The page allocation performance requirements of different workloads are often different. So, we need to tune the PCP (Per-CPU Pageset) high on each CPU automatically to optimize the page allocation performance.
The list of patches in series is as follows,
[1/9] mm, pcp: avoid to drain PCP when process exit [2/9] cacheinfo: calculate per-CPU data cache size [3/9] mm, pcp: reduce lock contention for draining high-order pages [4/9] mm: restrict the pcp batch scale factor to avoid too long latency [5/9] mm, page_alloc: scale the number of pages that are batch allocated [6/9] mm: add framework for PCP high auto-tuning [7/9] mm: tune PCP high automatically [8/9] mm, pcp: decrease PCP high if free pages < high watermark [9/9] mm, pcp: reduce detecting time of consecutive high order page freeing
Patch [1/9], [2/9], [3/9] optimize the PCP draining for consecutive high-order pages freeing.
Patch [4/9], [5/9] optimize batch freeing and allocating.
Patch [6/9], [7/9], [8/9] implement and optimize a PCP high auto-tuning method.
Patch [9/9] optimize the PCP draining for consecutive high order page freeing based on PCP high auto-tuning.
The test results for patches with performance impact are as follows,
kbuild ======
On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service.
build time lock contend% free_high alloc_zone ---------- ---------- --------- ---------- base 100.0 14.0 100.0 100.0 patch1 99.5 12.8 19.5 95.6 patch3 99.4 12.6 7.1 95.6 patch5 98.6 11.0 8.1 97.1 patch7 95.1 0.5 2.8 15.6 patch9 95.0 1.0 8.8 20.0
The PCP draining optimization (patch [1/9], [3/9]) and PCP batch allocation optimization (patch [5/9]) reduces zone lock contention a little. The PCP high auto-tuning (patch [7/9], [9/9]) reduces build time visibly. Where the tuning target: the number of pages allocated from zone reduces greatly. So, the zone contention cycles% reduces greatly.
With PCP tuning patches (patch [7/9], [9/9]), the average used memory during test increases up to 18.4% because more pages are cached in PCP. But at the end of the test, the number of the used memory decreases to the same level as that of the base patch. That is, the pages cached in PCP will be released to zone after not being used actively.
netperf SCTP_STREAM_MANY ========================
On a 2-socket Intel server with 128 logical CPU, we tested SCTP_STREAM_MANY test case of netperf test suite with 64-pair processes.
score lock contend% free_high alloc_zone cache miss rate% ----- ---------- --------- ---------- ---------------- base 100.0 2.1 100.0 100.0 1.3 patch1 99.4 2.1 99.4 99.4 1.3 patch3 106.4 1.3 13.3 106.3 1.3 patch5 106.0 1.2 13.2 105.9 1.3 patch7 103.4 1.9 6.7 90.3 7.6 patch9 108.6 1.3 13.7 108.6 1.3
The PCP draining optimization (patch [1/9]+[3/9]) improves performance. The PCP high auto-tuning (patch [7/9]) reduces performance a little because PCP draining cannot be triggered in time sometimes. So, the cache miss rate% increases. The further PCP draining optimization (patch [9/9]) based on PCP tuning restore the performance.
lmbench3 UNIX (AF_UNIX) =======================
On a 2-socket Intel server with 128 logical CPU, we tested UNIX (AF_UNIX socket) test case of lmbench3 test suite with 16-pair processes.
score lock contend% free_high alloc_zone cache miss rate% ----- ---------- --------- ---------- ---------------- base 100.0 51.4 100.0 100.0 0.2 patch1 116.8 46.1 69.5 104.3 0.2 patch3 199.1 21.3 7.0 104.9 0.2 patch5 200.0 20.8 7.1 106.9 0.3 patch7 191.6 19.9 6.8 103.8 2.8 patch9 193.4 21.7 7.0 104.7 2.1
The PCP draining optimization (patch [1/9], [3/9]) improves performance much. The PCP tuning (patch [7/9]) reduces performance a little because PCP draining cannot be triggered in time sometimes. The further PCP draining optimization (patch [9/9]) based on PCP tuning restores the performance partly.
The patchset adds several fields in struct per_cpu_pages. The struct layout before/after the patchset is as follows,
base ====
struct per_cpu_pages { spinlock_t lock; /* 0 4 */ int count; /* 4 4 */ int high; /* 8 4 */ int batch; /* 12 4 */ short int free_factor; /* 16 2 */ short int expire; /* 18 2 */
/* XXX 4 bytes hole, try to pack */
struct list_head lists[13]; /* 24 208 */
/* size: 256, cachelines: 4, members: 7 */ /* sum members: 228, holes: 1, sum holes: 4 */ /* padding: 24 */ } __attribute__((__aligned__(64)));
patched =======
struct per_cpu_pages { spinlock_t lock; /* 0 4 */ int count; /* 4 4 */ int high; /* 8 4 */ int high_min; /* 12 4 */ int high_max; /* 16 4 */ int batch; /* 20 4 */ u8 flags; /* 24 1 */ u8 alloc_factor; /* 25 1 */ u8 expire; /* 26 1 */
/* XXX 1 byte hole, try to pack */
short int free_count; /* 28 2 */
/* XXX 2 bytes hole, try to pack */
struct list_head lists[13]; /* 32 208 */
/* size: 256, cachelines: 4, members: 11 */ /* sum members: 237, holes: 2, sum holes: 3 */ /* padding: 16 */ } __attribute__((__aligned__(64)));
The size of the struct doesn't changed with the patchset.
Huang Ying (9): mm, pcp: avoid to drain PCP when process exit cacheinfo: calculate size of per-CPU data cache slice mm, pcp: reduce lock contention for draining high-order pages mm: restrict the pcp batch scale factor to avoid too long latency mm, page_alloc: scale the number of pages that are batch allocated mm: add framework for PCP high auto-tuning mm: tune PCP high automatically mm, pcp: decrease PCP high if free pages < high watermark mm, pcp: reduce detecting time of consecutive high order page freeing
drivers/base/cacheinfo.c | 51 ++++++- include/linux/cacheinfo.h | 1 + include/linux/gfp.h | 2 + include/linux/mmzone.h | 27 +++- mm/Kconfig | 11 ++ mm/page_alloc.c | 310 +++++++++++++++++++++++++++++++------- mm/vmstat.c | 8 +- 7 files changed, 345 insertions(+), 65 deletions(-)
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit ca71fe1ad9221a89c6a25f49159c600d9e598ae1 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
Patch series "mm: PCP high auto-tuning", v3.
The page allocation performance requirements of different workloads are often different. So, we need to tune the PCP (Per-CPU Pageset) high on each CPU automatically to optimize the page allocation performance.
The list of patches in series is as follows,
[1/9] mm, pcp: avoid to drain PCP when process exit [2/9] cacheinfo: calculate per-CPU data cache size [3/9] mm, pcp: reduce lock contention for draining high-order pages [4/9] mm: restrict the pcp batch scale factor to avoid too long latency [5/9] mm, page_alloc: scale the number of pages that are batch allocated [6/9] mm: add framework for PCP high auto-tuning [7/9] mm: tune PCP high automatically [8/9] mm, pcp: decrease PCP high if free pages < high watermark [9/9] mm, pcp: reduce detecting time of consecutive high order page freeing
Patch [1/9], [2/9], [3/9] optimize the PCP draining for consecutive high-order pages freeing.
Patch [4/9], [5/9] optimize batch freeing and allocating.
Patch [6/9], [7/9], [8/9] implement and optimize a PCP high auto-tuning method.
Patch [9/9] optimize the PCP draining for consecutive high order page freeing based on PCP high auto-tuning.
The test results for patches with performance impact are as follows,
kbuild ======
On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service.
build time lock contend% free_high alloc_zone ---------- ---------- --------- ---------- base 100.0 14.0 100.0 100.0 patch1 99.5 12.8 19.5 95.6 patch3 99.4 12.6 7.1 95.6 patch5 98.6 11.0 8.1 97.1 patch7 95.1 0.5 2.8 15.6 patch9 95.0 1.0 8.8 20.0
The PCP draining optimization (patch [1/9], [3/9]) and PCP batch allocation optimization (patch [5/9]) reduces zone lock contention a little. The PCP high auto-tuning (patch [7/9], [9/9]) reduces build time visibly. Where the tuning target: the number of pages allocated from zone reduces greatly. So, the zone contention cycles% reduces greatly.
