* Re: [PATCH 1/3] memory tiering: hot page selection with hint page fault latency
@ 2022-04-09 4:00 kernel test robot
0 siblings, 0 replies; 4+ messages in thread
From: kernel test robot @ 2022-04-09 4:00 UTC (permalink / raw)
To: kbuild
[-- Attachment #1: Type: text/plain, Size: 10347 bytes --]
CC: kbuild-all(a)lists.01.org
BCC: lkp(a)intel.com
In-Reply-To: <20220408071222.219689-2-ying.huang@intel.com>
References: <20220408071222.219689-2-ying.huang@intel.com>
TO: Huang Ying <ying.huang@intel.com>
Hi Huang,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on tip/sched/core]
[also build test WARNING on mcgrof/sysctl-next linus/master v5.18-rc1 next-20220408]
[cannot apply to hnaz-mm/master]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]
url: https://github.com/intel-lab-lkp/linux/commits/Huang-Ying/memory-tiering-hot-page-selection/20220408-151410
base: https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git 089c02ae2771a14af2928c59c56abfb9b885a8d7
:::::: branch date: 21 hours ago
:::::: commit date: 21 hours ago
config: i386-randconfig-m021 (https://download.01.org/0day-ci/archive/20220409/202204091107.3i51JkZh-lkp(a)intel.com/config)
compiler: gcc-11 (Debian 11.2.0-19) 11.2.0
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
smatch warnings:
mm/memory.c:4456 do_numa_page() warn: bitwise AND condition is false here
vim +4456 mm/memory.c
9532fec118d485e Mel Gorman 2012-11-15 4397
2b7403035459c75 Souptick Joarder 2018-08-23 4398 static vm_fault_t do_numa_page(struct vm_fault *vmf)
d10e63f29488b0f Mel Gorman 2012-10-25 4399 {
82b0f8c39a3869b Jan Kara 2016-12-14 4400 struct vm_area_struct *vma = vmf->vma;
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4401 struct page *page = NULL;
98fa15f34cb3798 Anshuman Khandual 2019-03-05 4402 int page_nid = NUMA_NO_NODE;
90572890d202527 Peter Zijlstra 2013-10-07 4403 int last_cpupid;
cbee9f88ec1b8dd Peter Zijlstra 2012-10-25 4404 int target_nid;
04a8645304500be Aneesh Kumar K.V 2019-03-05 4405 pte_t pte, old_pte;
288bc54949fc262 Aneesh Kumar K.V 2017-02-24 4406 bool was_writable = pte_savedwrite(vmf->orig_pte);
6688cc05473b36a Peter Zijlstra 2013-10-07 4407 int flags = 0;
d10e63f29488b0f Mel Gorman 2012-10-25 4408
d10e63f29488b0f Mel Gorman 2012-10-25 4409 /*
d10e63f29488b0f Mel Gorman 2012-10-25 4410 * The "pte" at this point cannot be used safely without
d10e63f29488b0f Mel Gorman 2012-10-25 4411 * validation through pte_unmap_same(). It's of NUMA type but
d10e63f29488b0f Mel Gorman 2012-10-25 4412 * the pfn may be screwed if the read is non atomic.
d10e63f29488b0f Mel Gorman 2012-10-25 4413 */
82b0f8c39a3869b Jan Kara 2016-12-14 4414 vmf->ptl = pte_lockptr(vma->vm_mm, vmf->pmd);
82b0f8c39a3869b Jan Kara 2016-12-14 4415 spin_lock(vmf->ptl);
cee216a696b2004 Aneesh Kumar K.V 2017-02-24 4416 if (unlikely(!pte_same(*vmf->pte, vmf->orig_pte))) {
82b0f8c39a3869b Jan Kara 2016-12-14 4417 pte_unmap_unlock(vmf->pte, vmf->ptl);
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4418 goto out;
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4419 }
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4420
b99a342d4f11a54 Huang Ying 2021-04-29 4421 /* Get the normal PTE */
b99a342d4f11a54 Huang Ying 2021-04-29 4422 old_pte = ptep_get(vmf->pte);
04a8645304500be Aneesh Kumar K.V 2019-03-05 4423 pte = pte_modify(old_pte, vma->vm_page_prot);
d10e63f29488b0f Mel Gorman 2012-10-25 4424
82b0f8c39a3869b Jan Kara 2016-12-14 4425 page = vm_normal_page(vma, vmf->address, pte);
b99a342d4f11a54 Huang Ying 2021-04-29 4426 if (!page)
b99a342d4f11a54 Huang Ying 2021-04-29 4427 goto out_map;
d10e63f29488b0f Mel Gorman 2012-10-25 4428
e81c48024f43b4a Kirill A. Shutemov 2016-01-15 4429 /* TODO: handle PTE-mapped THP */
b99a342d4f11a54 Huang Ying 2021-04-29 4430 if (PageCompound(page))
b99a342d4f11a54 Huang Ying 2021-04-29 4431 goto out_map;
e81c48024f43b4a Kirill A. Shutemov 2016-01-15 4432
6688cc05473b36a Peter Zijlstra 2013-10-07 4433 /*
bea66fbd11af1ca Mel Gorman 2015-03-25 4434 * Avoid grouping on RO pages in general. RO pages shouldn't hurt as
bea66fbd11af1ca Mel Gorman 2015-03-25 4435 * much anyway since they can be in shared cache state. This misses
bea66fbd11af1ca Mel Gorman 2015-03-25 4436 * the case where a mapping is writable but the process never writes
bea66fbd11af1ca Mel Gorman 2015-03-25 4437 * to it but pte_write gets cleared during protection updates and
bea66fbd11af1ca Mel Gorman 2015-03-25 4438 * pte_dirty has unpredictable behaviour between PTE scan updates,
bea66fbd11af1ca Mel Gorman 2015-03-25 4439 * background writeback, dirty balancing and application behaviour.
