linux-doc.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Nico Pache <npache@redhat.com>
To: David Hildenbrand <david@redhat.com>
Cc: linux-mm@kvack.org, linux-doc@vger.kernel.org,
	 linux-kernel@vger.kernel.org,
	linux-trace-kernel@vger.kernel.org,  ziy@nvidia.com,
	baolin.wang@linux.alibaba.com, lorenzo.stoakes@oracle.com,
	 Liam.Howlett@oracle.com, ryan.roberts@arm.com, dev.jain@arm.com,
	 corbet@lwn.net, rostedt@goodmis.org, mhiramat@kernel.org,
	 mathieu.desnoyers@efficios.com, akpm@linux-foundation.org,
	baohua@kernel.org,  willy@infradead.org, peterx@redhat.com,
	wangkefeng.wang@huawei.com,  usamaarif642@gmail.com,
	sunnanyong@huawei.com, vishal.moola@gmail.com,
	 thomas.hellstrom@linux.intel.com, yang@os.amperecomputing.com,
	 kirill.shutemov@linux.intel.com, aarcange@redhat.com,
	raquini@redhat.com,  anshuman.khandual@arm.com,
	catalin.marinas@arm.com, tiwai@suse.de,  will@kernel.org,
	dave.hansen@linux.intel.com, jack@suse.cz, cl@gentwo.org,
	 jglisse@google.com, surenb@google.com, zokeefe@google.com,
	hannes@cmpxchg.org,  rientjes@google.com, mhocko@suse.com,
	rdunlap@infradead.org, hughd@google.com
Subject: Re: [PATCH v9 06/14] khugepaged: introduce collapse_scan_bitmap for mTHP support
Date: Thu, 17 Jul 2025 01:23:42 -0600	[thread overview]
Message-ID: <CAA1CXcAFyQDZ_n2TLSLVCY0dqt0tudwaOceb+i9tDOLNo9rzZw@mail.gmail.com> (raw)
In-Reply-To: <82cd57c2-d72f-423a-8dbc-d9b64d1d469b@redhat.com>

On Wed, Jul 16, 2025 at 8:03 AM David Hildenbrand <david@redhat.com> wrote:
>
> On 14.07.25 02:31, Nico Pache wrote:
> > khugepaged scans anons PMD ranges for potential collapse to a hugepage.
> > To add mTHP support we use this scan to instead record chunks of utilized
> > sections of the PMD.
> >
> > collapse_scan_bitmap uses a stack struct to recursively scan a bitmap
> > that represents chunks of utilized regions. We can then determine what
> > mTHP size fits best and in the following patch, we set this bitmap while
> > scanning the anon PMD. A minimum collapse order of 2 is used as this is
> > the lowest order supported by anon memory.
> >
> > max_ptes_none is used as a scale to determine how "full" an order must
> > be before being considered for collapse.
> >
> > When attempting to collapse an order that has its order set to "always"
> > lets always collapse to that order in a greedy manner without
> > considering the number of bits set.
> >
> > Signed-off-by: Nico Pache <npache@redhat.com>
>
> Any reason this should not be squashed into the actual mTHP collapse patch?
I wanted to keep them seperate to conceptually separate the bitmap and
collapse, but given youre the second person to point this out, I went
ahead and squashed them and their commit messages to be one commit.
>
> In particular
>
> a) The locking changes look weird without the bigger context
>
> b) The compiler complains about unused functions
>
> > ---
> >   include/linux/khugepaged.h |  4 ++
> >   mm/khugepaged.c            | 94 ++++++++++++++++++++++++++++++++++----
> >   2 files changed, 89 insertions(+), 9 deletions(-)
> >
> > diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h
> > index ff6120463745..0f957711a117 100644
> > --- a/include/linux/khugepaged.h
> > +++ b/include/linux/khugepaged.h
> > @@ -1,6 +1,10 @@
> >   /* SPDX-License-Identifier: GPL-2.0 */
> >   #ifndef _LINUX_KHUGEPAGED_H
> >   #define _LINUX_KHUGEPAGED_H
> > +#define KHUGEPAGED_MIN_MTHP_ORDER    2
> > +#define KHUGEPAGED_MIN_MTHP_NR       (1<<KHUGEPAGED_MIN_MTHP_ORDER)
>
> "1 << "
Ah there is a mix of these ("1<<" vs "1 << ") being used across the
kernel. Fixed it thank you.
