Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
Message-ID: <de6529b0-9360-cf4a-4233-ee1498b44176@linux.ibm.com>
Date:   Mon, 4 Apr 2022 19:40:24 +0530
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.7.0
Subject: Re: [PATCH] arm64: mm: fix pmd_leaf()
Content-Language: en-US
To:     Muchun Song <songmuchun@bytedance.com>,
        Will Deacon <will@kernel.org>
Cc:     Catalin Marinas <catalin.marinas@arm.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Anshuman Khandual <anshuman.khandual@arm.com>,
        steven.price@arm.com, lengxujun2007@126.com,
        Arnd Bergmann <arnd@arndb.de>,
        LAK <linux-arm-kernel@lists.infradead.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Muchun Song <smuchun@gmail.com>,
        Xiongchun duan <duanxiongchun@bytedance.com>,
        Qian Cai <quic_qiancai@quicinc.com>
References: <20220403024928.4125-1-songmuchun@bytedance.com>
 <20220404091957.GA22790@willie-the-truck>
 <CAMZfGtU=4aFmbwyyh+5=M3iaqox7X7yaoPP+r=WJO9dReaNb-g@mail.gmail.com>
From:   Aneesh Kumar K V <aneesh.kumar@linux.ibm.com>
In-Reply-To: <CAMZfGtU=4aFmbwyyh+5=M3iaqox7X7yaoPP+r=WJO9dReaNb-g@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0
Precedence: bulk

On 4/4/22 5:10 PM, Muchun Song wrote:
> On Mon, Apr 4, 2022 at 5:20 PM Will Deacon <will@kernel.org> wrote:
>>
>> On Sun, Apr 03, 2022 at 10:49:28AM +0800, Muchun Song wrote:
>>> 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.
>>>
>>> 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>
>>> Signed-off-by: Muchun Song <songmuchun@bytedance.com>
>>> ---
>>>   arch/arm64/include/asm/pgtable.h | 2 +-
>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
>>> index 94e147e5456c..09eaae46a19b 100644
>>> --- a/arch/arm64/include/asm/pgtable.h
>>> +++ b/arch/arm64/include/asm/pgtable.h
>>> @@ -535,7 +535,7 @@ extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
>>>                                 PMD_TYPE_TABLE)
>>>   #define pmd_sect(pmd)                ((pmd_val(pmd) & PMD_TYPE_MASK) == \
>>>                                 PMD_TYPE_SECT)
>>> -#define pmd_leaf(pmd)                pmd_sect(pmd)
>>> +#define pmd_leaf(pmd)                (pmd_present(pmd) && !(pmd_val(pmd) & PMD_TABLE_BIT))
>>>   #define pmd_bad(pmd)         (!pmd_table(pmd))
>>>
>>>   #define pmd_leaf_size(pmd)   (pmd_cont(pmd) ? CONT_PMD_SIZE : PMD_SIZE)
>>
>> A bunch of the users of pmd_leaf() already check pmd_present() -- is it
>> documented that we need to handle this check inside the macro? afaict x86
>> doesn't do this either.


ppc64 also doesn't do a pmd_present check.

>>
> 
> arm64 is different from x86 here. pmd_leaf() could return true for
> the none pmd without pmd_present() check, the check of
> pmd_present() aims to exclude the pmd_none() case.  However,
> it could work properly on x86 without pmd_present() or pmd_none().
> So we don't see pmd_present() or pmd_none() check in pmd_leaf().
> For this reason, I think this check is necessary.
> 
> BTW, there are some users of pmd_leaf() (e.g. apply_to_pmd_range,
> walk_pmd_range, ptdump_pmd_entry) which do not check pmd_present()
> or pmd_none() before the call of pmd_leaf().  So it is also necessary
> to add the check.
> 


I would expect pmd_leaf check to return true, if the said pmd page table 
entry can point to a leaf page table entry which can also be a not 
present page table entry?

-aneesh