bea66fbd11af1ca Mel Gorman 2015-03-25 4440 */
b99a342d4f11a54 Huang Ying 2021-04-29 4441 if (!was_writable)
6688cc05473b36a Peter Zijlstra 2013-10-07 4442 flags |= TNF_NO_GROUP;
6688cc05473b36a Peter Zijlstra 2013-10-07 4443
dabe1d992414a64 Rik van Riel 2013-10-07 4444 /*
dabe1d992414a64 Rik van Riel 2013-10-07 4445 * Flag if the page is shared between multiple address spaces. This
dabe1d992414a64 Rik van Riel 2013-10-07 4446 * is later used when determining whether to group tasks together
dabe1d992414a64 Rik van Riel 2013-10-07 4447 */
dabe1d992414a64 Rik van Riel 2013-10-07 4448 if (page_mapcount(page) > 1 && (vma->vm_flags & VM_SHARED))
dabe1d992414a64 Rik van Riel 2013-10-07 4449 flags |= TNF_SHARED;
dabe1d992414a64 Rik van Riel 2013-10-07 4450
8191acbd30c73e4 Mel Gorman 2013-10-07 4451 page_nid = page_to_nid(page);
e4e875b64dea3af Huang Ying 2022-04-08 4452 /*
e4e875b64dea3af Huang Ying 2022-04-08 4453 * In memory tiering mode, cpupid of slow memory page is used
e4e875b64dea3af Huang Ying 2022-04-08 4454 * to record page access time. So use default value.
e4e875b64dea3af Huang Ying 2022-04-08 4455 */
e4e875b64dea3af Huang Ying 2022-04-08 @4456 if ((sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) &&
e4e875b64dea3af Huang Ying 2022-04-08 4457 !node_is_toptier(page_nid))
e4e875b64dea3af Huang Ying 2022-04-08 4458 last_cpupid = (-1 & LAST_CPUPID_MASK);
e4e875b64dea3af Huang Ying 2022-04-08 4459 else
e4e875b64dea3af Huang Ying 2022-04-08 4460 last_cpupid = page_cpupid_last(page);
82b0f8c39a3869b Jan Kara 2016-12-14 4461 target_nid = numa_migrate_prep(page, vma, vmf->address, page_nid,
bae473a423f65e4 Kirill A. Shutemov 2016-07-26 4462 &flags);
98fa15f34cb3798 Anshuman Khandual 2019-03-05 4463 if (target_nid == NUMA_NO_NODE) {
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4464 put_page(page);
b99a342d4f11a54 Huang Ying 2021-04-29 4465 goto out_map;
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4466 }
b99a342d4f11a54 Huang Ying 2021-04-29 4467 pte_unmap_unlock(vmf->pte, vmf->ptl);
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4468
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4469 /* Migrate to the requested node */
bf90ac198e30d24 Wang Qing 2021-04-29 4470 if (migrate_misplaced_page(page, vma, target_nid)) {
8191acbd30c73e4 Mel Gorman 2013-10-07 4471 page_nid = target_nid;
6688cc05473b36a Peter Zijlstra 2013-10-07 4472 flags |= TNF_MIGRATED;
b99a342d4f11a54 Huang Ying 2021-04-29 4473 } else {
074c238177a75f5 Mel Gorman 2015-03-25 4474 flags |= TNF_MIGRATE_FAIL;
b99a342d4f11a54 Huang Ying 2021-04-29 4475 vmf->pte = pte_offset_map(vmf->pmd, vmf->address);
b99a342d4f11a54 Huang Ying 2021-04-29 4476 spin_lock(vmf->ptl);
b99a342d4f11a54 Huang Ying 2021-04-29 4477 if (unlikely(!pte_same(*vmf->pte, vmf->orig_pte))) {
b99a342d4f11a54 Huang Ying 2021-04-29 4478 pte_unmap_unlock(vmf->pte, vmf->ptl);
b99a342d4f11a54 Huang Ying 2021-04-29 4479 goto out;
b99a342d4f11a54 Huang Ying 2021-04-29 4480 }
b99a342d4f11a54 Huang Ying 2021-04-29 4481 goto out_map;
b99a342d4f11a54 Huang Ying 2021-04-29 4482 }
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4483
4daae3b4b9e49b7 Mel Gorman 2012-11-02 4484 out:
98fa15f34cb3798 Anshuman Khandual 2019-03-05 4485 if (page_nid != NUMA_NO_NODE)
6688cc05473b36a Peter Zijlstra 2013-10-07 4486 task_numa_fault(last_cpupid, page_nid, 1, flags);
d10e63f29488b0f Mel Gorman 2012-10-25 4487 return 0;
b99a342d4f11a54 Huang Ying 2021-04-29 4488 out_map:
b99a342d4f11a54 Huang Ying 2021-04-29 4489 /*
b99a342d4f11a54 Huang Ying 2021-04-29 4490 * Make it present again, depending on how arch implements
b99a342d4f11a54 Huang Ying 2021-04-29 4491 * non-accessible ptes, some can allow access by kernel mode.
b99a342d4f11a54 Huang Ying 2021-04-29 4492 */
b99a342d4f11a54 Huang Ying 2021-04-29 4493 old_pte = ptep_modify_prot_start(vma, vmf->address, vmf->pte);
b99a342d4f11a54 Huang Ying 2021-04-29 4494 pte = pte_modify(old_pte, vma->vm_page_prot);
b99a342d4f11a54 Huang Ying 2021-04-29 4495 pte = pte_mkyoung(pte);
b99a342d4f11a54 Huang Ying 2021-04-29 4496 if (was_writable)
b99a342d4f11a54 Huang Ying 2021-04-29 4497 pte = pte_mkwrite(pte);
b99a342d4f11a54 Huang Ying 2021-04-29 4498 ptep_modify_prot_commit(vma, vmf->address, vmf->pte, old_pte, pte);
b99a342d4f11a54 Huang Ying 2021-04-29 4499 update_mmu_cache(vma, vmf->address, vmf->pte);
b99a342d4f11a54 Huang Ying 2021-04-29 4500 pte_unmap_unlock(vmf->pte, vmf->ptl);
b99a342d4f11a54 Huang Ying 2021-04-29 4501 goto out;
d10e63f29488b0f Mel Gorman 2012-10-25 4502 }
d10e63f29488b0f Mel Gorman 2012-10-25 4503
--
0-DAY CI Kernel Test Service
https://01.org/lkp
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH 1/3] memory tiering: hot page selection with hint page fault latency
2022-04-14 13:23 ` Jagdish Gediya
@ 2022-04-15 2:42 ` ying.huang
0 siblings, 0 replies; 4+ messages in thread
From: ying.huang @ 2022-04-15 2:42 UTC (permalink / raw)
To: Jagdish Gediya
Cc: Peter Zijlstra, Mel Gorman, Andrew Morton, linux-mm,
linux-kernel, Michal Hocko, Rik van Riel, Dave Hansen, Yang Shi,
Zi Yan, Wei Xu, osalvador, Shakeel Butt, Zhong Jiang
Hi, Jagdish,
On Thu, 2022-04-14 at 18:53 +0530, Jagdish Gediya wrote:
> On Fri, Apr 08, 2022 at 03:12:20PM +0800, Huang Ying wrote:
[snip]
> > +
> > +static int numa_hint_fault_latency(struct page *page)
> > +{
> > + int last_time, time;
> > +
> > + time = jiffies_to_msecs(jiffies);
> > + last_time = xchg_page_access_time(page, time);
> > +
> > + return (time - last_time) & PAGE_ACCESS_TIME_MASK;
>
> This code can possibly consider cold page as hot,
>
> Assume,
>
> LAST_CPUPID_SHIFT = 12
> PAGE_ACCESS_TIME_BUCKETS = 0
> sysctl_numa_balancing_hot_threshold = 1000
>
> Assume while changing pte,
> jiffies_to_msecs(jiffies) = 0xAABB0100
>
> So value saved in page->flags will be lowest 12 bits of 0xAABB0100
> which is 0x100.
