linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: akpm@linux-foundation.org, hdanton@sina.com,
	huang.ying.caritas@gmail.com, linux-mm@kvack.org,
	mhocko@suse.com, minchan@kernel.org, mm-commits@vger.kernel.org,
	oleksiy.avramchenko@sonymobile.com, rostedt@goodmis.org,
	torvalds@linux-foundation.org, urezki@gmail.com,
	willy@infradead.org
Subject: [patch 095/200] mm/vmalloc: rework the drain logic
Date: Mon, 14 Dec 2020 19:08:49 -0800	[thread overview]
Message-ID: <20201215030849.1tGt5v8fS%akpm@linux-foundation.org> (raw)
In-Reply-To: <20201214190237.a17b70ae14f129e2dca3d204@linux-foundation.org>

From: "Uladzislau Rezki (Sony)" <urezki@gmail.com>
Subject: mm/vmalloc: rework the drain logic

A current "lazy drain" model suffers from at least two issues.

First one is related to the unsorted list of vmap areas, thus in order to
identify the [min:max] range of areas to be drained, it requires a full
list scan.  What is a time consuming if the list is too long.

Second one and as a next step is about merging all fragments with a free
space.  What is also a time consuming because it has to iterate over
entire list which holds outstanding lazy areas.

See below the "preemptirqsoff" tracer that illustrates a high latency.  It
is ~24676us.  Our workloads like audio and video are effected by such long
latency:

<snip>
  tracer: preemptirqsoff

  preemptirqsoff latency trace v1.1.5 on 4.9.186-perf+
  --------------------------------------------------------------------
  latency: 24676 us, #4/4, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 P:8)
     -----------------
     | task: crtc_commit:112-261 (uid:0 nice:0 policy:1 rt_prio:16)
     -----------------
   => started at: __purge_vmap_area_lazy
   => ended at:   __purge_vmap_area_lazy

                   _------=> CPU#
                  / _-----=> irqs-off
                 | / _----=> need-resched
                 || / _---=> hardirq/softirq
                 ||| / _--=> preempt-depth
                 |||| /     delay
   cmd     pid   ||||| time  |   caller
      \   /      |||||  \    |   /
crtc_com-261     1...1    1us*: _raw_spin_lock <-__purge_vmap_area_lazy
[...]
crtc_com-261     1...1 24675us : _raw_spin_unlock <-__purge_vmap_area_lazy
crtc_com-261     1...1 24677us : trace_preempt_on <-__purge_vmap_area_lazy
crtc_com-261     1...1 24683us : <stack trace>
 => free_vmap_area_noflush
 => remove_vm_area
 => __vunmap
 => vfree
 => drm_property_free_blob
 => drm_mode_object_unreference
 => drm_property_unreference_blob
 => __drm_atomic_helper_crtc_destroy_state
 => sde_crtc_destroy_state
 => drm_atomic_state_default_clear
 => drm_atomic_state_clear
 => drm_atomic_state_free
 => complete_commit
 => _msm_drm_commit_work_cb
 => kthread_worker_fn
 => kthread
 => ret_from_fork
<snip>

To address those two issues we can redesign a purging of the outstanding
lazy areas.  Instead of queuing vmap areas to the list, we replace it by
the separate rb-tree.  In hat case an area is located in the tree/list in
ascending order.  It will give us below advantages:

a) Outstanding vmap areas are merged creating bigger coalesced blocks,
   thus it becomes less fragmented.

b) It is possible to calculate a flush range [min:max] without scanning
   all elements.  It is O(1) access time or complexity;

c) The final merge of areas with the rb-tree that represents a free
   space is faster because of (a).  As a result the lock contention is
   also reduced.

Link: https://lkml.kernel.org/r/20201116220033.1837-2-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sonymobile.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: huang ying <huang.ying.caritas@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 include/linux/vmalloc.h |    8 +--
 mm/vmalloc.c            |   90 +++++++++++++++++++++-----------------
 2 files changed, 53 insertions(+), 45 deletions(-)

--- a/include/linux/vmalloc.h~mm-vmalloc-rework-the-drain-logic
+++ a/include/linux/vmalloc.h
@@ -72,16 +72,14 @@ struct vmap_area {
 	struct list_head list;          /* address sorted list */
 
 	/*
-	 * The following three variables can be packed, because
-	 * a vmap_area object is always one of the three states:
+	 * The following two variables can be packed, because
+	 * a vmap_area object can be either:
 	 *    1) in "free" tree (root is vmap_area_root)
-	 *    2) in "busy" tree (root is free_vmap_area_root)
-	 *    3) in purge list  (head is vmap_purge_list)
+	 *    2) or "busy" tree (root is free_vmap_area_root)
 	 */
 	union {
 		unsigned long subtree_max_size; /* in "free" tree */
 		struct vm_struct *vm;           /* in "busy" tree */
-		struct llist_node purge_list;   /* in purge list */
 	};
 };
 
