Skip to content

V5.10 #4

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
elitetest1 wants to merge 74 commits into from
Closed

V5.10 #4

elitetest1 wants to merge 74 commits into from

Conversation

@elitetest1
Copy link
Member

@elitetest1 elitetest1 commented Nov 27, 2025

No description provided.

justin-he and others added 30 commits November 27, 2025 14:47
@justin-he @FPSensor
BACKPORT: arm64: cpufeature: introduce helper cpu_has_hw_af()
17ee1ab 
We unconditionally set the HW_AFDBM capability and only enable it on
CPUs which really have the feature. But sometimes we need to know
whether this cpu has the capability of HW AF. So decouple AF from
DBM by a new helper cpu_has_hw_af().

If later we noticed a potential performance issue on this path, we can
turn it into a static label as with other CPU features.

Link: https://lore.kernel.org/r/20191011140939.6115-2-justin.he@arm.com/
Change-Id: Ifed339bfe6c86ada6e45eddefa6aa010a7fe465d
Signed-off-by: Jia He <justin.he@arm.com>
Suggested-by: Suzuki Poulose <Suzuki.Poulose@arm.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
(cherry picked from commit 47d7b15b88f96a90694cfc607d0717d62dff6c45)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@justin-he @FPSensor
UPSTREAM: arm64: mm: implement arch_faults_on_old_pte() on arm64
e876bc0 
On arm64 without hardware Access Flag, copying from user will fail because
the pte is old and cannot be marked young. So we always end up with zeroed
page after fork() + CoW for pfn mappings. We don't always have a
hardware-managed Access Flag on arm64.

Hence implement arch_faults_on_old_pte on arm64 to indicate that it might
cause page fault when accessing old pte.

Link: https://lore.kernel.org/r/20191011140939.6115-3-justin.he@arm.com/
Change-Id: Ica230a882941cb17a1acdd6c6c9f3ec1b3f494b3
Signed-off-by: Jia He <justin.he@arm.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
(cherry picked from commit 6af31226d0394691f5562eca0134262bb935fa9c)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@justin-he @FPSensor
UPSTREAM: x86/mm: implement arch_faults_on_old_pte() stub on x86
176e2dd 
arch_faults_on_old_pte is a helper to indicate that it might cause page
fault when accessing old pte. But on x86, there is feature to setting
pte access flag by hardware. Hence implement an overriding stub which
always returns false.

Link: https://lore.kernel.org/r/20191011140939.6115-4-justin.he@arm.com/
Change-Id: I69cd5568e648cdf098269958de9e8bcedd603df4
Signed-off-by: Jia He <justin.he@arm.com>
Suggested-by: Will Deacon <will@kernel.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
(cherry picked from commit f2c4e5970cece75a895fcc45f0cd66b5a5ec0819)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
Signed-off-by: Vaisakh Murali <mvaisakh@statixos.com>
@FPSensor
UPSTREAM: mm: add generic p?d_leaf() macros
29b2bf5 
Patch series "Generic page walk and ptdump", v17.

Many architectures current have a debugfs file for dumping the kernel page
tables.  Currently each architecture has to implement custom functions for
this because the details of walking the page tables used by the kernel are
different between architectures.

This series extends the capabilities of walk_page_range() so that it can
deal with the page tables of the kernel (which have no VMAs and can
contain larger huge pages than exist for user space).  A generic PTDUMP
implementation is the implemented making use of the new functionality of
walk_page_range() and finally arm64 and x86 are switch to using it,
removing the custom table walkers.

To enable a generic page table walker to walk the unusual mappings of the
kernel we need to implement a set of functions which let us know when the
walker has reached the leaf entry.  After a suggestion from Will Deacon
I've chosen the name p?d_leaf() as this (hopefully) describes the purpose
(and is a new name so has no historic baggage).  Some architectures have
p?d_large macros but this is easily confused with "large pages".

This series ends with a generic PTDUMP implemention for arm64 and x86.

Mostly this is a clean up and there should be very little functional
change.  The exceptions are:

* arm64 PTDUMP debugfs now displays pages which aren't present (patch 22).

* arm64 has the ability to efficiently process KASAN pages (which
  previously only x86 implemented).  This means that the combination of
  KASAN and DEBUG_WX is now useable.

This patch (of 23):

Exposing the pud/pgd levels of the page tables to walk_page_range() means
we may come across the exotic large mappings that come with large areas of
contiguous memory (such as the kernel's linear map).

For architectures that don't provide all p?d_leaf() macros, provide
generic do nothing default that are suitable where there cannot be leaf
pages at that level.  Futher patches will add implementations for
individual architectures.

The name p?d_leaf() is chosen to minimize the confusion with existing uses
of "large" pages and "huge" pages which do not necessary mean that the
entry is a leaf (for example it may be a set of contiguous entries that
only take 1 TLB slot).  For the purpose of walking the page tables we
don't need to know how it will be represented in the TLB, but we do need
to know for sure if it is a leaf of the tree.

Link: http://lkml.kernel.org/r/20191218162402.45610-2-steven.price@arm.com
Change-Id: Ibc89e157616134076969b8daa9e9ce104d6c0900
Signed-off-by: Steven Price <steven.price@arm.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Morse <james.morse@arm.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will@kernel.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: "Liang, Kan" <kan.liang@linux.intel.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Alexandre Ghiti <alex@ghiti.fr>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: James Hogan <jhogan@kernel.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Burton <paul.burton@mips.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Zong Li <zong.li@sifive.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 93fab1b22ef7c4abcbc760ce4432762b02e7f3d1)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
UPSTREAM: arm64: mm: add p?d_leaf() definitions
3973df4 
walk_page_range() is going to be allowed to walk page tables other than
those of user space.  For this it needs to know when it has reached a
'leaf' entry in the page tables.  This information will be provided by the
p?d_leaf() functions/macros.

For arm64, we already have p?d_sect() macros which we can reuse for
p?d_leaf().

pud_sect() is defined as a dummy function when CONFIG_PGTABLE_LEVELS < 3
or CONFIG_ARM64_64K_PAGES is defined.  However when the kernel is
configured this way then architecturally it isn't allowed to have a large
page at this level, and any code using these page walking macros is
implicitly relying on the page size/number of levels being the same as the
kernel.  So it is safe to reuse this for p?d_leaf() as it is an
architectural restriction.

Link: http://lkml.kernel.org/r/20191218162402.45610-5-steven.price@arm.com
Change-Id: Ic031474949c96e9d3d191dd93f232cc20603032d
Signed-off-by: Steven Price <steven.price@arm.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Alexandre Ghiti <alex@ghiti.fr>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Hogan <jhogan@kernel.org>
Cc: James Morse <james.morse@arm.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: "Liang, Kan" <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Burton <paul.burton@mips.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Zong Li <zong.li@sifive.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 8aa82df3c123129025a364d8f823929cc488b834)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: arm64: mm: fix p?d_leaf()
8df1424 
The pmd_leaf() is used to test a leaf mapped PMD, however, it misses
the PROT_NONE mapped PMD on arm64.  Fix it.  A real world issue [1]
caused by this was reported by Qian Cai. Also fix pud_leaf().

Link: https://patchwork.kernel.org/comment/24798260/ [1]
Fixes: 8aa82df3c123 ("arm64: mm: add p?d_leaf() definitions")
Reported-by: Qian Cai <quic_qiancai@quicinc.com>
Change-Id: I3b632d653c4e00a8825e1cc626d935ca5cbac194
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Link: https://lore.kernel.org/r/20220422060033.48711-1-songmuchun@bytedance.com
Signed-off-by: Will Deacon <will@kernel.org>
(cherry picked from commit 23bc8f69f0eceecbb87c3801d2e48827d2dca92b)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: pagewalk: add p4d_entry() and pgd_entry()
576ff1f 
pgd_entry() and pud_entry() were removed by commit 0b1fbfe
("mm/pagewalk: remove pgd_entry() and pud_entry()") because there were no
users.  We're about to add users so reintroduce them, along with
p4d_entry() as we now have 5 levels of tables.

Note that commit a00cc7d ("mm, x86: add support for PUD-sized
transparent hugepages") already re-added pud_entry() but with different
semantics to the other callbacks.  This commit reverts the semantics back
to match the other callbacks.

To support hmm.c which now uses the new semantics of pud_entry() a new
member ('action') of struct mm_walk is added which allows the callbacks to
either descend (ACTION_SUBTREE, the default), skip (ACTION_CONTINUE) or
repeat the callback (ACTION_AGAIN).  hmm.c is then updated to call
pud_trans_huge_lock() itself and make use of the splitting/retry logic of
the core code.

After this change pud_entry() is called for all entries, not just
transparent huge pages.

[arnd@arndb.de: fix unused variable warning]
 Link: http://lkml.kernel.org/r/20200107204607.1533842-1-arnd@arndb.de
Link: http://lkml.kernel.org/r/20191218162402.45610-12-steven.price@arm.com
Change-Id: Id8578aab7de0f670db45d32cb3752993b81464be
Signed-off-by: Steven Price <steven.price@arm.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Alexandre Ghiti <alex@ghiti.fr>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Hogan <jhogan@kernel.org>
Cc: James Morse <james.morse@arm.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: "Liang, Kan" <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Burton <paul.burton@mips.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zong Li <zong.li@sifive.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 3afc423632a194d7d6afef34e4bb98f804cd071d)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@jgunthorpe @FPSensor
BACKPORT: mm/hmm: do not call hmm_vma_walk_hole() while holding a spi…
c484e65 
…nlock

This eventually calls into handle_mm_fault() which is a sleeping function.
Release the lock first.

hmm_vma_walk_hole() does not touch the contents of the PUD, so it does not
need the lock.

Link: https://lore.kernel.org/r/20200311183506.3997-4-jgg@ziepe.ca/
Fixes: 3afc423632a1 ("mm: pagewalk: add p4d_entry() and pgd_entry()")
Cc: Steven Price <steven.price@arm.com>
Reviewed-by: Ralph Campbell <rcampbell@nvidia.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Change-Id: I4dd851a2fcf242877ae7acd082f3521e7202d258
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
(cherry picked from commit 05fc1df95e5dc09802813bab9c1e718f1e419d93)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@RichardWeiYang @FPSensor
UPSTREAM: mm/vmscan.c: use update_lru_size() in update_lru_sizes()
24963d8 
We already defined the helper update_lru_size().

Let's use this to reduce code duplication.

Change-Id: Icae4274c4f4c9bdfd70355bb32eb2fe1800a79c1
Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Baoquan He <bhe@redhat.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Link: http://lkml.kernel.org/r/20200331221550.1011-1-richard.weiyang@gmail.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit a892cb6b977ffe209683809e5e9d627656d20aa8)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@JoonsooKim @FPSensor
BACKPORT: mm/swapcache: support to handle the shadow entries
adb2d95 
Workingset detection for anonymous page will be implemented in the
following patch and it requires to store the shadow entries into the
swapcache.  This patch implements an infrastructure to store the shadow
entry in the swapcache.

Change-Id: Id9c2ac3af7ae73f95ba06cdda5e28aa678954d08
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Link: http://lkml.kernel.org/r/1595490560-15117-5-git-send-email-iamjoonsoo.kim@lge.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 3852f6768ede542ed48b9077bedf482c7ecb6327)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: remove superfluous __ClearPageActive()
952b1d9 
To activate a page, mark_page_accessed() always holds a reference on it.
It either gets a new reference when adding a page to
lru_pvecs.activate_page or reuses an existing one it previously got when
it added a page to lru_pvecs.lru_add.  So it doesn't call SetPageActive()
on a page that doesn't have any reference left.  Therefore, the race is
impossible these days (I didn't brother to dig into its history).

For other paths, namely reclaim and migration, a reference count is always
held while calling SetPageActive() on a page.

SetPageSlabPfmemalloc() also uses SetPageActive(), but it's irrelevant to
LRU pages.

Change-Id: If818ad43f98960f6b397dae7c2dd555bd4469854
Signed-off-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Qian Cai <cai@lca.pw>
Link: http://lkml.kernel.org/r/20200818184704.3625199-2-yuzhao@google.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 6f4dd8de4835563de9bae797ce1d7a13465a7a7d)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: use self-explanatory macros rather than "2"
c1e5728 
Change-Id: I4bb60eb4191c2a378c3d7c5a8c8170eb72d598ae
Signed-off-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Link: http://lkml.kernel.org/r/20200831175042.3527153-2-yuzhao@google.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit ed0173733dd468883198c3136284394320b8fad6)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm/vmscan.c: use add_page_to_lru_list()
c8d1642 
Patch series "mm: lru related cleanups", v2.

The cleanups are intended to reduce the verbosity in lru list operations
and make them less error-prone.  A typical example would be how the
patches change __activate_page():

 static void __activate_page(struct page *page, struct lruvec *lruvec)
 {
 	if (!PageActive(page) && !PageUnevictable(page)) {
-		int lru = page_lru_base_type(page);
 		int nr_pages = thp_nr_pages(page);

-		del_page_from_lru_list(page, lruvec, lru);
+		del_page_from_lru_list(page, lruvec);
 		SetPageActive(page);
-		lru += LRU_ACTIVE;
-		add_page_to_lru_list(page, lruvec, lru);
+		add_page_to_lru_list(page, lruvec);
 		trace_mm_lru_activate(page);

There are a few more places like __activate_page() and they are
unnecessarily repetitive in terms of figuring out which list a page should
be added onto or deleted from.  And with the duplicated code removed, they
are easier to read, IMO.

Patch 1 to 5 basically cover the above.  Patch 6 and 7 make code more
robust by improving bug reporting.  Patch 8, 9 and 10 take care of some
dangling helpers left in header files.

This patch (of 10):

There is add_page_to_lru_list(), and move_pages_to_lru() should reuse it,
not duplicate it.

Link: https://lkml.kernel.org/r/20210122220600.906146-1-yuzhao@google.com
Link: https://lore.kernel.org/linux-mm/20201207220949.830352-2-yuzhao@google.com/
Link: https://lkml.kernel.org/r/20210122220600.906146-2-yuzhao@google.com
Change-Id: Ic8de94d421c2477a077249a7c337e3fafdd41ff3
Signed-off-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Alex Shi <alex.shi@linux.alibaba.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Roman Gushchin <guro@fb.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 42895ea73bcd37c4a79e4c9f681ab8b82243c7f7)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: include/linux/mm_inline.h: shuffle lru list addition and de…
bd3b314 
…letion functions

These functions will call page_lru() in the following patches.  Move them
below page_lru() to avoid the forward declaration.

