[PATCH v3] powerpc/hugetlb: Disable gigantic hugepages if fadump is active

[Sourabh Jain]

Hello Christophe,


On 27/01/25 13:23, Christophe Leroy wrote:
>
>
> Le 25/01/2025 à 11:49, Sourabh Jain a écrit :
>> The fadump kernel boots with limited memory solely to collect the kernel
>> core dump. Having gigantic hugepages in the fadump kernel is of no use.
>> Many times, the fadump kernel encounters OOM (Out of Memory) issues if
>> gigantic hugepages are allocated.
>>
>> To address this, disable gigantic hugepages if fadump is active by
>> returning early from arch_hugetlb_valid_size() using
>> hugepages_supported(). hugepages_supported() returns false if fadump is
>> active.
>
> Maybe you could explain how that's done. As far as I can see there is 
> no powerpc specific version of hugepages_supported(), so it relies on 
> hpage_shift being 0, but I was not able to quickly spot where that is 
> done.


hugepages_supported() is defined for powerpc in
arch/powerpc/include/asm/hugetlb.h

static inline bool hugepages_supported(void)
{
     if (hugetlb_disabled)
         return false;

     return HPAGE_SHIFT != 0;
}

Bu yes I will add a point about how it is disable for fadump kernel in
the commit message.


>
>>
>> Returning early from arch_hugetlb_valid_size() not only disables
>> gigantic hugepages but also avoids unnecessary hstate initialization for
>> every hugepage size supported by the platform.
>>
>> kernel logs related to hugepages with this patch included:
>> kernel argument passed: hugepagesz=1G hugepages=1
>>
>> First kernel: gigantic hugepage got allocated
>> ==============================================
>>
>> dmesg | grep -i "hugetlb"
>> -------------------------
>> HugeTLB: registered 1.00 GiB page size, pre-allocated 1 pages
>> HugeTLB: 0 KiB vmemmap can be freed for a 1.00 GiB page
>> HugeTLB: registered 2.00 MiB page size, pre-allocated 0 pages
>> HugeTLB: 0 KiB vmemmap can be freed for a 2.00 MiB page
>>
>> $ cat /proc/meminfo | grep -i "hugetlb"
>> -------------------------------------
>> Hugetlb:         1048576 kB
>>
>> Fadump kernel: gigantic hugepage not allocated
>> ===============================================
>>
>> dmesg | grep -i "hugetlb"
>> -------------------------
>> [    0.000000] HugeTLB: unsupported hugepagesz=1G
>> [    0.000000] HugeTLB: hugepages=1 does not follow a valid 
>> hugepagesz, ignoring
>> [    0.706375] HugeTLB support is disabled!
>> [    0.773530] hugetlbfs: disabling because there are no supported 
>> hugepage sizes
>>
>> $ cat /proc/meminfo | grep -i "hugetlb"
>> ----------------------------------
>> <Nothing>
>>
>> Cc: Christophe Leroy <christophe.leroy at csgroup.eu>
>> Cc: Hari Bathini <hbathini at linux.ibm.com>
>> Cc: Madhavan Srinivasan <maddy at linux.ibm.com>
>> Cc: Mahesh Salgaonkar <mahesh at linux.ibm.com>
>> Cc: Michael Ellerman <mpe at ellerman.id.au>
>> Cc: Ritesh Harjani (IBM)" <ritesh.list at gmail.com>
>> Signed-off-by: Sourabh Jain <sourabhjain at linux.ibm.com>
>
> Reviewed-by: Christophe Leroy <christophe.leroy at csgroup.eu>


Thank you for the Review!

- Sourabh Jain


>
>> ---
>>
>> Changelog:
>>
>> v1:
>> https://lore.kernel.org/all/20250121150419.1342794-1-sourabhjain@linux.ibm.com/ 
>>
>>
>> v2:
>> https://lore.kernel.org/all/20250124103220.111303-1-sourabhjain@linux.ibm.com/ 
>>
>>   - disable gigantic hugepage in arch code, arch_hugetlb_valid_size()
>>
>> v3:
>>   - Do not modify the initialization of the shift variable
>>
>> ---
>>   arch/powerpc/mm/hugetlbpage.c | 3 +++
>>   1 file changed, 3 insertions(+)
>>
>> diff --git a/arch/powerpc/mm/hugetlbpage.c 
>> b/arch/powerpc/mm/hugetlbpage.c
>> index 6b043180220a..88cfd182db4e 100644
>> --- a/arch/powerpc/mm/hugetlbpage.c
>> +++ b/arch/powerpc/mm/hugetlbpage.c
>> @@ -138,6 +138,9 @@ bool __init arch_hugetlb_valid_size(unsigned long 
>> size)
>>       int shift = __ffs(size);
>>       int mmu_psize;
>>   +    if (!hugepages_supported())
>> +        return false;
>> +
>>       /* Check that it is a page size supported by the hardware and
>>        * that it fits within pagetable and slice limits. */
>>       if (size <= PAGE_SIZE || !is_power_of_2(size))
>