Received-SPF: pass (google.com: domain of linux-kernel+bounces-34628-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Message-ID: <829fb129-f643-4960-a2da-cd38e5ee8f39@linux.dev>
Date: Tue, 23 Jan 2024 10:12:18 +0800
Precedence: bulk
MIME-Version: 1.0
Subject: Re: [PATCH v4 6/7] hugetlb: parallelize 2M hugetlb allocation and
 initialization
Content-Language: en-US
To: Muchun Song <muchun.song@linux.dev>
Cc: David Hildenbrand <david@redhat.com>, David Rientjes
 <rientjes@google.com>, Mike Kravetz <mike.kravetz@oracle.com>,
 Andrew Morton <akpm@linux-foundation.org>,
 Tim Chen <tim.c.chen@linux.intel.com>, Linux-MM <linux-mm@kvack.org>,
 Gang Li <gang.li@linux.dev>, LKML <linux-kernel@vger.kernel.org>,
 Gang Li <ligang.bdlg@bytedance.com>
References: <20240118123911.88833-1-gang.li@linux.dev>
 <20240118123911.88833-7-gang.li@linux.dev>
 <ddf37da4-4cbc-478a-be9b-3060b0aebc90@linux.dev>
 <14e38e95-2bc6-4571-b502-4e3954b4bcc4@linux.dev>
 <849D7EA4-BCF4-4587-8A78-F3B35B63EAE9@linux.dev>
From: Gang Li <gang.li@linux.dev>
In-Reply-To: <849D7EA4-BCF4-4587-8A78-F3B35B63EAE9@linux.dev>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit

On 2024/1/22 19:30, Muchun Song wrote:
>> On Jan 22, 2024, at 18:12, Gang Li <gang.li@linux.dev> wrote:
>>
>> On 2024/1/22 15:10, Muchun Song wrote:> On 2024/1/18 20:39, Gang Li wrote:
>>>> +static void __init hugetlb_alloc_node(unsigned long start, unsigned long end, void *arg)
>>>>    {
>>>> -    unsigned long i;
>>>> +    struct hstate *h = (struct hstate *)arg;
>>>> +    int i, num = end - start;
>>>> +    nodemask_t node_alloc_noretry;
>>>> +    unsigned long flags;
>>>> +    int next_node = 0;
>>> This should be first_online_node which may be not zero.
>>
>> That's right. Thanks!
>>
>>>> -    for (i = 0; i < h->max_huge_pages; ++i) {
>>>> -        if (!alloc_bootmem_huge_page(h, NUMA_NO_NODE))
>>>> +    /* Bit mask controlling how hard we retry per-node allocations.*/
>>>> +    nodes_clear(node_alloc_noretry);
>>>> +
>>>> +    for (i = 0; i < num; ++i) {
>>>> +        struct folio *folio = alloc_pool_huge_folio(h, &node_states[N_MEMORY],
>>>> +                        &node_alloc_noretry, &next_node);
>>>> +        if (!folio)
>>>>                break;
>>>> +        spin_lock_irqsave(&hugetlb_lock, flags);
>>>> I suspect there will more contention on this lock when parallelizing.
>>
>> In the worst case, there are only 'numa node number' of threads in
>> contention. And in my testing, it doesn't degrade performance, but
>> rather improves performance due to the reduced granularity.
> 
> So, the performance does not change if you move the lock out of
> loop?
>

If we move the lock out of loop, then multi-threading becomes 
single-threading, which definitely reduces performance.

```
+       spin_lock_irqsave(&hugetlb_lock, flags);
         for (i = 0; i < num; ++i) {
                 struct folio *folio = alloc_pool_huge_folio(h, 
&node_states[N_MEMORY],
                                                 &node_alloc_noretry, 
&next_node);
                 if (!folio)
                         break;
-               spin_lock_irqsave(&hugetlb_lock, flags);
                 __prep_account_new_huge_page(h, folio_nid(folio));
                 enqueue_hugetlb_folio(h, folio);
-               spin_unlock_irqrestore(&hugetlb_lock, flags);
                 cond_resched();
         }
+       spin_unlock_irqrestore(&hugetlb_lock, flags);
  }
```