public inbox for linux-mm@kvack.org
 help / color / mirror / Atom feed
From: Kefeng Wang <wangkefeng.wang@huawei.com>
To: <jane.chu@oracle.com>, Andrew Morton <akpm@linux-foundation.org>,
	David Hildenbrand <david@redhat.com>,
	Oscar Salvador <osalvador@suse.de>,
	Muchun Song <muchun.song@linux.dev>
Cc: <sidhartha.kumar@oracle.com>, Zi Yan <ziy@nvidia.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	Brendan Jackman <jackmanb@google.com>,
	Johannes Weiner <hannes@cmpxchg.org>, <linux-mm@kvack.org>
Subject: Re: [PATCH v2 9/9] mm: hugetlb: allocate frozen pages in alloc_gigantic_folio()
Date: Tue, 9 Sep 2025 15:33:51 +0800	[thread overview]
Message-ID: <c4fa23a3-3674-4a82-b29c-ee3aae39d8a1@huawei.com> (raw)
In-Reply-To: <cfaf16a0-00dc-4624-b379-17d6b1b5ad02@oracle.com>



On 2025/9/9 9:48, jane.chu@oracle.com wrote:
> 
> On 9/2/2025 5:48 AM, Kefeng Wang wrote:
> [..]
>> -
>>   static inline bool subpool_is_free(struct hugepage_subpool *spool)
>>   {
>>       if (spool->count)
>> @@ -1472,43 +1462,20 @@ static int hstate_next_node_to_free(struct 
>> hstate *h, nodemask_t *nodes_allowed)
>>           nr_nodes--)
>>   #ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
>> -#ifdef CONFIG_CONTIG_ALLOC
> 
> Why do you remove the CONTIG_ALLOC config switch?
> If it's not selected on x86, gigantic_pages_init() won't be invoked.

alloc_gigantic_folio() stub is useless, we have all stub function for no
CONTIG_ALLOC(CMA will enabled CONTIG_ALLOC), alloc_gigantic_folio()
will return NULL without CONFIG_CONTIG_ALLOC, right?


> 
>>   static struct folio *alloc_gigantic_folio(int order, gfp_t gfp_mask,
>>           int nid, nodemask_t *nodemask)
>>   {
>>       struct folio *folio;
>> -    bool retried = false;
>> -
>> -retry:
>> -    folio = hugetlb_cma_alloc_folio(order, gfp_mask, nid, nodemask);
>> -    if (!folio) {
>> -        if (hugetlb_cma_exclusive_alloc())
>> -            return NULL;
>> -
>> -        folio = folio_alloc_gigantic(order, gfp_mask, nid, nodemask);
>> -        if (!folio)
>> -            return NULL;
>> -    }
>> -    if (folio_ref_freeze(folio, 1))
>> +    folio = hugetlb_cma_alloc_frozen_folio(order, gfp_mask, nid, 
>> nodemask);
>> +    if (folio)
>>           return folio;
>> -    pr_warn("HugeTLB: unexpected refcount on PFN %lu\n", 
>> folio_pfn(folio));
>> -    hugetlb_free_folio(folio);
>> -    if (!retried) {
>> -        retried = true;
>> -        goto retry;
>> -    }
>> -    return NULL;
>> -}
>> +    if (hugetlb_cma_exclusive_alloc())
>> +        return NULL;
>> -#else /* !CONFIG_CONTIG_ALLOC */
>> -static struct folio *alloc_gigantic_folio(int order, gfp_t gfp_mask, 
>> int nid,
>> -                      nodemask_t *nodemask)
>> -{
>> -    return NULL;
>> +    return folio_alloc_frozen_gigantic(order, gfp_mask, nid, nodemask);
>>   }
>> -#endif /* CONFIG_CONTIG_ALLOC */
>>   #else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE */
> The rest look okay.
> 
> thanks,
> -jane
> 
> 



