This introduces the new flag MMF_VM_MERGE_ANY flag. When this flag is set, kernel samepage merging (ksm) gets enabled for all vma's of a process.

Signed-off-by: Stefan Roesch shr@devkernel.io --- include/linux/sched/coredump.h | 1 + 1 file changed, 1 insertion(+)

diff --git a/include/linux/sched/coredump.h b/include/linux/sched/coredump.h index 8270ad7ae14c..0e0de5b7469f 100644 --- a/include/linux/sched/coredump.h +++ b/include/linux/sched/coredump.h @@ -86,4 +86,5 @@ static inline int get_dumpable(struct mm_struct *mm) #define MMF_INIT_MASK (MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK |\ MMF_DISABLE_THP_MASK)

+#define MMF_VM_MERGE_ANY 29 #endif /* _LINUX_SCHED_COREDUMP_H */

This adds the flag parameter to the __ksm_exit() call. This allows to distinguish if this call is for an prctl or madvise invocation.

Signed-off-by: Stefan Roesch shr@devkernel.io --- include/linux/ksm.h | 8 +++++--- mm/ksm.c | 8 ++++++-- 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/include/linux/ksm.h b/include/linux/ksm.h index 50e6b56092f3..d38a05a36298 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -18,8 +18,8 @@ #ifdef CONFIG_KSM int ksm_madvise(struct vm_area_struct *vma, unsigned long start, unsigned long end, int advice, unsigned long *vm_flags); -void __ksm_exit(struct mm_struct *mm); int __ksm_enter(struct mm_struct *mm, int flag); +void __ksm_exit(struct mm_struct *mm, int flag);

static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) { @@ -32,8 +32,10 @@ static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm)

static inline void ksm_exit(struct mm_struct *mm) { - if (test_bit(MMF_VM_MERGEABLE, &mm->flags)) - __ksm_exit(mm); + if (test_bit(MMF_VM_MERGE_ANY, &mm->flags)) + __ksm_exit(mm, MMF_VM_MERGE_ANY); + else if (test_bit(MMF_VM_MERGEABLE, &mm->flags)) + __ksm_exit(mm, MMF_VM_MERGEABLE); }

/* diff --git a/mm/ksm.c b/mm/ksm.c index d84a244fe224..83796328574c 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -2563,12 +2563,16 @@ int __ksm_enter(struct mm_struct *mm, int flag) return 0; }

-void __ksm_exit(struct mm_struct *mm) +void __ksm_exit(struct mm_struct *mm, int flag) { struct ksm_mm_slot *mm_slot; struct mm_slot *slot; int easy_to_free = 0;

+ if (!(current->flags & PF_EXITING) && flag == MMF_VM_MERGE_ANY && + test_bit(MMF_VM_MERGE_ANY, &mm->flags)) + clear_bit(MMF_VM_MERGE_ANY, &mm->flags); + /* * This process is exiting: if it's straightforward (as is the * case when ksmd was never running), free mm_slot immediately. @@ -2595,7 +2599,7 @@ void __ksm_exit(struct mm_struct *mm)

if (easy_to_free) { mm_slot_free(mm_slot_cache, mm_slot); - clear_bit(MMF_VM_MERGEABLE, &mm->flags); + clear_bit(flag, &mm->flags); mmdrop(mm); } else if (mm_slot) { mmap_write_lock(mm);

This adds the ability to disable ksm for a process if ksm has been enabled for the process.

Signed-off-by: Stefan Roesch shr@devkernel.io --- mm/ksm.c | 25 ++++++++++++++++++++++++- 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/mm/ksm.c b/mm/ksm.c index 967eda719fab..5fa6b46dfa3b 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -2588,6 +2588,27 @@ int __ksm_enter(struct mm_struct *mm, int flag) return 0; }

+static void unmerge_vmas(struct mm_struct *mm) +{ + struct vm_area_struct *vma; + struct vma_iterator vmi; + + vma_iter_init(&vmi, mm, 0); + + mmap_read_lock(mm); + for_each_vma(vmi, vma) { + if (vma->vm_flags & VM_MERGEABLE) { + unsigned long flags = vma->vm_flags; + + if (ksm_madvise(vma, vma->vm_start, vma->vm_end, MADV_UNMERGEABLE, &flags)) + continue; + + vma->vm_flags = flags; + } + } + mmap_read_unlock(mm); +} + void __ksm_exit(struct mm_struct *mm, int flag) { struct ksm_mm_slot *mm_slot; @@ -2595,8 +2616,10 @@ void __ksm_exit(struct mm_struct *mm, int flag) int easy_to_free = 0;

if (!(current->flags & PF_EXITING) && flag == MMF_VM_MERGE_ANY && - test_bit(MMF_VM_MERGE_ANY, &mm->flags)) + test_bit(MMF_VM_MERGE_ANY, &mm->flags)) { clear_bit(MMF_VM_MERGE_ANY, &mm->flags); + unmerge_vmas(mm); + }

/* * This process is exiting: if it's straightforward (as is the

This adds the following tracepoints to ksm: - start / stop scan - ksm enter / exit - merge a page - merge a page with ksm - remove a page - remove a rmap item