>
> Assume when numa_hint_fault_latency() gets called,
> time = jiffies_to_msecs(jiffies) = 0xAACC0100
>
> So, time = 0xAACC0100, and last_time = 0x100,
> time - last_time = 0xAACC0100 - 0x100 = 0xAACC0000
> 0xAACC0000 & PAGE_ACCESS_TIME_MASK = 0xAACC0000 & ((1 << 12) - 1) = 0
>
> so the return value of this function is 0, the code will consider it as
> hot page but it is cold page because actual difference is
> 0xAACC0100 - 0xAABB0100 = 110000 ms
>
Yes. This is possible.
> There may be more such scenarios. What do you think?
The algorithm just works statistically correct. That is, for really hot
pages, their hint page fault latency will be short and we can promote it
when they are accessed. For cold pages, it's still possible for them to
be identified as hot pages. But the possibility is much lower than that
of the hot pages.
We can try to improve further here. But as the first step, I want to
keep the algorithm as simple as possible. Then we can try improve it
step by step and show benefit in each step to justify the further
optimization.
> > +}
> > +
Best Regards,
Huang, Ying
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH 1/3] memory tiering: hot page selection with hint page fault latency
2022-04-08 7:12 ` [PATCH 1/3] memory tiering: hot page selection with hint page fault latency Huang Ying
@ 2022-04-14 13:23 ` Jagdish Gediya
2022-04-15 2:42 ` ying.huang
0 siblings, 1 reply; 4+ messages in thread
From: Jagdish Gediya @ 2022-04-14 13:23 UTC (permalink / raw)
To: Huang Ying
Cc: Peter Zijlstra, Mel Gorman, Andrew Morton, linux-mm,
linux-kernel, Michal Hocko, Rik van Riel, Dave Hansen, Yang Shi,
Zi Yan, Wei Xu, osalvador, Shakeel Butt, Zhong Jiang
On Fri, Apr 08, 2022 at 03:12:20PM +0800, Huang Ying wrote:
> To optimize page placement in a memory tiering system with NUMA
> balancing, the hot pages in the slow memory node need to be
> identified. Essentially, the original NUMA balancing implementation
> selects the mostly recently accessed (MRU) pages as the hot pages.
> But this isn't a very good algorithm to identify the hot pages.
>
> So, in this patch we implemented a better hot page selection
> algorithm. Which is based on NUMA balancing page table scanning and
> hint page fault as follows,
>
> - When the page tables of the processes are scanned to change PTE/PMD
> to be PROT_NONE, the current time is recorded in struct page as scan
> time.
>
> - When the page is accessed, hint page fault will occur. The scan
> time is gotten from the struct page. And The hint page fault
> latency is defined as
>
> hint page fault time - scan time
>
> The shorter the hint page fault latency of a page is, the higher the
> probability of their access frequency to be higher. So the hint page
> fault latency is a good estimation of the page hot/cold.
>
> But it's hard to find some extra space in struct page to hold the scan
> time. Fortunately, we can reuse some bits used by the original NUMA
> balancing.
>
> NUMA balancing uses some bits in struct page to store the page
> accessing CPU and PID (referring to page_cpupid_xchg_last()). Which
> is used by the multi-stage node selection algorithm to avoid to
> migrate pages shared accessed by the NUMA nodes back and forth. But
> for pages in the slow memory node, even if they are shared accessed by
> multiple NUMA nodes, as long as the pages are hot, they need to be
> promoted to the fast memory node. So the accessing CPU and PID
> information are unnecessary for the slow memory pages. We can reuse
> these bits in struct page to record the scan time for them. For the
> fast memory pages, these bits are used as before.
>
> For the hot threshold, the default value is 1 second, which works well
> in our performance test. All pages with hint page fault latency < the
> threshold will be considered hot. A debugfs interface is also
> provided to adjust the hot threshold.
>
> The downside of the above method is that the response time to the
> workload hot spot changing may be much longer. For example,
>
> - A previous cold memory area becomes hot
>
> - The hint page fault will be triggered. But the hint page fault
> latency isn't shorter than the hot threshold. So the pages will
> not be promoted.
>
> - When the memory area is scanned again, maybe after a scan period,
> the hint page fault latency measured will be shorter than the hot
> threshold and the pages will be promoted.
>
> To mitigate this,
>
> - If there are enough free space in the fast memory node, the hot
> threshold will not be used, all pages will be promoted upon the hint
> page fault for fast response.
>
> - If fast response is more important for system performance, the
> administrator can set a higher hot threshold.
>
> Thanks Zhong Jiang reported and tested the fix for a bug when
> disabling memory tiering mode dynamically.