--- a/mm/vmalloc.c~mm-vmalloc-rework-the-drain-logic
+++ a/mm/vmalloc.c
@@ -413,10 +413,13 @@ static DEFINE_SPINLOCK(vmap_area_lock);
 static DEFINE_SPINLOCK(free_vmap_area_lock);
 /* Export for kexec only */
 LIST_HEAD(vmap_area_list);
-static LLIST_HEAD(vmap_purge_list);
 static struct rb_root vmap_area_root = RB_ROOT;
 static bool vmap_initialized __read_mostly;
 
+static struct rb_root purge_vmap_area_root = RB_ROOT;
+static LIST_HEAD(purge_vmap_area_list);
+static DEFINE_SPINLOCK(purge_vmap_area_lock);
+
 /*
  * This kmem_cache is used for vmap_area objects. Instead of
  * allocating from slab we reuse an object from this cache to
@@ -820,10 +823,17 @@ insert:
 	if (!merged)
 		link_va(va, root, parent, link, head);
 
-	/*
-	 * Last step is to check and update the tree.
-	 */
-	augment_tree_propagate_from(va);
+	return va;
+}
+
+static __always_inline struct vmap_area *
+merge_or_add_vmap_area_augment(struct vmap_area *va,
+	struct rb_root *root, struct list_head *head)
+{
+	va = merge_or_add_vmap_area(va, root, head);
+	if (va)
+		augment_tree_propagate_from(va);
+
 	return va;
 }
 
@@ -1138,7 +1148,7 @@ static void free_vmap_area(struct vmap_a
 	 * Insert/Merge it back to the free tree/list.
 	 */
 	spin_lock(&free_vmap_area_lock);
-	merge_or_add_vmap_area(va, &free_vmap_area_root, &free_vmap_area_list);
+	merge_or_add_vmap_area_augment(va, &free_vmap_area_root, &free_vmap_area_list);
 	spin_unlock(&free_vmap_area_lock);
 }
 
@@ -1326,32 +1336,32 @@ void set_iounmap_nonlazy(void)
 static bool __purge_vmap_area_lazy(unsigned long start, unsigned long end)
 {
 	unsigned long resched_threshold;
-	struct llist_node *valist;
-	struct vmap_area *va;
-	struct vmap_area *n_va;
+	struct list_head local_pure_list;
+	struct vmap_area *va, *n_va;
 
 	lockdep_assert_held(&vmap_purge_lock);
 
-	valist = llist_del_all(&vmap_purge_list);
-	if (unlikely(valist == NULL))
+	spin_lock(&purge_vmap_area_lock);
+	purge_vmap_area_root = RB_ROOT;
+	list_replace_init(&purge_vmap_area_list, &local_pure_list);
+	spin_unlock(&purge_vmap_area_lock);
+
+	if (unlikely(list_empty(&local_pure_list)))
 		return false;
 
-	/*
-	 * TODO: to calculate a flush range without looping.
-	 * The list can be up to lazy_max_pages() elements.
-	 */
-	llist_for_each_entry(va, valist, purge_list) {
-		if (va->va_start < start)
-			start = va->va_start;
-		if (va->va_end > end)
-			end = va->va_end;
-	}
+	start = min(start,
+		list_first_entry(&local_pure_list,
+			struct vmap_area, list)->va_start);
+
+	end = max(end,
+		list_last_entry(&local_pure_list,
+			struct vmap_area, list)->va_end);
 
 	flush_tlb_kernel_range(start, end);
 	resched_threshold = lazy_max_pages() << 1;
 
 	spin_lock(&free_vmap_area_lock);
-	llist_for_each_entry_safe(va, n_va, valist, purge_list) {
+	list_for_each_entry_safe(va, n_va, &local_pure_list, list) {
 		unsigned long nr = (va->va_end - va->va_start) >> PAGE_SHIFT;
 		unsigned long orig_start = va->va_start;
 		unsigned long orig_end = va->va_end;
@@ -1361,8 +1371,8 @@ static bool __purge_vmap_area_lazy(unsig
 		 * detached and there is no need to "unlink" it from
 		 * anything.
 		 */
-		va = merge_or_add_vmap_area(va, &free_vmap_area_root,
-					    &free_vmap_area_list);
+		va = merge_or_add_vmap_area_augment(va, &free_vmap_area_root,
+				&free_vmap_area_list);
 
 		if (!va)
 			continue;
@@ -1419,9 +1429,15 @@ static void free_vmap_area_noflush(struc
 	nr_lazy = atomic_long_add_return((va->va_end - va->va_start) >>
 				PAGE_SHIFT, &vmap_lazy_nr);
 
-	/* After this point, we may free va at any time */
-	llist_add(&va->purge_list, &vmap_purge_list);
+	/*
+	 * Merge or place it to the purge tree/list.
+	 */
+	spin_lock(&purge_vmap_area_lock);
+	merge_or_add_vmap_area(va,
+		&purge_vmap_area_root, &purge_vmap_area_list);
+	spin_unlock(&purge_vmap_area_lock);
 