Link: https://lore.kernel.org/linux-mm/20201207220949.830352-3-yuzhao@google.com/
Link: https://lkml.kernel.org/r/20210122220600.906146-3-yuzhao@google.com
Change-Id: I3f0d2411e98a92acd82e31d31139ac5ed6d7c1c6
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit f90d8191ac864df33b1898bc7edc54eaa24e22bc)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: don't pass "enum lru_list" to lru list addition functions
615ec35 
The "enum lru_list" parameter to add_page_to_lru_list() and
add_page_to_lru_list_tail() is redundant in the sense that it can
be extracted from the "struct page" parameter by page_lru().

A caveat is that we need to make sure PageActive() or
PageUnevictable() is correctly set or cleared before calling
these two functions. And they are indeed.

Link: https://lore.kernel.org/linux-mm/20201207220949.830352-4-yuzhao@google.com/
Link: https://lkml.kernel.org/r/20210122220600.906146-4-yuzhao@google.com
Change-Id: If050c0d8d802e9290c37aabeb693b00a714cfb70
Signed-off-by: Yu Zhao <yuzhao@google.com>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 3a9c9788a3149d9745b7eb2eae811e57ef3b127c)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm/swap.c: don't pass "enum lru_list" to trace_mm_lru_inser…
99c6936 
…tion()

The parameter is redundant in the sense that it can be extracted
from the "struct page" parameter by page_lru() correctly.

Link: https://lore.kernel.org/linux-mm/20201207220949.830352-5-yuzhao@google.com/
Link: https://lkml.kernel.org/r/20210122220600.906146-5-yuzhao@google.com
Change-Id: Ib2bcf361ce6ca4575c705869e7a5a21d6530f4fe
Signed-off-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 861404536a3af3c39f1b10959a40def3d8efa2dd)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm/swap.c: don't pass "enum lru_list" to del_page_from_lru_…
c5437ab 
…list()

The parameter is redundant in the sense that it can be potentially
extracted from the "struct page" parameter by page_lru(). We need to
make sure that existing PageActive() or PageUnevictable() remains
until the function returns. A few places don't conform, and simple
reordering fixes them.

This patch may have left page_off_lru() seemingly odd, and we'll take
care of it in the next patch.

Link: https://lore.kernel.org/linux-mm/20201207220949.830352-6-yuzhao@google.com/
Link: https://lkml.kernel.org/r/20210122220600.906146-6-yuzhao@google.com
Change-Id: I0a59491e861e60d7e4856a8232e131f81273c8f9
Signed-off-by: Yu Zhao <yuzhao@google.com>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 46ae6b2cc2a47904a368d238425531ea91f3a2a5)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: add __clear_page_lru_flags() to replace page_off_lru()
0820317 
Similar to page_off_lru(), the new function does non-atomic clearing
of PageLRU() in addition to PageActive() and PageUnevictable(), on a
page that has no references left.

If PageActive() and PageUnevictable() are both set, refuse to clear
either and leave them to bad_page(). This is a behavior change that
is meant to help debug.

Link: https://lore.kernel.org/linux-mm/20201207220949.830352-7-yuzhao@google.com/
Link: https://lkml.kernel.org/r/20210122220600.906146-7-yuzhao@google.com
Change-Id: I7b25e4b8e8ae2b926678da62afb20560801d5be2
Signed-off-by: Yu Zhao <yuzhao@google.com>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 875601796267214f286d3581fe74f2805d060fe8)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: VM_BUG_ON lru page flags
af4c284 
Move scattered VM_BUG_ONs to two essential places that cover all
lru list additions and deletions.

Link: https://lore.kernel.org/linux-mm/20201207220949.830352-8-yuzhao@google.com/
Link: https://lkml.kernel.org/r/20210122220600.906146-8-yuzhao@google.com
Change-Id: I7c8682617398a8eb145607788d2ead33726d8f08
Signed-off-by: Yu Zhao <yuzhao@google.com>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit bc7112719e1e80e4208eef3fc9bd8d2b6c263e7d)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: x86, arm64: add arch_has_hw_pte_young()
76386b8 
Patch series "Multi-Gen LRU Framework", v14.

What's new
==========
1. OpenWrt, in addition to Android, Arch Linux Zen, Armbian, ChromeOS,
   Liquorix, post-factum and XanMod, is now shipping MGLRU on 5.15.
2. Fixed long-tailed direct reclaim latency seen on high-memory (TBs)
   machines. The old direct reclaim backoff, which tries to enforce a
   minimum fairness among all eligible memcgs, over-swapped by about
   (total_mem>>DEF_PRIORITY)-nr_to_reclaim. The new backoff, which
   pulls the plug on swapping once the target is met, trades some
   fairness for curtailed latency:
   https://lore.kernel.org/r/20220918080010.2920238-10-yuzhao@google.com/
3. Fixed minior build warnings and conflicts. More comments and nits.

TLDR
====
The current page reclaim is too expensive in terms of CPU usage and it
often makes poor choices about what to evict. This patchset offers an
alternative solution that is performant, versatile and
straightforward.

Patchset overview
=================
The design and implementation overview is in patch 14:
https://lore.kernel.org/r/20220918080010.2920238-15-yuzhao@google.com/

01. mm: x86, arm64: add arch_has_hw_pte_young()
02. mm: x86: add CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG
Take advantage of hardware features when trying to clear the accessed
bit in many PTEs.

03. mm/vmscan.c: refactor shrink_node()
04. Revert "include/linux/mm_inline.h: fold __update_lru_size() into
    its sole caller"
Minor refactors to improve readability for the following patches.

05. mm: multi-gen LRU: groundwork
Adds the basic data structure and the functions that insert pages to
and remove pages from the multi-gen LRU (MGLRU) lists.

06. mm: multi-gen LRU: minimal implementation
A minimal implementation without optimizations.

07. mm: multi-gen LRU: exploit locality in rmap
Exploits spatial locality to improve efficiency when using the rmap.

08. mm: multi-gen LRU: support page table walks
Further exploits spatial locality by optionally scanning page tables.

09. mm: multi-gen LRU: optimize multiple memcgs
Optimizes the overall performance for multiple memcgs running mixed
types of workloads.

10. mm: multi-gen LRU: kill switch
Adds a kill switch to enable or disable MGLRU at runtime.

11. mm: multi-gen LRU: thrashing prevention
12. mm: multi-gen LRU: debugfs interface
Provide userspace with features like thrashing prevention, working set
estimation and proactive reclaim.

13. mm: multi-gen LRU: admin guide
14. mm: multi-gen LRU: design doc
Add an admin guide and a design doc.

Benchmark results
=================
Independent lab results
-----------------------
Based on the popularity of searches [01] and the memory usage in
Google's public cloud, the most popular open-source memory-hungry
applications, in alphabetical order, are:
      Apache Cassandra      Memcached
      Apache Hadoop         MongoDB
      Apache Spark          PostgreSQL
      MariaDB (MySQL)       Redis

An independent lab evaluated MGLRU with the most widely used benchmark
suites for the above applications. They posted 960 data points along
with kernel metrics and perf profiles collected over more than 500
hours of total benchmark time. Their final reports show that, with 95%
confidence intervals (CIs), the above applications all performed
significantly better for at least part of their benchmark matrices.

On 5.14:
1. Apache Spark [02] took 95% CIs [9.28, 11.19]% and [12.20, 14.93]%
   less wall time to sort three billion random integers, respectively,
   under the medium- and the high-concurrency conditions, when
   overcommitting memory. There were no statistically significant
   changes in wall time for the rest of the benchmark matrix.
2. MariaDB [03] achieved 95% CIs [5.24, 10.71]% and [20.22, 25.97]%
   more transactions per minute (TPM), respectively, under the medium-
   and the high-concurrency conditions, when overcommitting memory.
   There were no statistically significant changes in TPM for the rest
   of the benchmark matrix.
3. Memcached [04] achieved 95% CIs [23.54, 32.25]%, [20.76, 41.61]%
   and [21.59, 30.02]% more operations per second (OPS), respectively,
   for sequential access, random access and Gaussian (distribution)
   access, when THP=always; 95% CIs [13.85, 15.97]% and
   [23.94, 29.92]% more OPS, respectively, for random access and
   Gaussian access, when THP=never. There were no statistically
   significant changes in OPS for the rest of the benchmark matrix.
4. MongoDB [05] achieved 95% CIs [2.23, 3.44]%, [6.97, 9.73]% and
   [2.16, 3.55]% more operations per second (OPS), respectively, for
   exponential (distribution) access, random access and Zipfian
   (distribution) access, when underutilizing memory; 95% CIs
   [8.83, 10.03]%, [21.12, 23.14]% and [5.53, 6.46]% more OPS,
   respectively, for exponential access, random access and Zipfian
   access, when overcommitting memory.

On 5.15:
5. Apache Cassandra [06] achieved 95% CIs [1.06, 4.10]%, [1.94, 5.43]%
   and [4.11, 7.50]% more operations per second (OPS), respectively,
   for exponential (distribution) access, random access and Zipfian
   (distribution) access, when swap was off; 95% CIs [0.50, 2.60]%,
   [6.51, 8.77]% and [3.29, 6.75]% more OPS, respectively, for
   exponential access, random access and Zipfian access, when swap was
   on.
6. Apache Hadoop [07] took 95% CIs [5.31, 9.69]% and [2.02, 7.86]%
   less average wall time to finish twelve parallel TeraSort jobs,
   respectively, under the medium- and the high-concurrency
   conditions, when swap was on. There were no statistically
   significant changes in average wall time for the rest of the
   benchmark matrix.
7. PostgreSQL [08] achieved 95% CI [1.75, 6.42]% more transactions per
   minute (TPM) under the high-concurrency condition, when swap was
   off; 95% CIs [12.82, 18.69]% and [22.70, 46.86]% more TPM,
   respectively, under the medium- and the high-concurrency
   conditions, when swap was on. There were no statistically
   significant changes in TPM for the rest of the benchmark matrix.
8. Redis [09] achieved 95% CIs [0.58, 5.94]%, [6.55, 14.58]% and
   [11.47, 19.36]% more total operations per second (OPS),
   respectively, for sequential access, random access and Gaussian
   (distribution) access, when THP=always; 95% CIs [1.27, 3.54]%,
   [10.11, 14.81]% and [8.75, 13.64]% more total OPS, respectively,
   for sequential access, random access and Gaussian access, when
   THP=never.

Our lab results
---------------
To supplement the above results, we ran the following benchmark suites
on 5.16-rc7 and found no regressions [10].
      fs_fio_bench_hdd_mq      pft
      fs_lmbench               pgsql-hammerdb
      fs_parallelio            redis
      fs_postmark              stream
      hackbench                sysbenchthread
      kernbench                tpcc_spark
      memcached                unixbench
      multichase               vm-scalability
      mutilate                 will-it-scale
      nginx

[01] https://trends.google.com
[02] https://lore.kernel.org/r/20211102002002.92051-1-bot@edi.works/
[03] https://lore.kernel.org/r/20211009054315.47073-1-bot@edi.works/
[04] https://lore.kernel.org/r/20211021194103.65648-1-bot@edi.works/
[05] https://lore.kernel.org/r/20211109021346.50266-1-bot@edi.works/
[06] https://lore.kernel.org/r/20211202062806.80365-1-bot@edi.works/
[07] https://lore.kernel.org/r/20211209072416.33606-1-bot@edi.works/
[08] https://lore.kernel.org/r/20211218071041.24077-1-bot@edi.works/
[09] https://lore.kernel.org/r/20211122053248.57311-1-bot@edi.works/
[10] https://lore.kernel.org/r/20220104202247.2903702-1-yuzhao@google.com/

Read-world applications
=======================
Third-party testimonials
------------------------
Konstantin reported [11]:
   I have Archlinux with 8G RAM + zswap + swap. While developing, I
   have lots of apps opened such as multiple LSP-servers for different
   langs, chats, two browsers, etc... Usually, my system gets quickly
   to a point of SWAP-storms, where I have to kill LSP-servers,
   restart browsers to free memory, etc, otherwise the system lags
   heavily and is barely usable.

   1.5 day ago I migrated from 5.11.15 kernel to 5.12 + the LRU
   patchset, and I started up by opening lots of apps to create memory
   pressure, and worked for a day like this. Till now I had not a
   single SWAP-storm, and mind you I got 3.4G in SWAP. I was never
   getting to the point of 3G in SWAP before without a single
   SWAP-storm.

Vaibhav from IBM reported [12]:
   In a synthetic MongoDB Benchmark, seeing an average of ~19%
   throughput improvement on POWER10(Radix MMU + 64K Page Size) with
   MGLRU patches on top of 5.16 kernel for MongoDB + YCSB across
   three different request distributions, namely, Exponential, Uniform
   and Zipfan.

Shuang from U of Rochester reported [13]:
   With the MGLRU, fio achieved 95% CIs [38.95, 40.26]%, [4.12, 6.64]%
   and [9.26, 10.36]% higher throughput, respectively, for random
   access, Zipfian (distribution) access and Gaussian (distribution)
   access, when the average number of jobs per CPU is 1; 95% CIs
   [42.32, 49.15]%, [9.44, 9.89]% and [20.99, 22.86]% higher
   throughput, respectively, for random access, Zipfian access and
   Gaussian access, when the average number of jobs per CPU is 2.

Daniel from Michigan Tech reported [14]:
   With Memcached allocating ~100GB of byte-addressable Optante,
   performance improvement in terms of throughput (measured as queries
   per second) was about 10% for a series of workloads.

Large-scale deployments
-----------------------
We've rolled out MGLRU to tens of millions of ChromeOS users and
about a million Android users. Google's fleetwide profiling [15] shows
an overall 40% decrease in kswapd CPU usage, in addition to
improvements in other UX metrics, e.g., an 85% decrease in the number
of low-memory kills at the 75th percentile and an 18% decrease in
app launch time at the 50th percentile.

The downstream kernels that have been using MGLRU include:
1. Android [16]
2. Arch Linux Zen [17]
3. Armbian [18]
4. ChromeOS [19]
5. Liquorix [20]
6. OpenWrt [21]
7. post-factum [22]
8. XanMod [23]

[11] https://lore.kernel.org/r/140226722f2032c86301fbd326d91baefe3d7d23.camel@yandex.ru/
[12] https://lore.kernel.org/r/87czj3mux0.fsf@vajain21.in.ibm.com/
[13] https://lore.kernel.org/r/20220105024423.26409-1-szhai2@cs.rochester.edu/
[14] https://lore.kernel.org/r/CA+4-3vksGvKd18FgRinxhqHetBS1hQekJE2gwco8Ja-bJWKtFw@mail.gmail.com/
[15] https://dl.acm.org/doi/10.1145/2749469.2750392
[16] https://android.com
[17] https://archlinux.org
[18] https://armbian.com
[19] https://chromium.org
[20] https://liquorix.net
[21] https://openwrt.org
[22] https://codeberg.org/pf-kernel
[23] https://xanmod.org

Summary
=======
The facts are:
1. The independent lab results and the real-world applications
   indicate substantial improvements; there are no known regressions.
2. Thrashing prevention, working set estimation and proactive reclaim
   work out of the box; there are no equivalent solutions.
3. There is a lot of new code; no smaller changes have been
   demonstrated similar effects.