  reply	other threads:[~2025-09-09  7:34 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-02 12:48 [PATCH v2 0/9] mm: hugetlb: cleanup and allocate frozen hugetlb folio Kefeng Wang
2025-09-02 12:48 ` [PATCH v2 1/9] mm: hugetlb: convert to use more alloc_fresh_hugetlb_folio() Kefeng Wang
2025-09-08  9:21   ` Oscar Salvador
2025-09-08 12:59     ` Kefeng Wang
2025-09-09  0:54   ` Zi Yan
2025-09-02 12:48 ` [PATCH v2 2/9] mm: hugetlb: convert to account_new_hugetlb_folio() Kefeng Wang
2025-09-08  9:26   ` Oscar Salvador
2025-09-08 13:20     ` Kefeng Wang
2025-09-08 13:38       ` Oscar Salvador
2025-09-08 13:40   ` Oscar Salvador
2025-09-09  7:04     ` Kefeng Wang
2025-09-09  0:59   ` Zi Yan
2025-09-02 12:48 ` [PATCH v2 3/9] mm: hugetlb: directly pass order when allocate a hugetlb folio Kefeng Wang
2025-09-08  9:29   ` Oscar Salvador
2025-09-09  1:11   ` Zi Yan
2025-09-09  7:11     ` Kefeng Wang
2025-09-02 12:48 ` [PATCH v2 4/9] mm: hugetlb: remove struct hstate from init_new_hugetlb_folio() Kefeng Wang
2025-09-08  9:31   ` Oscar Salvador
2025-09-09  1:13   ` Zi Yan
2025-09-02 12:48 ` [PATCH v2 5/9] mm: hugeltb: check NUMA_NO_NODE in only_alloc_fresh_hugetlb_folio() Kefeng Wang
2025-09-08  9:34   ` Oscar Salvador
2025-09-09  1:16   ` Zi Yan
2025-09-02 12:48 ` [PATCH v2 6/9] mm: page_alloc: add alloc_contig_frozen_pages() Kefeng Wang
2025-09-09  0:21   ` jane.chu
2025-09-09  1:44   ` Zi Yan
2025-09-09  7:29     ` Kefeng Wang
2025-09-09  8:11   ` Oscar Salvador
2025-09-09 18:55   ` Matthew Wilcox
2025-09-09 19:08     ` Zi Yan
2025-09-10  2:05       ` Kefeng Wang
2025-09-02 12:48 ` [PATCH v2 7/9] mm: cma: add alloc flags for __cma_alloc() Kefeng Wang
2025-09-09  0:19   ` jane.chu
2025-09-09  2:03   ` Zi Yan
2025-09-09  8:05   ` Oscar Salvador
2025-09-02 12:48 ` [PATCH v2 8/9] mm: cma: add __cma_release() Kefeng Wang
2025-09-09  0:15   ` jane.chu
2025-09-02 12:48 ` [PATCH v2 9/9] mm: hugetlb: allocate frozen pages in alloc_gigantic_folio() Kefeng Wang
2025-09-09  1:48   ` jane.chu
2025-09-09  7:33     ` Kefeng Wang [this message]
2025-09-09  2:02   ` Zi Yan
2025-09-09  7:34     ` Kefeng Wang
2025-09-02 13:51 ` [PATCH v2 0/9] mm: hugetlb: cleanup and allocate frozen hugetlb folio Oscar Salvador

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c4fa23a3-3674-4a82-b29c-ee3aae39d8a1@huawei.com \
    --to=wangkefeng.wang@huawei.com \
    --cc=akpm@linux-foundation.org \
    --cc=david@redhat.com \
    --cc=hannes@cmpxchg.org \
    --cc=jackmanb@google.com \
    --cc=jane.chu@oracle.com \
    --cc=linux-mm@kvack.org \
    --cc=muchun.song@linux.dev \
    --cc=osalvador@suse.de \
    --cc=sidhartha.kumar@oracle.com \
    --cc=vbabka@suse.cz \
    --cc=ziy@nvidia.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox