Re: [PATCH v3 2/3] vmalloc: Optimize vfree

[Muhammad Usama Anjum <usama.anjum@arm.com>]

On 25/03/2026 4:16 pm, Uladzislau Rezki wrote:
> On Wed, Mar 25, 2026 at 03:02:14PM +0000, Muhammad Usama Anjum wrote:
>> On 25/03/2026 8:56 am, Uladzislau Rezki wrote:
>>> On Tue, Mar 24, 2026 at 10:55:55AM -0400, Zi Yan wrote:
>>>> On 24 Mar 2026, at 9:35, Muhammad Usama Anjum wrote:
>>>>
>>>>> From: Ryan Roberts <ryan.roberts@arm.com>
>>>>>
>>>>> Whenever vmalloc allocates high order pages (e.g. for a huge mapping) it
>>>>> must immediately split_page() to order-0 so that it remains compatible
>>>>> with users that want to access the underlying struct page.
>>>>> Commit a06157804399 ("mm/vmalloc: request large order pages from buddy
>>>>> allocator") recently made it much more likely for vmalloc to allocate
>>>>> high order pages which are subsequently split to order-0.
>>>>>
>>>>> Unfortunately this had the side effect of causing performance
>>>>> regressions for tight vmalloc/vfree loops (e.g. test_vmalloc.ko
>>>>> benchmarks). See Closes: tag. This happens because the high order pages
>>>>> must be gotten from the buddy but then because they are split to
>>>>> order-0, when they are freed they are freed to the order-0 pcp.
>>>>> Previously allocation was for order-0 pages so they were recycled from
>>>>> the pcp.
>>>>>
>>>>> It would be preferable if when vmalloc allocates an (e.g.) order-3 page
>>>>> that it also frees that order-3 page to the order-3 pcp, then the
>>>>> regression could be removed.
>>>>>
>>>>> So let's do exactly that; use the new __free_contig_range() API to
>>>>> batch-free contiguous ranges of pfns. This not only removes the
>>>>> regression, but significantly improves performance of vfree beyond the
>>>>> baseline.
>>>>>
>>>>> A selection of test_vmalloc benchmarks running on arm64 server class
>>>>> system. mm-new is the baseline. Commit a06157804399 ("mm/vmalloc: request
>>>>> large order pages from buddy allocator") was added in v6.19-rc1 where we
>>>>> see regressions. Then with this change performance is much better. (>0
>>>>> is faster, <0 is slower, (R)/(I) = statistically significant
>>>>> Regression/Improvement):
>>>>>
>>>>> +-----------------+----------------------------------------------------------+-------------------+--------------------+
>>>>> | Benchmark       | Result Class                                             |   mm-new          |  this series       |
>>>>> +=================+==========================================================+===================+====================+
>>>>> | micromm/vmalloc | fix_align_alloc_test: p:1, h:0, l:500000 (usec)          |        1331843.33 |         (I) 67.17% |
>>>>> |                 | fix_size_alloc_test: p:1, h:0, l:500000 (usec)           |         415907.33 |             -5.14% |
>>>>> |                 | fix_size_alloc_test: p:4, h:0, l:500000 (usec)           |         755448.00 |         (I) 53.55% |
>>>>> |                 | fix_size_alloc_test: p:16, h:0, l:500000 (usec)          |        1591331.33 |         (I) 57.26% |
>>>>> |                 | fix_size_alloc_test: p:16, h:1, l:500000 (usec)          |        1594345.67 |         (I) 68.46% |
>>>>> |                 | fix_size_alloc_test: p:64, h:0, l:100000 (usec)          |        1071826.00 |         (I) 79.27% |
>>>>> |                 | fix_size_alloc_test: p:64, h:1, l:100000 (usec)          |        1018385.00 |         (I) 84.17% |
>>>>> |                 | fix_size_alloc_test: p:256, h:0, l:100000 (usec)         |        3970899.67 |         (I) 77.01% |
>>>>> |                 | fix_size_alloc_test: p:256, h:1, l:100000 (usec)         |        3821788.67 |         (I) 89.44% |
>>>>> |                 | fix_size_alloc_test: p:512, h:0, l:100000 (usec)         |        7795968.00 |         (I) 82.67% |
>>>>> |                 | fix_size_alloc_test: p:512, h:1, l:100000 (usec)         |        6530169.67 |        (I) 118.09% |
>>>>> |                 | full_fit_alloc_test: p:1, h:0, l:500000 (usec)           |         626808.33 |             -0.98% |
>>>>> |                 | kvfree_rcu_1_arg_vmalloc_test: p:1, h:0, l:500000 (usec) |         532145.67 |             -1.68% |
>>>>> |                 | kvfree_rcu_2_arg_vmalloc_test: p:1, h:0, l:500000 (usec) |         537032.67 |             -0.96% |
>>>>> |                 | long_busy_list_alloc_test: p:1, h:0, l:500000 (usec)     |        8805069.00 |         (I) 74.58% |