Our options, accordingly, are:
1. Given the amount of evidence, the reported improvements will likely
   materialize for a wide range of workloads.
2. Gauging the interest from the past discussions, the new features
   will likely be put to use for both personal computers and data
   centers.
3. Based on Google's track record, the new code will likely be well
   maintained in the long term. It'd be more difficult if not
   impossible to achieve similar effects with other approaches.

This patch (of 14):

Some architectures automatically set the accessed bit in PTEs, e.g., x86
and arm64 v8.2.  On architectures that do not have this capability,
clearing the accessed bit in a PTE usually triggers a page fault following
the TLB miss of this PTE (to emulate the accessed bit).

Being aware of this capability can help make better decisions, e.g.,
whether to spread the work out over a period of time to reduce bursty page
faults when trying to clear the accessed bit in many PTEs.

Note that theoretically this capability can be unreliable, e.g.,
hotplugged CPUs might be different from builtin ones.  Therefore it should
not be used in architecture-independent code that involves correctness,
e.g., to determine whether TLB flushes are required (in combination with
the accessed bit).

Link: https://lkml.kernel.org/r/20220918080010.2920238-1-yuzhao@google.com
Link: https://lkml.kernel.org/r/20220918080010.2920238-2-yuzhao@google.com
Change-Id: I7c94aa3ffeb0a8e570c2d7db15183f87658d0141
Signed-off-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Barry Song <baohua@kernel.org>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Acked-by: Will Deacon <will@kernel.org>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit e1fd09e3d1dd4a1a8b3b33bc1fd647eee9f4e475)
[Kalesh Singh - Fix trivial conflict in arch/arm64/include/asm/pgtable.h]
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: x86: add CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG
70e9663 
Some architectures support the accessed bit in non-leaf PMD entries, e.g.,
x86 sets the accessed bit in a non-leaf PMD entry when using it as part of
linear address translation [1].  Page table walkers that clear the
accessed bit may use this capability to reduce their search space.

Note that:
1. Although an inline function is preferable, this capability is added
   as a configuration option for consistency with the existing macros.
2. Due to the little interest in other varieties, this capability was
   only tested on Intel and AMD CPUs.

Thanks to the following developers for their efforts [2][3].
  Randy Dunlap <rdunlap@infradead.org>
  Stephen Rothwell <sfr@canb.auug.org.au>

[1]: Intel 64 and IA-32 Architectures Software Developer's Manual
     Volume 3 (June 2021), section 4.8
[2] https://lore.kernel.org/r/bfdcc7c8-922f-61a9-aa15-7e7250f04af7@infradead.org/
[3] https://lore.kernel.org/r/20220413151513.5a0d7a7e@canb.auug.org.au/

Link: https://lkml.kernel.org/r/20220918080010.2920238-3-yuzhao@google.com
Change-Id: Iccf98138153b8d466c393232df80187dd3687036
Signed-off-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Barry Song <baohua@kernel.org>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit eed9a328aa1ae6ac1edaa026957e6882f57de0dd)
[Kalesh Singh - Fix trivial conflict in arch/Kconfig]
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: multi-gen LRU: groundwork
e3ae36f 
Evictable pages are divided into multiple generations for each lruvec.
The youngest generation number is stored in lrugen->max_seq for both
anon and file types as they are aged on an equal footing. The oldest
generation numbers are stored in lrugen->min_seq[] separately for anon
and file types as clean file pages can be evicted regardless of swap
constraints. These three variables are monotonically increasing.

Generation numbers are truncated into order_base_2(MAX_NR_GENS+1) bits
in order to fit into the gen counter in page->flags. Each truncated
generation number is an index to lrugen->lists[]. The sliding window
technique is used to track at least MIN_NR_GENS and at most
MAX_NR_GENS generations. The gen counter stores a value within [1,
MAX_NR_GENS] while a page is on one of lrugen->lists[]. Otherwise it
stores 0.

There are two conceptually independent procedures: "the aging", which
produces young generations, and "the eviction", which consumes old
generations.  They form a closed-loop system, i.e., "the page reclaim".
Both procedures can be invoked from userspace for the purposes of working
set estimation and proactive reclaim.  These techniques are commonly used
to optimize job scheduling (bin packing) in data centers [1][2].

To avoid confusion, the terms "hot" and "cold" will be applied to the
multi-gen LRU, as a new convention; the terms "active" and "inactive" will
be applied to the active/inactive LRU, as usual.

The protection of hot pages and the selection of cold pages are based
on page access channels and patterns. There are two access channels:
one through page tables and the other through file descriptors. The
protection of the former channel is by design stronger because:
1. The uncertainty in determining the access patterns of the former
   channel is higher due to the approximation of the accessed bit.
2. The cost of evicting the former channel is higher due to the TLB
   flushes required and the likelihood of encountering the dirty bit.
3. The penalty of underprotecting the former channel is higher because
   applications usually do not prepare themselves for major page
   faults like they do for blocked I/O. E.g., GUI applications
   commonly use dedicated I/O threads to avoid blocking rendering
   threads.

There are also two access patterns: one with temporal locality and the
other without.  For the reasons listed above, the former channel is
assumed to follow the former pattern unless VM_SEQ_READ or VM_RAND_READ is
present; the latter channel is assumed to follow the latter pattern unless
outlying refaults have been observed [3][4].

The next patch will address the "outlying refaults".  Three macros, i.e.,
LRU_REFS_WIDTH, LRU_REFS_PGOFF and LRU_REFS_MASK, used later are added in
this patch to make the entire patchset less diffy.

A page is added to the youngest generation on faulting.  The aging needs
to check the accessed bit at least twice before handing this page over to
the eviction.  The first check takes care of the accessed bit set on the
initial fault; the second check makes sure this page has not been used
since then.  This protocol, AKA second chance, requires a minimum of two
generations, hence MIN_NR_GENS.

[1] https://dl.acm.org/doi/10.1145/3297858.3304053
[2] https://dl.acm.org/doi/10.1145/3503222.3507731
[3] https://lwn.net/Articles/495543/
[4] https://lwn.net/Articles/815342/

Link: https://lkml.kernel.org/r/20220918080010.2920238-6-yuzhao@google.com
Change-Id: I7b24d1e9d263e4eb2c2ee23f2eb143824fcb5201
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit ec1c86b25f4bdd9dce6436c0539d2a6ae676e1c4)
[ Resolve conflicts in mm/memory.c, mm/memcontrol.c, mm/Kconfig,
include/linux/mm_inline.h]
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
Signed-off-by: Vaisakh Murali <mvaisakh@statixos.com>
@FPSensor
BACKPORT: mm: multi-gen LRU: minimal implementation
a374c74 
To avoid confusion, the terms "promotion" and "demotion" will be applied
to the multi-gen LRU, as a new convention; the terms "activation" and
"deactivation" will be applied to the active/inactive LRU, as usual.

The aging produces young generations.  Given an lruvec, it increments
max_seq when max_seq-min_seq+1 approaches MIN_NR_GENS.  The aging promotes
hot pages to the youngest generation when it finds them accessed through
page tables; the demotion of cold pages happens consequently when it
increments max_seq.  Promotion in the aging path does not involve any LRU
list operations, only the updates of the gen counter and
lrugen->nr_pages[]; demotion, unless as the result of the increment of
max_seq, requires LRU list operations, e.g., lru_deactivate_fn().  The
aging has the complexity O(nr_hot_pages), since it is only interested in
hot pages.

The eviction consumes old generations.  Given an lruvec, it increments
min_seq when lrugen->lists[] indexed by min_seq%MAX_NR_GENS becomes empty.
A feedback loop modeled after the PID controller monitors refaults over
anon and file types and decides which type to evict when both types are
available from the same generation.

The protection of pages accessed multiple times through file descriptors
takes place in the eviction path.  Each generation is divided into
multiple tiers.  A page accessed N times through file descriptors is in
tier order_base_2(N).  Tiers do not have dedicated lrugen->lists[], only
bits in page->flags.  The aforementioned feedback loop also monitors
refaults over all tiers and decides when to protect pages in which tiers
(N>1), using the first tier (N=0,1) as a baseline.  The first tier
contains single-use unmapped clean pages, which are most likely the best
choices.  In contrast to promotion in the aging path, the protection of a
page in the eviction path is achieved by moving this page to the next
generation, i.e., min_seq+1, if the feedback loop decides so.  This
approach has the following advantages:

1. It removes the cost of activation in the buffered access path by
   inferring whether pages accessed multiple times through file
   descriptors are statistically hot and thus worth protecting in the
   eviction path.
2. It takes pages accessed through page tables into account and avoids
   overprotecting pages accessed multiple times through file
   descriptors. (Pages accessed through page tables are in the first
   tier, since N=0.)
3. More tiers provide better protection for pages accessed more than
   twice through file descriptors, when under heavy buffered I/O
   workloads.

Server benchmark results:
  Single workload:
    fio (buffered I/O): +[30, 32]%
                IOPS         BW
      5.19-rc1: 2673k        10.2GiB/s
      patch1-6: 3491k        13.3GiB/s

  Single workload:
    memcached (anon): -[4, 6]%
                Ops/sec      KB/sec
      5.19-rc1: 1161501.04   45177.25
      patch1-6: 1106168.46   43025.04

  Configurations:
    CPU: two Xeon 6154
    Mem: total 256G

    Node 1 was only used as a ram disk to reduce the variance in the
    results.

    patch drivers/block/brd.c <<EOF
    99,100c99,100
    < 	gfp_flags = GFP_NOIO | __GFP_ZERO | __GFP_HIGHMEM;
    < 	page = alloc_page(gfp_flags);
    ---
    > 	gfp_flags = GFP_NOIO | __GFP_ZERO | __GFP_HIGHMEM | __GFP_THISNODE;
    > 	page = alloc_pages_node(1, gfp_flags, 0);
    EOF

    cat >>/etc/systemd/system.conf <<EOF
    CPUAffinity=numa
    NUMAPolicy=bind
    NUMAMask=0
    EOF

    cat >>/etc/memcached.conf <<EOF
    -m 184320
    -s /var/run/memcached/memcached.sock
    -a 0766
    -t 36
    -B binary
    EOF

    cat fio.sh
    modprobe brd rd_nr=1 rd_size=113246208
    swapoff -a
    mkfs.ext4 /dev/ram0
    mount -t ext4 /dev/ram0 /mnt

    mkdir /sys/fs/cgroup/user.slice/test
    echo 38654705664 >/sys/fs/cgroup/user.slice/test/memory.max
    echo $$ >/sys/fs/cgroup/user.slice/test/cgroup.procs
    fio -name=mglru --numjobs=72 --directory=/mnt --size=1408m \
      --buffered=1 --ioengine=io_uring --iodepth=128 \
      --iodepth_batch_submit=32 --iodepth_batch_complete=32 \
      --rw=randread --random_distribution=random --norandommap \
      --time_based --ramp_time=10m --runtime=5m --group_reporting

    cat memcached.sh
    modprobe brd rd_nr=1 rd_size=113246208
    swapoff -a
    mkswap /dev/ram0
    swapon /dev/ram0

    memtier_benchmark -S /var/run/memcached/memcached.sock \
      -P memcache_binary -n allkeys --key-minimum=1 \
      --key-maximum=65000000 --key-pattern=P:P -c 1 -t 36 \
      --ratio 1:0 --pipeline 8 -d 2000

    memtier_benchmark -S /var/run/memcached/memcached.sock \
      -P memcache_binary -n allkeys --key-minimum=1 \
      --key-maximum=65000000 --key-pattern=R:R -c 1 -t 36 \
      --ratio 0:1 --pipeline 8 --randomize --distinct-client-seed

Client benchmark results:
  kswapd profiles:
    5.19-rc1
      40.33%  page_vma_mapped_walk (overhead)
      21.80%  lzo1x_1_do_compress (real work)
       7.53%  do_raw_spin_lock
       3.95%  _raw_spin_unlock_irq
       2.52%  vma_interval_tree_iter_next
       2.37%  page_referenced_one
       2.28%  vma_interval_tree_subtree_search
       1.97%  anon_vma_interval_tree_iter_first
       1.60%  ptep_clear_flush
       1.06%  __zram_bvec_write

    patch1-6
      39.03%  lzo1x_1_do_compress (real work)
      18.47%  page_vma_mapped_walk (overhead)
       6.74%  _raw_spin_unlock_irq
       3.97%  do_raw_spin_lock
       2.49%  ptep_clear_flush
       2.48%  anon_vma_interval_tree_iter_first
       1.92%  page_referenced_one
       1.88%  __zram_bvec_write
       1.48%  memmove
       1.31%  vma_interval_tree_iter_next

  Configurations:
    CPU: single Snapdragon 7c
    Mem: total 4G

    ChromeOS MemoryPressure [1]

[1] https://chromium.googlesource.com/chromiumos/platform/tast-tests/

Link: https://lkml.kernel.org/r/20220918080010.2920238-7-yuzhao@google.com
Change-Id: I30b26b3086ce1879b83b96eb265f8f0dcb16a1fb
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit ac35a490237446b71e3b4b782b1596967edd0aa8)
[Resolve confilcts in mm/Kconfig, mm/swap.c, mm/vmscan.c
 Remove isolation race check in isolate_page() - TestClearPageLRU()
 not available in 5.10 ]
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>

WIP: DO NOT SUBMIT: swap refault / shadow fixup

I've commented out the bits so we don't hit the panic

Change-Id: Ia1340031d028201e74469c2e8b8e61a2426b1278
Signed-off-by: Vaisakh Murali <mvaisakh@statixos.com>
@FPSensor
BACKPORT: mm: multi-gen LRU: exploit locality in rmap
bdd8505 
Searching the rmap for PTEs mapping each page on an LRU list (to test and
clear the accessed bit) can be expensive because pages from different VMAs
(PA space) are not cache friendly to the rmap (VA space).  For workloads
mostly using mapped pages, searching the rmap can incur the highest CPU
cost in the reclaim path.