+	/* After this point, we may free va at any time */
 	if (unlikely(nr_lazy > lazy_max_pages()))
 		try_purge_vmap_area_lazy();
 }
@@ -3351,8 +3367,8 @@ recovery:
 	while (area--) {
 		orig_start = vas[area]->va_start;
 		orig_end = vas[area]->va_end;
-		va = merge_or_add_vmap_area(vas[area], &free_vmap_area_root,
-					    &free_vmap_area_list);
+		va = merge_or_add_vmap_area_augment(vas[area], &free_vmap_area_root,
+				&free_vmap_area_list);
 		if (va)
 			kasan_release_vmalloc(orig_start, orig_end,
 				va->va_start, va->va_end);
@@ -3401,8 +3417,8 @@ err_free_shadow:
 	for (area = 0; area < nr_vms; area++) {
 		orig_start = vas[area]->va_start;
 		orig_end = vas[area]->va_end;
-		va = merge_or_add_vmap_area(vas[area], &free_vmap_area_root,
-					    &free_vmap_area_list);
+		va = merge_or_add_vmap_area_augment(vas[area], &free_vmap_area_root,
+				&free_vmap_area_list);
 		if (va)
 			kasan_release_vmalloc(orig_start, orig_end,
 				va->va_start, va->va_end);
@@ -3482,18 +3498,15 @@ static void show_numa_info(struct seq_fi
 
 static void show_purge_info(struct seq_file *m)
 {
-	struct llist_node *head;
 	struct vmap_area *va;
 
-	head = READ_ONCE(vmap_purge_list.first);
-	if (head == NULL)
-		return;
-
-	llist_for_each_entry(va, head, purge_list) {
+	spin_lock(&purge_vmap_area_lock);
+	list_for_each_entry(va, &purge_vmap_area_list, list) {
 		seq_printf(m, "0x%pK-0x%pK %7ld unpurged vm_area\n",
 			(void *)va->va_start, (void *)va->va_end,
 			va->va_end - va->va_start);
 	}
+	spin_unlock(&purge_vmap_area_lock);
 }
 
 static int s_show(struct seq_file *m, void *p)
@@ -3551,10 +3564,7 @@ static int s_show(struct seq_file *m, vo
 	seq_putc(m, '\n');
 
 	/*
-	 * As a final step, dump "unpurged" areas. Note,
-	 * that entire "/proc/vmallocinfo" output will not
-	 * be address sorted, because the purge list is not
-	 * sorted.
+	 * As a final step, dump "unpurged" areas.
 	 */
 	if (list_is_last(&va->list, &vmap_area_list))
 		show_purge_info(m);
_


  parent reply	other threads:[~2020-12-15  3:08 UTC|newest]

