| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <dbb631dc-32d3-4b2d-a23c-968c7a61001c@nvidia.com>
Date: Mon, 15 Sep 2025 13:54:01 +1000
From: Balbir Singh <balbirs@...dia.com>
To: David Hildenbrand <david@...hat.com>, linux-kernel@...r.kernel.org,
linux-mm@...ck.org
Cc: damon@...ts.linux.dev, dri-devel@...ts.freedesktop.org,
Andrew Morton <akpm@...ux-foundation.org>, Zi Yan <ziy@...dia.com>,
Joshua Hahn <joshua.hahnjy@...il.com>, Rakie Kim <rakie.kim@...com>,
Byungchul Park <byungchul@...com>, Gregory Price <gourry@...rry.net>,
Ying Huang <ying.huang@...ux.alibaba.com>,
Alistair Popple <apopple@...dia.com>, Oscar Salvador <osalvador@...e.de>,
Lorenzo Stoakes <lorenzo.stoakes@...cle.com>,
Baolin Wang <baolin.wang@...ux.alibaba.com>,
"Liam R. Howlett" <Liam.Howlett@...cle.com>, Nico Pache <npache@...hat.com>,
Ryan Roberts <ryan.roberts@....com>, Dev Jain <dev.jain@....com>,
Barry Song <baohua@...nel.org>, Lyude Paul <lyude@...hat.com>,
Danilo Krummrich <dakr@...nel.org>, David Airlie <airlied@...il.com>,
Simona Vetter <simona@...ll.ch>, Ralph Campbell <rcampbell@...dia.com>,
Mika Penttilä <mpenttil@...hat.com>,
Matthew Brost <matthew.brost@...el.com>,
Francois Dugast <francois.dugast@...el.com>
Subject: Re: [v5 04/15] mm/huge_memory: implement device-private THP splitting
On 9/11/25 22:31, David Hildenbrand wrote:
> On 08.09.25 02:04, Balbir Singh wrote:
>> Add support for splitting device-private THP folios, enabling fallback
>> to smaller page sizes when large page allocation or migration fails.
>>
>> Key changes:
>> - split_huge_pmd(): Handle device-private PMD entries during splitting
>> - Preserve RMAP_EXCLUSIVE semantics for anonymous exclusive folios
>> - Skip RMP_USE_SHARED_ZEROPAGE for device-private entries as they
>> don't support shared zero page semantics
>>
>> Cc: Andrew Morton <akpm@...ux-foundation.org>
>> Cc: David Hildenbrand <david@...hat.com>
>> Cc: Zi Yan <ziy@...dia.com>
>> Cc: Joshua Hahn <joshua.hahnjy@...il.com>
>> Cc: Rakie Kim <rakie.kim@...com>
>> Cc: Byungchul Park <byungchul@...com>
>> Cc: Gregory Price <gourry@...rry.net>
>> Cc: Ying Huang <ying.huang@...ux.alibaba.com>
>> Cc: Alistair Popple <apopple@...dia.com>
>> Cc: Oscar Salvador <osalvador@...e.de>
>> Cc: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
>> Cc: Baolin Wang <baolin.wang@...ux.alibaba.com>
>> Cc: "Liam R. Howlett" <Liam.Howlett@...cle.com>
>> Cc: Nico Pache <npache@...hat.com>
>> Cc: Ryan Roberts <ryan.roberts@....com>
>> Cc: Dev Jain <dev.jain@....com>
>> Cc: Barry Song <baohua@...nel.org>
>> Cc: Lyude Paul <lyude@...hat.com>
>> Cc: Danilo Krummrich <dakr@...nel.org>
>> Cc: David Airlie <airlied@...il.com>
>> Cc: Simona Vetter <simona@...ll.ch>
>> Cc: Ralph Campbell <rcampbell@...dia.com>
>> Cc: Mika Penttilä <mpenttil@...hat.com>
>> Cc: Matthew Brost <matthew.brost@...el.com>
>> Cc: Francois Dugast <francois.dugast@...el.com>
>>
>> Signed-off-by: Balbir Singh <balbirs@...dia.com>
>> ---
>> mm/huge_memory.c | 129 +++++++++++++++++++++++++++++++++--------------