This patch exploits spatial locality to reduce the trips into the rmap.
When shrink_page_list() walks the rmap and finds a young PTE, a new
function lru_gen_look_around() scans at most BITS_PER_LONG-1 adjacent
PTEs.  On finding another young PTE, it clears the accessed bit and
updates the gen counter of the page mapped by this PTE to
(max_seq%MAX_NR_GENS)+1.

Server benchmark results:
  Single workload:
    fio (buffered I/O): no change

  Single workload:
    memcached (anon): +[3, 5]%
                Ops/sec      KB/sec
      patch1-6: 1106168.46   43025.04
      patch1-7: 1147696.57   44640.29

  Configurations:
    no change

Client benchmark results:
  kswapd profiles:
    patch1-6
      39.03%  lzo1x_1_do_compress (real work)
      18.47%  page_vma_mapped_walk (overhead)
       6.74%  _raw_spin_unlock_irq
       3.97%  do_raw_spin_lock
       2.49%  ptep_clear_flush
       2.48%  anon_vma_interval_tree_iter_first
       1.92%  page_referenced_one
       1.88%  __zram_bvec_write
       1.48%  memmove
       1.31%  vma_interval_tree_iter_next

    patch1-7
      48.16%  lzo1x_1_do_compress (real work)
       8.20%  page_vma_mapped_walk (overhead)
       7.06%  _raw_spin_unlock_irq
       2.92%  ptep_clear_flush
       2.53%  __zram_bvec_write
       2.11%  do_raw_spin_lock
       2.02%  memmove
       1.93%  lru_gen_look_around
       1.56%  free_unref_page_list
       1.40%  memset

  Configurations:
    no change

Link: https://lkml.kernel.org/r/20220918080010.2920238-8-yuzhao@google.com
Change-Id: Iac405b6d42e2e3f632b6748368f61202c164f1ad
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Barry Song <baohua@kernel.org>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit 018ee47f14893d500131dfca2ff9f3ff8ebd4ed2)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: multi-gen LRU: support page table walks
4ba4764 
To further exploit spatial locality, the aging prefers to walk page tables
to search for young PTEs and promote hot pages.  A kill switch will be
added in the next patch to disable this behavior.  When disabled, the
aging relies on the rmap only.

NB: this behavior has nothing similar with the page table scanning in the
2.4 kernel [1], which searches page tables for old PTEs, adds cold pages
to swapcache and unmaps them.

To avoid confusion, the term "iteration" specifically means the traversal
of an entire mm_struct list; the term "walk" will be applied to page
tables and the rmap, as usual.

An mm_struct list is maintained for each memcg, and an mm_struct follows
its owner task to the new memcg when this task is migrated.  Given an
lruvec, the aging iterates lruvec_memcg()->mm_list and calls
walk_page_range() with each mm_struct on this list to promote hot pages
before it increments max_seq.

When multiple page table walkers iterate the same list, each of them gets
a unique mm_struct; therefore they can run concurrently.  Page table
walkers ignore any misplaced pages, e.g., if an mm_struct was migrated,
pages it left in the previous memcg will not be promoted when its current
memcg is under reclaim.  Similarly, page table walkers will not promote
pages from nodes other than the one under reclaim.

This patch uses the following optimizations when walking page tables:
1. It tracks the usage of mm_struct's between context switches so that
   page table walkers can skip processes that have been sleeping since
   the last iteration.
2. It uses generational Bloom filters to record populated branches so
   that page table walkers can reduce their search space based on the
   query results, e.g., to skip page tables containing mostly holes or
   misplaced pages.
3. It takes advantage of the accessed bit in non-leaf PMD entries when
   CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG=y.
4. It does not zigzag between a PGD table and the same PMD table
   spanning multiple VMAs. IOW, it finishes all the VMAs within the
   range of the same PMD table before it returns to a PGD table. This
   improves the cache performance for workloads that have large
   numbers of tiny VMAs [2], especially when CONFIG_PGTABLE_LEVELS=5.

Server benchmark results:
  Single workload:
    fio (buffered I/O): no change

  Single workload:
    memcached (anon): +[8, 10]%
                Ops/sec      KB/sec
      patch1-7: 1147696.57   44640.29
      patch1-8: 1245274.91   48435.66

  Configurations:
    no change

Client benchmark results:
  kswapd profiles:
    patch1-7
      48.16%  lzo1x_1_do_compress (real work)
       8.20%  page_vma_mapped_walk (overhead)
       7.06%  _raw_spin_unlock_irq
       2.92%  ptep_clear_flush
       2.53%  __zram_bvec_write
       2.11%  do_raw_spin_lock
       2.02%  memmove
       1.93%  lru_gen_look_around
       1.56%  free_unref_page_list
       1.40%  memset

    patch1-8
      49.44%  lzo1x_1_do_compress (real work)
       6.19%  page_vma_mapped_walk (overhead)
       5.97%  _raw_spin_unlock_irq
       3.13%  get_pfn_page
       2.85%  ptep_clear_flush
       2.42%  __zram_bvec_write
       2.08%  do_raw_spin_lock
       1.92%  memmove
       1.44%  alloc_zspage
       1.36%  memset

  Configurations:
    no change

Thanks to the following developers for their efforts [3].
  kernel test robot <lkp@intel.com>

[1] https://lwn.net/Articles/23732/
[2] https://llvm.org/docs/ScudoHardenedAllocator.html
[3] https://lore.kernel.org/r/202204160827.ekEARWQo-lkp@intel.com/

Link: https://lkml.kernel.org/r/20220918080010.2920238-9-yuzhao@google.com
Change-Id: I7ed3daf288e664e15bfd34991a77467a19a4e39a
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit bd74fdaea146029e4fa12c6de89adbe0779348a9)
[ Resolve conflicts in include/linux/memcontrol.h,
  include/linux/mm_types.h ]
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
Signed-off-by: Vaisakh Murali <mvaisakh@statixos.com>
@FPSensor
BACKPORT: mm: multi-gen LRU: optimize multiple memcgs
caecc25 
When multiple memcgs are available, it is possible to use generations as a
frame of reference to make better choices and improve overall performance
under global memory pressure.  This patch adds a basic optimization to
select memcgs that can drop single-use unmapped clean pages first.  Doing
so reduces the chance of going into the aging path or swapping, which can
be costly.

A typical example that benefits from this optimization is a server running
mixed types of workloads, e.g., heavy anon workload in one memcg and heavy
buffered I/O workload in the other.

Though this optimization can be applied to both kswapd and direct reclaim,
it is only added to kswapd to keep the patchset manageable.  Later
improvements may cover the direct reclaim path.

While ensuring certain fairness to all eligible memcgs, proportional scans
of individual memcgs also require proper backoff to avoid overshooting
their aggregate reclaim target by too much.  Otherwise it can cause high
direct reclaim latency.  The conditions for backoff are:

1. At low priorities, for direct reclaim, if aging fairness or direct
   reclaim latency is at risk, i.e., aging one memcg multiple times or
   swapping after the target is met.
2. At high priorities, for global reclaim, if per-zone free pages are
   above respective watermarks.

Server benchmark results:
  Mixed workloads:
    fio (buffered I/O): +[19, 21]%
                IOPS         BW
      patch1-8: 1880k        7343MiB/s
      patch1-9: 2252k        8796MiB/s

    memcached (anon): +[119, 123]%
                Ops/sec      KB/sec
      patch1-8: 862768.65    33514.68
      patch1-9: 1911022.12   74234.54

  Mixed workloads:
    fio (buffered I/O): +[75, 77]%
                IOPS         BW
      5.19-rc1: 1279k        4996MiB/s
      patch1-9: 2252k        8796MiB/s

    memcached (anon): +[13, 15]%
                Ops/sec      KB/sec
      5.19-rc1: 1673524.04   65008.87
      patch1-9: 1911022.12   74234.54

  Configurations:
    (changes since patch 6)

    cat mixed.sh
    modprobe brd rd_nr=2 rd_size=56623104

    swapoff -a
    mkswap /dev/ram0
    swapon /dev/ram0

    mkfs.ext4 /dev/ram1
    mount -t ext4 /dev/ram1 /mnt

    memtier_benchmark -S /var/run/memcached/memcached.sock \
      -P memcache_binary -n allkeys --key-minimum=1 \
      --key-maximum=50000000 --key-pattern=P:P -c 1 -t 36 \
      --ratio 1:0 --pipeline 8 -d 2000

    fio -name=mglru --numjobs=36 --directory=/mnt --size=1408m \
      --buffered=1 --ioengine=io_uring --iodepth=128 \
      --iodepth_batch_submit=32 --iodepth_batch_complete=32 \
      --rw=randread --random_distribution=random --norandommap \
      --time_based --ramp_time=10m --runtime=90m --group_reporting &
    pid=$!

    sleep 200

    memtier_benchmark -S /var/run/memcached/memcached.sock \
      -P memcache_binary -n allkeys --key-minimum=1 \
      --key-maximum=50000000 --key-pattern=R:R -c 1 -t 36 \
      --ratio 0:1 --pipeline 8 --randomize --distinct-client-seed

    kill -INT $pid
    wait

Client benchmark results:
  no change (CONFIG_MEMCG=n)

Link: https://lkml.kernel.org/r/20220918080010.2920238-10-yuzhao@google.com
Change-Id: I590780e6381d577d800d4c4551a702047fc31cc7
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit f76c83378851f8e70f032848c4e61203f39480e4)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: multi-gen LRU: kill switch
f58e11b 
Add /sys/kernel/mm/lru_gen/enabled as a kill switch. Components that
can be disabled include:
  0x0001: the multi-gen LRU core
  0x0002: walking page table, when arch_has_hw_pte_young() returns
          true
  0x0004: clearing the accessed bit in non-leaf PMD entries, when
          CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG=y
  [yYnN]: apply to all the components above
E.g.,
  echo y >/sys/kernel/mm/lru_gen/enabled
  cat /sys/kernel/mm/lru_gen/enabled
  0x0007
  echo 5 >/sys/kernel/mm/lru_gen/enabled
  cat /sys/kernel/mm/lru_gen/enabled
  0x0005

NB: the page table walks happen on the scale of seconds under heavy memory
pressure, in which case the mmap_lock contention is a lesser concern,
compared with the LRU lock contention and the I/O congestion.  So far the
only well-known case of the mmap_lock contention happens on Android, due
to Scudo [1] which allocates several thousand VMAs for merely a few
hundred MBs.  The SPF and the Maple Tree also have provided their own
assessments [2][3].  However, if walking page tables does worsen the
mmap_lock contention, the kill switch can be used to disable it.  In this
case the multi-gen LRU will suffer a minor performance degradation, as
shown previously.

Clearing the accessed bit in non-leaf PMD entries can also be disabled,
since this behavior was not tested on x86 varieties other than Intel and
AMD.

[1] https://source.android.com/devices/tech/debug/scudo
[2] https://lore.kernel.org/r/20220128131006.67712-1-michel@lespinasse.org/
[3] https://lore.kernel.org/r/20220426150616.3937571-1-Liam.Howlett@oracle.com/

Link: https://lkml.kernel.org/r/20220918080010.2920238-11-yuzhao@google.com
Change-Id: If3116e6698cc6967b6992c2017962fac6c2d3a11
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit 354ed597442952fb680c9cafc7e4eb8a76f9514c)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
UPSTREAM: mm: multi-gen LRU: thrashing prevention
63bcb52 
Add /sys/kernel/mm/lru_gen/min_ttl_ms for thrashing prevention, as
requested by many desktop users [1].

When set to value N, it prevents the working set of N milliseconds from
getting evicted.  The OOM killer is triggered if this working set cannot
be kept in memory.  Based on the average human detectable lag (~100ms),
N=1000 usually eliminates intolerable lags due to thrashing.  Larger
values like N=3000 make lags less noticeable at the risk of premature OOM
kills.

Compared with the size-based approach [2], this time-based approach
has the following advantages:

1. It is easier to configure because it is agnostic to applications
   and memory sizes.
2. It is more reliable because it is directly wired to the OOM killer.

[1] https://lore.kernel.org/r/Ydza%2FzXKY9ATRoh6@google.com/
[2] https://lore.kernel.org/r/20101028191523.GA14972@google.com/

Link: https://lkml.kernel.org/r/20220918080010.2920238-12-yuzhao@google.com
Change-Id: I52c14154a55c3e131d6e43fc623b3030cfa435ec
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit 1332a809d95a4fc763cabe5ecb6d4fb6a6d941b2)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
BACKPORT: mm: multi-gen LRU: debugfs interface
9d4824e 
Add /sys/kernel/debug/lru_gen for working set estimation and proactive
reclaim.  These techniques are commonly used to optimize job scheduling
(bin packing) in data centers [1][2].

Compared with the page table-based approach and the PFN-based
approach, this lruvec-based approach has the following advantages:
1. It offers better choices because it is aware of memcgs, NUMA nodes,
   shared mappings and unmapped page cache.
2. It is more scalable because it is O(nr_hot_pages), whereas the
   PFN-based approach is O(nr_total_pages).

Add /sys/kernel/debug/lru_gen_full for debugging.

[1] https://dl.acm.org/doi/10.1145/3297858.3304053
[2] https://dl.acm.org/doi/10.1145/3503222.3507731

Link: https://lkml.kernel.org/r/20220918080010.2920238-13-yuzhao@google.com
Change-Id: I2d07e743e7ed139fb2ba16d5f2c1a5f32f238ccc
Signed-off-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Qi Zheng <zhengqi.arch@bytedance.com>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit d6c3af7d8a2ba5602c28841248c551a712ac50f5)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
@FPSensor
UPSTREAM: mm: multi-gen LRU: admin guide
ab34e8b 
Add an admin guide.

Link: https://lkml.kernel.org/r/20220918080010.2920238-14-yuzhao@google.com
Change-Id: Ia4dba47e8231eda4f0e76fb8969df7291a9bfe7c
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Brian Geffon <bgeffon@google.com>
Acked-by: Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Acked-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Steven Barrett <steven@liquorix.net>
Acked-by: Suleiman Souhlal <suleiman@google.com>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>
Tested-by: Daniel Byrne <djbyrne@mtu.edu>
Tested-by: Donald Carr <d@chaos-reins.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Tested-by: Konstantin Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Shuang Zhai <szhai2@cs.rochester.edu>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michael Larabel <Michael@MichaelLarabel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit 07017acb06012d250fb68930e809257e6694d324)
Bug: 261052246
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
dianders and others added 24 commits November 27, 2025 15:20
@dianders @FPSensor
UPSTREAM: blk-mq: Add blk_mq_delay_run_hw_queues() API call
e21fd0e 
We have:
* blk_mq_run_hw_queue()
* blk_mq_delay_run_hw_queue()
* blk_mq_run_hw_queues()

...but not blk_mq_delay_run_hw_queues(), presumably because nobody
needed it before now.  Since we need it for a later patch in this
series, add it.