With PCP tuning patches (patch [7/9], [9/9]), the average used memory during test increases up to 18.4% because more pages are cached in PCP. But at the end of the test, the number of the used memory decreases to the same level as that of the base patch. That is, the pages cached in PCP will be released to zone after not being used actively.
netperf SCTP_STREAM_MANY ========================
On a 2-socket Intel server with 128 logical CPU, we tested SCTP_STREAM_MANY test case of netperf test suite with 64-pair processes.
score lock contend% free_high alloc_zone cache miss rate% ----- ---------- --------- ---------- ---------------- base 100.0 2.1 100.0 100.0 1.3 patch1 99.4 2.1 99.4 99.4 1.3 patch3 106.4 1.3 13.3 106.3 1.3 patch5 106.0 1.2 13.2 105.9 1.3 patch7 103.4 1.9 6.7 90.3 7.6 patch9 108.6 1.3 13.7 108.6 1.3
The PCP draining optimization (patch [1/9]+[3/9]) improves performance. The PCP high auto-tuning (patch [7/9]) reduces performance a little because PCP draining cannot be triggered in time sometimes. So, the cache miss rate% increases. The further PCP draining optimization (patch [9/9]) based on PCP tuning restore the performance.
lmbench3 UNIX (AF_UNIX) =======================
On a 2-socket Intel server with 128 logical CPU, we tested UNIX (AF_UNIX socket) test case of lmbench3 test suite with 16-pair processes.
score lock contend% free_high alloc_zone cache miss rate% ----- ---------- --------- ---------- ---------------- base 100.0 51.4 100.0 100.0 0.2 patch1 116.8 46.1 69.5 104.3 0.2 patch3 199.1 21.3 7.0 104.9 0.2 patch5 200.0 20.8 7.1 106.9 0.3 patch7 191.6 19.9 6.8 103.8 2.8 patch9 193.4 21.7 7.0 104.7 2.1
The PCP draining optimization (patch [1/9], [3/9]) improves performance much. The PCP tuning (patch [7/9]) reduces performance a little because PCP draining cannot be triggered in time sometimes. The further PCP draining optimization (patch [9/9]) based on PCP tuning restores the performance partly.
The patchset adds several fields in struct per_cpu_pages. The struct layout before/after the patchset is as follows,
base ====
struct per_cpu_pages { spinlock_t lock; /* 0 4 */ int count; /* 4 4 */ int high; /* 8 4 */ int batch; /* 12 4 */ short int free_factor; /* 16 2 */ short int expire; /* 18 2 */
/* XXX 4 bytes hole, try to pack */
struct list_head lists[13]; /* 24 208 */
/* size: 256, cachelines: 4, members: 7 */ /* sum members: 228, holes: 1, sum holes: 4 */ /* padding: 24 */ } __attribute__((__aligned__(64)));
patched =======
struct per_cpu_pages { spinlock_t lock; /* 0 4 */ int count; /* 4 4 */ int high; /* 8 4 */ int high_min; /* 12 4 */ int high_max; /* 16 4 */ int batch; /* 20 4 */ u8 flags; /* 24 1 */ u8 alloc_factor; /* 25 1 */ u8 expire; /* 26 1 */
/* XXX 1 byte hole, try to pack */
short int free_count; /* 28 2 */
/* XXX 2 bytes hole, try to pack */
struct list_head lists[13]; /* 32 208 */
/* size: 256, cachelines: 4, members: 11 */ /* sum members: 237, holes: 2, sum holes: 3 */ /* padding: 16 */ } __attribute__((__aligned__(64)));
The size of the struct doesn't changed with the patchset.
This patch (of 9):
In commit f26b3fa04611 ("mm/page_alloc: limit number of high-order pages on PCP during bulk free"), the PCP (Per-CPU Pageset) will be drained when PCP is mostly used for high-order pages freeing to improve the cache-hot pages reusing between page allocation and freeing CPUs.
But, the PCP draining mechanism may be triggered unexpectedly when process exits. With some customized trace point, it was found that PCP draining (free_high == true) was triggered with the order-1 page freeing with the following call stack,
=> free_unref_page_commit => free_unref_page => __mmdrop => exit_mm => do_exit => do_group_exit => __x64_sys_exit_group => do_syscall_64
Checking the source code, this is the page table PGD freeing (mm_free_pgd()). It's a order-1 page freeing if CONFIG_PAGE_TABLE_ISOLATION=y. Which is a common configuration for security.
Just before that, page freeing with the following call stack was found,
=> free_unref_page_commit => free_unref_page_list => release_pages => tlb_batch_pages_flush => tlb_finish_mmu => exit_mmap => __mmput => exit_mm => do_exit => do_group_exit => __x64_sys_exit_group => do_syscall_64
So, when a process exits,
- a large number of user pages of the process will be freed without page allocation, it's highly possible that pcp->free_factor becomes > 0. In fact, this is expected behavior to improve process exit performance.
- after freeing all user pages, the PGD will be freed, which is a order-1 page freeing, PCP will be drained.
All in all, when a process exits, it's high possible that the PCP will be drained. This is an unexpected behavior.
To avoid this, in the patch, the PCP draining will only be triggered for 2 consecutive high-order page freeing.
On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service. With the patch, the cycles% of the spinlock contention (mostly for zone lock) decreases from 14.0% to 12.8% (with PCP size == 367). The number of PCP draining for high order pages freeing (free_high) decreases 80.5%.
This helps network workload too for reduced zone lock contention. On a 2-socket Intel server with 128 logical CPU, with the patch, the network bandwidth of the UNIX (AF_UNIX) test case of lmbench test suite with 16-pair processes increase 16.8%. The cycles% of the spinlock contention (mostly for zone lock) decreases from 51.4% to 46.1%. The number of PCP draining for high order pages freeing (free_high) decreases 30.5%. The cache miss rate keeps 0.2%.
Link: https://lkml.kernel.org/r/20231016053002.756205-1-ying.huang@intel.com Link: https://lkml.kernel.org/r/20231016053002.756205-2-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Acked-by: Mel Gorman mgorman@techsingularity.net Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Michal Hocko mhocko@suse.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Cc: Sudeep Holla sudeep.holla@arm.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- include/linux/mmzone.h | 12 +++++++++++- mm/page_alloc.c | 11 ++++++++--- 2 files changed, 19 insertions(+), 4 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 4106fbc5b4b3..19c40a6f7e45 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -676,12 +676,22 @@ enum zone_watermarks { #define high_wmark_pages(z) (z->_watermark[WMARK_HIGH] + z->watermark_boost) #define wmark_pages(z, i) (z->_watermark[i] + z->watermark_boost)
+/* + * Flags used in pcp->flags field. + * + * PCPF_PREV_FREE_HIGH_ORDER: a high-order page is freed in the + * previous page freeing. To avoid to drain PCP for an accident + * high-order page freeing. + */ +#define PCPF_PREV_FREE_HIGH_ORDER BIT(0) + struct per_cpu_pages { spinlock_t lock; /* Protects lists field */ int count; /* number of pages in the list */ int high; /* high watermark, emptying needed */ int batch; /* chunk size for buddy add/remove */ - short free_factor; /* batch scaling factor during free */ + u8 flags; /* protected by pcp->lock */ + u8 free_factor; /* batch scaling factor during free */ #ifdef CONFIG_NUMA short expire; /* When 0, remote pagesets are drained */ #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 85741403948f..89fb899f2d67 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2370,7 +2370,7 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, { int high; int pindex; - bool free_high; + bool free_high = false;
__count_vm_events(PGFREE, 1 << order); pindex = order_to_pindex(migratetype, order); @@ -2383,8 +2383,13 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, * freeing without allocation. The remainder after bulk freeing * stops will be drained from vmstat refresh context. */ - free_high = (pcp->free_factor && order && order <= PAGE_ALLOC_COSTLY_ORDER); - + if (order && order <= PAGE_ALLOC_COSTLY_ORDER) { + free_high = (pcp->free_factor && + (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER)); + pcp->flags |= PCPF_PREV_FREE_HIGH_ORDER; + } else if (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER) { + pcp->flags &= ~PCPF_PREV_FREE_HIGH_ORDER; + } high = nr_pcp_high(pcp, zone, free_high); if (pcp->count >= high) { free_pcppages_bulk(zone, nr_pcp_free(pcp, high, free_high), pcp, pindex);
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit 94a3bfe4073cd88b05f7fb201ea7bf9dfa2cf5d5 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
This can be used to estimate the size of the data cache slice that can be used by one CPU under ideal circumstances. Both DATA caches and UNIFIED caches are used in calculation. So, the users need to consider the impact of the code cache usage.