>
> Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Michal Hocko <mhocko@suse.com>
> Cc: Rik van Riel <riel@surriel.com>
> Cc: Mel Gorman <mgorman@techsingularity.net>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Dave Hansen <dave.hansen@linux.intel.com>
> Cc: Yang Shi <shy828301@gmail.com>
> Cc: Zi Yan <ziy@nvidia.com>
> Cc: Wei Xu <weixugc@google.com>
> Cc: osalvador <osalvador@suse.de>
> Cc: Shakeel Butt <shakeelb@google.com>
> Cc: Zhong Jiang <zhongjiang-ali@linux.alibaba.com>
> Cc: linux-kernel@vger.kernel.org
> Cc: linux-mm@kvack.org
> ---
> include/linux/mm.h | 30 ++++++++++++++++++
> kernel/sched/debug.c | 1 +
> kernel/sched/fair.c | 74 ++++++++++++++++++++++++++++++++++++++++++++
> kernel/sched/sched.h | 1 +
> mm/huge_memory.c | 13 ++++++--
> mm/memory.c | 11 ++++++-
> mm/migrate.c | 12 +++++++
> mm/mprotect.c | 8 ++++-
> 8 files changed, 145 insertions(+), 5 deletions(-)
>
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index e34edb775334..455a3d0e699d 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1311,6 +1311,18 @@ static inline int folio_nid(const struct folio *folio)
> }
>
> #ifdef CONFIG_NUMA_BALANCING
> +/* page access time bits needs to hold at least 4 seconds */
> +#define PAGE_ACCESS_TIME_MIN_BITS 12
> +#if LAST_CPUPID_SHIFT < PAGE_ACCESS_TIME_MIN_BITS
> +#define PAGE_ACCESS_TIME_BUCKETS \
> + (PAGE_ACCESS_TIME_MIN_BITS - LAST_CPUPID_SHIFT)
> +#else
> +#define PAGE_ACCESS_TIME_BUCKETS 0
> +#endif
> +
> +#define PAGE_ACCESS_TIME_MASK \
> + (LAST_CPUPID_MASK << PAGE_ACCESS_TIME_BUCKETS)
> +
> static inline int cpu_pid_to_cpupid(int cpu, int pid)
> {
> return ((cpu & LAST__CPU_MASK) << LAST__PID_SHIFT) | (pid & LAST__PID_MASK);
> @@ -1346,6 +1358,11 @@ static inline bool __cpupid_match_pid(pid_t task_pid, int cpupid)
> return (task_pid & LAST__PID_MASK) == cpupid_to_pid(cpupid);
> }
>
> +static inline bool check_cpupid(int cpupid)
> +{
> + return cpupid_to_cpu(cpupid) < nr_cpu_ids;
> +}
> +
> #define cpupid_match_pid(task, cpupid) __cpupid_match_pid(task->pid, cpupid)
> #ifdef LAST_CPUPID_NOT_IN_PAGE_FLAGS
> static inline int page_cpupid_xchg_last(struct page *page, int cpupid)
> @@ -1374,12 +1391,25 @@ static inline void page_cpupid_reset_last(struct page *page)
> page->flags |= LAST_CPUPID_MASK << LAST_CPUPID_PGSHIFT;
> }
> #endif /* LAST_CPUPID_NOT_IN_PAGE_FLAGS */
> +
> +static inline int xchg_page_access_time(struct page *page, int time)
> +{
> + int last_time;
> +
> + last_time = page_cpupid_xchg_last(page, time >> PAGE_ACCESS_TIME_BUCKETS);
> + return last_time << PAGE_ACCESS_TIME_BUCKETS;
> +}
> #else /* !CONFIG_NUMA_BALANCING */
> static inline int page_cpupid_xchg_last(struct page *page, int cpupid)
> {
> return page_to_nid(page); /* XXX */
> }
>
> +static inline int xchg_page_access_time(struct page *page, int time)
> +{
> + return 0;
> +}
> +
> static inline int page_cpupid_last(struct page *page)
> {
> return page_to_nid(page); /* XXX */
> diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
> index bb3d63bdf4ae..ad63dbfc54f1 100644
> --- a/kernel/sched/debug.c
> +++ b/kernel/sched/debug.c
> @@ -333,6 +333,7 @@ static __init int sched_init_debug(void)
> debugfs_create_u32("scan_period_min_ms", 0644, numa, &sysctl_numa_balancing_scan_period_min);
> debugfs_create_u32("scan_period_max_ms", 0644, numa, &sysctl_numa_balancing_scan_period_max);
> debugfs_create_u32("scan_size_mb", 0644, numa, &sysctl_numa_balancing_scan_size);
> + debugfs_create_u32("hot_threshold_ms", 0644, numa, &sysctl_numa_balancing_hot_threshold);
> #endif
>
> debugfs_create_file("debug", 0444, debugfs_sched, NULL, &sched_debug_fops);
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index d4bd299d67ab..cb130ea46c71 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -1058,6 +1058,9 @@ unsigned int sysctl_numa_balancing_scan_size = 256;
> /* Scan @scan_size MB every @scan_period after an initial @scan_delay in ms */
> unsigned int sysctl_numa_balancing_scan_delay = 1000;
>
> +/* The page with hint page fault latency < threshold in ms is considered hot */
> +unsigned int sysctl_numa_balancing_hot_threshold = 1000;
> +
> struct numa_group {
> refcount_t refcount;
>
> @@ -1400,6 +1403,37 @@ static inline unsigned long group_weight(struct task_struct *p, int nid,
> return 1000 * faults / total_faults;
> }
>
> +static bool pgdat_free_space_enough(struct pglist_data *pgdat)
> +{
> + int z;
> + unsigned long enough_mark;
> +
> + enough_mark = max(1UL * 1024 * 1024 * 1024 >> PAGE_SHIFT,
> + pgdat->node_present_pages >> 4);
> + for (z = pgdat->nr_zones - 1; z >= 0; z--) {
> + struct zone *zone = pgdat->node_zones + z;
> +
> + if (!populated_zone(zone))
> + continue;
> +
> + if (zone_watermark_ok(zone, 0,
> + high_wmark_pages(zone) + enough_mark,
> + ZONE_MOVABLE, 0))
> + return true;
> + }
> + return false;
> +}
> +
> +static int numa_hint_fault_latency(struct page *page)
> +{
> + int last_time, time;
> +
> + time = jiffies_to_msecs(jiffies);
> + last_time = xchg_page_access_time(page, time);
> +
> + return (time - last_time) & PAGE_ACCESS_TIME_MASK;
This code can possibly consider cold page as hot,
Assume,
LAST_CPUPID_SHIFT = 12
PAGE_ACCESS_TIME_BUCKETS = 0
sysctl_numa_balancing_hot_threshold = 1000
Assume while changing pte,
jiffies_to_msecs(jiffies) = 0xAABB0100
So value saved in page->flags will be lowest 12 bits of 0xAABB0100
which is 0x100.