>>>>> |                 | pcpu_alloc_test: p:1, h:0, l:500000 (usec)               |         500824.67 |              4.35% |
>>>>> |                 | random_size_align_alloc_test: p:1, h:0, l:500000 (usec)  |        1637554.67 |         (I) 76.99% |
>>>>> |                 | random_size_alloc_test: p:1, h:0, l:500000 (usec)        |        4556288.67 |         (I) 72.23% |
>>>>> |                 | vm_map_ram_test: p:1, h:0, l:500000 (usec)               |         107371.00 |             -0.70% |
>>>>> +-----------------+----------------------------------------------------------+-------------------+--------------------+
>>>>>
>>>>> Fixes: a06157804399 ("mm/vmalloc: request large order pages from buddy allocator")
>>>>> Closes: https://lore.kernel.org/all/66919a28-bc81-49c9-b68f-dd7c73395a0d@arm.com/
>>>>> Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
>>>>> Co-developed-by: Muhammad Usama Anjum <usama.anjum@arm.com>
>>>>> Signed-off-by: Muhammad Usama Anjum <usama.anjum@arm.com>
>>>>> ---
>>>>> Changes since v2:
>>>>> - Remove BUG_ON in favour of simple implementation as this has never
>>>>>   been seen to output any bug in the past as well
>>>>> - Move the free loop to separate function, free_pages_bulk()
>>>>> - Update stats, lruvec_stat in separate loop
>>>>>
>>>>> Changes since v1:
>>>>> - Rebase on mm-new
>>>>> - Rerun benchmarks
>>>>>
>>>>> Made-with: Cursor
>>>>> ---
>>>>>  include/linux/gfp.h |  2 ++
>>>>>  mm/page_alloc.c     | 23 +++++++++++++++++++++++
>>>>>  mm/vmalloc.c        | 16 +++++-----------
>>>>>  3 files changed, 30 insertions(+), 11 deletions(-)
>>>>>
>>>>> diff --git a/include/linux/gfp.h b/include/linux/gfp.h
>>>>> index 7c1f9da7c8e56..71f9097ab99a0 100644
>>>>> --- a/include/linux/gfp.h
>>>>> +++ b/include/linux/gfp.h
>>>>> @@ -239,6 +239,8 @@ unsigned long alloc_pages_bulk_noprof(gfp_t gfp, int preferred_nid,
>>>>>  				struct page **page_array);
>>>>>  #define __alloc_pages_bulk(...)			alloc_hooks(alloc_pages_bulk_noprof(__VA_ARGS__))
>>>>>
>>>>> +void free_pages_bulk(struct page **page_array, unsigned long nr_pages);
>>>>> +
>>>>>  unsigned long alloc_pages_bulk_mempolicy_noprof(gfp_t gfp,
>>>>>  				unsigned long nr_pages,
>>>>>  				struct page **page_array);
>>>>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
>>>>> index eedce9a30eb7e..250cc07e547b8 100644
>>>>> --- a/mm/page_alloc.c
>>>>> +++ b/mm/page_alloc.c
>>>>> @@ -5175,6 +5175,29 @@ unsigned long alloc_pages_bulk_noprof(gfp_t gfp, int preferred_nid,
>>>>>  }
>>>>>  EXPORT_SYMBOL_GPL(alloc_pages_bulk_noprof);
>>>>>
>>>>> +void free_pages_bulk(struct page **page_array, unsigned long nr_pages)
>>>>> +{
>>>>> +	unsigned long start_pfn = 0, pfn;
>>>>> +	unsigned long i, nr_contig = 0;
>>>>> +
>>>>> +	for (i = 0; i < nr_pages; i++) {
>>>>> +		pfn = page_to_pfn(page_array[i]);
>>>>> +		if (!nr_contig) {
>>>>> +			start_pfn = pfn;
>>>>> +			nr_contig = 1;
>>>>> +		} else if (start_pfn + nr_contig != pfn) {
>>>>> +			__free_contig_range(start_pfn, nr_contig);
>>>>> +			start_pfn = pfn;
>>>>> +			nr_contig = 1;
>>>>> +			cond_resched();
>>>>
>>> It will cause schedule while atomic. Have you checked that
>>> __free_contig_range() also can sleep? Of so then we are aligned, if not
>>> probably we should remove it.
>> Sorry, I didn't get it. How does having cond_resched() in this function
>> affects __free_contig_range()?
>>
> It is not. What i am asking is about:
> 
> <snip>
> spin_lock();
> free_pages_bulk()
> ...
> <snip>
> 
> so this is not allowed because there is cond_resched() call. We
> can remove it and make it possible to invoke free_pages_bulk() under
> spin-lock, __but__ only if for example other calls do not sleep:
> 
> __free_contig_range()
> memdesc_section()
> free_prepared_contig_range()
> ...
> 
>>
>> The current user of this function is only vfree() which is sleepable.
>>
> I know. But this function can be used by others soon or later.
> 
> Another option is add a comment, saying that it is only for sleepable
> contexts.
Thank you for detailed response. I can move cond_resched() to vfree() and make
free_pages_bulk() allowed to be called form sleepable context. But I feel the
current implementation is better to avoid latency spikes. I'll put explicit
comment that this function can only be called from sleepable contexts.

Thanks,
Usama

> 
> --
> Uladzislau Rezki