Change-Id: If6b8f67f554b967b9c3699c2a79052c2806164dc
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: ralph950412 <ralph950412@gmail.com>
@jankara @FPSensor
FROMGIT: blk-mq: Improve performance of non-mq IO schedulers with mul…
6884ac1 
…tiple HW queues

Currently when non-mq aware IO scheduler (BFQ, mq-deadline) is used for
a queue with multiple HW queues, the performance it rather bad. The
problem is that these IO schedulers use queue-wide locking and their
dispatch function does not respect the hctx it is passed in and returns
any request it finds appropriate. Thus locality of request access is
broken and dispatch from multiple CPUs just contends on IO scheduler
locks. For these IO schedulers there's little point in dispatching from
multiple CPUs. Instead dispatch always only from a single CPU to limit
contention.

Below is a comparison of dbench runs on XFS filesystem where the storage
is a raid card with 64 HW queues and to it attached a single rotating
disk. BFQ is used as IO scheduler:

      clients           MQ                     SQ             MQ-Patched
Amean 1      39.12 (0.00%)       43.29 * -10.67%*       36.09 *   7.74%*
Amean 2     128.58 (0.00%)      101.30 *  21.22%*       96.14 *  25.23%*
Amean 4     577.42 (0.00%)      494.47 *  14.37%*      508.49 *  11.94%*
Amean 8     610.95 (0.00%)      363.86 *  40.44%*      362.12 *  40.73%*
Amean 16    391.78 (0.00%)      261.49 *  33.25%*      282.94 *  27.78%*
Amean 32    324.64 (0.00%)      267.71 *  17.54%*      233.00 *  28.23%*
Amean 64    295.04 (0.00%)      253.02 *  14.24%*      242.37 *  17.85%*
Amean 512 10281.61 (0.00%)    10211.16 *   0.69%*    10447.53 *  -1.61%*

Numbers are times so lower is better. MQ is stock 5.10-rc6 kernel. SQ is
the same kernel with megaraid_sas.host_tagset_enable=0 so that the card
advertises just a single HW queue. MQ-Patched is a kernel with this
patch applied.

You can see multiple hardware queues heavily hurt performance in
combination with BFQ. The patch restores the performance.

Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
BUG: 187357408
Change-Id: I53645eb48cb308cd3af81a1c5e718a6abec6a1f9
(cherry picked from commit fa56cac78af68bd93734c290a0ffd0716e871dba git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@linfeng2999 @FPSensor
FROMGIT: bfq/mq-deadline: remove redundant check for passthrough request
dcc687f 
Since commit 01e99aeca39796003 'blk-mq: insert passthrough request into
hctx->dispatch directly', passthrough request should not appear in
IO-scheduler any more, so blk_rq_is_passthrough checking in addon IO
schedulers is redundant.

(Notes: this patch passes generic IO load test with hdds under SAS
controller and hdds under AHCI controller but obviously not covers all.
Not sure if passthrough request can still escape into IO scheduler from
blk_mq_sched_insert_requests, which is used by blk_mq_flush_plug_list and
has lots of indirect callers.)

Signed-off-by: Lin Feng <linf@wangsu.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
BUG: 187357408
Change-Id: I97d85c38e584add44399295f3839994b694bc9ca
(cherry picked from commit 0856faaa220759a4fe4334f5c57a8661c94c14ce git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/Kconfig: Make the BLK_WBT and BLK_WBT_MQ entries conse…
65f3b2b 
…cutive

These entries were consecutive at the time of their introduction but are no
longer consecutive. Make these again consecutive. Additionally, modify the
help text since it refers to blk-mq and since the legacy block layer has
been removed.

Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I568383377a3244efba9748adf0a2e90bd7660bb2
(cherry picked from commit fdc250ea26e44066d690bbe65a03fab512af0699 git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/blk-cgroup: Swap the blk_throtl_init() and blk_iolaten…
32c5ab7 
…cy_init() calls

Before adding more calls in this function, simplify the error path.

Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Cc: Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I8568b87d1bebbd3841e42a79b7efe2d0a1bff2bc
(cherry picked from commit f1a7f539c2720906fb10be0af3514b034e1a9fee git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/blk-rq-qos: Move a function from a header file into a …
8e6bbd6 
…C file

rq_qos_id_to_name() is only used in blk-mq-debugfs.c so move that function
into in blk-mq-debugfs.c.

Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: If03083a13917bc2f88b6df7151e033a11ab1bc50
(cherry picked from commit f1a7f539c2720906fb10be0af3514b034e1a9fee git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block: Introduce the ioprio rq-qos policy
5328df5 
Introduce an rq-qos policy that assigns an I/O priority to requests based
on blk-cgroup configuration settings. This policy has the following
advantages over the ioprio_set() system call:
- This policy is cgroup based so it has all the advantages of cgroups.
- While ioprio_set() does not affect page cache writeback I/O, this rq-qos
  controller affects page cache writeback I/O for filesystems that support
  assiociating a cgroup with writeback I/O. See also
  Documentation/admin-guide/cgroup-v2.rst.

Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: If51e608ad37ee7a3f57b507bb17900dcfcb263ed
(cherry picked from commit ee9d2a55c960f152b5710078bbe399a4c51eb0a9 git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Add several comments
175e823 
Make the code easier to read by adding more comments.

Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: If62eb600614d2883d72ee3bd7e7859ae66b24512
(cherry picked from commit 16c3afdb127bbff7d3552e076e568281765674b7 git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Add two lockdep_assert_held() statements
ba71aa9 
Document the locking strategy by adding two lockdep_assert_held()
statements.

Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: Ie8cf0b0ae208c9cc87731a9c6d7df5e5e59332d5
(cherry picked from commit 91831ddfd7c6e3df9857526a76cfa88673ec0637 git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Remove two local variables
19b36b6 
Make __dd_dispatch_request() easier to read by removing two local
variables.

Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I5567f7d02a2c628efb437058a1c103c7b123747a
(cherry picked from commit f005b6ff19d2a961a2c3ae9c5f49d48fda143469 git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Rename dd_init_queue() and dd_exit_queue()
d66412a 
Change "queue" into "sched" to make the function names reflect better the
purpose of these functions.

Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I30825b379146dbaef4ff3f85148b2e788667a77c
(cherry picked from commit a6e57fe5ab09c250fc741294e6321270a4364fec git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Improve compile-time argument checking
575b937 
Modern compilers complain if an out-of-range value is passed to a function
argument that has an enumeration type. Let the compiler detect out-of-range
data direction arguments instead of verifying the data_dir argument at
runtime.

Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I4ad8c106a86d17f3010e12e172702e77eca61e80
(cherry picked from commit d9baee13f8cf66a8fac9ec67fdb85ce419fcce3a git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Improve the sysfs show and store macros
f5db22f 
Define separate macros for integers and jiffies to improve readability.
Use sysfs_emit() and kstrtoint() instead of sprintf() and simple_strtol().
The former functions are the recommended functions.

Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I4e0fd35124cd0319fcace0d1d5e3c113b60a213c
(cherry picked from commit d9baee13f8cf66a8fac9ec67fdb85ce419fcce3a git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
BACKPORT: FROMGIT: block/mq-deadline: Reserve 25% of scheduler tags f…
873ec1f 
…or synchronous requests

For interactive workloads it is important that synchronous requests are
not delayed. Hence reserve 25% of scheduler tags for synchronous requests.
This patch still allows asynchronous requests to fill the hardware queues
since blk_mq_init_sched() makes sure that the number of scheduler requests
is the double of the hardware queue depth. From blk_mq_init_sched():

	q->nr_requests = 2 * min_t(unsigned int, q->tag_set->queue_depth,
				   BLKDEV_MAX_RQ);

Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: Ib9cd753a39c8e5f5c45908001d69334130ef2067
(cherry picked from commit c970bc8292aaaf6f2d333d612e657df3a99f417c git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
Signed-off-by: ralph950412 <ralph950412@gmail.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Micro-optimize the batching algorithm
213b6b9 
When dispatching the first request of a batch, the deadline_move_request()
call clears .next_rq[] for the opposite data direction. .next_rq[] is not
restored when changing data direction. Fix this by not clearing .next_rq[]
and by keeping track of the data direction of a batch in a variable instead.

This patch is a micro-optimization because:
- The number of deadline_next_request() calls for the read direction is
  halved.
- The number of times that deadline_next_request() returns NULL is reduced.

Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I582e99603a5443d75cf2b18a5daa2c93b5c66de3
(cherry picked from commit ea0fd2a525436ab5b9ada0f1953b0c0a29357311 git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Add I/O priority support
97dc0ad 
Maintain one dispatch list and one FIFO list per I/O priority class: RT, BE
and IDLE. Maintain statistics for each priority level. Split the debugfs
attributes per priority level as follows:

$ ls /sys/kernel/debug/block/.../sched/
async_depth  dispatch2        read_next_rq      write2_fifo_list
batching     read0_fifo_list  starved           write_next_rq
dispatch0    read1_fifo_list  write0_fifo_list
dispatch1    read2_fifo_list  write1_fifo_list

Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I60451cfdb416ad27601dc3ffb4eb307fa6ff783f
(cherry picked from commit 5b701a6e040ff8626ecf29ac06de9689efc00754 git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Track I/O statistics
d3aafd3 
Track I/O statistics per I/O priority and export these statistics to
debugfs. These statistics help developers of the deadline scheduler.

Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I8e91693dc1d015060737fa2fc15f5f2ebee2530c
(cherry picked from commit 9dc236caf2518c1e434be7a4f8fae60fb0be506a git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@bvanassche @FPSensor
BACKPORT: FROMGIT: block/mq-deadline: Add cgroup support
ca6d9ea 
Maintain statistics per cgroup and export these to user space. These
statistics are essential for verifying whether the proper I/O priorities
have been assigned to requests. An example of the statistics data with
this patch applied:

$ cat /sys/fs/cgroup/io.stat
11:2 rbytes=0 wbytes=0 rios=3 wios=0 dbytes=0 dios=0 [NONE] dispatched=0 inserted=0 merged=171 [RT] dispatched=0 inserted=0 merged=0 [BE] dispatched=0 inserted=0 merged=0 [IDLE] dispatched=0 inserted=0 merged=0
8:32 rbytes=2142720 wbytes=0 rios=105 wios=0 dbytes=0 dios=0 [NONE] dispatched=0 inserted=0 merged=171 [RT] dispatched=0 inserted=0 merged=0 [BE] dispatched=0 inserted=0 merged=0 [IDLE] dispatched=0 inserted=0 merged=0

Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
BUG: 187357408
Change-Id: I8d976c62ba2c0397cbb18076f3e61d5ab246cbcf
(cherry picked from commit f5dc926252cb31739809f7d27a8cbc9941b4d36d git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
Signed-off-by: ralph950412 <ralph950412@gmail.com>
@bvanassche @FPSensor
FROMGIT: block/mq-deadline: Prioritize high-priority requests
15357bf 
While one or more requests with a certain I/O priority are pending, do not
dispatch lower priority requests. Dispatch lower priority requests anyway
after the "aging" time has expired.

This patch has been tested as follows:

modprobe scsi_debug ndelay=1000000 max_queue=16 &&
sd='' &&
while [ -z "$sd" ]; do
  sd=/dev/$(basename /sys/bus/pseudo/drivers/scsi_debug/adapter*/host*/target*/*/block/*)
done &&
echo $((100*1000)) > /sys/block/$sd/queue/iosched/aging_expire &&
cd /sys/fs/cgroup/blkio/ &&
echo $$ >cgroup.procs &&
echo restrict-to-be >blkio.prio.class &&
mkdir -p hipri &&
cd hipri &&
echo none-to-rt >blkio.prio.class &&
{ max-iops -a1 -d32 -j1 -e mq-deadline $sd >& ~/low-pri.txt & } &&
echo $$ >cgroup.procs &&
max-iops -a1 -d32 -j1 -e mq-deadline $sd >& ~/hi-pri.txt

Result:
* 11000 IOPS for the high-priority job
*    40 IOPS for the low-priority job

If the aging expiry time is changed from 100s into 0, the IOPS results change
into 6712 and 6796 IOPS.

The max-iops script is a script that runs fio with the following arguments:
--bs=4K --gtod_reduce=1 --ioengine=libaio --ioscheduler=${arg_e} --runtime=60
--norandommap --rw=read --thread --buffered=0 --numjobs=${arg_j}
--iodepth=${arg_d} --iodepth_batch_submit=${arg_a}
--iodepth_batch_complete=$((arg_d / 2)) --name=${positional_argument_1}
--filename=${positional_argument_1}

Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Change-Id: I99a0674b018d096ec96bbfa3008eedcfda5013da
BUG: 187357408
(cherry picked from commit 40d5d42992b0de3ae7961735ea15eef5bd385ebf git://git.kernel.dk/linux-block/ for-5.14/block)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@YangYang866 @FPSensor
ANDROID: GKI: Add ANDROID_OEM_DATA in struct blk_mq_ctx
b212035 
Add ANDROID_OEM_DATA for implement of oem gki

Bug: 188749221
Change-Id: Ide8378a898de01a34d8ca3c34472844cd4ffa71c
Signed-off-by: Yang Yang <yang.yang@vivo.com>
@YangYang866 @FPSensor
ANDROID: GKI: Add ANDROID_OEM_DATA in struct blk_mq_tags
3fc1251 
Add ANDROID_OEM_DATA for implement of oem gki

Bug: 188749221
Change-Id: I1feba2334aa34e3bc46eb9d0217118485405beb4
Signed-off-by: Yang Yang <yang.yang@vivo.com>
@YangYang866 @FPSensor
ANDROID: GKI: enable CONFIG_BLK_CGROUP_IOCOST
f5ee609 
Enable CONFIG_BLK_CGROUP_IOCOST to help control IO resources.

Bug: 188749221
Change-Id: I611b3ff5929d0a998fa6241967887803636b7588
Signed-off-by: Yang Yang <yang.yang@vivo.com>
@FPSensor
ANDROID: Enable the ioprio cgroup controller
d7d89ab 
Enable this controller such that it can be used in performance tests.

Bug: 186902601
Change-Id: Icd863fec49e3f928b992834e07ae0660ba57ea65
Signed-off-by: Bart Van Assche <bvanassche@google.com>
@FPSensor
arm64: configs: regenerate defconfig
8ae1a9f 
Signed-off-by: FPSensor <gkartyt@gmail.com>
@elitetest1 elitetest1 closed this Nov 27, 2025
elitetest1 pushed a commit that referenced this pull request Dec 11, 2025
@gregkh
devcoredump: Fix circular locking dependency with devcd->mutex.
c21831f 
[ Upstream commit a91c8096590bd7801a26454789f2992094fe36da ]

The original code causes a circular locking dependency found by lockdep.