Because the cache inclusive/non-inclusive information isn't available now, we just use the size of the per-CPU slice of LLC to make the result more predictable across architectures. This may be improved when more cache information is available in the future.
A brute-force algorithm to iterate all online CPUs is used to avoid to allocate an extra cpumask, especially in offline callback.
Link: https://lkml.kernel.org/r/20231016053002.756205-3-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Acked-by: Mel Gorman mgorman@techsingularity.net Cc: Sudeep Holla sudeep.holla@arm.com Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Michal Hocko mhocko@suse.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- drivers/base/cacheinfo.c | 49 ++++++++++++++++++++++++++++++++++++++- include/linux/cacheinfo.h | 1 + 2 files changed, 49 insertions(+), 1 deletion(-)
diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c index cbae8be1fe52..585c66fce9d9 100644 --- a/drivers/base/cacheinfo.c +++ b/drivers/base/cacheinfo.c @@ -898,6 +898,48 @@ static int cache_add_dev(unsigned int cpu) return rc; }
+/* + * Calculate the size of the per-CPU data cache slice. This can be + * used to estimate the size of the data cache slice that can be used + * by one CPU under ideal circumstances. UNIFIED caches are counted + * in addition to DATA caches. So, please consider code cache usage + * when use the result. + * + * Because the cache inclusive/non-inclusive information isn't + * available, we just use the size of the per-CPU slice of LLC to make + * the result more predictable across architectures. + */ +static void update_per_cpu_data_slice_size_cpu(unsigned int cpu) +{ + struct cpu_cacheinfo *ci; + struct cacheinfo *llc; + unsigned int nr_shared; + + if (!last_level_cache_is_valid(cpu)) + return; + + ci = ci_cacheinfo(cpu); + llc = per_cpu_cacheinfo_idx(cpu, cache_leaves(cpu) - 1); + + if (llc->type != CACHE_TYPE_DATA && llc->type != CACHE_TYPE_UNIFIED) + return; + + nr_shared = cpumask_weight(&llc->shared_cpu_map); + if (nr_shared) + ci->per_cpu_data_slice_size = llc->size / nr_shared; +} + +static void update_per_cpu_data_slice_size(bool cpu_online, unsigned int cpu) +{ + unsigned int icpu; + + for_each_online_cpu(icpu) { + if (!cpu_online && icpu == cpu) + continue; + update_per_cpu_data_slice_size_cpu(icpu); + } +} + static int cacheinfo_cpu_online(unsigned int cpu) { int rc = detect_cache_attributes(cpu); @@ -906,7 +948,11 @@ static int cacheinfo_cpu_online(unsigned int cpu) return rc; rc = cache_add_dev(cpu); if (rc) - free_cache_attributes(cpu); + goto err; + update_per_cpu_data_slice_size(true, cpu); + return 0; +err: + free_cache_attributes(cpu); return rc; }
@@ -916,6 +962,7 @@ static int cacheinfo_cpu_pre_down(unsigned int cpu) cpu_cache_sysfs_exit(cpu);
free_cache_attributes(cpu); + update_per_cpu_data_slice_size(false, cpu); return 0; }
diff --git a/include/linux/cacheinfo.h b/include/linux/cacheinfo.h index a5cfd44fab45..d504eb4b49ab 100644 --- a/include/linux/cacheinfo.h +++ b/include/linux/cacheinfo.h @@ -73,6 +73,7 @@ struct cacheinfo {
struct cpu_cacheinfo { struct cacheinfo *info_list; + unsigned int per_cpu_data_slice_size; unsigned int num_levels; unsigned int num_leaves; bool cpu_map_populated;
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit 362d37a106dd3f6431b2fdd91d9208b0d023b50d category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
In commit f26b3fa04611 ("mm/page_alloc: limit number of high-order pages on PCP during bulk free"), the PCP (Per-CPU Pageset) will be drained when PCP is mostly used for high-order pages freeing to improve the cache-hot pages reusing between page allocating and freeing CPUs.
On system with small per-CPU data cache slice, pages shouldn't be cached before draining to guarantee cache-hot. But on a system with large per-CPU data cache slice, some pages can be cached before draining to reduce zone lock contention.
So, in this patch, instead of draining without any caching, "pcp->batch" pages will be cached in PCP before draining if the size of the per-CPU data cache slice is more than "3 * batch".
In theory, if the size of per-CPU data cache slice is more than "2 * batch", we can reuse cache-hot pages between CPUs. But considering the other usage of cache (code, other data accessing, etc.), "3 * batch" is used.
Note: "3 * batch" is chosen to make sure the optimization works on recent x86_64 server CPUs. If you want to increase it, please check whether it breaks the optimization.
On a 2-socket Intel server with 128 logical CPU, with the patch, the network bandwidth of the UNIX (AF_UNIX) test case of lmbench test suite with 16-pair processes increase 70.5%. The cycles% of the spinlock contention (mostly for zone lock) decreases from 46.1% to 21.3%. The number of PCP draining for high order pages freeing (free_high) decreases 89.9%. The cache miss rate keeps 0.2%.
Link: https://lkml.kernel.org/r/20231016053002.756205-4-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Acked-by: Mel Gorman mgorman@techsingularity.net Cc: Sudeep Holla sudeep.holla@arm.com Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Michal Hocko mhocko@suse.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- drivers/base/cacheinfo.c | 2 ++ include/linux/gfp.h | 1 + include/linux/mmzone.h | 6 ++++++ mm/page_alloc.c | 38 +++++++++++++++++++++++++++++++++++++- 4 files changed, 46 insertions(+), 1 deletion(-)
diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c index 585c66fce9d9..f1e79263fe61 100644 --- a/drivers/base/cacheinfo.c +++ b/drivers/base/cacheinfo.c @@ -950,6 +950,7 @@ static int cacheinfo_cpu_online(unsigned int cpu) if (rc) goto err; update_per_cpu_data_slice_size(true, cpu); + setup_pcp_cacheinfo(); return 0; err: free_cache_attributes(cpu); @@ -963,6 +964,7 @@ static int cacheinfo_cpu_pre_down(unsigned int cpu)
free_cache_attributes(cpu); update_per_cpu_data_slice_size(false, cpu); + setup_pcp_cacheinfo(); return 0; }
diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 665f06675c83..665edc11fb9f 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -325,6 +325,7 @@ void drain_all_pages(struct zone *zone); void drain_local_pages(struct zone *zone);
void page_alloc_init_late(void); +void setup_pcp_cacheinfo(void);
/* * gfp_allowed_mask is set to GFP_BOOT_MASK during early boot to restrict what diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 19c40a6f7e45..cdff247e8c6f 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -682,8 +682,14 @@ enum zone_watermarks { * PCPF_PREV_FREE_HIGH_ORDER: a high-order page is freed in the * previous page freeing. To avoid to drain PCP for an accident * high-order page freeing. + * + * PCPF_FREE_HIGH_BATCH: preserve "pcp->batch" pages in PCP before + * draining PCP for consecutive high-order pages freeing without + * allocation if data cache slice of CPU is large enough. To reduce + * zone lock contention and keep cache-hot pages reusing. */ #define PCPF_PREV_FREE_HIGH_ORDER BIT(0) +#define PCPF_FREE_HIGH_BATCH BIT(1)
struct per_cpu_pages { spinlock_t lock; /* Protects lists field */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 89fb899f2d67..c573e7736798 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -52,6 +52,7 @@ #include <linux/psi.h> #include <linux/khugepaged.h> #include <linux/delayacct.h> +#include <linux/cacheinfo.h> #include <asm/div64.h> #include "internal.h" #include "shuffle.h" @@ -2385,7 +2386,9 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, */ if (order && order <= PAGE_ALLOC_COSTLY_ORDER) { free_high = (pcp->free_factor && - (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER)); + (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER) && + (!(pcp->flags & PCPF_FREE_HIGH_BATCH) || + pcp->count >= READ_ONCE(pcp->batch))); pcp->flags |= PCPF_PREV_FREE_HIGH_ORDER; } else if (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER) { pcp->flags &= ~PCPF_PREV_FREE_HIGH_ORDER; @@ -5418,6 +5421,39 @@ static void zone_pcp_update(struct zone *zone, int cpu_online) mutex_unlock(&pcp_batch_high_lock); }
+static void zone_pcp_update_cacheinfo(struct zone *zone) +{ + int cpu; + struct per_cpu_pages *pcp; + struct cpu_cacheinfo *cci; + + for_each_online_cpu(cpu) { + pcp = per_cpu_ptr(zone->per_cpu_pageset, cpu); + cci = get_cpu_cacheinfo(cpu); + /* + * If data cache slice of CPU is large enough, "pcp->batch" + * pages can be preserved in PCP before draining PCP for + * consecutive high-order pages freeing without allocation. + * This can reduce zone lock contention without hurting + * cache-hot pages sharing. + */ + spin_lock(&pcp->lock); + if ((cci->per_cpu_data_slice_size >> PAGE_SHIFT) > 3 * pcp->batch) + pcp->flags |= PCPF_FREE_HIGH_BATCH; + else + pcp->flags &= ~PCPF_FREE_HIGH_BATCH; + spin_unlock(&pcp->lock); + } +} + +void setup_pcp_cacheinfo(void) +{ + struct zone *zone; + + for_each_populated_zone(zone) + zone_pcp_update_cacheinfo(zone); +} + /* * Allocate per cpu pagesets and initialize them. * Before this call only boot pagesets were available.