>
> > +#define MAX_MTHP_BITMAP_SIZE  (1 << (ilog2(MAX_PTRS_PER_PTE) - KHUGEPAGED_MIN_MTHP_ORDER))
> > +#define MTHP_BITMAP_SIZE  (1 << (HPAGE_PMD_ORDER - KHUGEPAGED_MIN_MTHP_ORDER))
> >
> >   extern unsigned int khugepaged_max_ptes_none __read_mostly;
> >   #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> > diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> > index ee54e3c1db4e..59b2431ca616 100644
> > --- a/mm/khugepaged.c
> > +++ b/mm/khugepaged.c
> > @@ -94,6 +94,11 @@ static DEFINE_READ_MOSTLY_HASHTABLE(mm_slots_hash, MM_SLOTS_HASH_BITS);
> >
> >   static struct kmem_cache *mm_slot_cache __ro_after_init;
> >
> > +struct scan_bit_state {
> > +     u8 order;
> > +     u16 offset;
> > +};
> > +
> >   struct collapse_control {
> >       bool is_khugepaged;
> >
> > @@ -102,6 +107,18 @@ struct collapse_control {
> >
> >       /* nodemask for allocation fallback */
> >       nodemask_t alloc_nmask;
> > +
> > +     /*
> > +      * bitmap used to collapse mTHP sizes.
> > +      * 1bit = order KHUGEPAGED_MIN_MTHP_ORDER mTHP
> > +      */
> > +     DECLARE_BITMAP(mthp_bitmap, MAX_MTHP_BITMAP_SIZE);
> > +     DECLARE_BITMAP(mthp_bitmap_temp, MAX_MTHP_BITMAP_SIZE);
> > +     struct scan_bit_state mthp_bitmap_stack[MAX_MTHP_BITMAP_SIZE];
> > +};
> > +
> > +struct collapse_control khugepaged_collapse_control = {
> > +     .is_khugepaged = true,
> >   };
> >
> >   /**
> > @@ -838,10 +855,6 @@ static void khugepaged_alloc_sleep(void)
> >       remove_wait_queue(&khugepaged_wait, &wait);
> >   }
> >
> > -struct collapse_control khugepaged_collapse_control = {
> > -     .is_khugepaged = true,
> > -};
> > -
> >   static bool collapse_scan_abort(int nid, struct collapse_control *cc)
> >   {
> >       int i;
> > @@ -1115,7 +1128,8 @@ static int alloc_charge_folio(struct folio **foliop, struct mm_struct *mm,
> >
> >   static int collapse_huge_page(struct mm_struct *mm, unsigned long address,
> >                             int referenced, int unmapped,
> > -                           struct collapse_control *cc)
> > +                           struct collapse_control *cc, bool *mmap_locked,
> > +                               u8 order, u16 offset)
>
> Indent broken.
Fixed!
>
> >   {
> >       LIST_HEAD(compound_pagelist);
> >       pmd_t *pmd, _pmd;
> > @@ -1134,8 +1148,12 @@ static int collapse_huge_page(struct mm_struct *mm, unsigned long address,
> >        * The allocation can take potentially a long time if it involves
> >        * sync compaction, and we do not need to hold the mmap_lock during
> >        * that. We will recheck the vma after taking it again in write mode.
> > +      * If collapsing mTHPs we may have already released the read_lock.