Thread overview: 215+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-15  3:02 incoming Andrew Morton
2020-12-15  3:03 ` [patch 001/200] kthread: add kthread_work tracepoints Andrew Morton
2020-12-15  3:03 ` [patch 002/200] kthread_worker: document CPU hotplug handling Andrew Morton
2020-12-15  3:03 ` [patch 003/200] uapi: move constants from <linux/kernel.h> to <linux/const.h> Andrew Morton
2020-12-15  3:03 ` [patch 004/200] ide/falcon: remove in_interrupt() usage Andrew Morton
2020-12-15  3:03 ` [patch 005/200] ide: remove BUG_ON(in_interrupt() || irqs_disabled()) from ide_unregister() Andrew Morton
2020-12-15  3:03 ` [patch 006/200] fs/ntfs: remove unused varibles Andrew Morton
2020-12-15  3:03 ` [patch 007/200] fs/ntfs: remove unused variable attr_len Andrew Morton
2020-12-15  3:03 ` [patch 008/200] fs/ocfs2/cluster/tcp.c: remove unneeded break Andrew Morton
2020-12-15  3:03 ` [patch 009/200] ocfs2: ratelimit the 'max lookup times reached' notice Andrew Morton
2020-12-15  3:03 ` [patch 010/200] arch/Kconfig: fix spelling mistakes Andrew Morton
2020-12-15  3:03 ` [patch 011/200] mm/slab_common.c: use list_for_each_entry in dump_unreclaimable_slab() Andrew Morton
2020-12-15  3:03 ` [patch 012/200] mm: slab: clarify krealloc()'s behavior with __GFP_ZERO Andrew Morton
2020-12-15 14:30   ` Christian König
2020-12-15 19:08     ` Andy Shevchenko
2020-12-16  7:47       ` Christian König
2020-12-16 11:23         ` Andy Shevchenko
2020-12-15  3:03 ` [patch 013/200] mm: slab: provide krealloc_array() Andrew Morton
2020-12-15  3:03 ` [patch 014/200] ALSA: pcm: use krealloc_array() Andrew Morton
2020-12-15  3:04 ` [patch 015/200] vhost: vringh: " Andrew Morton
2020-12-15  3:04 ` [patch 016/200] pinctrl: " Andrew Morton
2020-12-15  3:04 ` [patch 017/200] edac: ghes: " Andrew Morton
2020-12-15  3:04 ` [patch 018/200] drm: atomic: " Andrew Morton
2020-12-15  3:04 ` [patch 019/200] hwtracing: intel: " Andrew Morton
2020-12-15  3:04 ` [patch 020/200] dma-buf: " Andrew Morton
2020-12-15  7:42   ` Christian König
2020-12-15  3:04 ` [patch 021/200] mm, slab, slub: clear the slab_cache field when freeing page Andrew Morton
2020-12-15  3:04 ` [patch 022/200] mm/slab: rerform init_on_free earlier Andrew Morton
2020-12-15 16:45   ` Alexander Potapenko
2020-12-15 20:46     ` Alexander Popov
2020-12-15 21:08       ` Alexander Popov
2020-12-15  3:04 ` [patch 023/200] mm, slub: use kmem_cache_debug_flags() in deactivate_slab() Andrew Morton
2020-12-15  3:04 ` [patch 024/200] mm/slub: let number of online CPUs determine the slub page order Andrew Morton
2020-12-15  3:04 ` [patch 025/200] device-dax/kmem: use struct_size() Andrew Morton
2020-12-15  3:04 ` [patch 026/200] mm: fix page_owner initializing issue for arm32 Andrew Morton
2020-12-15  3:04 ` [patch 027/200] mm/page_owner: record timestamp and pid Andrew Morton
2020-12-15  3:04 ` [patch 028/200] mm/filemap/c: break generic_file_buffered_read up into multiple functions Andrew Morton
2020-12-15  3:04 ` [patch 029/200] mm/filemap.c: generic_file_buffered_read() now uses find_get_pages_contig Andrew Morton
2020-12-15  3:04 ` [patch 030/200] mm/truncate: add parameter explanation for invalidate_mapping_pagevec Andrew Morton
2020-12-15  3:05 ` [patch 031/200] mm/filemap.c: remove else after a return Andrew Morton
2020-12-15  3:05 ` [patch 032/200] mm/gup_benchmark: rename to mm/gup_test Andrew Morton
2020-12-15  3:05 ` [patch 033/200] selftests/vm: use a common gup_test.h Andrew Morton
2020-12-15  3:05 ` [patch 034/200] selftests/vm: rename run_vmtests --> run_vmtests.sh Andrew Morton
2020-12-15  3:05 ` [patch 035/200] selftests/vm: minor cleanup: Makefile and gup_test.c Andrew Morton
2020-12-15  3:05 ` [patch 036/200] selftests/vm: only some gup_test items are really benchmarks Andrew Morton
2020-12-15  3:05 ` [patch 037/200] selftests/vm: gup_test: introduce the dump_pages() sub-test Andrew Morton
2020-12-15  3:05 ` [patch 038/200] selftests/vm: run_vmtests.sh: update and clean up gup_test invocation Andrew Morton
2020-12-15  3:05 ` [patch 039/200] selftests/vm: hmm-tests: remove the libhugetlbfs dependency Andrew Morton
2020-12-15  3:05 ` [patch 040/200] selftests/vm: 2x speedup for run_vmtests.sh Andrew Morton
2020-12-15  3:05 ` [patch 041/200] mm/gup_test.c: mark gup_test_init as __init function Andrew Morton