======================================================
WARNING: possible circular locking dependency detected
6.16.0-rc6-lgci-xe-xe-pw-151626v3+ #1 Tainted: G S   U
------------------------------------------------------
xe_fault_inject/5091 is trying to acquire lock:
ffff888156815688 ((work_completion)(&(&devcd->del_wk)->work)){+.+.}-{0:0}, at: __flush_work+0x25d/0x660

but task is already holding lock:

ffff888156815620 (&devcd->mutex){+.+.}-{3:3}, at: dev_coredump_put+0x3f/0xa0
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #2 (&devcd->mutex){+.+.}-{3:3}:
       mutex_lock_nested+0x4e/0xc0
       devcd_data_write+0x27/0x90
       sysfs_kf_bin_write+0x80/0xf0
       kernfs_fop_write_iter+0x169/0x220
       vfs_write+0x293/0x560
       ksys_write+0x72/0xf0
       __x64_sys_write+0x19/0x30
       x64_sys_call+0x2bf/0x2660
       do_syscall_64+0x93/0xb60
       entry_SYSCALL_64_after_hwframe+0x76/0x7e
-> #1 (kn->active#236){++++}-{0:0}:
       kernfs_drain+0x1e2/0x200
       __kernfs_remove+0xae/0x400
       kernfs_remove_by_name_ns+0x5d/0xc0
       remove_files+0x54/0x70
       sysfs_remove_group+0x3d/0xa0
       sysfs_remove_groups+0x2e/0x60
       device_remove_attrs+0xc7/0x100
       device_del+0x15d/0x3b0
       devcd_del+0x19/0x30
       process_one_work+0x22b/0x6f0
       worker_thread+0x1e8/0x3d0
       kthread+0x11c/0x250
       ret_from_fork+0x26c/0x2e0
       ret_from_fork_asm+0x1a/0x30
-> #0 ((work_completion)(&(&devcd->del_wk)->work)){+.+.}-{0:0}:
       __lock_acquire+0x1661/0x2860
       lock_acquire+0xc4/0x2f0
       __flush_work+0x27a/0x660
       flush_delayed_work+0x5d/0xa0
       dev_coredump_put+0x63/0xa0
       xe_driver_devcoredump_fini+0x12/0x20 [xe]
       devm_action_release+0x12/0x30
       release_nodes+0x3a/0x120
       devres_release_all+0x8a/0xd0
       device_unbind_cleanup+0x12/0x80
       device_release_driver_internal+0x23a/0x280
       device_driver_detach+0x14/0x20
       unbind_store+0xaf/0xc0
       drv_attr_store+0x21/0x50
       sysfs_kf_write+0x4a/0x80
       kernfs_fop_write_iter+0x169/0x220
       vfs_write+0x293/0x560
       ksys_write+0x72/0xf0
       __x64_sys_write+0x19/0x30
       x64_sys_call+0x2bf/0x2660
       do_syscall_64+0x93/0xb60
       entry_SYSCALL_64_after_hwframe+0x76/0x7e
other info that might help us debug this:
Chain exists of: (work_completion)(&(&devcd->del_wk)->work) --> kn->active#236 --> &devcd->mutex
 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  lock(&devcd->mutex);
                               lock(kn->active#236);
                               lock(&devcd->mutex);
  lock((work_completion)(&(&devcd->del_wk)->work));
 *** DEADLOCK ***
5 locks held by xe_fault_inject/5091:
 #0: ffff8881129f9488 (sb_writers#5){.+.+}-{0:0}, at: ksys_write+0x72/0xf0
 #1: ffff88810c755078 (&of->mutex#2){+.+.}-{3:3}, at: kernfs_fop_write_iter+0x123/0x220
 #2: ffff8881054811a0 (&dev->mutex){....}-{3:3}, at: device_release_driver_internal+0x55/0x280
 #3: ffff888156815620 (&devcd->mutex){+.+.}-{3:3}, at: dev_coredump_put+0x3f/0xa0
 #4: ffffffff8359e020 (rcu_read_lock){....}-{1:2}, at: __flush_work+0x72/0x660
stack backtrace:
CPU: 14 UID: 0 PID: 5091 Comm: xe_fault_inject Tainted: G S   U              6.16.0-rc6-lgci-xe-xe-pw-151626v3+ #1 PREEMPT_{RT,(lazy)}
Tainted: [S]=CPU_OUT_OF_SPEC, [U]=USER
Hardware name: Micro-Star International Co., Ltd. MS-7D25/PRO Z690-A DDR4(MS-7D25), BIOS 1.10 12/13/2021
Call Trace:
 <TASK>
 dump_stack_lvl+0x91/0xf0
 dump_stack+0x10/0x20
 print_circular_bug+0x285/0x360
 check_noncircular+0x135/0x150
 ? register_lock_class+0x48/0x4a0
 __lock_acquire+0x1661/0x2860
 lock_acquire+0xc4/0x2f0
 ? __flush_work+0x25d/0x660
 ? mark_held_locks+0x46/0x90
 ? __flush_work+0x25d/0x660
 __flush_work+0x27a/0x660
 ? __flush_work+0x25d/0x660
 ? trace_hardirqs_on+0x1e/0xd0
 ? __pfx_wq_barrier_func+0x10/0x10
 flush_delayed_work+0x5d/0xa0
 dev_coredump_put+0x63/0xa0
 xe_driver_devcoredump_fini+0x12/0x20 [xe]
 devm_action_release+0x12/0x30
 release_nodes+0x3a/0x120
 devres_release_all+0x8a/0xd0
 device_unbind_cleanup+0x12/0x80
 device_release_driver_internal+0x23a/0x280
 ? bus_find_device+0xa8/0xe0
 device_driver_detach+0x14/0x20
 unbind_store+0xaf/0xc0
 drv_attr_store+0x21/0x50
 sysfs_kf_write+0x4a/0x80
 kernfs_fop_write_iter+0x169/0x220
 vfs_write+0x293/0x560
 ksys_write+0x72/0xf0
 __x64_sys_write+0x19/0x30
 x64_sys_call+0x2bf/0x2660
 do_syscall_64+0x93/0xb60
 ? __f_unlock_pos+0x15/0x20
 ? __x64_sys_getdents64+0x9b/0x130
 ? __pfx_filldir64+0x10/0x10
 ? do_syscall_64+0x1a2/0xb60
 ? clear_bhb_loop+0x30/0x80
 ? clear_bhb_loop+0x30/0x80
 entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x76e292edd574
Code: c7 00 16 00 00 00 b8 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 80 3d d5 ea 0e 00 00 74 13 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 54 c3 0f 1f 00 55 48 89 e5 48 83 ec 20 48 89
RSP: 002b:00007fffe247a828 EFLAGS: 00000202 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 000076e292edd574
RDX: 000000000000000c RSI: 00006267f6306063 RDI: 000000000000000b
RBP: 000000000000000c R08: 000076e292fc4b20 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000202 R12: 00006267f6306063
R13: 000000000000000b R14: 00006267e6859c00 R15: 000076e29322a000
 </TASK>
xe 0000:03:00.0: [drm] Xe device coredump has been deleted.

Fixes: 01daccf74832 ("devcoredump : Serialize devcd_del work")
Cc: Mukesh Ojha <quic_mojha@quicinc.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Johannes Berg <johannes@sipsolutions.net>
Cc: Rafael J. Wysocki <rafael@kernel.org>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: linux-kernel@vger.kernel.org
Cc: stable@vger.kernel.org # v6.1+
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
Cc: Matthew Brost <matthew.brost@intel.com>
Acked-by: Mukesh Ojha <mukesh.ojha@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250723142416.1020423-1-dev@lankhorst.se
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ replaced disable_delayed_work_sync() with cancel_delayed_work_sync() ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
elitetest1 pushed a commit that referenced this pull request Dec 11, 2025
@Vudentz @gregkh
Bluetooth: SCO: Fix UAF on sco_conn_free
5770713 
[ Upstream commit ecb9a843be4d6fd710d7026e359f21015a062572 ]

BUG: KASAN: slab-use-after-free in sco_conn_free net/bluetooth/sco.c:87 [inline]
BUG: KASAN: slab-use-after-free in kref_put include/linux/kref.h:65 [inline]
BUG: KASAN: slab-use-after-free in sco_conn_put+0xdd/0x410
net/bluetooth/sco.c:107
Write of size 8 at addr ffff88811cb96b50 by task kworker/u17:4/352

CPU: 1 UID: 0 PID: 352 Comm: kworker/u17:4 Not tainted
6.17.0-rc5-g717368f83676 #4 PREEMPT(voluntary)
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
Workqueue: hci13 hci_cmd_sync_work
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x10b/0x170 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0x191/0x550 mm/kasan/report.c:482
 kasan_report+0xc4/0x100 mm/kasan/report.c:595
 sco_conn_free net/bluetooth/sco.c:87 [inline]
 kref_put include/linux/kref.h:65 [inline]
 sco_conn_put+0xdd/0x410 net/bluetooth/sco.c:107
 sco_connect_cfm+0xb4/0xae0 net/bluetooth/sco.c:1441
 hci_connect_cfm include/net/bluetooth/hci_core.h:2082 [inline]
 hci_conn_failed+0x20a/0x2e0 net/bluetooth/hci_conn.c:1313
 hci_conn_unlink+0x55f/0x810 net/bluetooth/hci_conn.c:1121
 hci_conn_del+0xb6/0x1110 net/bluetooth/hci_conn.c:1147
 hci_abort_conn_sync+0x8c5/0xbb0 net/bluetooth/hci_sync.c:5689
 hci_cmd_sync_work+0x281/0x380 net/bluetooth/hci_sync.c:332
 process_one_work kernel/workqueue.c:3236 [inline]
 process_scheduled_works+0x77e/0x1040 kernel/workqueue.c:3319
 worker_thread+0xbee/0x1200 kernel/workqueue.c:3400
 kthread+0x3c7/0x870 kernel/kthread.c:463
 ret_from_fork+0x13a/0x1e0 arch/x86/kernel/process.c:148
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
 </TASK>

Allocated by task 31370:
 kasan_save_stack mm/kasan/common.c:47 [inline]
 kasan_save_track+0x30/0x70 mm/kasan/common.c:68
 poison_kmalloc_redzone mm/kasan/common.c:388 [inline]
 __kasan_kmalloc+0x82/0x90 mm/kasan/common.c:405
 kasan_kmalloc include/linux/kasan.h:260 [inline]
 __do_kmalloc_node mm/slub.c:4382 [inline]
 __kmalloc_noprof+0x22f/0x390 mm/slub.c:4394
 kmalloc_noprof include/linux/slab.h:909 [inline]
 sk_prot_alloc+0xae/0x220 net/core/sock.c:2239
 sk_alloc+0x34/0x5a0 net/core/sock.c:2295
 bt_sock_alloc+0x3c/0x330 net/bluetooth/af_bluetooth.c:151
 sco_sock_alloc net/bluetooth/sco.c:562 [inline]
 sco_sock_create+0xc0/0x350 net/bluetooth/sco.c:593
 bt_sock_create+0x161/0x3b0 net/bluetooth/af_bluetooth.c:135
 __sock_create+0x3ad/0x780 net/socket.c:1589
 sock_create net/socket.c:1647 [inline]
 __sys_socket_create net/socket.c:1684 [inline]
 __sys_socket+0xd5/0x330 net/socket.c:1731
 __do_sys_socket net/socket.c:1745 [inline]
 __se_sys_socket net/socket.c:1743 [inline]
 __x64_sys_socket+0x7a/0x90 net/socket.c:1743
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xc7/0x240 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 31374:
 kasan_save_stack mm/kasan/common.c:47 [inline]
 kasan_save_track+0x30/0x70 mm/kasan/common.c:68
 kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:576
 poison_slab_object mm/kasan/common.c:243 [inline]
 __kasan_slab_free+0x3d/0x50 mm/kasan/common.c:275
 kasan_slab_free include/linux/kasan.h:233 [inline]
 slab_free_hook mm/slub.c:2428 [inline]
 slab_free mm/slub.c:4701 [inline]
 kfree+0x199/0x3b0 mm/slub.c:4900
 sk_prot_free net/core/sock.c:2278 [inline]
 __sk_destruct+0x4aa/0x630 net/core/sock.c:2373
 sco_sock_release+0x2ad/0x300 net/bluetooth/sco.c:1333
 __sock_release net/socket.c:649 [inline]
 sock_close+0xb8/0x230 net/socket.c:1439
 __fput+0x3d1/0x9e0 fs/file_table.c:468
 task_work_run+0x206/0x2a0 kernel/task_work.c:227
 get_signal+0x1201/0x1410 kernel/signal.c:2807
 arch_do_signal_or_restart+0x34/0x740 arch/x86/kernel/signal.c:337
 exit_to_user_mode_loop+0x68/0xc0 kernel/entry/common.c:40
 exit_to_user_mode_prepare include/linux/irq-entry-common.h:225 [inline]
 syscall_exit_to_user_mode_work include/linux/entry-common.h:175 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:210 [inline]
 do_syscall_64+0x1dd/0x240 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Reported-by: cen zhang <zzzccc427@gmail.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
elitetest1 pushed a commit that referenced this pull request Jan 9, 2026
@elitetest1
ANDROID: vmscan: Support multiple kswapd threads per node
f714a03 
Page replacement is handled in the Linux Kernel in one of two ways:

1) Asynchronously via kswapd
2) Synchronously, via direct reclaim

At page allocation time the allocating task is immediately given a page
from the zone free list allowing it to go right back to work doing
whatever it was doing; Probably directly or indirectly executing business
logic.

Just prior to satisfying the allocation, free pages is checked to see if
it has reached the zone low watermark and if so, kswapd is awakened.
Kswapd will start scanning pages looking for inactive pages to evict to
make room for new page allocations. The work of kswapd allows tasks to
continue allocating memory from their respective zone free list without
incurring any delay.

When the demand for free pages exceeds the rate that kswapd tasks can
supply them, page allocation works differently. Once the allocating task
finds that the number of free pages is at or below the zone min watermark,
the task will no longer pull pages from the free list. Instead, the task
will run the same CPU-bound routines as kswapd to satisfy its own
allocation by scanning and evicting pages. This is called a direct reclaim.

The time spent performing a direct reclaim can be substantial, often
taking tens to hundreds of milliseconds for small order0 allocations to
half a second or more for order9 huge-page allocations. In fact, kswapd is
not actually required on a linux system. It exists for the sole purpose of
optimizing performance by preventing direct reclaims.