> >        */
> > -     mmap_read_unlock(mm);
> > +     if (*mmap_locked) {
> > +             mmap_read_unlock(mm);
> > +             *mmap_locked = false;
> > +     }
> >
> >       result = alloc_charge_folio(&folio, mm, cc, HPAGE_PMD_ORDER);
> >       if (result != SCAN_SUCCEED)
> > @@ -1272,12 +1290,72 @@ static int collapse_huge_page(struct mm_struct *mm, unsigned long address,
> >   out_up_write:
> >       mmap_write_unlock(mm);
> >   out_nolock:
> > +     *mmap_locked = false;
> >       if (folio)
> >               folio_put(folio);
> >       trace_mm_collapse_huge_page(mm, result == SCAN_SUCCEED, result);
> >       return result;
> >   }
> >
> > +/* Recursive function to consume the bitmap */
> > +static int collapse_scan_bitmap(struct mm_struct *mm, unsigned long address,
> > +                     int referenced, int unmapped, struct collapse_control *cc,
> > +                     bool *mmap_locked, unsigned long enabled_orders)
> > +{
> > +     u8 order, next_order;
> > +     u16 offset, mid_offset;
> > +     int num_chunks;
> > +     int bits_set, threshold_bits;
> > +     int top = -1;
> > +     int collapsed = 0;
> > +     int ret;
> > +     struct scan_bit_state state;
> > +     bool is_pmd_only = (enabled_orders == (1 << HPAGE_PMD_ORDER));
> > +
> > +     cc->mthp_bitmap_stack[++top] = (struct scan_bit_state)
> > +             { HPAGE_PMD_ORDER - KHUGEPAGED_MIN_MTHP_ORDER, 0 };
> > +
> > +     while (top >= 0) {
> > +             state = cc->mthp_bitmap_stack[top--];
> > +             order = state.order + KHUGEPAGED_MIN_MTHP_ORDER;
> > +             offset = state.offset;
> > +             num_chunks = 1 << (state.order);
> > +             // Skip mTHP orders that are not enabled
>
>
> /* */
>
> Same applies to the other instances.
Thank you! Fixed all instances
>
> --
> Cheers,
>
> David / dhildenb
>


  reply	other threads:[~2025-07-17  7:24 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-14  0:31 [PATCH v9 00/14] khugepaged: mTHP support Nico Pache
2025-07-14  0:31 ` [PATCH v9 01/14] khugepaged: rename hpage_collapse_* to collapse_* Nico Pache
2025-07-15 15:39   ` David Hildenbrand
2025-07-16 14:29   ` Liam R. Howlett
2025-07-16 15:20     ` David Hildenbrand
2025-07-17  7:21     ` Nico Pache
2025-07-25 16:43   ` Lorenzo Stoakes
2025-07-25 22:35     ` Nico Pache
2025-07-14  0:31 ` [PATCH v9 02/14] introduce collapse_single_pmd to unify khugepaged and madvise_collapse Nico Pache
2025-07-15 15:53   ` David Hildenbrand
2025-07-23  1:56     ` Nico Pache
2025-07-16 15:12   ` Liam R. Howlett
2025-07-23  1:55     ` Nico Pache
2025-07-14  0:31 ` [PATCH v9 03/14] khugepaged: generalize hugepage_vma_revalidate for mTHP support Nico Pache
2025-07-15 15:55   ` David Hildenbrand
2025-07-14  0:31 ` [PATCH v9 04/14] khugepaged: generalize alloc_charge_folio() Nico Pache
2025-07-16 13:46   ` David Hildenbrand
2025-07-17  7:22     ` Nico Pache
2025-07-14  0:31 ` [PATCH v9 05/14] khugepaged: generalize __collapse_huge_page_* for mTHP support Nico Pache
2025-07-16 13:52   ` David Hildenbrand
2025-07-17  7:22     ` Nico Pache
2025-07-16 14:02   ` David Hildenbrand
2025-07-17  7:23     ` Nico Pache
2025-07-17 15:54     ` Lorenzo Stoakes
2025-07-25 16:09   ` Lorenzo Stoakes
2025-07-25 22:37     ` Nico Pache
2025-07-14  0:31 ` [PATCH v9 06/14] khugepaged: introduce collapse_scan_bitmap " Nico Pache
2025-07-16 14:03   ` David Hildenbrand
2025-07-17  7:23     ` Nico Pache [this message]
2025-07-16 15:38   ` Liam R. Howlett
2025-07-17  7:24     ` Nico Pache
2025-07-14  0:32 ` [PATCH v9 07/14] khugepaged: add " Nico Pache
2025-07-14  0:32 ` [PATCH v9 08/14] khugepaged: skip collapsing mTHP to smaller orders Nico Pache
2025-07-16 14:32   ` David Hildenbrand
2025-07-17  7:24     ` Nico Pache
2025-07-14  0:32 ` [PATCH v9 09/14] khugepaged: avoid unnecessary mTHP collapse attempts Nico Pache
2025-07-18  2:14   ` Baolin Wang
2025-07-18 22:34     ` Nico Pache
2025-07-14  0:32 ` [PATCH v9 10/14] khugepaged: allow khugepaged to check all anonymous mTHP orders Nico Pache
2025-07-16 15:28   ` David Hildenbrand
2025-07-17  7:25     ` Nico Pache
2025-07-18  8:40       ` David Hildenbrand
2025-07-14  0:32 ` [PATCH v9 11/14] khugepaged: kick khugepaged for enabling none-PMD-sized mTHPs Nico Pache
2025-07-14  0:32 ` [PATCH v9 12/14] khugepaged: improve tracepoints for mTHP orders Nico Pache
2025-07-22 15:39   ` David Hildenbrand
2025-07-14  0:32 ` [PATCH v9 13/14] khugepaged: add per-order mTHP khugepaged stats Nico Pache
2025-07-18  5:04   ` Baolin Wang
2025-07-18 21:00     ` Nico Pache
2025-07-19  4:42       ` Baolin Wang
2025-07-14  0:32 ` [PATCH v9 14/14] Documentation: mm: update the admin guide for mTHP collapse Nico Pache
2025-07-15  0:39 ` [PATCH v9 00/14] khugepaged: mTHP support Andrew Morton

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAA1CXcAFyQDZ_n2TLSLVCY0dqt0tudwaOceb+i9tDOLNo9rzZw@mail.gmail.com \
    --to=npache@redhat.com \
    --cc=Liam.Howlett@oracle.com \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=anshuman.khandual@arm.com \
    --cc=baohua@kernel.org \
    --cc=baolin.wang@linux.alibaba.com \
    --cc=catalin.marinas@arm.com \
    --cc=cl@gentwo.org \
    --cc=corbet@lwn.net \
    --cc=dave.hansen@linux.intel.com \
    --cc=david@redhat.com \
    --cc=dev.jain@arm.com \
    --cc=hannes@cmpxchg.org \
    --cc=hughd@google.com \
    --cc=jack@suse.cz \
    --cc=jglisse@google.com \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mhiramat@kernel.org \
    --cc=mhocko@suse.com \
    --cc=peterx@redhat.com \
    --cc=raquini@redhat.com \
    --cc=rdunlap@infradead.org \
    --cc=rientjes@google.com \
    --cc=rostedt@goodmis.org \
    --cc=ryan.roberts@arm.com \
    --cc=sunnanyong@huawei.com \
    --cc=surenb@google.com \
    --cc=thomas.hellstrom@linux.intel.com \
    --cc=tiwai@suse.de \
    --cc=usamaarif642@gmail.com \
    --cc=vishal.moola@gmail.com \
    --cc=wangkefeng.wang@huawei.com \
    --cc=will@kernel.org \
    --cc=willy@infradead.org \
    --cc=yang@os.amperecomputing.com \
    --cc=ziy@nvidia.com \
    --cc=zokeefe@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).