Received-SPF: pass (google.com: domain of linux-kernel+bounces-159660-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Message-ID: <35210542-1309-4330-9367-e4a597be59fc@arm.com>
Date: Fri, 26 Apr 2024 08:48:07 +0100
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v1] mm: Fix race between __split_huge_pmd_locked() and
 GUP-fast
Content-Language: en-GB
To: Zi Yan <ziy@nvidia.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
 "Aneesh Kumar K.V" <aneesh.kumar@kernel.org>, linux-mm@kvack.org,
 linux-kernel@vger.kernel.org, Anshuman Khandual <anshuman.khandual@arm.com>,
 Huacai Chen <chenhuacai@kernel.org>
References: <20240425170704.3379492-1-ryan.roberts@arm.com>
 <922F6794-1C32-4862-8A67-90D4F7DBC474@nvidia.com>
From: Ryan Roberts <ryan.roberts@arm.com>
In-Reply-To: <922F6794-1C32-4862-8A67-90D4F7DBC474@nvidia.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

On 25/04/2024 19:58, Zi Yan wrote:
> +Anshuman, who changed pmd_present() semantics. See:
> https://lore.kernel.org/all/1599627183-14453-2-git-send-email-anshuman.khandual@arm.com/ and commit b65399f6111b ("arm64/mm: Change
> THP helpers to comply with generic MM semantics")
> 
> On 25 Apr 2024, at 13:07, Ryan Roberts wrote:
> 
>> __split_huge_pmd_locked() can be called for a present THP, devmap or
>> (non-present) migration entry. It calls pmdp_invalidate()
>> unconditionally on the pmdp and only determines if it is present or not
>> based on the returned old pmd. This is a problem for the migration entry
>> case because pmd_mkinvalid(), called by pmdp_invalidate() must only be
>> called for a present pmd.
>>
>> On arm64 at least, pmd_mkinvalid() will mark the pmd such that any
>> future call to pmd_present() will return true. And therefore any
> 
> But pmd_mkinvalid() on x86 does not behave so. Maybe we should fix
> pmd_mkinvalid() on arm64 by not setting PMD_PRESENT_INVALID when the
> entry is invalid already. And add a test in mm/debug_vm_pgtable.c.

Yes, we *could* make pmd_mkinvalid() defensive. But we don't do that for the
other getters/setters (e.g. pte_mkwrite()). So not sure why we would want that
here. Ultimately it makes no semantic sense to invalidate a non-present pmd.

See my other mail for excessive detail.

Thanks,
Ryan

> 
> I notice that x86, risc-v, mips behave the same. loongarch also
> has _PAGE_PRESENT_INVALID bit set during pmd_mkinvalid(), but its
> pmd_present() makes sure _PAGE_HUEG is set before checks _PAGE_PRESENT_INVALID.
> So it is not a problem for loongarch. Add Huacai to confirm this.
> 
> Maybe pmd_present() on arm64 can do that too?>
>> lockless pgtable walker could see the migration entry pmd in this state
>> and start interpretting the fields as if it were present, leading to
>> BadThings (TM). GUP-fast appears to be one such lockless pgtable walker.
>> I suspect the same is possible on other architectures.
>>
>> Fix this by only calling pmdp_invalidate() for a present pmd. And for
>> good measure let's add a warning to the generic implementation of
>> pmdp_invalidate(). I've manually reviewed all other
>> pmdp_invalidate[_ad]() call sites and believe all others to be
>> conformant.
>>
>> This is a theoretical bug found during code review. I don't have any
>> test case to trigger it in practice.
>>
>> Fixes: 84c3fc4e9c56 ("mm: thp: check pmd migration entry in common path")
>> Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
>> ---
>>
>> Applies on top of v6.9-rc5. Passes all the mm selftests on arm64.
>>
>> Thanks,
>> Ryan
>>
>>
>>  mm/huge_memory.c     | 5 +++--
>>  mm/pgtable-generic.c | 2 ++
>>  2 files changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
>> index 89f58c7603b2..80939ad00718 100644
>> --- a/mm/huge_memory.c
>> +++ b/mm/huge_memory.c
>> @@ -2513,12 +2513,12 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>>  	 * for this pmd), then we flush the SMP TLB and finally we write the
>>  	 * non-huge version of the pmd entry with pmd_populate.
>>  	 */
>> -	old_pmd = pmdp_invalidate(vma, haddr, pmd);
>>
>> -	pmd_migration = is_pmd_migration_entry(old_pmd);
>> +	pmd_migration = is_pmd_migration_entry(*pmd);
>>  	if (unlikely(pmd_migration)) {
>>  		swp_entry_t entry;
>>
>> +		old_pmd = *pmd;
>>  		entry = pmd_to_swp_entry(old_pmd);
>>  		page = pfn_swap_entry_to_page(entry);
>>  		write = is_writable_migration_entry(entry);
>> @@ -2529,6 +2529,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>>  		soft_dirty = pmd_swp_soft_dirty(old_pmd);
>>  		uffd_wp = pmd_swp_uffd_wp(old_pmd);
>>  	} else {
>> +		old_pmd = pmdp_invalidate(vma, haddr, pmd);
>>  		page = pmd_page(old_pmd);
>>  		folio = page_folio(page);
>>  		if (pmd_dirty(old_pmd)) {
>> diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
>> index 4fcd959dcc4d..74e34ea90656 100644
>> --- a/mm/pgtable-generic.c
>> +++ b/mm/pgtable-generic.c
>> @@ -198,6 +198,7 @@ pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
>>  pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
>>  		     pmd_t *pmdp)
>>  {
>> +	VM_WARN_ON(!pmd_present(*pmdp));
>>  	pmd_t old = pmdp_establish(vma, address, pmdp, pmd_mkinvalid(*pmdp));
>>  	flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
>>  	return old;
>> @@ -208,6 +209,7 @@ pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
>>  pmd_t pmdp_invalidate_ad(struct vm_area_struct *vma, unsigned long address,
>>  			 pmd_t *pmdp)
>>  {
>> +	VM_WARN_ON(!pmd_present(*pmdp));  	return pmdp_invalidate(vma, address, pmdp);
>>  }
>>  #endif
>> --
>> 2.25.1
> 
> 
> --
> Best Regards,
> Yan, Zi