2020-12-15  3:05 ` [patch 042/200] mm/gup_test: GUP_TEST depends on DEBUG_FS Andrew Morton
2020-12-15  3:05 ` [patch 043/200] mm/gup: reorganize internal_get_user_pages_fast() Andrew Morton
2020-12-15  3:05 ` [patch 044/200] mm/gup: prevent gup_fast from racing with COW during fork Andrew Morton
2020-12-15  3:05 ` [patch 045/200] mm/gup: remove the vma allocation from gup_longterm_locked() Andrew Morton
2020-12-15  3:05 ` [patch 046/200] mm/gup: combine put_compound_head() and unpin_user_page() Andrew Morton
2020-12-15  3:05 ` [patch 047/200] mm: handle zone device pages in release_pages() Andrew Morton
2020-12-15  3:05 ` [patch 048/200] mm/swapfile.c: use helper function swap_count() in add_swap_count_continuation() Andrew Morton
2020-12-15  3:06 ` [patch 049/200] mm/swap_state: skip meaningless swap cache readahead when ra_info.win == 0 Andrew Morton
2020-12-15  3:06 ` [patch 050/200] mm/swapfile.c: remove unnecessary out label in __swap_duplicate() Andrew Morton
2020-12-15  3:06 ` [patch 051/200] mm/swapfile.c: use memset to fill the swap_map with SWAP_HAS_CACHE Andrew Morton
2020-12-15  3:06 ` [patch 052/200] mm: remove pagevec_lookup_range_nr_tag() Andrew Morton
2020-12-15  3:06 ` [patch 053/200] mm/shmem.c: make shmem_mapping() inline Andrew Morton
2020-12-15  3:06 ` [patch 054/200] tmpfs: fix Documentation nits Andrew Morton
2020-12-15  3:06 ` [patch 055/200] mm: memcontrol: add file_thp, shmem_thp to memory.stat Andrew Morton
2020-12-15  3:06 ` [patch 056/200] mm: memcontrol: remove unused mod_memcg_obj_state() Andrew Morton
2020-12-15  3:06 ` [patch 057/200] mm: memcontrol: eliminate redundant check in __mem_cgroup_insert_exceeded() Andrew Morton
2020-12-15  3:06 ` [patch 058/200] mm: memcg/slab: fix return of child memcg objcg for root memcg Andrew Morton
2020-12-15  3:06 ` [patch 059/200] mm: memcg/slab: fix use after free in obj_cgroup_charge Andrew Morton
2020-12-15  3:06 ` [patch 060/200] mm/rmap: always do TTU_IGNORE_ACCESS Andrew Morton
2020-12-15  3:06 ` [patch 061/200] mm/memcg: update page struct member in comments Andrew Morton
2020-12-15  3:06 ` [patch 062/200] mm: memcg: fix obsolete code comments Andrew Morton
2020-12-15  3:06 ` [patch 063/200] mm: memcg: deprecate the non-hierarchical mode Andrew Morton
2020-12-15  3:06 ` [patch 064/200] docs: cgroup-v1: reflect the deprecation of " Andrew Morton
2020-12-15  3:06 ` [patch 065/200] cgroup: remove obsoleted broken_hierarchy and warned_broken_hierarchy Andrew Morton
2020-12-15  3:06 ` [patch 066/200] mm/page_counter: use page_counter_read in page_counter_set_max Andrew Morton
2020-12-15  3:07 ` [patch 067/200] mm: memcg: remove obsolete memcg_has_children() Andrew Morton
2020-12-15  3:07 ` [patch 068/200] mm: memcg/slab: rename *_lruvec_slab_state to *_lruvec_kmem_state Andrew Morton
2020-12-15  3:07 ` [patch 069/200] mm: memcontrol: sssign boolean values to a bool variable Andrew Morton
2020-12-15  3:07 ` [patch 070/200] mm/memcg: remove incorrect comment Andrew Morton
2020-12-15  3:07 ` [patch 071/200] mm: move lruvec stats update functions to vmstat.h Andrew Morton
2020-12-15  3:07 ` [patch 072/200] mm: memcontrol: account pagetables per node Andrew Morton
2020-12-15  3:07 ` [patch 073/200] xen/unpopulated-alloc: consolidate pgmap manipulation Andrew Morton
2020-12-15  3:07 ` [patch 074/200] kselftests: vm: add mremap tests Andrew Morton
2020-12-15  3:07 ` [patch 075/200] mm: speedup mremap on 1GB or larger regions Andrew Morton
2020-12-15 19:52   ` Linus Torvalds
2020-12-15 23:16     ` Kalesh Singh
2020-12-15  3:07 ` [patch 076/200] arm64: mremap speedup - enable HAVE_MOVE_PUD Andrew Morton
2020-12-15  3:07 ` [patch 077/200] x86: mremap speedup - Enable HAVE_MOVE_PUD Andrew Morton
2020-12-15  3:07 ` [patch 078/200] mm: cleanup: remove unused tsk arg from __access_remote_vm Andrew Morton
2020-12-15  3:07 ` [patch 079/200] mm/mapping_dirty_helpers: enhance the kernel-doc markups Andrew Morton
2020-12-15  3:07 ` [patch 080/200] mm/page_vma_mapped.c: add colon to fix kernel-doc markups error for check_pte Andrew Morton
2020-12-15  3:07 ` [patch 081/200] mm: mmap_lock: add tracepoints around lock acquisition Andrew Morton
2020-12-15  3:07 ` [patch 082/200] sparc: fix handling of page table constructor failure Andrew Morton
2020-12-15  3:08 ` [patch 083/200] mm: move free_unref_page to mm/internal.h Andrew Morton
2020-12-15  3:08 ` [patch 084/200] mm/mremap: account memory on do_munmap() failure Andrew Morton
2020-12-15  3:08 ` [patch 085/200] mm/mremap: for MREMAP_DONTUNMAP check security_vm_enough_memory_mm() Andrew Morton
2020-12-15  3:08 ` [patch 086/200] mremap: don't allow MREMAP_DONTUNMAP on special_mappings and aio Andrew Morton
2020-12-28 17:59   ` Brian Geffon
2020-12-15  3:08 ` [patch 087/200] vm_ops: rename .split() callback to .may_split() Andrew Morton
2020-12-15  3:08 ` [patch 088/200] mremap: check if it's possible to split original vma Andrew Morton
2020-12-15  3:08 ` [patch 089/200] mm: forbid splitting special mappings Andrew Morton
2020-12-15  3:08 ` [patch 090/200] mm: track mmu notifiers in fs_reclaim_acquire/release Andrew Morton
2020-12-15  3:08 ` [patch 091/200] mm: extract might_alloc() debug check Andrew Morton
2020-12-15  3:08 ` [patch 092/200] locking/selftests: add testcases for fs_reclaim Andrew Morton
2020-12-15  3:08 ` [patch 093/200] mm/vmalloc.c:__vmalloc_area_node(): avoid 32-bit overflow Andrew Morton
2020-12-15  3:08 ` [patch 094/200] mm/vmalloc: use free_vm_area() if an allocation fails Andrew Morton
2020-12-15  3:08 ` Andrew Morton [this message]
2020-12-15  3:08 ` [patch 096/200] mm/vmalloc: add 'align' parameter explanation for pvm_determine_end_from_reverse Andrew Morton
2020-12-15  3:08 ` [patch 097/200] mm/vmalloc.c: remove unnecessary return statement Andrew Morton
2020-12-15  3:08 ` [patch 098/200] mm/vmalloc: Fix unlock order in s_stop() Andrew Morton
2020-12-15  3:09 ` [patch 099/200] docs/vm: remove unused 3 items explanation for /proc/vmstat Andrew Morton
2020-12-15  3:09 ` [patch 100/200] mm/vmalloc.c: fix kasan shadow poisoning size Andrew Morton
2020-12-15  3:09 ` [patch 101/200] workqueue: kasan: record workqueue stack Andrew Morton
2020-12-15  3:09 ` [patch 102/200] kasan: print " Andrew Morton
2020-12-15  3:09 ` [patch 103/200] lib/test_kasan.c: add workqueue test case Andrew Morton
2020-12-15  3:09 ` [patch 104/200] kasan: update documentation for generic kasan Andrew Morton
2020-12-15  3:09 ` [patch 105/200] lkdtm: disable KASAN for rodata.o Andrew Morton
2020-12-15  3:09 ` [patch 106/200] alpha: switch from DISCONTIGMEM to SPARSEMEM Andrew Morton
2020-12-15  3:09 ` [patch 107/200] ia64: remove custom __early_pfn_to_nid() Andrew Morton
2020-12-15  3:09 ` [patch 108/200] ia64: remove 'ifdef CONFIG_ZONE_DMA32' statements Andrew Morton
2020-12-15  3:09 ` [patch 109/200] ia64: discontig: paging_init(): remove local max_pfn calculation Andrew Morton
2020-12-15  3:09 ` [patch 110/200] ia64: split virtual map initialization out of paging_init() Andrew Morton
2020-12-15  3:09 ` [patch 111/200] ia64: forbid using VIRTUAL_MEM_MAP with FLATMEM Andrew Morton
2020-12-15  3:09 ` [patch 112/200] ia64: make SPARSEMEM default and disable DISCONTIGMEM Andrew Morton
2020-12-15  3:09 ` [patch 113/200] arm: remove CONFIG_ARCH_HAS_HOLES_MEMORYMODEL Andrew Morton
2020-12-15  3:09 ` [patch 114/200] arm, arm64: move free_unused_memmap() to generic mm Andrew Morton
2020-12-15  3:10 ` [patch 115/200] arc: use FLATMEM with freeing of unused memory map instead of DISCONTIGMEM Andrew Morton
2020-12-15  3:10 ` [patch 116/200] m68k/mm: make node data and node setup depend on CONFIG_DISCONTIGMEM Andrew Morton
2020-12-15  3:10 ` [patch 117/200] m68k/mm: enable use of generic memory_model.h for !DISCONTIGMEM Andrew Morton
2020-12-15  3:10 ` [patch 118/200] m68k: deprecate DISCONTIGMEM Andrew Morton
2020-12-15  3:10 ` [patch 119/200] mm: introduce debug_pagealloc_{map,unmap}_pages() helpers Andrew Morton
2020-12-15  3:10 ` [patch 120/200] PM: hibernate: make direct map manipulations more explicit Andrew Morton
2020-12-15  3:10 ` [patch 121/200] arch, mm: restore dependency of __kernel_map_pages() on DEBUG_PAGEALLOC Andrew Morton
2020-12-15  3:10 ` [patch 122/200] arch, mm: make kernel_page_present() always available Andrew Morton
2020-12-15  3:10 ` [patch 123/200] mm, page_alloc: clean up pageset high and batch update Andrew Morton
2020-12-15  3:10 ` [patch 124/200] mm, page_alloc: calculate pageset high and batch once per zone Andrew Morton
2020-12-15  3:10 ` [patch 125/200] mm, page_alloc: remove setup_pageset() Andrew Morton
2020-12-15  3:10 ` [patch 126/200] mm, page_alloc: simplify pageset_update() Andrew Morton
