linux-kernel - Re: [PATCH v2 10/11] mm: thp: always enable mTHP support

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <f7596f9b-5796-452e-82e1-f8cc6e042eef@linux.alibaba.com>
Date: Wed, 11 Feb 2026 09:12:41 +0800
From: Baolin Wang <baolin.wang@...ux.alibaba.com>
To: Luiz Capitulino <luizcap@...hat.com>, linux-kernel@...r.kernel.org,
 linux-mm@...ck.org, david@...nel.org
Cc: ryan.roberts@....com, akpm@...ux-foundation.org,
 lorenzo.stoakes@...cle.com
Subject: Re: [PATCH v2 10/11] mm: thp: always enable mTHP support



On 2/10/26 9:28 PM, Luiz Capitulino wrote:
> On 2026-02-10 04:56, Baolin Wang wrote:
>>
>>
>> On 2/10/26 6:14 AM, Luiz Capitulino wrote:
>>> If PMD-sized pages are not supported on an architecture (ie. the
>>> arch implements arch_has_pmd_leaves() and it returns false) then the
>>> current code disables all THP, including mTHP.
>>>
>>> This commit fixes this by allowing mTHP to be always enabled for all
>>> archs. When PMD-sized pages are not supported, its sysfs entry won't be
>>> created and their mapping will be disallowed at page-fault time.
>>>
>>> Similarly, this commit implements the following changes for shmem:
>>>
>>>   - In shmem_allowable_huge_orders(): drop the pgtable_has_pmd_leaves()
>>>     check so that mTHP sizes are considered
>>>   - In shmem_alloc_and_add_folio(): don't consider PMD and PUD orders
>>>     when PMD-sized pages are not supported by the CPU
>>>
>>> Signed-off-by: Luiz Capitulino <luizcap@...hat.com>
>>> ---
>>>   mm/huge_memory.c | 11 +++++++----
>>>   mm/shmem.c       |  4 +++-
>>>   2 files changed, 10 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
>>> index 1e5ea2e47f79..882331592928 100644
>>> --- a/mm/huge_memory.c
>>> +++ b/mm/huge_memory.c
>>> @@ -115,6 +115,9 @@ unsigned long __thp_vma_allowable_orders(struct 
>>> vm_area_struct *vma,
>>>       else
>>>           supported_orders = THP_ORDERS_ALL_FILE_DEFAULT;
>>> +    if (!pgtable_has_pmd_leaves())
>>> +        supported_orders &= ~(BIT(PMD_ORDER) | BIT(PUD_ORDER));
>>> +
>>>       orders &= supported_orders;
>>>       if (!orders)
>>>           return 0;
>>> @@ -122,7 +125,7 @@ unsigned long __thp_vma_allowable_orders(struct 
>>> vm_area_struct *vma,
>>>       if (!vma->vm_mm)        /* vdso */
>>>           return 0;
>>> -    if (!pgtable_has_pmd_leaves() || vma_thp_disabled(vma, vm_flags, 
>>> forced_collapse))
>>> +    if (vma_thp_disabled(vma, vm_flags, forced_collapse))
>>>           return 0;
>>>       /* khugepaged doesn't collapse DAX vma, but page fault is fine. */
>>> @@ -806,6 +809,9 @@ static int __init hugepage_init_sysfs(struct 
>>> kobject **hugepage_kobj)
>>>       }
>>>       orders = THP_ORDERS_ALL_ANON | THP_ORDERS_ALL_FILE_DEFAULT;
>>> +    if (!pgtable_has_pmd_leaves())
>>> +        orders &= ~(BIT(PMD_ORDER) | BIT(PUD_ORDER));
>>
>> I think you should also handle the 'huge_anon_orders_inherit' setting 
>> in this function if pgtable_has_pmd_leaves() returns false. Shmem as 
>> well.
>>
>> if (!anon_orders_configured)
>>      huge_anon_orders_inherit = BIT(PMD_ORDER);
> 
> Good catch. So, would you agree that should set it to BIT(PMD_ORDER - 1)
> in this case?

 From the documentation:
"
By default, PMD-sized hugepages have enabled="inherit" and all other
hugepage sizes have enabled="never".
"

So if pgtable_has_pmd_leaves() returns false, IMO, we should just skip 
setting the PMD-sized order for huge_anon_orders_inherit. What I mean is:

if (!anon_orders_configured && pgtable_has_pmd_leaves())
	huge_anon_orders_inherit = BIT(PMD_ORDER);