[PATCH] mm/hugetlb: early exit from hugetlb_pages_alloc_boot() when max_huge_pages=0

[PATCH] mm/hugetlb: early exit from hugetlb_pages_alloc_boot() when max_huge_pages=0

[lirongqing]

From: Li RongQing <lirongqing@baidu.com>

Optimize hugetlb_pages_alloc_boot() to return immediately when
max_huge_pages is 0, saving cycles when hugepages aren't configured
in the kernel command line.

Signed-off-by: Li RongQing <lirongqing@baidu.com>
---
 mm/hugetlb.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 753f99b..514fab5 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -3654,6 +3654,9 @@ static void __init hugetlb_hstate_alloc_pages(struct hstate *h)
 		return;
 	}
 
+	if (!h->max_huge_pages)
+		return;
+
 	/* do node specific alloc */
 	if (hugetlb_hstate_alloc_pages_specific_nodes(h))
 		return;
-- 
2.9.4

Re: [PATCH] mm/hugetlb: early exit from hugetlb_pages_alloc_boot() when max_huge_pages=0

[David Hildenbrand]

On 14.08.25 10:29, lirongqing wrote:
> From: Li RongQing <lirongqing@baidu.com>
> 
> Optimize hugetlb_pages_alloc_boot() to return immediately when
> max_huge_pages is 0, saving cycles when hugepages aren't configured
> in the kernel command line.

Do we really care?

-- 
Cheers

David / dhildenb

RE: [外部邮件] Re: [PATCH] mm/hugetlb: early exit from hugetlb_pages_alloc_boot() when max_huge_pages=0

[Li,Rongqing]

> On 14.08.25 10:29, lirongqing wrote:
> > From: Li RongQing <lirongqing@baidu.com>
> >
> > Optimize hugetlb_pages_alloc_boot() to return immediately when
> > max_huge_pages is 0, saving cycles when hugepages aren't configured in
> > the kernel command line.
> 
> Do we really care?
> 

I find this when I see the boot log, I think this log is unnecessary if user did not configure hugetlbfs in kernel cmdline

kernel: HugeTLB: allocation took 0ms with hugepage_allocation_threads=32


Br
-Li

> --
> Cheers
> 
> David / dhildenb

Re: [外部邮件] Re: [PATCH] mm/hugetlb: early exit from hugetlb_pages_alloc_boot() when max_huge_pages=0

[David Hildenbrand]

On 14.08.25 11:12, Li,Rongqing wrote:
>> On 14.08.25 10:29, lirongqing wrote:
>>> From: Li RongQing <lirongqing@baidu.com>
>>>
>>> Optimize hugetlb_pages_alloc_boot() to return immediately when
>>> max_huge_pages is 0, saving cycles when hugepages aren't configured in
>>> the kernel command line.
>>
>> Do we really care?
>>
> 
> I find this when I see the boot log, I think this log is unnecessary if user did not configure hugetlbfs in kernel cmdline
> 
> kernel: HugeTLB: allocation took 0ms with hugepage_allocation_threads=32

That's valuable information to add to the patch description :)

-- 
Cheers

David / dhildenb