When memory shortfall is sufficient to trigger direct reclaims, they can
occur in any task that is running on the system. A single aggressive
memory allocating task can set the stage for collateral damage to occur in
small tasks that rarely allocate additional memory. Consider the impact of
injecting an additional 100ms of latency when nscd allocates memory to
facilitate caching of a DNS query.

The presence of direct reclaims 10 years ago was a fairly reliable
indicator that too much was being asked of a Linux system. Kswapd was
likely wasting time scanning pages that were ineligible for eviction.
Adding RAM or reducing the working set size would usually make the problem
go away. Since then hardware has evolved to bring a new struggle for
kswapd. Storage speeds have increased by orders of magnitude while CPU
clock speeds stayed the same or even slowed down in exchange for more
cores per package. This presents a throughput problem for a single
threaded kswapd that will get worse with each generation of new hardware.

Test Details

NOTE: The tests below were run with shadow entries disabled. See the
associated patch and cover letter for details

The tests below were designed with the assumption that a kswapd bottleneck
is best demonstrated using filesystem reads. This way, the inactive list
will be full of clean pages, simplifying the analysis and allowing kswapd
to achieve the highest possible steal rate. Maximum steal rates for kswapd
are likely to be the same or lower for any other mix of page types on the
system.

Tests were run on a 2U Oracle X7-2L with 52 Intel Xeon Skylake 2GHz cores,
756GB of RAM and 8 x 3.6 TB NVMe Solid State Disk drives. Each drive has
an XFS file system mounted separately as /d0 through /d7. SSD drives
require multiple concurrent streams to show their potential, so I created
eleven 250GB zero-filled files on each drive so that I could test with
parallel reads.

The test script runs in multiple stages. At each stage, the number of dd
tasks run concurrently is increased by 2. I did not include all of the
test output for brevity.

During each stage dd tasks are launched to read from each drive in a round
robin fashion until the specified number of tasks for the stage has been
reached. Then iostat, vmstat and top are started in the background with 10
second intervals. After five minutes, all of the dd tasks are killed and
the iostat, vmstat and top output is parsed in order to report the
following:

CPU consumption
- sy - aggregate kernel mode CPU consumption from vmstat output. The value
       doesn't tend to fluctuate much so I just grab the highest value.
       Each sample is averaged over 10 seconds
- dd_cpu - for all of the dd tasks averaged across the top samples since
           there is a lot of variation.

Throughput
- in Kbytes
- Command is iostat -x -d 10 -g total

This first test performs reads using O_DIRECT in order to show the maximum
throughput that can be obtained using these drives. It also demonstrates
how rapidly throughput scales as the number of dd tasks are increased.

The dd command for this test looks like this:

Command Used: dd iflag=direct if=/d${i}/$n of=/dev/null bs=4M

Test #1: Direct IO
dd sy dd_cpu throughput
6  0  2.33   14726026.40
10 1  2.95   19954974.80
16 1  2.63   24419689.30
22 1  2.63   25430303.20
28 1  2.91   26026513.20
34 1  2.53   26178618.00
40 1  2.18   26239229.20
46 1  1.91   26250550.40
52 1  1.69   26251845.60
58 1  1.54   26253205.60
64 1  1.43   26253780.80
70 1  1.31   26254154.80
76 1  1.21   26253660.80
82 1  1.12   26254214.80
88 1  1.07   26253770.00
90 1  1.04   26252406.40

Throughput was close to peak with only 22 dd tasks. Very little system CPU
was consumed as expected as the drives DMA directly into the user address
space when using direct IO.

In this next test, the iflag=direct option is removed and we only run the
test until the pgscan_kswapd from /proc/vmstat starts to increment. At
that point metrics are parsed and reported and the pagecache contents are
dropped prior to the next test. Lather, rinse, repeat.

Test #2: standard file system IO, no page replacement
dd sy dd_cpu throughput
6  2  28.78  5134316.40
10 3  31.40  8051218.40
16 5  34.73  11438106.80
22 7  33.65  14140596.40
28 8  31.24  16393455.20
34 10 29.88  09e9cd8.60
40 11 28.33  19644159.60
46 11 25.05  20802497.60
52 13 26.92  22092370.00
58 13 23.29  22884881.20
64 14 23.12  23452248.80
70 15 22.40  23916468.00
76 16 22.06  24328737.20
82 17 20.97  24718693.20
88 16 18.57  25149404.40
90 16 18.31  25245565.60

Each read has to pause after the buffer in kernel space is populated while
those pages are added to the pagecache and copied into the user address
space. For this reason, more parallel streams are required to achieve peak
throughput. The copy operation consumes substantially more CPU than direct
IO as expected.

The next test measures throughput after kswapd starts running. This is the
same test only we wait for kswapd to wake up before we start collecting
metrics. The script actually keeps track of a few things that were not
mentioned earlier. It tracks direct reclaims and page scans by watching
the metrics in /proc/vmstat. CPU consumption for kswapd is tracked the
same way it is tracked for dd.

Since the test is 100% reads, you can assume that the page steal rate for
kswapd and direct reclaims is almost identical to the scan rate.

Test #3: 1 kswapd thread per node
dd sy dd_cpu kswapd0 kswapd1 throughput  dr    pgscan_kswapd pgscan_direct
10 4  26.07  28.56   27.03   7355924.40  0     459316976     0
16 7  34.94  69.33   69.66   10867895.20 0     872661643     0
22 10 36.03  93.99   99.33   13130613.60 489   1037654473    11268334
28 10 30.34  95.90   98.60   14601509.60 671   1182591373    15429142
34 14 34.77  97.50   99.23   16468012.00 10850 1069005644    249839515
40 17 36.32  91.49   97.11   17335987.60 18903 975417728     434467710
46 19 38.40  90.54   91.61   17705394.40 25369 855737040     582427973
52 22 40.88  83.97   83.70   17607680.40 31250 709532935     724282458
58 25 40.89  82.19   80.14   17976905.60 35060 657796473     804117540
64 28 41.77  73.49   75.20   18001910.00 39073 561813658     895289337
70 33 45.51  63.78   64.39   17061897.20 44523 379465571     1020726436
76 36 46.95  57.96   60.32   16964459.60 47717 291299464     1093172384
82 39 47.16  55.43   56.16   16949956.00 49479 247071062     1134163008
88 42 47.41  53.75   47.62   16930911.20 51521 195449924     1180442208
90 43 47.18  51.40   50.59   16864428.00 51618 190758156     1183203901

In the previous test where kswapd was not involved, the system-wide kernel
mode CPU consumption with 90 dd tasks was 16%. In this test CPU consumption
with 90 tasks is at 43%. With 52 cores, and two kswapd tasks (one per NUMA
node), kswapd can only be responsible for a little over 4% of the increase.
The rest is likely caused by 51,618 direct reclaims that scanned 1.2
billion pages over the five minute time period of the test.

Same test, more kswapd tasks:

Test #4: 4 kswapd threads per node
dd sy dd_cpu kswapd0 kswapd1 throughput  dr    pgscan_kswapd pgscan_direct
10 5  27.09  16.65   14.17   7842605.60  0     459105291     0
16 10 37.12  26.02   24.85   11352920.40 15    920527796     358515
22 11 36.94  37.13   35.82   13771869.60 0     1132169011     0
28 13 35.23  48.43   46.86   16089746.00 0     1312902070     0
34 15 33.37  53.02   55.69   18314856.40 0     1476169080     0
40 19 35.90  69.60   64.41   19836126.80 0     1629999149     0
46 22 36.82  88.55   57.20   20740216.40 0     1708478106     0
52 24 34.38  93.76   68.34   21758352.00 0     1794055559     0
58 24 30.51  79.20   82.33   22735594.00 0     1872794397     0
64 26 30.21  97.12   76.73   23302203.60 176   1916593721     4206821
70 33 32.92  92.91   92.87   23776588.00 3575  1817685086     85574159
76 37 31.62  91.20   89.83   24308196.80 4752  1812262569     113981763
82 29 25.53  93.23   92.33   24802791.20 306   2032093122     7350704
88 43 37.12  76.18   77.01   25145694.40 20310 1253204719     487048202
90 42 38.56  73.90   74.57   22516787.60 22774 1193637495     545463615

By increasing the number of kswapd threads, throughput increased by ~50%
while kernel mode CPU utilization decreased or stayed the same, likely due
to a decrease in the number of parallel tasks at any given time doing page
replacement.

Signed-off-by: Buddy Lumpkin <buddy.lumpkin@oracle.com>
Bug: 171351667
Link: https://lore.kernel.org/lkml/1522661062-39745-1-git-send-email-buddy.lumpkin@oracle.com
[charante@codeaurora.org]: Changes made to select number of kswapds through uapi
Change-Id: I8425cab7f40cbeaf65af0ea118c1a9ac7da0930e
Signed-off-by: Charan Teja Reddy <charante@codeaurora.org>
Flopster101 pushed a commit that referenced this pull request Jan 9, 2026
@Flopster101
ANDROID: vmscan: Support multiple kswapd threads per node
f4e1655 
Page replacement is handled in the Linux Kernel in one of two ways:

1) Asynchronously via kswapd
2) Synchronously, via direct reclaim

At page allocation time the allocating task is immediately given a page
from the zone free list allowing it to go right back to work doing
whatever it was doing; Probably directly or indirectly executing business
logic.

Just prior to satisfying the allocation, free pages is checked to see if
it has reached the zone low watermark and if so, kswapd is awakened.
Kswapd will start scanning pages looking for inactive pages to evict to
make room for new page allocations. The work of kswapd allows tasks to
continue allocating memory from their respective zone free list without
incurring any delay.

When the demand for free pages exceeds the rate that kswapd tasks can
supply them, page allocation works differently. Once the allocating task
finds that the number of free pages is at or below the zone min watermark,
the task will no longer pull pages from the free list. Instead, the task
will run the same CPU-bound routines as kswapd to satisfy its own
allocation by scanning and evicting pages. This is called a direct reclaim.

The time spent performing a direct reclaim can be substantial, often
taking tens to hundreds of milliseconds for small order0 allocations to
half a second or more for order9 huge-page allocations. In fact, kswapd is
not actually required on a linux system. It exists for the sole purpose of
optimizing performance by preventing direct reclaims.

When memory shortfall is sufficient to trigger direct reclaims, they can
occur in any task that is running on the system. A single aggressive
memory allocating task can set the stage for collateral damage to occur in
small tasks that rarely allocate additional memory. Consider the impact of
injecting an additional 100ms of latency when nscd allocates memory to
facilitate caching of a DNS query.

The presence of direct reclaims 10 years ago was a fairly reliable
indicator that too much was being asked of a Linux system. Kswapd was
likely wasting time scanning pages that were ineligible for eviction.
Adding RAM or reducing the working set size would usually make the problem
go away. Since then hardware has evolved to bring a new struggle for
kswapd. Storage speeds have increased by orders of magnitude while CPU
clock speeds stayed the same or even slowed down in exchange for more
cores per package. This presents a throughput problem for a single
threaded kswapd that will get worse with each generation of new hardware.

Test Details

NOTE: The tests below were run with shadow entries disabled. See the
associated patch and cover letter for details

The tests below were designed with the assumption that a kswapd bottleneck
is best demonstrated using filesystem reads. This way, the inactive list
will be full of clean pages, simplifying the analysis and allowing kswapd
to achieve the highest possible steal rate. Maximum steal rates for kswapd
are likely to be the same or lower for any other mix of page types on the
system.

Tests were run on a 2U Oracle X7-2L with 52 Intel Xeon Skylake 2GHz cores,
756GB of RAM and 8 x 3.6 TB NVMe Solid State Disk drives. Each drive has
an XFS file system mounted separately as /d0 through /d7. SSD drives
require multiple concurrent streams to show their potential, so I created
eleven 250GB zero-filled files on each drive so that I could test with
parallel reads.

The test script runs in multiple stages. At each stage, the number of dd
tasks run concurrently is increased by 2. I did not include all of the
test output for brevity.

During each stage dd tasks are launched to read from each drive in a round
robin fashion until the specified number of tasks for the stage has been
reached. Then iostat, vmstat and top are started in the background with 10
second intervals. After five minutes, all of the dd tasks are killed and
the iostat, vmstat and top output is parsed in order to report the
following:

CPU consumption
- sy - aggregate kernel mode CPU consumption from vmstat output. The value
       doesn't tend to fluctuate much so I just grab the highest value.
       Each sample is averaged over 10 seconds
- dd_cpu - for all of the dd tasks averaged across the top samples since
           there is a lot of variation.

Throughput
- in Kbytes
- Command is iostat -x -d 10 -g total

This first test performs reads using O_DIRECT in order to show the maximum
throughput that can be obtained using these drives. It also demonstrates
how rapidly throughput scales as the number of dd tasks are increased.

The dd command for this test looks like this:

Command Used: dd iflag=direct if=/d${i}/$n of=/dev/null bs=4M

Test #1: Direct IO
dd sy dd_cpu throughput
6  0  2.33   14726026.40
10 1  2.95   19954974.80
16 1  2.63   24419689.30
22 1  2.63   25430303.20
28 1  2.91   26026513.20
34 1  2.53   26178618.00
40 1  2.18   26239229.20
46 1  1.91   26250550.40
52 1  1.69   26251845.60
58 1  1.54   26253205.60
64 1  1.43   26253780.80
70 1  1.31   26254154.80
76 1  1.21   26253660.80
82 1  1.12   26254214.80
88 1  1.07   26253770.00
90 1  1.04   26252406.40

Throughput was close to peak with only 22 dd tasks. Very little system CPU
was consumed as expected as the drives DMA directly into the user address
space when using direct IO.

In this next test, the iflag=direct option is removed and we only run the
test until the pgscan_kswapd from /proc/vmstat starts to increment. At
that point metrics are parsed and reported and the pagecache contents are
dropped prior to the next test. Lather, rinse, repeat.

Test #2: standard file system IO, no page replacement
dd sy dd_cpu throughput
6  2  28.78  5134316.40
10 3  31.40  8051218.40
16 5  34.73  11438106.80
22 7  33.65  14140596.40
28 8  31.24  16393455.20
34 10 29.88  09e9cd8.60
40 11 28.33  19644159.60
46 11 25.05  20802497.60
52 13 26.92  22092370.00
58 13 23.29  22884881.20
64 14 23.12  23452248.80
70 15 22.40  23916468.00
76 16 22.06  24328737.20
82 17 20.97  24718693.20
88 16 18.57  25149404.40
90 16 18.31  25245565.60

Each read has to pause after the buffer in kernel space is populated while
those pages are added to the pagecache and copied into the user address
space. For this reason, more parallel streams are required to achieve peak
throughput. The copy operation consumes substantially more CPU than direct
IO as expected.