Assume when numa_hint_fault_latency() gets called,
time = jiffies_to_msecs(jiffies) = 0xAACC0100
So, time = 0xAACC0100, and last_time = 0x100,
time - last_time = 0xAACC0100 - 0x100 = 0xAACC0000
0xAACC0000 & PAGE_ACCESS_TIME_MASK = 0xAACC0000 & ((1 << 12) - 1) = 0
so the return value of this function is 0, the code will consider it as
hot page but it is cold page because actual difference is
0xAACC0100 - 0xAABB0100 = 110000 ms
There may be more such scenarios. What do you think?
> +}
> +
> bool should_numa_migrate_memory(struct task_struct *p, struct page * page,
> int src_nid, int dst_cpu)
> {
> @@ -1407,9 +1441,38 @@ bool should_numa_migrate_memory(struct task_struct *p, struct page * page,
> int dst_nid = cpu_to_node(dst_cpu);
> int last_cpupid, this_cpupid;
>
> + /*
> + * The pages in slow memory node should be migrated according
> + * to hot/cold instead of accessing CPU node.
> + */
> + if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING &&
> + !node_is_toptier(src_nid)) {
> + struct pglist_data *pgdat;
> + unsigned long latency, th;
> +
> + pgdat = NODE_DATA(dst_nid);
> + if (pgdat_free_space_enough(pgdat))
> + return true;
> +
> + th = sysctl_numa_balancing_hot_threshold;
> + latency = numa_hint_fault_latency(page);
> + if (latency >= th)
> + return false;
> +
> + return true;
> + }
> +
> this_cpupid = cpu_pid_to_cpupid(dst_cpu, current->pid);
> last_cpupid = page_cpupid_xchg_last(page, this_cpupid);
>
> + /*
> + * The cpupid may be invalid when NUMA_BALANCING_MEMORY_TIERING
> + * is disabled dynamically.
> + */
> + if (!(sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) &&
> + !node_is_toptier(src_nid) && !check_cpupid(last_cpupid))
> + return false;
> +
> /*
> * Allow first faults or private faults to migrate immediately early in
> * the lifetime of a task. The magic number 4 is based on waiting for
> @@ -2642,6 +2705,17 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags)
> if (!p->mm)
> return;
>
> + /*
> + * NUMA faults statistics are unnecessary for the slow memory node.
> + *
> + * And, the cpupid may be invalid when NUMA_BALANCING_MEMORY_TIERING
> + * is disabled dynamically.
> + */
> + if (!node_is_toptier(mem_node) &&
> + (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING ||
> + !check_cpupid(last_cpupid)))
> + return;
> +
> /* Allocate buffer to track faults on a per-node basis */
> if (unlikely(!p->numa_faults)) {
> int size = sizeof(*p->numa_faults) *
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index 58263f90c559..86ce60d3d472 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -2408,6 +2408,7 @@ extern unsigned int sysctl_numa_balancing_scan_delay;
> extern unsigned int sysctl_numa_balancing_scan_period_min;
> extern unsigned int sysctl_numa_balancing_scan_period_max;
> extern unsigned int sysctl_numa_balancing_scan_size;
> +extern unsigned int sysctl_numa_balancing_hot_threshold;
> #endif
>
> #ifdef CONFIG_SCHED_HRTICK
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 2fe38212e07c..3a715eeeebb5 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -1401,7 +1401,7 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf)
> struct page *page;
> unsigned long haddr = vmf->address & HPAGE_PMD_MASK;
> int page_nid = NUMA_NO_NODE;
> - int target_nid, last_cpupid = -1;
> + int target_nid, last_cpupid = (-1 & LAST_CPUPID_MASK);
> bool migrated = false;
> bool was_writable = pmd_savedwrite(oldpmd);
> int flags = 0;
> @@ -1422,7 +1422,8 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf)
> flags |= TNF_NO_GROUP;
>
> page_nid = page_to_nid(page);
> - last_cpupid = page_cpupid_last(page);
> + if (node_is_toptier(page_nid))
> + last_cpupid = page_cpupid_last(page);
> target_nid = numa_migrate_prep(page, vma, haddr, page_nid,
> &flags);
>
> @@ -1740,6 +1741,7 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
>
> if (prot_numa) {
> struct page *page;
> + bool toptier;
> /*
> * Avoid trapping faults against the zero page. The read-only
> * data is likely to be read-cached on the local CPU and
> @@ -1752,13 +1754,18 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
> goto unlock;
>
> page = pmd_page(*pmd);
> + toptier = node_is_toptier(page_to_nid(page));
> /*
> * Skip scanning top tier node if normal numa
> * balancing is disabled
> */
> if (!(sysctl_numa_balancing_mode & NUMA_BALANCING_NORMAL) &&
> - node_is_toptier(page_to_nid(page)))
> + toptier)
> goto unlock;
> +
> + if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING &&
> + !toptier)
> + xchg_page_access_time(page, jiffies_to_msecs(jiffies));
> }
> /*
> * In case prot_numa, we are under mmap_read_lock(mm). It's critical
> diff --git a/mm/memory.c b/mm/memory.c
> index b370ed118767..a8ac15ce7a75 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -74,6 +74,7 @@
> #include <linux/perf_event.h>
> #include <linux/ptrace.h>
> #include <linux/vmalloc.h>
> +#include <linux/sched/sysctl.h>
>
> #include <trace/events/kmem.h>
>
> @@ -4455,8 +4456,16 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf)
> if (page_mapcount(page) > 1 && (vma->vm_flags & VM_SHARED))
> flags |= TNF_SHARED;
>
> - last_cpupid = page_cpupid_last(page);
> page_nid = page_to_nid(page);
> + /*
> + * In memory tiering mode, cpupid of slow memory page is used
> + * to record page access time. So use default value.
