[PATCH v2] mm/hugetlb: use kvmalloc_objs for hugetlb_fault_mutex_table

[PATCH v2] mm/hugetlb: use kvmalloc_objs for hugetlb_fault_mutex_table

[Dheeraj Kumar Srivastava]

On systems/VMs with a large number of CPUs (e.g. 4096), hugetlb_init()
computes num_fault_mutexes = roundup_pow_of_two(8 * num_possible_cpus())
= 32768.  The subsequent kmalloc for the mutex table can exceed
MAX_PAGE_ORDER when struct mutex is enlarged by debug options like
CONFIG_DEBUG_MUTEXES and CONFIG_DEBUG_LOCK_ALLOC, resulting in:

  WARNING: at __alloc_frozen_pages_noprof (order > MAX_PAGE_ORDER)
  kernel BUG at mm/hugetlb.c (BUG_ON(!hugetlb_fault_mutex_table))
  Kernel panic - not syncing: Fatal exception

Switch to kvmalloc_objs() so the allocation falls back to vmalloc when
the contiguous physical allocation is too large.  The table is only
accessed by index, so virtual contiguity is sufficient.

Fixes: 8382d914ebf7 ("mm, hugetlb: improve page-fault scalability")
Signed-off-by: Dheeraj Kumar Srivastava <dheerajkumar.srivastava@amd.com>
---
 mm/hugetlb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index cea25773a6c9..6fc5c0bdb500 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -4177,7 +4177,7 @@ static int __init hugetlb_init(void)
 	num_fault_mutexes = 1;
 #endif
 	hugetlb_fault_mutex_table =
-		kmalloc_objs(struct mutex, num_fault_mutexes);
+		kvmalloc_objs(struct mutex, num_fault_mutexes);
 	BUG_ON(!hugetlb_fault_mutex_table);
 
 	for (i = 0; i < num_fault_mutexes; i++)
-- 
2.25.1

Re: [PATCH v2] mm/hugetlb: use kvmalloc_objs for hugetlb_fault_mutex_table

[Dheeraj Kumar Srivastava]

Hi all,

On 10/6/2026 9:31 PM, Dheeraj Kumar Srivastava wrote:
> On systems/VMs with a large number of CPUs (e.g. 4096), hugetlb_init()
> computes num_fault_mutexes = roundup_pow_of_two(8 * num_possible_cpus())
> = 32768.  The subsequent kmalloc for the mutex table can exceed
> MAX_PAGE_ORDER when struct mutex is enlarged by debug options like
> CONFIG_DEBUG_MUTEXES and CONFIG_DEBUG_LOCK_ALLOC, resulting in:
> 
>    WARNING: at __alloc_frozen_pages_noprof (order > MAX_PAGE_ORDER)
>    kernel BUG at mm/hugetlb.c (BUG_ON(!hugetlb_fault_mutex_table))
>    Kernel panic - not syncing: Fatal exception
> 
> Switch to kvmalloc_objs() so the allocation falls back to vmalloc when
> the contiguous physical allocation is too large.  The table is only
> accessed by index, so virtual contiguity is sufficient.

Sorry, I forgot to include the "Changes since v1" change log.

Changes since v1:
-> Use kvmalloc_objs() instead of kvmalloc_array().

Thanks
Dheeraj

> 
> Fixes: 8382d914ebf7 ("mm, hugetlb: improve page-fault scalability")
> Signed-off-by: Dheeraj Kumar Srivastava <dheerajkumar.srivastava@amd.com>
> ---
>   mm/hugetlb.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index cea25773a6c9..6fc5c0bdb500 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -4177,7 +4177,7 @@ static int __init hugetlb_init(void)
>   	num_fault_mutexes = 1;
>   #endif
>   	hugetlb_fault_mutex_table =
> -		kmalloc_objs(struct mutex, num_fault_mutexes);
> +		kvmalloc_objs(struct mutex, num_fault_mutexes);
>   	BUG_ON(!hugetlb_fault_mutex_table);
>   
>   	for (i = 0; i < num_fault_mutexes; i++)