>> 1 file changed, 91 insertions(+), 38 deletions(-)
>>
>> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
>> index 337d8e3dd837..b720870c04b2 100644
>> --- a/mm/huge_memory.c
>> +++ b/mm/huge_memory.c
>> @@ -2880,16 +2880,19 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>> struct page *page;
>> pgtable_t pgtable;
>> pmd_t old_pmd, _pmd;
>> - bool young, write, soft_dirty, pmd_migration = false, uffd_wp = false;
>> - bool anon_exclusive = false, dirty = false;
>> + bool young, write, soft_dirty, uffd_wp = false;
>> + bool anon_exclusive = false, dirty = false, present = false;
>> unsigned long addr;
>> pte_t *pte;
>> int i;
>> + swp_entry_t swp_entry;
>> VM_BUG_ON(haddr & ~HPAGE_PMD_MASK);
>> VM_BUG_ON_VMA(vma->vm_start > haddr, vma);
>> VM_BUG_ON_VMA(vma->vm_end < haddr + HPAGE_PMD_SIZE, vma);
>> - VM_BUG_ON(!is_pmd_migration_entry(*pmd) && !pmd_trans_huge(*pmd));
>> +
>> + VM_WARN_ON(!is_pmd_migration_entry(*pmd) && !pmd_trans_huge(*pmd) &&
>> + !is_pmd_device_private_entry(*pmd));
>>
>
> Indentation. But I do wonder if we want a helper to do a more
> efficient
>
The indentation at my end is fine, do you mean you want to see the conditions aligned?
> is_pmd_migration_entry() || is_pmd_device_private_entry()
>
> If only I could come up with a good name ... any ideas?
>
> is_non_present_folio_entry()
>
> maybe?
>
There is is_pfn_swap_entry(), but that includes hwpoison entries as well.
> Well, there is device-exclusive .... but that would not be reachable on these paths yet, ever.
>
>
>> count_vm_event(THP_SPLIT_PMD);
>> @@ -2937,18 +2940,43 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>> return __split_huge_zero_page_pmd(vma, haddr, pmd);
>> }
>> - pmd_migration = is_pmd_migration_entry(*pmd);
>> - if (unlikely(pmd_migration)) {
>> - swp_entry_t entry;
>> + present = pmd_present(*pmd);
>> + if (unlikely(!present)) {
>
> I hate this whole function. But maybe in this case it's better
> to just have here
>
> if (is_pmd_migration_entry(old_pmd)) {
>
> } else if (is_pmd_device_private_entry(old_pmd)) {
>
> There is not much shared code and the helps reduce the indentation level.
>
We can definitely try this
>> + swp_entry = pmd_to_swp_entry(*pmd);
>> old_pmd = *pmd;
>> - entry = pmd_to_swp_entry(old_pmd);
>> - page = pfn_swap_entry_to_page(entry);
>> - write = is_writable_migration_entry(entry);
>> - if (PageAnon(page))
>> - anon_exclusive = is_readable_exclusive_migration_entry(entry);
>> - young = is_migration_entry_young(entry);
>> - dirty = is_migration_entry_dirty(entry);
>> +
>> + folio = pfn_swap_entry_folio(swp_entry);
>> + VM_WARN_ON(!is_migration_entry(swp_entry) &&
>> + !is_device_private_entry(swp_entry));
>
> Indentation.
Same as above, checkpatch.pl does not seem to complain
>
>> + page = pfn_swap_entry_to_page(swp_entry);
>> +
>> + if (is_pmd_migration_entry(old_pmd)) {
>> + write = is_writable_migration_entry(swp_entry);
>> + if (PageAnon(page))
>> + anon_exclusive =
>> + is_readable_exclusive_migration_entry(
>> + swp_entry);
>
> Single line please, this is unreadable.