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit 52166607ecc980391b1fffbce0be3074a96d0c7b category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
In page allocator, PCP (Per-CPU Pageset) is refilled and drained in batches to increase page allocation throughput, reduce page allocation/freeing latency per page, and reduce zone lock contention. But too large batch size will cause too long maximal allocation/freeing latency, which may punish arbitrary users. So the default batch size is chosen carefully (in zone_batchsize(), the value is 63 for zone > 1GB) to avoid that.
In commit 3b12e7e97938 ("mm/page_alloc: scale the number of pages that are batch freed"), the batch size will be scaled for large number of page freeing to improve page freeing performance and reduce zone lock contention. Similar optimization can be used for large number of pages allocation too.
To find out a suitable max batch scale factor (that is, max effective batch size), some tests and measurement on some machines were done as follows.
A set of debug patches are implemented as follows,
- Set PCP high to be 2 * batch to reduce the effect of PCP high
- Disable free batch size scaling to get the raw performance.
- The code with zone lock held is extracted from rmqueue_bulk() and free_pcppages_bulk() to 2 separate functions to make it easy to measure the function run time with ftrace function_graph tracer.
- The batch size is hard coded to be 63 (default), 127, 255, 511, 1023, 2047, 4095.
Then will-it-scale/page_fault1 is used to generate the page allocation/freeing workload. The page allocation/freeing throughput (page/s) is measured via will-it-scale. The page allocation/freeing average latency (alloc/free latency avg, in us) and allocation/freeing latency at 99 percentile (alloc/free latency 99%, in us) are measured with ftrace function_graph tracer.
The test results are as follows,
Sapphire Rapids Server ====================== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 513633.4 2.33 3.57 2.67 6.83 127 517616.7 4.35 6.65 4.22 13.03 255 520822.8 8.29 13.32 7.52 25.24 511 524122.0 15.79 23.42 14.02 49.35 1023 525980.5 30.25 44.19 25.36 94.88 2047 526793.6 59.39 84.50 45.22 140.81
Ice Lake Server =============== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 620210.3 2.21 3.68 2.02 4.35 127 627003.0 4.09 6.86 3.51 8.28 255 630777.5 7.70 13.50 6.17 15.97 511 633651.5 14.85 22.62 11.66 31.08 1023 637071.1 28.55 42.02 20.81 54.36 2047 638089.7 56.54 84.06 39.28 91.68
Cascade Lake Server =================== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 404706.7 3.29 5.03 3.53 4.75 127 422475.2 6.12 9.09 6.36 8.76 255 411522.2 11.68 16.97 10.90 16.39 511 428124.1 22.54 31.28 19.86 32.25 1023 414718.4 43.39 62.52 40.00 66.33 2047 429848.7 86.64 120.34 71.14 106.08
Commet Lake Desktop =================== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- -------------
63 795183.13 2.18 3.55 2.03 3.05 127 803067.85 3.91 6.56 3.85 5.52 255 812771.10 7.35 10.80 7.14 10.20 511 817723.48 14.17 27.54 13.43 30.31 1023 818870.19 27.72 40.10 27.89 46.28
Coffee Lake Desktop =================== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 510542.8 3.13 4.40 2.48 3.43 127 514288.6 5.97 7.89 4.65 6.04 255 516889.7 11.86 15.58 8.96 12.55 511 519802.4 23.10 28.81 16.95 26.19 1023 520802.7 45.30 52.51 33.19 45.95 2047 519997.1 90.63 104.00 65.26 81.74
From the above data, to restrict the allocation/freeing latency to be less than 100 us in most times, the max batch scale factor needs to be less than or equal to 5.
Although it is reasonable to use 5 as max batch scale factor for the systems tested, there are also slower systems. Where smaller value should be used to constrain the page allocation/freeing latency.
So, in this patch, a new kconfig option (PCP_BATCH_SCALE_MAX) is added to set the max batch scale factor. Whose default value is 5, and users can reduce it when necessary.
Link: https://lkml.kernel.org/r/20231016053002.756205-5-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Acked-by: Andrew Morton akpm@linux-foundation.org Acked-by: Mel Gorman mgorman@techsingularity.net Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Michal Hocko mhocko@suse.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Cc: Sudeep Holla sudeep.holla@arm.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- mm/Kconfig | 11 +++++++++++ mm/page_alloc.c | 2 +- 2 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/mm/Kconfig b/mm/Kconfig index 264a2df5ecf5..ece4f2847e2b 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -704,6 +704,17 @@ config HUGETLB_PAGE_SIZE_VARIABLE config CONTIG_ALLOC def_bool (MEMORY_ISOLATION && COMPACTION) || CMA
+config PCP_BATCH_SCALE_MAX + int "Maximum scale factor of PCP (Per-CPU pageset) batch allocate/free" + default 5 + range 0 6 + help + In page allocator, PCP (Per-CPU pageset) is refilled and drained in + batches. The batch number is scaled automatically to improve page + allocation/free throughput. But too large scale factor may hurt + latency. This option sets the upper limit of scale factor to limit + the maximum latency. + config PHYS_ADDR_T_64BIT def_bool 64BIT
diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c573e7736798..49486ed22cf8 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2340,7 +2340,7 @@ static int nr_pcp_free(struct per_cpu_pages *pcp, int high, bool free_high) * freeing of pages without any allocation. */ batch <<= pcp->free_factor; - if (batch < max_nr_free) + if (batch < max_nr_free && pcp->free_factor < CONFIG_PCP_BATCH_SCALE_MAX) pcp->free_factor++; batch = clamp(batch, min_nr_free, max_nr_free);
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit c0a242394cb980bd00e1f61dc8aacb453d2bbe6a category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
When a task is allocating a large number of order-0 pages, it may acquire the zone->lock multiple times allocating pages in batches. This may unnecessarily contend on the zone lock when allocating very large number of pages. This patch adapts the size of the batch based on the recent pattern to scale the batch size for subsequent allocations.
On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service. With the patch, the cycles% of the spinlock contention (mostly for zone lock) decreases from 12.6% to 11.0% (with PCP size == 367).