The next test measures throughput after kswapd starts running. This is the
same test only we wait for kswapd to wake up before we start collecting
metrics. The script actually keeps track of a few things that were not
mentioned earlier. It tracks direct reclaims and page scans by watching
the metrics in /proc/vmstat. CPU consumption for kswapd is tracked the
same way it is tracked for dd.

Since the test is 100% reads, you can assume that the page steal rate for
kswapd and direct reclaims is almost identical to the scan rate.

Test #3: 1 kswapd thread per node
dd sy dd_cpu kswapd0 kswapd1 throughput  dr    pgscan_kswapd pgscan_direct
10 4  26.07  28.56   27.03   7355924.40  0     459316976     0
16 7  34.94  69.33   69.66   10867895.20 0     872661643     0
22 10 36.03  93.99   99.33   13130613.60 489   1037654473    11268334
28 10 30.34  95.90   98.60   14601509.60 671   1182591373    15429142
34 14 34.77  97.50   99.23   16468012.00 10850 1069005644    249839515
40 17 36.32  91.49   97.11   17335987.60 18903 975417728     434467710
46 19 38.40  90.54   91.61   17705394.40 25369 855737040     582427973
52 22 40.88  83.97   83.70   17607680.40 31250 709532935     724282458
58 25 40.89  82.19   80.14   17976905.60 35060 657796473     804117540
64 28 41.77  73.49   75.20   18001910.00 39073 561813658     895289337
70 33 45.51  63.78   64.39   17061897.20 44523 379465571     1020726436
76 36 46.95  57.96   60.32   16964459.60 47717 291299464     1093172384
82 39 47.16  55.43   56.16   16949956.00 49479 247071062     1134163008
88 42 47.41  53.75   47.62   16930911.20 51521 195449924     1180442208
90 43 47.18  51.40   50.59   16864428.00 51618 190758156     1183203901

In the previous test where kswapd was not involved, the system-wide kernel
mode CPU consumption with 90 dd tasks was 16%. In this test CPU consumption
with 90 tasks is at 43%. With 52 cores, and two kswapd tasks (one per NUMA
node), kswapd can only be responsible for a little over 4% of the increase.
The rest is likely caused by 51,618 direct reclaims that scanned 1.2
billion pages over the five minute time period of the test.

Same test, more kswapd tasks:

Test #4: 4 kswapd threads per node
dd sy dd_cpu kswapd0 kswapd1 throughput  dr    pgscan_kswapd pgscan_direct
10 5  27.09  16.65   14.17   7842605.60  0     459105291     0
16 10 37.12  26.02   24.85   11352920.40 15    920527796     358515
22 11 36.94  37.13   35.82   13771869.60 0     1132169011     0
28 13 35.23  48.43   46.86   16089746.00 0     1312902070     0
34 15 33.37  53.02   55.69   18314856.40 0     1476169080     0
40 19 35.90  69.60   64.41   19836126.80 0     1629999149     0
46 22 36.82  88.55   57.20   20740216.40 0     1708478106     0
52 24 34.38  93.76   68.34   21758352.00 0     1794055559     0
58 24 30.51  79.20   82.33   22735594.00 0     1872794397     0
64 26 30.21  97.12   76.73   23302203.60 176   1916593721     4206821
70 33 32.92  92.91   92.87   23776588.00 3575  1817685086     85574159
76 37 31.62  91.20   89.83   24308196.80 4752  1812262569     113981763
82 29 25.53  93.23   92.33   24802791.20 306   2032093122     7350704
88 43 37.12  76.18   77.01   25145694.40 20310 1253204719     487048202
90 42 38.56  73.90   74.57   22516787.60 22774 1193637495     545463615

By increasing the number of kswapd threads, throughput increased by ~50%
while kernel mode CPU utilization decreased or stayed the same, likely due
to a decrease in the number of parallel tasks at any given time doing page
replacement.

Signed-off-by: Buddy Lumpkin <buddy.lumpkin@oracle.com>
Bug: 171351667
Link: https://lore.kernel.org/lkml/1522661062-39745-1-git-send-email-buddy.lumpkin@oracle.com
[charante@codeaurora.org]: Changes made to select number of kswapds through uapi
Change-Id: I8425cab7f40cbeaf65af0ea118c1a9ac7da0930e
Signed-off-by: Charan Teja Reddy <charante@codeaurora.org>
Flopster101 pushed a commit that referenced this pull request Jan 9, 2026
@Flopster101
ANDROID: vmscan: Support multiple kswapd threads per node
94b93a3 
Page replacement is handled in the Linux Kernel in one of two ways:

1) Asynchronously via kswapd
2) Synchronously, via direct reclaim

At page allocation time the allocating task is immediately given a page
from the zone free list allowing it to go right back to work doing
whatever it was doing; Probably directly or indirectly executing business
logic.

Just prior to satisfying the allocation, free pages is checked to see if
it has reached the zone low watermark and if so, kswapd is awakened.
Kswapd will start scanning pages looking for inactive pages to evict to
make room for new page allocations. The work of kswapd allows tasks to
continue allocating memory from their respective zone free list without
incurring any delay.

When the demand for free pages exceeds the rate that kswapd tasks can
supply them, page allocation works differently. Once the allocating task
finds that the number of free pages is at or below the zone min watermark,
the task will no longer pull pages from the free list. Instead, the task
will run the same CPU-bound routines as kswapd to satisfy its own
allocation by scanning and evicting pages. This is called a direct reclaim.

The time spent performing a direct reclaim can be substantial, often
taking tens to hundreds of milliseconds for small order0 allocations to
half a second or more for order9 huge-page allocations. In fact, kswapd is
not actually required on a linux system. It exists for the sole purpose of
optimizing performance by preventing direct reclaims.

When memory shortfall is sufficient to trigger direct reclaims, they can
occur in any task that is running on the system. A single aggressive
memory allocating task can set the stage for collateral damage to occur in
small tasks that rarely allocate additional memory. Consider the impact of
injecting an additional 100ms of latency when nscd allocates memory to
facilitate caching of a DNS query.

The presence of direct reclaims 10 years ago was a fairly reliable
indicator that too much was being asked of a Linux system. Kswapd was
likely wasting time scanning pages that were ineligible for eviction.
Adding RAM or reducing the working set size would usually make the problem
go away. Since then hardware has evolved to bring a new struggle for
kswapd. Storage speeds have increased by orders of magnitude while CPU
clock speeds stayed the same or even slowed down in exchange for more
cores per package. This presents a throughput problem for a single
threaded kswapd that will get worse with each generation of new hardware.

Test Details

NOTE: The tests below were run with shadow entries disabled. See the
associated patch and cover letter for details

The tests below were designed with the assumption that a kswapd bottleneck
is best demonstrated using filesystem reads. This way, the inactive list
will be full of clean pages, simplifying the analysis and allowing kswapd
to achieve the highest possible steal rate. Maximum steal rates for kswapd
are likely to be the same or lower for any other mix of page types on the
system.

Tests were run on a 2U Oracle X7-2L with 52 Intel Xeon Skylake 2GHz cores,
756GB of RAM and 8 x 3.6 TB NVMe Solid State Disk drives. Each drive has
an XFS file system mounted separately as /d0 through /d7. SSD drives
require multiple concurrent streams to show their potential, so I created
eleven 250GB zero-filled files on each drive so that I could test with
parallel reads.

The test script runs in multiple stages. At each stage, the number of dd
tasks run concurrently is increased by 2. I did not include all of the
test output for brevity.

During each stage dd tasks are launched to read from each drive in a round
robin fashion until the specified number of tasks for the stage has been
reached. Then iostat, vmstat and top are started in the background with 10
second intervals. After five minutes, all of the dd tasks are killed and
the iostat, vmstat and top output is parsed in order to report the
following:

CPU consumption
- sy - aggregate kernel mode CPU consumption from vmstat output. The value
       doesn't tend to fluctuate much so I just grab the highest value.
       Each sample is averaged over 10 seconds
- dd_cpu - for all of the dd tasks averaged across the top samples since
           there is a lot of variation.

Throughput
- in Kbytes
- Command is iostat -x -d 10 -g total

This first test performs reads using O_DIRECT in order to show the maximum
throughput that can be obtained using these drives. It also demonstrates
how rapidly throughput scales as the number of dd tasks are increased.

The dd command for this test looks like this:

Command Used: dd iflag=direct if=/d${i}/$n of=/dev/null bs=4M

Test #1: Direct IO
dd sy dd_cpu throughput
6  0  2.33   14726026.40
10 1  2.95   19954974.80
16 1  2.63   24419689.30
22 1  2.63   25430303.20
28 1  2.91   26026513.20
34 1  2.53   26178618.00
40 1  2.18   26239229.20
46 1  1.91   26250550.40
52 1  1.69   26251845.60
58 1  1.54   26253205.60
64 1  1.43   26253780.80
70 1  1.31   26254154.80
76 1  1.21   26253660.80
82 1  1.12   26254214.80
88 1  1.07   26253770.00
90 1  1.04   26252406.40

Throughput was close to peak with only 22 dd tasks. Very little system CPU
was consumed as expected as the drives DMA directly into the user address
space when using direct IO.

In this next test, the iflag=direct option is removed and we only run the
test until the pgscan_kswapd from /proc/vmstat starts to increment. At
that point metrics are parsed and reported and the pagecache contents are
dropped prior to the next test. Lather, rinse, repeat.

Test #2: standard file system IO, no page replacement
dd sy dd_cpu throughput
6  2  28.78  5134316.40
10 3  31.40  8051218.40
16 5  34.73  11438106.80
22 7  33.65  14140596.40
28 8  31.24  16393455.20
34 10 29.88  09e9cd8.60
40 11 28.33  19644159.60
46 11 25.05  20802497.60
52 13 26.92  22092370.00
58 13 23.29  22884881.20
64 14 23.12  23452248.80
70 15 22.40  23916468.00
76 16 22.06  24328737.20
82 17 20.97  24718693.20
88 16 18.57  25149404.40
90 16 18.31  25245565.60

Each read has to pause after the buffer in kernel space is populated while
those pages are added to the pagecache and copied into the user address
space. For this reason, more parallel streams are required to achieve peak
throughput. The copy operation consumes substantially more CPU than direct
IO as expected.

The next test measures throughput after kswapd starts running. This is the
same test only we wait for kswapd to wake up before we start collecting
metrics. The script actually keeps track of a few things that were not
mentioned earlier. It tracks direct reclaims and page scans by watching
the metrics in /proc/vmstat. CPU consumption for kswapd is tracked the
same way it is tracked for dd.

Since the test is 100% reads, you can assume that the page steal rate for
kswapd and direct reclaims is almost identical to the scan rate.

Test #3: 1 kswapd thread per node
dd sy dd_cpu kswapd0 kswapd1 throughput  dr    pgscan_kswapd pgscan_direct
10 4  26.07  28.56   27.03   7355924.40  0     459316976     0
16 7  34.94  69.33   69.66   10867895.20 0     872661643     0
22 10 36.03  93.99   99.33   13130613.60 489   1037654473    11268334
28 10 30.34  95.90   98.60   14601509.60 671   1182591373    15429142
34 14 34.77  97.50   99.23   16468012.00 10850 1069005644    249839515
40 17 36.32  91.49   97.11   17335987.60 18903 975417728     434467710
46 19 38.40  90.54   91.61   17705394.40 25369 855737040     582427973
52 22 40.88  83.97   83.70   17607680.40 31250 709532935     724282458
58 25 40.89  82.19   80.14   17976905.60 35060 657796473     804117540
64 28 41.77  73.49   75.20   18001910.00 39073 561813658     895289337
70 33 45.51  63.78   64.39   17061897.20 44523 379465571     1020726436
76 36 46.95  57.96   60.32   16964459.60 47717 291299464     1093172384
82 39 47.16  55.43   56.16   16949956.00 49479 247071062     1134163008
88 42 47.41  53.75   47.62   16930911.20 51521 195449924     1180442208
90 43 47.18  51.40   50.59   16864428.00 51618 190758156     1183203901

In the previous test where kswapd was not involved, the system-wide kernel
mode CPU consumption with 90 dd tasks was 16%. In this test CPU consumption
with 90 tasks is at 43%. With 52 cores, and two kswapd tasks (one per NUMA
node), kswapd can only be responsible for a little over 4% of the increase.
The rest is likely caused by 51,618 direct reclaims that scanned 1.2
billion pages over the five minute time period of the test.

Same test, more kswapd tasks:

Test #4: 4 kswapd threads per node
dd sy dd_cpu kswapd0 kswapd1 throughput  dr    pgscan_kswapd pgscan_direct
10 5  27.09  16.65   14.17   7842605.60  0     459105291     0
16 10 37.12  26.02   24.85   11352920.40 15    920527796     358515
22 11 36.94  37.13   35.82   13771869.60 0     1132169011     0
28 13 35.23  48.43   46.86   16089746.00 0     1312902070     0
34 15 33.37  53.02   55.69   18314856.40 0     1476169080     0
40 19 35.90  69.60   64.41   19836126.80 0     1629999149     0
46 22 36.82  88.55   57.20   20740216.40 0     1708478106     0
52 24 34.38  93.76   68.34   21758352.00 0     1794055559     0
58 24 30.51  79.20   82.33   22735594.00 0     1872794397     0
64 26 30.21  97.12   76.73   23302203.60 176   1916593721     4206821
70 33 32.92  92.91   92.87   23776588.00 3575  1817685086     85574159
76 37 31.62  91.20   89.83   24308196.80 4752  1812262569     113981763
82 29 25.53  93.23   92.33   24802791.20 306   2032093122     7350704
88 43 37.12  76.18   77.01   25145694.40 20310 1253204719     487048202
90 42 38.56  73.90   74.57   22516787.60 22774 1193637495     545463615

By increasing the number of kswapd threads, throughput increased by ~50%
while kernel mode CPU utilization decreased or stayed the same, likely due
to a decrease in the number of parallel tasks at any given time doing page
replacement.

Signed-off-by: Buddy Lumpkin <buddy.lumpkin@oracle.com>
Bug: 171351667
Link: https://lore.kernel.org/lkml/1522661062-39745-1-git-send-email-buddy.lumpkin@oracle.com
[charante@codeaurora.org]: Changes made to select number of kswapds through uapi
Change-Id: I8425cab7f40cbeaf65af0ea118c1a9ac7da0930e
Signed-off-by: Charan Teja Reddy <charante@codeaurora.org>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.