> + */
> + if ((sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) &&
> + !node_is_toptier(page_nid))
> + last_cpupid = (-1 & LAST_CPUPID_MASK);
> + else
> + last_cpupid = page_cpupid_last(page);
> target_nid = numa_migrate_prep(page, vma, vmf->address, page_nid,
> &flags);
> if (target_nid == NUMA_NO_NODE) {
> diff --git a/mm/migrate.c b/mm/migrate.c
> index dc84edfae842..e73f26dfeb38 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -534,6 +534,18 @@ void folio_migrate_flags(struct folio *newfolio, struct folio *folio)
> * future migrations of this same page.
> */
> cpupid = page_cpupid_xchg_last(&folio->page, -1);
> + /*
> + * If migrate between slow and fast memory node, reset cpupid,
> + * because that is used to record page access time in slow
> + * memory node
> + */
> + if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) {
> + bool f_toptier = node_is_toptier(page_to_nid(&folio->page));
> + bool t_toptier = node_is_toptier(page_to_nid(&newfolio->page));
> +
> + if (f_toptier != t_toptier)
> + cpupid = -1;
> + }
> page_cpupid_xchg_last(&newfolio->page, cpupid);
>
> folio_migrate_ksm(newfolio, folio);
> diff --git a/mm/mprotect.c b/mm/mprotect.c
> index b69ce7a7b2b7..e7cb90d84fac 100644
> --- a/mm/mprotect.c
> +++ b/mm/mprotect.c
> @@ -85,6 +85,7 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
> if (prot_numa) {
> struct page *page;
> int nid;
> + bool toptier;
>
> /* Avoid TLB flush if possible */
> if (pte_protnone(oldpte))
> @@ -114,14 +115,19 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
> nid = page_to_nid(page);
> if (target_node == nid)
> continue;
> + toptier = node_is_toptier(nid);
>
> /*
> * Skip scanning top tier node if normal numa
> * balancing is disabled
> */
> if (!(sysctl_numa_balancing_mode & NUMA_BALANCING_NORMAL) &&
> - node_is_toptier(nid))
> + toptier)
> continue;
> + if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING &&
> + !toptier)
> + xchg_page_access_time(page,
> + jiffies_to_msecs(jiffies));
> }
>
> oldpte = ptep_modify_prot_start(vma, addr, pte);
> --
> 2.30.2
>
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH 1/3] memory tiering: hot page selection with hint page fault latency
2022-04-08 7:12 [PATCH 0/3] memory tiering: hot page selection Huang Ying
@ 2022-04-08 7:12 ` Huang Ying
2022-04-14 13:23 ` Jagdish Gediya
0 siblings, 1 reply; 4+ messages in thread
From: Huang Ying @ 2022-04-08 7:12 UTC (permalink / raw)
To: Peter Zijlstra, Mel Gorman, Andrew Morton
Cc: linux-mm, linux-kernel, Huang Ying, Michal Hocko, Rik van Riel,
Dave Hansen, Yang Shi, Zi Yan, Wei Xu, osalvador, Shakeel Butt,
Zhong Jiang
To optimize page placement in a memory tiering system with NUMA
balancing, the hot pages in the slow memory node need to be
identified. Essentially, the original NUMA balancing implementation
selects the mostly recently accessed (MRU) pages as the hot pages.
But this isn't a very good algorithm to identify the hot pages.
So, in this patch we implemented a better hot page selection
algorithm. Which is based on NUMA balancing page table scanning and
hint page fault as follows,
- When the page tables of the processes are scanned to change PTE/PMD
to be PROT_NONE, the current time is recorded in struct page as scan
time.
- When the page is accessed, hint page fault will occur. The scan
time is gotten from the struct page. And The hint page fault
latency is defined as
hint page fault time - scan time
The shorter the hint page fault latency of a page is, the higher the
probability of their access frequency to be higher. So the hint page
fault latency is a good estimation of the page hot/cold.
But it's hard to find some extra space in struct page to hold the scan
time. Fortunately, we can reuse some bits used by the original NUMA
balancing.
NUMA balancing uses some bits in struct page to store the page
accessing CPU and PID (referring to page_cpupid_xchg_last()). Which
is used by the multi-stage node selection algorithm to avoid to
migrate pages shared accessed by the NUMA nodes back and forth. But
for pages in the slow memory node, even if they are shared accessed by
multiple NUMA nodes, as long as the pages are hot, they need to be
promoted to the fast memory node. So the accessing CPU and PID
information are unnecessary for the slow memory pages. We can reuse
these bits in struct page to record the scan time for them. For the
fast memory pages, these bits are used as before.
For the hot threshold, the default value is 1 second, which works well
in our performance test. All pages with hint page fault latency < the
threshold will be considered hot. A debugfs interface is also
provided to adjust the hot threshold.
The downside of the above method is that the response time to the
workload hot spot changing may be much longer. For example,
- A previous cold memory area becomes hot
- The hint page fault will be triggered. But the hint page fault
latency isn't shorter than the hot threshold. So the pages will
not be promoted.
- When the memory area is scanned again, maybe after a scan period,
the hint page fault latency measured will be shorter than the hot
threshold and the pages will be promoted.
To mitigate this,
- If there are enough free space in the fast memory node, the hot
threshold will not be used, all pages will be promoted upon the hint
page fault for fast response.
- If fast response is more important for system performance, the
administrator can set a higher hot threshold.
Thanks Zhong Jiang reported and tested the fix for a bug when
disabling memory tiering mode dynamically.
