch 085/118] sysctl.c: fix underflow value setting risk in vm_table Andrew Morton
2021-02-26 1:20 ` [patch 086/118] include/linux: remove repeated words Andrew Morton
2021-02-26 1:21 ` [patch 087/118] treewide: Miguel has moved Andrew Morton
2021-02-26 1:21 ` [patch 088/118] groups: use flexible-array member in struct group_info Andrew Morton
2021-02-26 1:21 ` [patch 089/118] groups: simplify struct group_info allocation Andrew Morton
2021-02-26 1:21 ` [patch 090/118] kernel: delete repeated words in comments Andrew Morton
2021-02-26 1:21 ` [patch 091/118] MAINTAINERS: add uapi directories to API/ABI section Andrew Morton
2021-02-27 19:18 ` Michael Kerrisk (man-pages)
2021-02-26 1:21 ` [patch 092/118] lib/genalloc.c: change return type to unsigned long for bitmap_set_ll Andrew Morton
2021-02-26 1:21 ` [patch 093/118] string.h: move fortified functions definitions in a dedicated header Andrew Morton
2021-02-26 1:21 ` [patch 094/118] lib: stackdepot: add support to configure STACK_HASH_SIZE Andrew Morton
2021-02-26 1:21 ` [patch 095/118] lib: stackdepot: add support to disable stack depot Andrew Morton
2021-02-26 1:21 ` [patch 096/118] lib: stackdepot: fix ignoring return value warning Andrew Morton
2021-02-26 1:21 ` [patch 097/118] lib/cmdline: remove an unneeded local variable in next_arg() Andrew Morton
2021-02-26 1:21 ` [patch 098/118] include/linux/bitops.h: spelling s/synomyn/synonym/ Andrew Morton
2021-02-26 1:21 ` [patch 099/118] checkpatch: improve blank line after declaration test Andrew Morton
2021-02-26 1:21 ` [patch 100/118] checkpatch: ignore warning designated initializers using NR_CPUS Andrew Morton
2021-02-26 1:21 ` [patch 101/118] checkpatch: trivial style fixes Andrew Morton
2021-02-26 1:21 ` [patch 102/118] checkpatch: prefer ftrace over function entry/exit printks Andrew Morton
2021-02-26 1:21 ` [patch 103/118] checkpatch: improve TYPECAST_INT_CONSTANT test message Andrew Morton
2021-02-26 1:21 ` [patch 104/118] checkpatch: add warning for avoiding .L prefix symbols in assembly files Andrew Morton
2021-02-26 1:22 ` [patch 105/118] checkpatch: add kmalloc_array_node to unnecessary OOM message check Andrew Morton
2021-02-26 1:22 ` [patch 106/118] checkpatch: don't warn about colon termination in linker scripts Andrew Morton
2021-02-26 1:22 ` [patch 107/118] checkpatch: do not apply "initialise globals to 0" check to BPF progs Andrew Morton
2021-02-26 1:22 ` [patch 108/118] init/version.c: remove Version_<LINUX_VERSION_CODE> symbol Andrew Morton
2021-02-26 1:22 ` [patch 109/118] init: clean up early_param_on_off() macro Andrew Morton
2021-02-26 1:22 ` [patch 110/118] init/Kconfig: fix a typo in CC_VERSION_TEXT help text Andrew Morton
2021-02-26 1:22 ` [patch 111/118] fs/coredump: use kmap_local_page() Andrew Morton
2021-02-26 1:22 ` [patch 112/118] seq_file: document how per-entry resources are managed Andrew Morton
2021-02-26 1:22 ` [patch 113/118] x86: fix seq_file iteration for pat/memtype.c Andrew Morton
2021-02-26 1:22 ` [patch 114/118] scripts/gdb: fix list_for_each Andrew Morton
2021-02-26 1:22 ` [patch 115/118] kgdb: fix to kill breakpoints on initmem after boot Andrew Morton
2021-02-26 1:22 ` [patch 116/118] ubsan: remove overflow checks Andrew Morton
2021-02-26 1:22 ` [patch 117/118] initramfs: panic with memory information Andrew Morton
2021-02-26 1:22 ` [patch 118/118] MIPS: make userspace mapping young by default Andrew Morton
2021-02-26 17:55 ` incoming Linus Torvalds
2021-02-26 19:16 ` incoming Andrew Morton
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210226011525.jSx19jby_%akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=dchinner@redhat.com \
--cc=hannes@cmpxchg.org \
--cc=hch@lst.de \
--cc=hughd@google.com \
--cc=jack@suse.cz \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-mm@kvack.org \
--cc=mm-commits@vger.kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=william.kucharski@oracle.com \
--cc=willy@infradead.org \
--cc=yang.shi@linux.alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).