>
Sure, will do, I double checked and it seems like 100 columns is fine as per checkpatch.pl
>> + young = is_migration_entry_young(swp_entry);
>> + dirty = is_migration_entry_dirty(swp_entry);
>> + } else if (is_pmd_device_private_entry(old_pmd)) {
>> + write = is_writable_device_private_entry(swp_entry);
>> + anon_exclusive = PageAnonExclusive(page);
>> + if (freeze && anon_exclusive &&
>> + folio_try_share_anon_rmap_pmd(folio, page))
>> + freeze = false;
>> + if (!freeze) {
>> + rmap_t rmap_flags = RMAP_NONE;
>> +
>> + folio_ref_add(folio, HPAGE_PMD_NR - 1);
>> + if (anon_exclusive)
>> + rmap_flags |= RMAP_EXCLUSIVE;
>> +
>> + folio_add_anon_rmap_ptes(folio, page, HPAGE_PMD_NR,
>> + vma, haddr, rmap_flags);
>> + }
>> + }
>> +
>> soft_dirty = pmd_swp_soft_dirty(old_pmd);
>> uffd_wp = pmd_swp_uffd_wp(old_pmd);
>> } else {
>> @@ -3034,30 +3062,49 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>> * Note that NUMA hinting access restrictions are not transferred to
>> * avoid any possibility of altering permissions across VMAs.
>> */
>> - if (freeze || pmd_migration) {
>> + if (freeze || !present) {
>
> Here too, I wonder if we should just handle device-private completely separately for now.
>
For all practical purposes they are inside the if statement. freeze and is_migration need to go together.
>> for (i = 0, addr = haddr; i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE) {
>> pte_t entry;
>> - swp_entry_t swp_entry;
>> -
>> - if (write)
>> - swp_entry = make_writable_migration_entry(
>> - page_to_pfn(page + i));
>> - else if (anon_exclusive)
>> - swp_entry = make_readable_exclusive_migration_entry(
>> - page_to_pfn(page + i));
>> - else
>> - swp_entry = make_readable_migration_entry(
>> - page_to_pfn(page + i));
>> - if (young)
>> - swp_entry = make_migration_entry_young(swp_entry);
>> - if (dirty)
>> - swp_entry = make_migration_entry_dirty(swp_entry);
>> - entry = swp_entry_to_pte(swp_entry);
>> - if (soft_dirty)
>> - entry = pte_swp_mksoft_dirty(entry);
>> - if (uffd_wp)
>> - entry = pte_swp_mkuffd_wp(entry);
>> -
>> + if (freeze || is_migration_entry(swp_entry)) {
>> + if (write)
>> + swp_entry = make_writable_migration_entry(
>> + page_to_pfn(page + i));
>> + else if (anon_exclusive)
>> + swp_entry = make_readable_exclusive_migration_entry(
>> + page_to_pfn(page + i));
>> + else
>> + swp_entry = make_readable_migration_entry(
>> + page_to_pfn(page + i));
>> + if (young)
>> + swp_entry = make_migration_entry_young(swp_entry);
>> + if (dirty)
>> + swp_entry = make_migration_entry_dirty(swp_entry);
>> + entry = swp_entry_to_pte(swp_entry);
>> + if (soft_dirty)
>> + entry = pte_swp_mksoft_dirty(entry);
>> + if (uffd_wp)
>> + entry = pte_swp_mkuffd_wp(entry);
>> + } else {
>> + /*
>> + * anon_exclusive was already propagated to the relevant
>> + * pages corresponding to the pte entries when freeze
>> + * is false.
>> + */
>> + if (write)
>> + swp_entry = make_writable_device_private_entry(
>> + page_to_pfn(page + i));
>> + else
>> + swp_entry = make_readable_device_private_entry(
>> + page_to_pfn(page + i));
>> + /*
>> + * Young and dirty bits are not progated via swp_entry
>> + */
>> + entry = swp_entry_to_pte(swp_entry);
>> + if (soft_dirty)
>> + entry = pte_swp_mksoft_dirty(entry);
>> + if (uffd_wp)
>> + entry = pte_swp_mkuffd_wp(entry);
>> + }
>> VM_WARN_ON(!pte_none(ptep_get(pte + i)));
>> set_pte_at(mm, addr, pte + i, entry);
>> }
>> @@ -3084,7 +3131,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>> }
>> pte_unmap(pte);
>> - if (!pmd_migration)
>> + if (!is_pmd_migration_entry(*pmd))
>> folio_remove_rmap_pmd(folio, page, vma);
>> if (freeze)
>> put_page(page);
>> @@ -3096,8 +3143,10 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>> void split_huge_pmd_locked(struct vm_area_struct *vma, unsigned long address,
>> pmd_t *pmd, bool freeze)
>> {
>> +
>
> Unrelated change.
>
Will do
Thanks!
Balbir
Powered by blists - more mailing lists