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Wei Xu <weixugc@google.com>
Cc: osalvador <osalvador@suse.de>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Zhong Jiang <zhongjiang-ali@linux.alibaba.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
---
include/linux/mm.h | 30 ++++++++++++++++++
kernel/sched/debug.c | 1 +
kernel/sched/fair.c | 74 ++++++++++++++++++++++++++++++++++++++++++++
kernel/sched/sched.h | 1 +
mm/huge_memory.c | 13 ++++++--
mm/memory.c | 11 ++++++-
mm/migrate.c | 12 +++++++
mm/mprotect.c | 8 ++++-
8 files changed, 145 insertions(+), 5 deletions(-)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index e34edb775334..455a3d0e699d 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1311,6 +1311,18 @@ static inline int folio_nid(const struct folio *folio)
}
#ifdef CONFIG_NUMA_BALANCING
+/* page access time bits needs to hold at least 4 seconds */
+#define PAGE_ACCESS_TIME_MIN_BITS 12
+#if LAST_CPUPID_SHIFT < PAGE_ACCESS_TIME_MIN_BITS
+#define PAGE_ACCESS_TIME_BUCKETS \
+ (PAGE_ACCESS_TIME_MIN_BITS - LAST_CPUPID_SHIFT)
+#else
+#define PAGE_ACCESS_TIME_BUCKETS 0
+#endif
+
+#define PAGE_ACCESS_TIME_MASK \
+ (LAST_CPUPID_MASK << PAGE_ACCESS_TIME_BUCKETS)
+
static inline int cpu_pid_to_cpupid(int cpu, int pid)
{
return ((cpu & LAST__CPU_MASK) << LAST__PID_SHIFT) | (pid & LAST__PID_MASK);
@@ -1346,6 +1358,11 @@ static inline bool __cpupid_match_pid(pid_t task_pid, int cpupid)
return (task_pid & LAST__PID_MASK) == cpupid_to_pid(cpupid);
}
+static inline bool check_cpupid(int cpupid)
+{
+ return cpupid_to_cpu(cpupid) < nr_cpu_ids;
+}
+
#define cpupid_match_pid(task, cpupid) __cpupid_match_pid(task->pid, cpupid)
#ifdef LAST_CPUPID_NOT_IN_PAGE_FLAGS
static inline int page_cpupid_xchg_last(struct page *page, int cpupid)
@@ -1374,12 +1391,25 @@ static inline void page_cpupid_reset_last(struct page *page)
page->flags |= LAST_CPUPID_MASK << LAST_CPUPID_PGSHIFT;
}
#endif /* LAST_CPUPID_NOT_IN_PAGE_FLAGS */
+
+static inline int xchg_page_access_time(struct page *page, int time)
+{
+ int last_time;
+
+ last_time = page_cpupid_xchg_last(page, time >> PAGE_ACCESS_TIME_BUCKETS);
+ return last_time << PAGE_ACCESS_TIME_BUCKETS;
+}
#else /* !CONFIG_NUMA_BALANCING */
static inline int page_cpupid_xchg_last(struct page *page, int cpupid)
{
return page_to_nid(page); /* XXX */
}
+static inline int xchg_page_access_time(struct page *page, int time)
+{
+ return 0;
+}
+
static inline int page_cpupid_last(struct page *page)
{
return page_to_nid(page); /* XXX */
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index bb3d63bdf4ae..ad63dbfc54f1 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -333,6 +333,7 @@ static __init int sched_init_debug(void)
debugfs_create_u32("scan_period_min_ms", 0644, numa, &sysctl_numa_balancing_scan_period_min);
debugfs_create_u32("scan_period_max_ms", 0644, numa, &sysctl_numa_balancing_scan_period_max);
debugfs_create_u32("scan_size_mb", 0644, numa, &sysctl_numa_balancing_scan_size);
+ debugfs_create_u32("hot_threshold_ms", 0644, numa, &sysctl_numa_balancing_hot_threshold);
#endif
debugfs_create_file("debug", 0444, debugfs_sched, NULL, &sched_debug_fops);
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index d4bd299d67ab..cb130ea46c71 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -1058,6 +1058,9 @@ unsigned int sysctl_numa_balancing_scan_size = 256;
/* Scan @scan_size MB every @scan_period after an initial @scan_delay in ms */
unsigned int sysctl_numa_balancing_scan_delay = 1000;
+/* The page with hint page fault latency < threshold in ms is considered hot */
+unsigned int sysctl_numa_balancing_hot_threshold = 1000;
+
struct numa_group {
refcount_t refcount;
@@ -1400,6 +1403,37 @@ static inline unsigned long group_weight(struct task_struct *p, int nid,
return 1000 * faults / total_faults;
}
+static bool pgdat_free_space_enough(struct pglist_data *pgdat)
+{
+ int z;
+ unsigned long enough_mark;
+
+ enough_mark = max(1UL * 1024 * 1024 * 1024 >> PAGE_SHIFT,
+ pgdat->node_present_pages >> 4);
+ for (z = pgdat->nr_zones - 1; z >= 0; z--) {
+ struct zone *zone = pgdat->node_zones + z;
+
+ if (!populated_zone(zone))
+ continue;
+
+ if (zone_watermark_ok(zone, 0,
+ high_wmark_pages(zone) + enough_mark,
+ ZONE_MOVABLE, 0))
+ return true;
+ }
+ return false;
+}
+
+static int numa_hint_fault_latency(struct page *page)
+{
+ int last_time, time;
+
+ time = jiffies_to_msecs(jiffies);
+ last_time = xchg_page_access_time(page, time);
+
+ return (time - last_time) & PAGE_ACCESS_TIME_MASK;
+}
+
bool should_numa_migrate_memory(struct task_struct *p, struct page * page,
int src_nid, int dst_cpu)
{
@@ -1407,9 +1441,38 @@ bool should_numa_migrate_memory(struct task_struct *p, struct page * page,
int dst_nid = cpu_to_node(dst_cpu);
int last_cpupid, this_cpupid;
+ /*
+ * The pages in slow memory node should be migrated according
+ * to hot/cold instead of accessing CPU node.
+ */
+ if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING &&
+ !node_is_toptier(src_nid)) {
+ struct pglist_data *pgdat;
+ unsigned long latency, th;
+
+ pgdat = NODE_DATA(dst_nid);
+ if (pgdat_free_space_enough(pgdat))
+ return true;
+
+ th = sysctl_numa_balancing_hot_threshold;
+ latency = numa_hint_fault_latency(page);
+ if (latency >= th)
+ return false;
+
+ return true;
+ }
+
this_cpupid = cpu_pid_to_cpupid(dst_cpu, current->pid);
last_cpupid = page_cpupid_xchg_last(page, this_cpupid);
+ /*
+ * The cpupid may be invalid when NUMA_BALANCING_MEMORY_TIERING
+ * is disabled dynamically.