Link: https://lkml.kernel.org/r/20231016053002.756205-6-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Suggested-by: Mel Gorman mgorman@techsingularity.net Acked-by: Mel Gorman mgorman@techsingularity.net Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Michal Hocko mhocko@suse.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Cc: Sudeep Holla sudeep.holla@arm.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- include/linux/mmzone.h | 3 ++- mm/page_alloc.c | 53 ++++++++++++++++++++++++++++++++++-------- 2 files changed, 45 insertions(+), 11 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index cdff247e8c6f..ba548ae20686 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -697,9 +697,10 @@ struct per_cpu_pages { int high; /* high watermark, emptying needed */ int batch; /* chunk size for buddy add/remove */ u8 flags; /* protected by pcp->lock */ + u8 alloc_factor; /* batch scaling factor during allocate */ u8 free_factor; /* batch scaling factor during free */ #ifdef CONFIG_NUMA - short expire; /* When 0, remote pagesets are drained */ + u8 expire; /* When 0, remote pagesets are drained */ #endif
/* Lists of pages, one per migrate type stored on the pcp-lists */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 49486ed22cf8..769874713837 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2373,6 +2373,12 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, int pindex; bool free_high = false;
+ /* + * On freeing, reduce the number of pages that are batch allocated. + * See nr_pcp_alloc() where alloc_factor is increased for subsequent + * allocations. + */ + pcp->alloc_factor >>= 1; __count_vm_events(PGFREE, 1 << order); pindex = order_to_pindex(migratetype, order); list_add(&page->pcp_list, &pcp->lists[pindex]); @@ -2679,6 +2685,42 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone, return page; }
+static int nr_pcp_alloc(struct per_cpu_pages *pcp, int order) +{ + int high, batch, max_nr_alloc; + + high = READ_ONCE(pcp->high); + batch = READ_ONCE(pcp->batch); + + /* Check for PCP disabled or boot pageset */ + if (unlikely(high < batch)) + return 1; + + /* + * Double the number of pages allocated each time there is subsequent + * allocation of order-0 pages without any freeing. + */ + if (!order) { + max_nr_alloc = max(high - pcp->count - batch, batch); + batch <<= pcp->alloc_factor; + if (batch <= max_nr_alloc && + pcp->alloc_factor < CONFIG_PCP_BATCH_SCALE_MAX) + pcp->alloc_factor++; + batch = min(batch, max_nr_alloc); + } + + /* + * Scale batch relative to order if batch implies free pages + * can be stored on the PCP. Batch can be 1 for small zones or + * for boot pagesets which should never store free pages as + * the pages may belong to arbitrary zones. + */ + if (batch > 1) + batch = max(batch >> order, 2); + + return batch; +} + /* Remove page from the per-cpu list, caller must protect the list */ static inline struct page *__rmqueue_pcplist(struct zone *zone, unsigned int order, @@ -2691,18 +2733,9 @@ struct page *__rmqueue_pcplist(struct zone *zone, unsigned int order,
do { if (list_empty(list)) { - int batch = READ_ONCE(pcp->batch); + int batch = nr_pcp_alloc(pcp, order); int alloced;
- /* - * Scale batch relative to order if batch implies - * free pages can be stored on the PCP. Batch can - * be 1 for small zones or for boot pagesets which - * should never store free pages as the pages may - * belong to arbitrary zones. - */ - if (batch > 1) - batch = max(batch >> order, 2); alloced = rmqueue_bulk(zone, order, batch, list, migratetype, alloc_flags);
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit 90b41691b9881376fe784e13b5766ec3676fdb55 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
The page allocation performance requirements of different workloads are usually different. So, we need to tune PCP (per-CPU pageset) high to optimize the workload page allocation performance. Now, we have a system wide sysctl knob (percpu_pagelist_high_fraction) to tune PCP high by hand. But, it's hard to find out the best value by hand. And one global configuration may not work best for the different workloads that run on the same system. One solution to these issues is to tune PCP high of each CPU automatically.
This patch adds the framework for PCP high auto-tuning. With it, pcp->high of each CPU will be changed automatically by tuning algorithm at runtime. The minimal high (pcp->high_min) is the original PCP high value calculated based on the low watermark pages. While the maximal high (pcp->high_max) is the PCP high value when percpu_pagelist_high_fraction sysctl knob is set to MIN_PERCPU_PAGELIST_HIGH_FRACTION. That is, the maximal pcp->high that can be set via sysctl knob by hand.
It's possible that PCP high auto-tuning doesn't work well for some workloads. So, when PCP high is tuned by hand via the sysctl knob, the auto-tuning will be disabled. The PCP high set by hand will be used instead.
This patch only adds the framework, so pcp->high will be set to pcp->high_min (original default) always. We will add actual auto-tuning algorithm in the following patches in the series.
Link: https://lkml.kernel.org/r/20231016053002.756205-7-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Acked-by: Mel Gorman mgorman@techsingularity.net Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Michal Hocko mhocko@suse.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Cc: Sudeep Holla sudeep.holla@arm.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- include/linux/mmzone.h | 5 ++- mm/page_alloc.c | 71 +++++++++++++++++++++++++++--------------- 2 files changed, 50 insertions(+), 26 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index ba548ae20686..ec3f7daedcc7 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -695,6 +695,8 @@ struct per_cpu_pages { spinlock_t lock; /* Protects lists field */ int count; /* number of pages in the list */ int high; /* high watermark, emptying needed */ + int high_min; /* min high watermark */ + int high_max; /* max high watermark */ int batch; /* chunk size for buddy add/remove */ u8 flags; /* protected by pcp->lock */ u8 alloc_factor; /* batch scaling factor during allocate */ @@ -854,7 +856,8 @@ struct zone { * the high and batch values are copied to individual pagesets for * faster access */ - int pageset_high; + int pageset_high_min; + int pageset_high_max; int pageset_batch;
#ifndef CONFIG_SPARSEMEM diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 769874713837..27c1c9ad1928 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2350,7 +2350,7 @@ static int nr_pcp_free(struct per_cpu_pages *pcp, int high, bool free_high) static int nr_pcp_high(struct per_cpu_pages *pcp, struct zone *zone, bool free_high) { - int high = READ_ONCE(pcp->high); + int high = READ_ONCE(pcp->high_min);
if (unlikely(!high || free_high)) return 0; @@ -2689,7 +2689,7 @@ static int nr_pcp_alloc(struct per_cpu_pages *pcp, int order) { int high, batch, max_nr_alloc;
- high = READ_ONCE(pcp->high); + high = READ_ONCE(pcp->high_min); batch = READ_ONCE(pcp->batch);
/* Check for PCP disabled or boot pageset */ @@ -5296,14 +5296,15 @@ static int zone_batchsize(struct zone *zone) }
static int percpu_pagelist_high_fraction; -static int zone_highsize(struct zone *zone, int batch, int cpu_online) +static int zone_highsize(struct zone *zone, int batch, int cpu_online, + int high_fraction) { #ifdef CONFIG_MMU int high; int nr_split_cpus; unsigned long total_pages;
- if (!percpu_pagelist_high_fraction) { + if (!high_fraction) { /* * By default, the high value of the pcp is based on the zone * low watermark so that if they are full then background @@ -5316,15 +5317,15 @@ static int zone_highsize(struct zone *zone, int batch, int cpu_online) * value is based on a fraction of the managed pages in the * zone. */ - total_pages = zone_managed_pages(zone) / percpu_pagelist_high_fraction; + total_pages = zone_managed_pages(zone) / high_fraction; }