2020-12-15  3:10 ` [patch 127/200] mm, page_alloc: cache pageset high and batch in struct zone Andrew Morton
2020-12-15  3:10 ` [patch 128/200] mm, page_alloc: move draining pcplists to page isolation users Andrew Morton
2020-12-15  3:10 ` [patch 129/200] mm, page_alloc: disable pcplists during memory offline Andrew Morton
2020-12-15  3:11 ` [patch 130/200] include/linux/page-flags.h: remove unused __[Set|Clear]PagePrivate Andrew Morton
2020-12-15  3:11 ` [patch 131/200] mm/page-flags: fix comment Andrew Morton
2020-12-15  3:11 ` [patch 132/200] mm/page_alloc: add __free_pages() documentation Andrew Morton
2020-12-15  3:11 ` [patch 133/200] mm/page_alloc: mark some symbols with static keyword Andrew Morton
2020-12-15  3:11 ` [patch 134/200] mm/page_alloc: clear all pages in post_alloc_hook() with init_on_alloc=1 Andrew Morton
2020-12-15  3:11 ` [patch 135/200] init/main: fix broken buffer_init when DEFERRED_STRUCT_PAGE_INIT set Andrew Morton
2020-12-15  3:11 ` [patch 136/200] mm: page_alloc: refactor setup_per_zone_lowmem_reserve() Andrew Morton
2020-12-15  3:11 ` [patch 137/200] mm/page_alloc: speed up the iteration of max_order Andrew Morton
2020-12-15  3:11 ` [patch 138/200] mm,hwpoison: drain pcplists before bailing out for non-buddy zero-refcount page Andrew Morton
2020-12-15  3:11 ` [patch 139/200] mm,hwpoison: take free pages off the buddy freelists Andrew Morton
2020-12-15  3:11 ` [patch 140/200] mm,hwpoison: drop unneeded pcplist draining Andrew Morton
2020-12-15  3:11 ` [patch 141/200] mm,hwpoison: refactor get_any_page Andrew Morton
2020-12-15  3:11 ` [patch 142/200] mm,hwpoison: disable pcplists before grabbing a refcount Andrew Morton
2020-12-15  3:11 ` [patch 143/200] mm,hwpoison: remove drain_all_pages from shake_page Andrew Morton
2020-12-15  3:11 ` [patch 144/200] mm,memory_failure: always pin the page in madvise_inject_error Andrew Morton
2020-12-15  3:11 ` [patch 145/200] mm,hwpoison: return -EBUSY when migration fails Andrew Morton
2020-12-15  3:11 ` [patch 146/200] mm/hugetlb.c: just use put_page_testzero() instead of page_count() Andrew Morton
2020-12-15  3:11 ` [patch 147/200] include/linux/huge_mm.h: remove extern keyword Andrew Morton
2020-12-15  3:12 ` [patch 148/200] khugepaged: add parameter explanations for kernel-doc markup Andrew Morton
2020-12-15  3:12 ` [patch 149/200] mm: hugetlb: fix type of delta parameter and related local variables in gather_surplus_pages() Andrew Morton
2020-12-15  3:12 ` [patch 150/200] mm,hugetlb: remove unneeded initialization Andrew Morton
2020-12-15  3:12 ` [patch 151/200] hugetlb: fix an error code in hugetlb_reserve_pages() Andrew Morton
2020-12-15  3:12 ` [patch 152/200] mm: don't wake kswapd prematurely when watermark boosting is disabled Andrew Morton
2020-12-15  3:12 ` [patch 153/200] mm/vmscan: drop unneeded assignment in kswapd() Andrew Morton
2020-12-15  3:12 ` [patch 154/200] mm/vmscan.c: remove the filename in the top of file comment Andrew Morton
2020-12-15  3:12 ` [patch 155/200] mm/page_isolation: do not isolate the max order page Andrew Morton
2020-12-15  3:12 ` [patch 156/200] z3fold: simplify freeing slots Andrew Morton
2020-12-15  3:12 ` [patch 157/200] z3fold: stricter locking and more careful reclaim Andrew Morton
2020-12-15  3:12 ` [patch 158/200] z3fold: remove preempt disabled sections for RT Andrew Morton
2020-12-15  3:12 ` [patch 159/200] mm/compaction: rename 'start_pfn' to 'iteration_start_pfn' in compact_zone() Andrew Morton
2020-12-15  3:12 ` [patch 160/200] mm/compaction: move compaction_suitable's comment to right place Andrew Morton
2020-12-15  3:12 ` [patch 161/200] mm/compaction: make defer_compaction and compaction_deferred static Andrew Morton
2020-12-15  3:12 ` [patch 162/200] mm/oom_kill: change comment and rename is_dump_unreclaim_slabs() Andrew Morton
2020-12-15  3:12 ` [patch 163/200] mm/migrate.c: fix comment spelling Andrew Morton
2020-12-15  3:12 ` [patch 164/200] mm/migrate.c: optimize migrate_vma_pages() mmu notifier Andrew Morton
2020-12-15  3:12 ` [patch 165/200] mm: support THPs in zero_user_segments Andrew Morton
2020-12-15  3:13 ` [patch 166/200] mm: truncate_complete_page() does not exist any more Andrew Morton
2020-12-15  3:13 ` [patch 167/200] mm: migrate: simplify the logic for handling permanent failure Andrew Morton