+ */
+ if (!(sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) &&
+ !node_is_toptier(src_nid) && !check_cpupid(last_cpupid))
+ return false;
+
/*
* Allow first faults or private faults to migrate immediately early in
* the lifetime of a task. The magic number 4 is based on waiting for
@@ -2642,6 +2705,17 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags)
if (!p->mm)
return;
+ /*
+ * NUMA faults statistics are unnecessary for the slow memory node.
+ *
+ * And, the cpupid may be invalid when NUMA_BALANCING_MEMORY_TIERING
+ * is disabled dynamically.
+ */
+ if (!node_is_toptier(mem_node) &&
+ (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING ||
+ !check_cpupid(last_cpupid)))
+ return;
+
/* Allocate buffer to track faults on a per-node basis */
if (unlikely(!p->numa_faults)) {
int size = sizeof(*p->numa_faults) *
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 58263f90c559..86ce60d3d472 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -2408,6 +2408,7 @@ extern unsigned int sysctl_numa_balancing_scan_delay;
extern unsigned int sysctl_numa_balancing_scan_period_min;
extern unsigned int sysctl_numa_balancing_scan_period_max;
extern unsigned int sysctl_numa_balancing_scan_size;
+extern unsigned int sysctl_numa_balancing_hot_threshold;
#endif
#ifdef CONFIG_SCHED_HRTICK
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 2fe38212e07c..3a715eeeebb5 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1401,7 +1401,7 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf)
struct page *page;
unsigned long haddr = vmf->address & HPAGE_PMD_MASK;
int page_nid = NUMA_NO_NODE;
- int target_nid, last_cpupid = -1;
+ int target_nid, last_cpupid = (-1 & LAST_CPUPID_MASK);
bool migrated = false;
bool was_writable = pmd_savedwrite(oldpmd);
int flags = 0;
@@ -1422,7 +1422,8 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf)
flags |= TNF_NO_GROUP;
page_nid = page_to_nid(page);
- last_cpupid = page_cpupid_last(page);
+ if (node_is_toptier(page_nid))
+ last_cpupid = page_cpupid_last(page);
target_nid = numa_migrate_prep(page, vma, haddr, page_nid,
&flags);
@@ -1740,6 +1741,7 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
if (prot_numa) {
struct page *page;
+ bool toptier;
/*
* Avoid trapping faults against the zero page. The read-only
* data is likely to be read-cached on the local CPU and
@@ -1752,13 +1754,18 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
goto unlock;
page = pmd_page(*pmd);
+ toptier = node_is_toptier(page_to_nid(page));
/*
* Skip scanning top tier node if normal numa
* balancing is disabled
*/
if (!(sysctl_numa_balancing_mode & NUMA_BALANCING_NORMAL) &&
- node_is_toptier(page_to_nid(page)))
+ toptier)
goto unlock;
+
+ if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING &&
+ !toptier)
+ xchg_page_access_time(page, jiffies_to_msecs(jiffies));
}
/*
* In case prot_numa, we are under mmap_read_lock(mm). It's critical
diff --git a/mm/memory.c b/mm/memory.c
index b370ed118767..a8ac15ce7a75 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -74,6 +74,7 @@
#include <linux/perf_event.h>
#include <linux/ptrace.h>
#include <linux/vmalloc.h>
+#include <linux/sched/sysctl.h>
#include <trace/events/kmem.h>
@@ -4455,8 +4456,16 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf)
if (page_mapcount(page) > 1 && (vma->vm_flags & VM_SHARED))
flags |= TNF_SHARED;
- last_cpupid = page_cpupid_last(page);
page_nid = page_to_nid(page);
+ /*
+ * In memory tiering mode, cpupid of slow memory page is used
+ * to record page access time. So use default value.
+ */
+ if ((sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) &&
+ !node_is_toptier(page_nid))
+ last_cpupid = (-1 & LAST_CPUPID_MASK);
+ else
+ last_cpupid = page_cpupid_last(page);
target_nid = numa_migrate_prep(page, vma, vmf->address, page_nid,
&flags);
if (target_nid == NUMA_NO_NODE) {
diff --git a/mm/migrate.c b/mm/migrate.c
index dc84edfae842..e73f26dfeb38 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -534,6 +534,18 @@ void folio_migrate_flags(struct folio *newfolio, struct folio *folio)
* future migrations of this same page.
*/
cpupid = page_cpupid_xchg_last(&folio->page, -1);
+ /*
+ * If migrate between slow and fast memory node, reset cpupid,
+ * because that is used to record page access time in slow
+ * memory node
+ */
+ if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) {
+ bool f_toptier = node_is_toptier(page_to_nid(&folio->page));
+ bool t_toptier = node_is_toptier(page_to_nid(&newfolio->page));
+
+ if (f_toptier != t_toptier)
+ cpupid = -1;
+ }
page_cpupid_xchg_last(&newfolio->page, cpupid);
folio_migrate_ksm(newfolio, folio);
diff --git a/mm/mprotect.c b/mm/mprotect.c
index b69ce7a7b2b7..e7cb90d84fac 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -85,6 +85,7 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
if (prot_numa) {
struct page *page;
int nid;
+ bool toptier;
/* Avoid TLB flush if possible */
if (pte_protnone(oldpte))
@@ -114,14 +115,19 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
nid = page_to_nid(page);
if (target_node == nid)
continue;
+ toptier = node_is_toptier(nid);
/*
* Skip scanning top tier node if normal numa
* balancing is disabled
*/
if (!(sysctl_numa_balancing_mode & NUMA_BALANCING_NORMAL) &&
- node_is_toptier(nid))
+ toptier)
continue;
+ if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING &&
+ !toptier)
+ xchg_page_access_time(page,
+ jiffies_to_msecs(jiffies));
}
oldpte = ptep_modify_prot_start(vma, addr, pte);
--
2.30.2
^ permalink raw reply related [flat|nested] 4+ messages in thread
end of thread, other threads:[~2022-04-15 2:42 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-04-09 4:00 [PATCH 1/3] memory tiering: hot page selection with hint page fault latency kernel test robot
-- strict thread matches above, loose matches on Subject: below --
2022-04-08 7:12 [PATCH 0/3] memory tiering: hot page selection Huang Ying
2022-04-08 7:12 ` [PATCH 1/3] memory tiering: hot page selection with hint page fault latency Huang Ying
2022-04-14 13:23 ` Jagdish Gediya
2022-04-15 2:42 ` ying.huang
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.