/* * Split the high value across all online CPUs local to the zone. Note * that early in boot that CPUs may not be online yet and that during * CPU hotplug that the cpumask is not yet updated when a CPU is being - * onlined. For memory nodes that have no CPUs, split pcp->high across - * all online CPUs to mitigate the risk that reclaim is triggered + * onlined. For memory nodes that have no CPUs, split the high value + * across all online CPUs to mitigate the risk that reclaim is triggered * prematurely due to pages stored on pcp lists. */ nr_split_cpus = cpumask_weight(cpumask_of_node(zone_to_nid(zone))) + cpu_online; @@ -5352,19 +5353,21 @@ static int zone_highsize(struct zone *zone, int batch, int cpu_online) * However, guaranteeing these relations at all times would require e.g. write * barriers here but also careful usage of read barriers at the read side, and * thus be prone to error and bad for performance. Thus the update only prevents - * store tearing. Any new users of pcp->batch and pcp->high should ensure they - * can cope with those fields changing asynchronously, and fully trust only the - * pcp->count field on the local CPU with interrupts disabled. + * store tearing. Any new users of pcp->batch, pcp->high_min and pcp->high_max + * should ensure they can cope with those fields changing asynchronously, and + * fully trust only the pcp->count field on the local CPU with interrupts + * disabled. * * mutex_is_locked(&pcp_batch_high_lock) required when calling this function * outside of boot time (or some other assurance that no concurrent updaters * exist). */ -static void pageset_update(struct per_cpu_pages *pcp, unsigned long high, - unsigned long batch) +static void pageset_update(struct per_cpu_pages *pcp, unsigned long high_min, + unsigned long high_max, unsigned long batch) { WRITE_ONCE(pcp->batch, batch); - WRITE_ONCE(pcp->high, high); + WRITE_ONCE(pcp->high_min, high_min); + WRITE_ONCE(pcp->high_max, high_max); }
static void per_cpu_pages_init(struct per_cpu_pages *pcp, struct per_cpu_zonestat *pzstats) @@ -5384,20 +5387,21 @@ static void per_cpu_pages_init(struct per_cpu_pages *pcp, struct per_cpu_zonesta * need to be as careful as pageset_update() as nobody can access the * pageset yet. */ - pcp->high = BOOT_PAGESET_HIGH; + pcp->high_min = BOOT_PAGESET_HIGH; + pcp->high_max = BOOT_PAGESET_HIGH; pcp->batch = BOOT_PAGESET_BATCH; pcp->free_factor = 0; }
-static void __zone_set_pageset_high_and_batch(struct zone *zone, unsigned long high, - unsigned long batch) +static void __zone_set_pageset_high_and_batch(struct zone *zone, unsigned long high_min, + unsigned long high_max, unsigned long batch) { struct per_cpu_pages *pcp; int cpu;
for_each_possible_cpu(cpu) { pcp = per_cpu_ptr(zone->per_cpu_pageset, cpu); - pageset_update(pcp, high, batch); + pageset_update(pcp, high_min, high_max, batch); } }
@@ -5407,19 +5411,34 @@ static void __zone_set_pageset_high_and_batch(struct zone *zone, unsigned long h */ static void zone_set_pageset_high_and_batch(struct zone *zone, int cpu_online) { - int new_high, new_batch; + int new_high_min, new_high_max, new_batch;
new_batch = max(1, zone_batchsize(zone)); - new_high = zone_highsize(zone, new_batch, cpu_online); + if (percpu_pagelist_high_fraction) { + new_high_min = zone_highsize(zone, new_batch, cpu_online, + percpu_pagelist_high_fraction); + /* + * PCP high is tuned manually, disable auto-tuning via + * setting high_min and high_max to the manual value. + */ + new_high_max = new_high_min; + } else { + new_high_min = zone_highsize(zone, new_batch, cpu_online, 0); + new_high_max = zone_highsize(zone, new_batch, cpu_online, + MIN_PERCPU_PAGELIST_HIGH_FRACTION); + }
- if (zone->pageset_high == new_high && + if (zone->pageset_high_min == new_high_min && + zone->pageset_high_max == new_high_max && zone->pageset_batch == new_batch) return;
- zone->pageset_high = new_high; + zone->pageset_high_min = new_high_min; + zone->pageset_high_max = new_high_max; zone->pageset_batch = new_batch;
- __zone_set_pageset_high_and_batch(zone, new_high, new_batch); + __zone_set_pageset_high_and_batch(zone, new_high_min, new_high_max, + new_batch); }
void __meminit setup_zone_pageset(struct zone *zone) @@ -5528,7 +5547,8 @@ __meminit void zone_pcp_init(struct zone *zone) */ zone->per_cpu_pageset = &boot_pageset; zone->per_cpu_zonestats = &boot_zonestats; - zone->pageset_high = BOOT_PAGESET_HIGH; + zone->pageset_high_min = BOOT_PAGESET_HIGH; + zone->pageset_high_max = BOOT_PAGESET_HIGH; zone->pageset_batch = BOOT_PAGESET_BATCH;
if (populated_zone(zone)) @@ -6430,13 +6450,14 @@ EXPORT_SYMBOL(free_contig_range); void zone_pcp_disable(struct zone *zone) { mutex_lock(&pcp_batch_high_lock); - __zone_set_pageset_high_and_batch(zone, 0, 1); + __zone_set_pageset_high_and_batch(zone, 0, 0, 1); __drain_all_pages(zone, true); }
void zone_pcp_enable(struct zone *zone) { - __zone_set_pageset_high_and_batch(zone, zone->pageset_high, zone->pageset_batch); + __zone_set_pageset_high_and_batch(zone, zone->pageset_high_min, + zone->pageset_high_max, zone->pageset_batch); mutex_unlock(&pcp_batch_high_lock); }
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit 51a755c56dc05a8b31ed28d24f28354946dc7529 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
The target to tune PCP high automatically is as follows,
- Minimize allocation/freeing from/to shared zone
- Minimize idle pages in PCP
- Minimize pages in PCP if the system free pages is too few
To reach these target, a tuning algorithm as follows is designed,
- When we refill PCP via allocating from the zone, increase PCP high. Because if we had larger PCP, we could avoid to allocate from the zone.
- In periodic vmstat updating kworker (via refresh_cpu_vm_stats()), decrease PCP high to try to free possible idle PCP pages.
- When page reclaiming is active for the zone, stop increasing PCP high in allocating path, decrease PCP high and free some pages in freeing path.
So, the PCP high can be tuned to the page allocating/freeing depth of workloads eventually.
One issue of the algorithm is that if the number of pages allocated is much more than that of pages freed on a CPU, the PCP high may become the maximal value even if the allocating/freeing depth is small. But this isn't a severe issue, because there are no idle pages in this case.
One alternative choice is to increase PCP high when we drain PCP via trying to free pages to the zone, but don't increase PCP high during PCP refilling. This can avoid the issue above. But if the number of pages allocated is much less than that of pages freed on a CPU, there will be many idle pages in PCP and it is hard to free these idle pages.
1/8 (>> 3) of PCP high will be decreased periodically. The value 1/8 is kind of arbitrary. Just to make sure that the idle PCP pages will be freed eventually.
On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service. With the patch, the build time decreases 3.5%. The cycles% of the spinlock contention (mostly for zone lock) decreases from 11.0% to 0.5%. The number of PCP draining for high order pages freeing (free_high) decreases 65.6%. The number of pages allocated from zone (instead of from PCP) decreases 83.9%.
Link: https://lkml.kernel.org/r/20231016053002.756205-8-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Suggested-by: Mel Gorman mgorman@techsingularity.net Suggested-by: Michal Hocko mhocko@suse.com Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Cc: Sudeep Holla sudeep.holla@arm.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- include/linux/gfp.h | 1 + mm/page_alloc.c | 119 ++++++++++++++++++++++++++++++++++---------- mm/vmstat.c | 8 +-- 3 files changed, 99 insertions(+), 29 deletions(-)
diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 665edc11fb9f..5b917e5b9350 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -320,6 +320,7 @@ extern void page_frag_free(void *addr); #define free_page(addr) free_pages((addr), 0)
void page_alloc_init_cpuhp(void); +int decay_pcp_high(struct zone *zone, struct per_cpu_pages *pcp); void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp); void drain_all_pages(struct zone *zone); void drain_local_pages(struct zone *zone); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 27c1c9ad1928..db2693dfa920 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2157,6 +2157,40 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order, return i; }
