Re: [PATCH 2/2] mm/vmalloc: free unused pages on vrealloc() shrink

[Shivam Kalra]

On 03/03/26 16:42, Shivam Kalra wrote:
> On 03/03/26 14:40, Alice Ryhl wrote:
>> On Mon, Mar 2, 2026 at 3:03 PM Shivam Kalra via B4 Relay
>> <devnull+shivamkalra98.zohomail.in@xxxxxxxxxx> wrote:
>>>
>>> From: Shivam Kalra <shivamkalra98@xxxxxxxxxxx>
>>>
>>> When vrealloc() shrinks an allocation and the new size crosses a page
>>> boundary, unmap and free the tail pages that are no longer needed. This
>>> reclaims physical memory that was previously wasted for the lifetime
>>> of the allocation.
>>>
>>> The heuristic is simple: always free when at least one full page becomes
>>> unused. Huge page allocations (page_order > 0) are skipped, as partial
>>> freeing would require splitting.
>>>
>>> The virtual address reservation (vm->size / vmap_area) is intentionally
>>> kept unchanged, preserving the address for potential future grow-in-place
>>> support.
>>>
>>> Fix the grow-in-place check to compare against vm->nr_pages rather than
>>> get_vm_area_size(), since the latter reflects the virtual reservation
>>> which does not shrink. Without this fix, a grow after shrink would
>>> access freed pages.
>>>
>>> Signed-off-by: Shivam Kalra <shivamkalra98@xxxxxxxxxxx>
>>> ---
>>>  mm/vmalloc.c | 25 ++++++++++++++++++++-----
>>>  1 file changed, 20 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/mm/vmalloc.c b/mm/vmalloc.c
>>> index 54e76a47e995..7a4c59422638 100644
>>> --- a/mm/vmalloc.c
>>> +++ b/mm/vmalloc.c
>>> @@ -4327,14 +4327,29 @@ void *vrealloc_node_align_noprof(const void *p, size_t size, unsigned long align
>>>                         goto need_realloc;
>>>         }
>>>
>>> -       /*
>>> -        * TODO: Shrink the vm_area, i.e. unmap and free unused pages. What
>>> -        * would be a good heuristic for when to shrink the vm_area?
>>> -        */
>>>         if (size <= old_size) {
>>> +               unsigned int new_nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT;
>>> +
>>>                 /* Zero out "freed" memory, potentially for future realloc. */
>>>                 if (want_init_on_free() || want_init_on_alloc(flags))
>>>                         memset((void *)p + size, 0, old_size - size);
>>> +
>>> +               /* Free tail pages when shrink crosses a page boundary. */
>>> +               if (new_nr_pages < vm->nr_pages &&
>>> +                   !vm_area_page_order(vm)) {
>>> +                       unsigned long addr = (unsigned long)p;
>>> +
>>> +                       vunmap_range(addr + (new_nr_pages << PAGE_SHIFT),
>>> +                                    addr + (vm->nr_pages << PAGE_SHIFT));
>>> +
>>> +                       kasan_poison_vmalloc(
>>> +                               (void *)(addr + (new_nr_pages << PAGE_SHIFT)),
>>> +                               (vm->nr_pages - new_nr_pages) << PAGE_SHIFT);
>>
>> There is a kasan_poison_vmalloc() call here.
>>
>>> +                       vmalloc_free_pages(vm, new_nr_pages, vm->nr_pages);
>>> +                       vm->nr_pages = new_nr_pages;
>>> +               }
>>> +
>>>                 vm->requested_size = size;
>>>                 kasan_poison_vmalloc(p + size, old_size - size);
>>
>> And there is a kasan_poison_vmalloc() call here.
>>
>> Furthermore, they seem to touch overlapping ranges. Perhaps the first
>> call can be dropped?
>>
>> Alice
> Thanks for the feedback Alice. I will do two things for v2.
> 1. Update the base-commit to the recent most commit on char-misc-next
> 2. Keep only a single `kasan_vrealloc` call after the rebase.
Correction: since this is mm, I will use the mm-new branch.