2020-12-15  3:13 ` [patch 168/200] mm: migrate: skip shared exec THP for NUMA balancing Andrew Morton
2020-12-15  3:13 ` [patch 169/200] mm: migrate: clean up migrate_prep{_local} Andrew Morton
2020-12-15  3:13 ` [patch 170/200] mm: migrate: return -ENOSYS if THP migration is unsupported Andrew Morton
2020-12-15  3:13 ` [patch 171/200] mm: migrate: remove unused parameter in migrate_vma_insert_page() Andrew Morton
2020-12-15  3:13 ` [patch 172/200] mm/cma.c: remove redundant cma_mutex lock Andrew Morton
2020-12-15  3:13 ` [patch 173/200] mm: cma: improve pr_debug log in cma_release() Andrew Morton
2020-12-15  3:13 ` [patch 174/200] mm, page_alloc: do not rely on the order of page_poison and init_on_alloc/free parameters Andrew Morton
2020-12-15  3:13 ` [patch 175/200] mm, page_poison: use static key more efficiently Andrew Morton
2020-12-15  3:13 ` [patch 176/200] kernel/power: allow hibernation with page_poison sanity checking Andrew Morton
2020-12-15  3:13 ` [patch 177/200] mm, page_poison: remove CONFIG_PAGE_POISONING_NO_SANITY Andrew Morton
2020-12-15  3:13 ` [patch 178/200] mm, page_poison: remove CONFIG_PAGE_POISONING_ZERO Andrew Morton
2020-12-15  3:13 ` [patch 179/200] userfaultfd: add UFFD_USER_MODE_ONLY Andrew Morton
2020-12-15  3:13 ` [patch 180/200] userfaultfd: add user-mode only option to unprivileged_userfaultfd sysctl knob Andrew Morton
2020-12-15  3:13 ` [patch 181/200] userfaultfd: selftests: make __{s,u}64 format specifiers portable Andrew Morton
2020-12-15  3:14 ` [patch 182/200] userfaultfd/selftests: always dump something in modes Andrew Morton
2020-12-15  3:14 ` [patch 183/200] userfaultfd/selftests: fix retval check for userfaultfd_open() Andrew Morton
2020-12-15  3:14 ` [patch 184/200] userfaultfd/selftests: hint the test runner on required privilege Andrew Morton
2020-12-15  3:14 ` [patch 185/200] mm/zswap: make struct kernel_param_ops definitions const Andrew Morton
2020-12-15  3:14 ` [patch 186/200] mm/zswap: fix passing zero to 'PTR_ERR' warning Andrew Morton
2020-12-15  3:14 ` [patch 187/200] mm/zswap: move to use crypto_acomp API for hardware acceleration Andrew Morton
2020-12-15  3:14 ` [patch 188/200] mm/zsmalloc.c: rework the list_add code in insert_zspage() Andrew Morton
2020-12-15  3:14 ` [patch 189/200] mm/process_vm_access: remove redundant initialization of iov_r Andrew Morton
2020-12-15  3:14 ` [patch 190/200] zram: support page writeback Andrew Morton
2020-12-15  3:14 ` [patch 191/200] zram: add stat to gather incompressible pages since zram set up Andrew Morton
2020-12-15  3:14 ` [patch 192/200] zram: break the strict dependency from lzo Andrew Morton
2020-12-15  3:14 ` [patch 193/200] mm: fix kernel-doc markups Andrew Morton
2020-12-15  3:14 ` [patch 194/200] mm: use sysfs_emit for struct kobject * uses Andrew Morton
2020-12-15  3:14 ` [patch 195/200] mm: huge_memory: convert remaining use of sprintf to sysfs_emit and neatening Andrew Morton
2020-12-15  3:14 ` [patch 196/200] mm:backing-dev: use sysfs_emit in macro defining functions Andrew Morton
2020-12-15  3:14 ` [patch 197/200] mm: shmem: convert shmem_enabled_show to use sysfs_emit_at Andrew Morton
2020-12-15  3:14 ` [patch 198/200] mm: slub: convert sysfs sprintf family to sysfs_emit/sysfs_emit_at Andrew Morton
2020-12-15  3:15 ` [patch 199/200] mm: fix fall-through warnings for Clang Andrew Morton
2020-12-15  3:15 ` [patch 200/200] mm: cleanup kstrto*() usage Andrew Morton
2020-12-15  3:25 ` incoming Linus Torvalds
2020-12-15  3:30   ` incoming Linus Torvalds
2020-12-15 14:04     ` incoming Konstantin Ryabitsev

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=20201215030849.1tGt5v8fS%akpm@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=hdanton@sina.com \
    --cc=huang.ying.caritas@gmail.com \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.com \
    --cc=minchan@kernel.org \
    --cc=mm-commits@vger.kernel.org \
    --cc=oleksiy.avramchenko@sonymobile.com \
    --cc=rostedt@goodmis.org \
    --cc=torvalds@linux-foundation.org \
    --cc=urezki@gmail.com \
    --cc=willy@infradead.org \
    --subject='Re: [patch 095/200] mm/vmalloc: rework the drain logic' \
    /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

This is a public inbox, see mirroring instructions
on how to clone and mirror all data and code used for this inbox