+/* + * Called from the vmstat counter updater to decay the PCP high. + * Return whether there are addition works to do. + */ +int decay_pcp_high(struct zone *zone, struct per_cpu_pages *pcp) +{ + int high_min, to_drain, batch; + int todo = 0; + + high_min = READ_ONCE(pcp->high_min); + batch = READ_ONCE(pcp->batch); + /* + * Decrease pcp->high periodically to try to free possible + * idle PCP pages. And, avoid to free too many pages to + * control latency. This caps pcp->high decrement too. + */ + if (pcp->high > high_min) { + pcp->high = max3(pcp->count - (batch << CONFIG_PCP_BATCH_SCALE_MAX), + pcp->high - (pcp->high >> 3), high_min); + if (pcp->high > high_min) + todo++; + } + + to_drain = pcp->count - pcp->high; + if (to_drain > 0) { + spin_lock(&pcp->lock); + free_pcppages_bulk(zone, to_drain, pcp, 0); + spin_unlock(&pcp->lock); + todo++; + } + + return todo; +} + #ifdef CONFIG_NUMA /* * Called from the vmstat counter updater to drain pagesets of this @@ -2318,14 +2352,13 @@ static bool free_unref_page_prepare(struct page *page, unsigned long pfn, return true; }
-static int nr_pcp_free(struct per_cpu_pages *pcp, int high, bool free_high) +static int nr_pcp_free(struct per_cpu_pages *pcp, int batch, int high, bool free_high) { int min_nr_free, max_nr_free; - int batch = READ_ONCE(pcp->batch);
- /* Free everything if batch freeing high-order pages. */ + /* Free as much as possible if batch freeing high-order pages. */ if (unlikely(free_high)) - return pcp->count; + return min(pcp->count, batch << CONFIG_PCP_BATCH_SCALE_MAX);
/* Check for PCP disabled or boot pageset */ if (unlikely(high < batch)) @@ -2340,7 +2373,7 @@ static int nr_pcp_free(struct per_cpu_pages *pcp, int high, bool free_high) * freeing of pages without any allocation. */ batch <<= pcp->free_factor; - if (batch < max_nr_free && pcp->free_factor < CONFIG_PCP_BATCH_SCALE_MAX) + if (batch <= max_nr_free && pcp->free_factor < CONFIG_PCP_BATCH_SCALE_MAX) pcp->free_factor++; batch = clamp(batch, min_nr_free, max_nr_free);
@@ -2348,28 +2381,48 @@ static int nr_pcp_free(struct per_cpu_pages *pcp, int high, bool free_high) }
static int nr_pcp_high(struct per_cpu_pages *pcp, struct zone *zone, - bool free_high) + int batch, bool free_high) { - int high = READ_ONCE(pcp->high_min); + int high, high_min, high_max; + + high_min = READ_ONCE(pcp->high_min); + high_max = READ_ONCE(pcp->high_max); + high = pcp->high = clamp(pcp->high, high_min, high_max);
- if (unlikely(!high || free_high)) + if (unlikely(!high)) return 0;
- if (!test_bit(ZONE_RECLAIM_ACTIVE, &zone->flags)) - return high; + if (unlikely(free_high)) { + pcp->high = max(high - (batch << CONFIG_PCP_BATCH_SCALE_MAX), + high_min); + return 0; + }
/* * If reclaim is active, limit the number of pages that can be * stored on pcp lists */ - return min(READ_ONCE(pcp->batch) << 2, high); + if (test_bit(ZONE_RECLAIM_ACTIVE, &zone->flags)) { + pcp->high = max(high - (batch << pcp->free_factor), high_min); + return min(batch << 2, pcp->high); + } + + if (pcp->count >= high && high_min != high_max) { + int need_high = (batch << pcp->free_factor) + batch; + + /* pcp->high should be large enough to hold batch freed pages */ + if (pcp->high < need_high) + pcp->high = clamp(need_high, high_min, high_max); + } + + return high; }
static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, struct page *page, int migratetype, unsigned int order) { - int high; + int high, batch; int pindex; bool free_high = false;
@@ -2384,6 +2437,7 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, list_add(&page->pcp_list, &pcp->lists[pindex]); pcp->count += 1 << order;
+ batch = READ_ONCE(pcp->batch); /* * As high-order pages other than THP's stored on PCP can contribute * to fragmentation, limit the number stored when PCP is heavily @@ -2394,14 +2448,15 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, free_high = (pcp->free_factor && (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER) && (!(pcp->flags & PCPF_FREE_HIGH_BATCH) || - pcp->count >= READ_ONCE(pcp->batch))); + pcp->count >= READ_ONCE(batch))); pcp->flags |= PCPF_PREV_FREE_HIGH_ORDER; } else if (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER) { pcp->flags &= ~PCPF_PREV_FREE_HIGH_ORDER; } - high = nr_pcp_high(pcp, zone, free_high); + high = nr_pcp_high(pcp, zone, batch, free_high); if (pcp->count >= high) { - free_pcppages_bulk(zone, nr_pcp_free(pcp, high, free_high), pcp, pindex); + free_pcppages_bulk(zone, nr_pcp_free(pcp, batch, high, free_high), + pcp, pindex); } }
@@ -2685,24 +2740,38 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone, return page; }
-static int nr_pcp_alloc(struct per_cpu_pages *pcp, int order) +static int nr_pcp_alloc(struct per_cpu_pages *pcp, struct zone *zone, int order) { - int high, batch, max_nr_alloc; + int high, base_batch, batch, max_nr_alloc; + int high_max, high_min;
- high = READ_ONCE(pcp->high_min); - batch = READ_ONCE(pcp->batch); + base_batch = READ_ONCE(pcp->batch); + high_min = READ_ONCE(pcp->high_min); + high_max = READ_ONCE(pcp->high_max); + high = pcp->high = clamp(pcp->high, high_min, high_max);
/* Check for PCP disabled or boot pageset */ - if (unlikely(high < batch)) + if (unlikely(high < base_batch)) return 1;
+ if (order) + batch = base_batch; + else + batch = (base_batch << pcp->alloc_factor); + /* - * Double the number of pages allocated each time there is subsequent - * allocation of order-0 pages without any freeing. + * If we had larger pcp->high, we could avoid to allocate from + * zone. */ + if (high_min != high_max && !test_bit(ZONE_RECLAIM_ACTIVE, &zone->flags)) + high = pcp->high = min(high + batch, high_max); + if (!order) { - max_nr_alloc = max(high - pcp->count - batch, batch); - batch <<= pcp->alloc_factor; + max_nr_alloc = max(high - pcp->count - base_batch, base_batch); + /* + * Double the number of pages allocated each time there is + * subsequent allocation of order-0 pages without any freeing. + */ if (batch <= max_nr_alloc && pcp->alloc_factor < CONFIG_PCP_BATCH_SCALE_MAX) pcp->alloc_factor++; @@ -2733,7 +2802,7 @@ struct page *__rmqueue_pcplist(struct zone *zone, unsigned int order,
do { if (list_empty(list)) { - int batch = nr_pcp_alloc(pcp, order); + int batch = nr_pcp_alloc(pcp, zone, order); int alloced;
alloced = rmqueue_bulk(zone, order, diff --git a/mm/vmstat.c b/mm/vmstat.c index 00e81e99c6ee..2f716ad14168 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -814,9 +814,7 @@ static int refresh_cpu_vm_stats(bool do_pagesets)
for_each_populated_zone(zone) { struct per_cpu_zonestat __percpu *pzstats = zone->per_cpu_zonestats; -#ifdef CONFIG_NUMA struct per_cpu_pages __percpu *pcp = zone->per_cpu_pageset; -#endif
for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) { int v; @@ -832,10 +830,12 @@ static int refresh_cpu_vm_stats(bool do_pagesets) #endif } } -#ifdef CONFIG_NUMA
if (do_pagesets) { cond_resched(); + + changes += decay_pcp_high(zone, this_cpu_ptr(pcp)); +#ifdef CONFIG_NUMA /* * Deal with draining the remote pageset of this * processor @@ -862,8 +862,8 @@ static int refresh_cpu_vm_stats(bool do_pagesets) drain_zone_pages(zone, this_cpu_ptr(pcp)); changes++; } - } #endif + } }
for_each_online_pgdat(pgdat) {
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit 57c0419c5f0ea2ccab8700895c8fac20ba1eb21f category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
One target of PCP is to minimize pages in PCP if the system free pages is too few. To reach that target, when page reclaiming is active for the zone (ZONE_RECLAIM_ACTIVE), we will stop increasing PCP high in allocating path, decrease PCP high and free some pages in freeing path. But this may be too late because the background page reclaiming may introduce latency for some workloads. So, in this patch, during page allocation we will detect whether the number of free pages of the zone is below high watermark. If so, we will stop increasing PCP high in allocating path, decrease PCP high and free some pages in freeing path. With this, we can reduce the possibility of the premature background page reclaiming caused by too large PCP.
The high watermark checking is done in allocating path to reduce the overhead in hotter freeing path.
