Re: [PATCH 2/3] mm/memory.c: Update local TLB if PTE entry exists

From: maobibo <maobibo@loongson.cn>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: "Thomas Bogendoerfer" <tsbogend@alpha.franken.de>,
	"Jiaxun Yang" <jiaxun.yang@flygoat.com>,
	"Huacai Chen" <chenhc@lemote.com>,
	"Paul Burton" <paulburton@kernel.org>,
	"Dmitry Korotin" <dkorotin@wavecomp.com>,
	"Philippe Mathieu-Daudé" <f4bug@amsat.org>,
	"Stafford Horne" <shorne@gmail.com>,
	"Steven Price" <steven.price@arm.com>,
	"Anshuman Khandual" <anshuman.khandual@arm.com>,
	linux-mips@vger.kernel.org, linux-kernel@vger.kernel.org,
	"Mike Rapoport" <rppt@linux.ibm.com>,
	"Sergei Shtylyov" <sergei.shtylyov@cogentembedded.com>,
	"Maciej W. Rozycki" <macro@wdc.com>,
	linux-mm@kvack.org
Subject: Re: [PATCH 2/3] mm/memory.c: Update local TLB if PTE entry exists
Date: Sat, 16 May 2020 17:42:55 +0800	[thread overview]
Message-ID: <61ebeb0a-89cf-834f-1d20-765a2ab43161@loongson.cn> (raw)
In-Reply-To: <20200515134046.cf107c6a13b9604c46ad71b8@linux-foundation.org>


On 05/16/2020 04:40 AM, Andrew Morton wrote:
> On Fri, 15 May 2020 12:10:08 +0800 Bibo Mao <maobibo@loongson.cn> wrote:
> 
>> If there are two threads hitting page fault at the same page,
>> one thread updates PTE entry and local TLB, the other can
>> update local tlb also, rather than give up and do page fault
>> again.
>>
>> ...
>>
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -1770,8 +1770,8 @@ static vm_fault_t insert_pfn(struct vm_area_struct *vma, unsigned long addr,
>>  			}
>>  			entry = pte_mkyoung(*pte);
>>  			entry = maybe_mkwrite(pte_mkdirty(entry), vma);
>> -			if (ptep_set_access_flags(vma, addr, pte, entry, 1))
>> -				update_mmu_cache(vma, addr, pte);
>> +			ptep_set_access_flags(vma, addr, pte, entry, 1);
>> +			update_mmu_cache(vma, addr, pte);
> 
> Presumably these changes mean that other architectures will run
> update_mmu_cache() more frequently than they used to.  How much more
> frequently, and what will be the impact of this change?  (Please fully
> explain all this in the changelog).
> 
It is only useful for those architects where software can update tlb, if the  function update_mmu_cache is used for other reason, it will bring out somewhat impact, and I will explain it in the changelog.

>>  		}
>>  		goto out_unlock;
>>  	}
>>
>> ...
>>
>> @@ -2463,7 +2462,8 @@ static inline bool cow_user_page(struct page *dst, struct page *src,
>>  		vmf->pte = pte_offset_map_lock(mm, vmf->pmd, addr, &vmf->ptl);
>>  		locked = true;
>>  		if (!likely(pte_same(*vmf->pte, vmf->orig_pte))) {
>> -			/* The PTE changed under us. Retry page fault. */
>> +			/* The PTE changed under us, update local tlb */
>> +			pdate_mmu_cache(vma, addr, vmf->pte);
> 
> Missing a 'u' there.  Which tells me this patch isn't the one which you
> tested!
> 
Sorry about it, I will refresh the patch and add modification about this obvious typo

regards
bibo, mao
>>  			ret = false;
>>  			goto pte_unlock;
>>  		}
>>
>> ...
>>