Link: https://lkml.kernel.org/r/20231016053002.756205-9-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Cc: Mel Gorman mgorman@techsingularity.net Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Michal Hocko mhocko@suse.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Cc: Sudeep Holla sudeep.holla@arm.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- include/linux/mmzone.h | 1 + mm/page_alloc.c | 33 +++++++++++++++++++++++++++++++-- 2 files changed, 32 insertions(+), 2 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index ec3f7daedcc7..c88770381aaf 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -1018,6 +1018,7 @@ enum zone_flags { * Cleared when kswapd is woken. */ ZONE_RECLAIM_ACTIVE, /* kswapd may be scanning the zone. */ + ZONE_BELOW_HIGH, /* zone is below high watermark. */ };
static inline unsigned long zone_managed_pages(struct zone *zone) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index db2693dfa920..728cbc56ac1d 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2407,7 +2407,13 @@ static int nr_pcp_high(struct per_cpu_pages *pcp, struct zone *zone, return min(batch << 2, pcp->high); }
- if (pcp->count >= high && high_min != high_max) { + if (high_min == high_max) + return high; + + if (test_bit(ZONE_BELOW_HIGH, &zone->flags)) { + pcp->high = max(high - (batch << pcp->free_factor), high_min); + high = max(pcp->count, high_min); + } else if (pcp->count >= high) { int need_high = (batch << pcp->free_factor) + batch;
/* pcp->high should be large enough to hold batch freed pages */ @@ -2457,6 +2463,10 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, if (pcp->count >= high) { free_pcppages_bulk(zone, nr_pcp_free(pcp, batch, high, free_high), pcp, pindex); + if (test_bit(ZONE_BELOW_HIGH, &zone->flags) && + zone_watermark_ok(zone, 0, high_wmark_pages(zone), + ZONE_MOVABLE, 0)) + clear_bit(ZONE_BELOW_HIGH, &zone->flags); } }
@@ -2763,7 +2773,7 @@ static int nr_pcp_alloc(struct per_cpu_pages *pcp, struct zone *zone, int order) * If we had larger pcp->high, we could avoid to allocate from * zone. */ - if (high_min != high_max && !test_bit(ZONE_RECLAIM_ACTIVE, &zone->flags)) + if (high_min != high_max && !test_bit(ZONE_BELOW_HIGH, &zone->flags)) high = pcp->high = min(high + batch, high_max);
if (!order) { @@ -3225,6 +3235,25 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags, } }
+ /* + * Detect whether the number of free pages is below high + * watermark. If so, we will decrease pcp->high and free + * PCP pages in free path to reduce the possibility of + * premature page reclaiming. Detection is done here to + * avoid to do that in hotter free path. + */ + if (test_bit(ZONE_BELOW_HIGH, &zone->flags)) + goto check_alloc_wmark; + + mark = high_wmark_pages(zone); + if (zone_watermark_fast(zone, order, mark, + ac->highest_zoneidx, alloc_flags, + gfp_mask)) + goto try_this_zone; + else + set_bit(ZONE_BELOW_HIGH, &zone->flags); + +check_alloc_wmark: mark = wmark_pages(zone, alloc_flags & ALLOC_WMARK_MASK); if (!zone_watermark_fast(zone, order, mark, ac->highest_zoneidx, alloc_flags,
From: Huang Ying ying.huang@intel.com
mainline inclusion from mainline-v6.7-rc1 commit 6ccdcb6d3a741c4e005ca6ffd4a62ddf8b5bead3 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8JXIR
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...
--------------------------------
In current PCP auto-tuning design, if the number of pages allocated is much more than that of pages freed on a CPU, the PCP high may become the maximal value even if the allocating/freeing depth is small, for example, in the sender of network workloads. If a CPU was used as sender originally, then it is used as receiver after context switching, we need to fill the whole PCP with maximal high before triggering PCP draining for consecutive high order freeing. This will hurt the performance of some network workloads.
To solve the issue, in this patch, we will track the consecutive page freeing with a counter in stead of relying on PCP draining. So, we can detect consecutive page freeing much earlier.
On a 2-socket Intel server with 128 logical CPU, we tested SCTP_STREAM_MANY test case of netperf test suite with 64-pair processes. With the patch, the network bandwidth improves 5.0%. This restores the performance drop caused by PCP auto-tuning.
Link: https://lkml.kernel.org/r/20231016053002.756205-10-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Cc: Mel Gorman mgorman@techsingularity.net Cc: Vlastimil Babka vbabka@suse.cz Cc: David Hildenbrand david@redhat.com Cc: Johannes Weiner jweiner@redhat.com Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Michal Hocko mhocko@suse.com Cc: Pavel Tatashin pasha.tatashin@soleen.com Cc: Matthew Wilcox willy@infradead.org Cc: Christoph Lameter cl@linux.com Cc: Arjan van de Ven arjan@linux.intel.com Cc: Sudeep Holla sudeep.holla@arm.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Ze Zuo zuoze1@huawei.com --- include/linux/mmzone.h | 2 +- mm/page_alloc.c | 27 +++++++++++++++------------ 2 files changed, 16 insertions(+), 13 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index c88770381aaf..57086c57b8e4 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -700,10 +700,10 @@ struct per_cpu_pages { int batch; /* chunk size for buddy add/remove */ u8 flags; /* protected by pcp->lock */ u8 alloc_factor; /* batch scaling factor during allocate */ - u8 free_factor; /* batch scaling factor during free */ #ifdef CONFIG_NUMA u8 expire; /* When 0, remote pagesets are drained */ #endif + short free_count; /* consecutive free count */
/* Lists of pages, one per migrate type stored on the pcp-lists */ struct list_head lists[NR_PCP_LISTS]; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 728cbc56ac1d..f5b61c1060d1 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2369,13 +2369,10 @@ static int nr_pcp_free(struct per_cpu_pages *pcp, int batch, int high, bool free max_nr_free = high - batch;
/* - * Double the number of pages freed each time there is subsequent - * freeing of pages without any allocation. + * Increase the batch number to the number of the consecutive + * freed pages to reduce zone lock contention. */ - batch <<= pcp->free_factor; - if (batch <= max_nr_free && pcp->free_factor < CONFIG_PCP_BATCH_SCALE_MAX) - pcp->free_factor++; - batch = clamp(batch, min_nr_free, max_nr_free); + batch = clamp_t(int, pcp->free_count, min_nr_free, max_nr_free);
return batch; } @@ -2403,7 +2400,9 @@ static int nr_pcp_high(struct per_cpu_pages *pcp, struct zone *zone, * stored on pcp lists */ if (test_bit(ZONE_RECLAIM_ACTIVE, &zone->flags)) { - pcp->high = max(high - (batch << pcp->free_factor), high_min); + int free_count = max_t(int, pcp->free_count, batch); + + pcp->high = max(high - free_count, high_min); return min(batch << 2, pcp->high); }
@@ -2411,10 +2410,12 @@ static int nr_pcp_high(struct per_cpu_pages *pcp, struct zone *zone, return high;
if (test_bit(ZONE_BELOW_HIGH, &zone->flags)) { - pcp->high = max(high - (batch << pcp->free_factor), high_min); + int free_count = max_t(int, pcp->free_count, batch); + + pcp->high = max(high - free_count, high_min); high = max(pcp->count, high_min); } else if (pcp->count >= high) { - int need_high = (batch << pcp->free_factor) + batch; + int need_high = pcp->free_count + batch;
/* pcp->high should be large enough to hold batch freed pages */ if (pcp->high < need_high) @@ -2451,7 +2452,7 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, * stops will be drained from vmstat refresh context. */ if (order && order <= PAGE_ALLOC_COSTLY_ORDER) { - free_high = (pcp->free_factor && + free_high = (pcp->free_count >= batch && (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER) && (!(pcp->flags & PCPF_FREE_HIGH_BATCH) || pcp->count >= READ_ONCE(batch))); @@ -2459,6 +2460,8 @@ static void free_unref_page_commit(struct zone *zone, struct per_cpu_pages *pcp, } else if (pcp->flags & PCPF_PREV_FREE_HIGH_ORDER) { pcp->flags &= ~PCPF_PREV_FREE_HIGH_ORDER; } + if (pcp->free_count < (batch << CONFIG_PCP_BATCH_SCALE_MAX)) + pcp->free_count += (1 << order); high = nr_pcp_high(pcp, zone, batch, free_high); if (pcp->count >= high) { free_pcppages_bulk(zone, nr_pcp_free(pcp, batch, high, free_high), @@ -2855,7 +2858,7 @@ static struct page *rmqueue_pcplist(struct zone *preferred_zone, * See nr_pcp_free() where free_factor is increased for subsequent * frees. */ - pcp->free_factor >>= 1; + pcp->free_count >>= 1; list = &pcp->lists[order_to_pindex(migratetype, order)]; page = __rmqueue_pcplist(zone, order, migratetype, alloc_flags, pcp, list); pcp_spin_unlock(pcp); @@ -5488,7 +5491,7 @@ static void per_cpu_pages_init(struct per_cpu_pages *pcp, struct per_cpu_zonesta pcp->high_min = BOOT_PAGESET_HIGH; pcp->high_max = BOOT_PAGESET_HIGH; pcp->batch = BOOT_PAGESET_BATCH; - pcp->free_factor = 0; + pcp->free_count = 0; }
static void __zone_set_pageset_high_and_batch(struct zone *zone, unsigned long high_min,
-
Ze Zuo