Signed-off-by: Stefan Roesch shr@devkernel.io --- MAINTAINERS | 1 + include/trace/events/ksm.h | 257 +++++++++++++++++++++++++++++++++++++ mm/ksm.c | 20 ++- 3 files changed, 276 insertions(+), 2 deletions(-) create mode 100644 include/trace/events/ksm.h

diff --git a/MAINTAINERS b/MAINTAINERS index 123216b76534..990a28bdc263 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -13482,6 +13482,7 @@ F: include/linux/memory_hotplug.h F: include/linux/mm.h F: include/linux/mmzone.h F: include/linux/pagewalk.h +F: include/trace/events/ksm.h F: mm/ F: tools/testing/selftests/vm/

diff --git a/include/trace/events/ksm.h b/include/trace/events/ksm.h new file mode 100644 index 000000000000..727d8f6dbc3f --- /dev/null +++ b/include/trace/events/ksm.h @@ -0,0 +1,257 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM ksm + +#if !defined(_TRACE_KSM_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_KSM_H + +#include <linux/tracepoint.h> + +/** + * ksm_scan_template - called for start / stop scan + * + * @seq: sequence number of scan + * @rmap_entries: actual number of rmap entries + * + * Allows to trace the start / stop of a ksm scan. + */ +DECLARE_EVENT_CLASS(ksm_scan_template, + + TP_PROTO(int seq, u32 rmap_entries), + + TP_ARGS(seq, rmap_entries), + + TP_STRUCT__entry( + __field(int, seq) + __field(u32, rmap_entries) + ), + + TP_fast_assign( + __entry->seq = seq; + __entry->rmap_entries = rmap_entries; + ), + + TP_printk("seq %d rmap size %d", + __entry->seq, __entry->rmap_entries) +); + +/** + * ksm_start_scan - called after a new ksm scan is started + * + * @seq: sequence number of scan + * @rmap_entries: actual number of rmap entries + * + * Allows to trace the start of a ksm scan. + */ +DEFINE_EVENT(ksm_scan_template, ksm_start_scan, + + TP_PROTO(int seq, u32 rmap_entries), + + TP_ARGS(seq, rmap_entries) +); + +/** + * ksm_stop_scan - called after a new ksm scan has completed + * + * @seq: sequence number of scan + * @rmap_entries: actual number of rmap entries + * + * Allows to trace the completion of a ksm scan. + */ +DEFINE_EVENT(ksm_scan_template, ksm_stop_scan, + + TP_PROTO(int seq, u32 rmap_entries), + + TP_ARGS(seq, rmap_entries) +); + +/** + * ksm_enter - called after a new process has been added / removed from ksm + * + * @mm: address of the mm object of the process + * @flag: specified flags + * + * Allows to trace the when a process has been added or removed from ksm. + */ +DECLARE_EVENT_CLASS(ksm_enter_exit_template, + + TP_PROTO(void *mm, unsigned int flag), + + TP_ARGS(mm, flag), + + TP_STRUCT__entry( + __field(void *, mm) + __field(unsigned int, flag) + ), + + TP_fast_assign( + __entry->mm = mm; + __entry->flag = flag; + ), + + TP_printk("mm %p flag %u", + __entry->mm, __entry->flag) +); + +/** + * ksm_enter - called after a new process has been added to ksm + * + * @mm: address of the mm object of the process + * @flag: specified flags + * + * Allows to trace the when a process has been added to ksm. + */ +DEFINE_EVENT(ksm_enter_exit_template, ksm_enter, + + TP_PROTO(void *mm, unsigned int flag), + + TP_ARGS(mm, flag) +); + +/** + * ksm_exit - called after a new process has been removed from ksm + * + * @mm: address of the mm object of the process + * @flag: specified flags + * + * Allows to trace the when a process has been removed from ksm. + */ +DEFINE_EVENT(ksm_enter_exit_template, ksm_exit, + + TP_PROTO(void *mm, unsigned int flag), + + TP_ARGS(mm, flag) +); + +/** + * ksm_merge_one_page - called after a page has been merged + * + * @pfn: page frame number of ksm page + * @rmap_item: address of rmap_item object + * @mm: address of the process mm struct + * @err: success + * + * Allows to trace the ksm merging of individual pages. + */ +TRACE_EVENT(ksm_merge_one_page, + + TP_PROTO(unsigned long pfn, void *rmap_item, void *mm, int err), + + TP_ARGS(pfn, rmap_item, mm, err), + + TP_STRUCT__entry( + __field(unsigned long, pfn) + __field(void *, rmap_item) + __field(void *, mm) + __field(int, err) + ), + + TP_fast_assign( + __entry->pfn = pfn; + __entry->rmap_item = rmap_item; + __entry->mm = mm; + __entry->err = err; + ), + + TP_printk("ksm pfn %lu rmap_item %p mm %p error %d", + __entry->pfn, __entry->rmap_item, __entry->mm, __entry->err) +); + +/** + * ksm_merge_with_ksm_page - called after a page has been merged with a ksm page + * + * @ksm_page: address ksm page + * @pfn: page frame number of ksm page + * @rmap_item: address of rmap_item object + * @mm: address of the mm object of the process + * @err: success + * + * Allows to trace the merging of a page with a ksm page. + */ +TRACE_EVENT(ksm_merge_with_ksm_page, + + TP_PROTO(void *ksm_page, unsigned long pfn, void *rmap_item, void *mm, int err), + + TP_ARGS(ksm_page, pfn, rmap_item, mm, err), + + TP_STRUCT__entry( + __field(void *, ksm_page) + __field(unsigned long, pfn) + __field(void *, rmap_item) + __field(void *, mm) + __field(int, err) + ), + + TP_fast_assign( + __entry->ksm_page = ksm_page; + __entry->pfn = pfn; + __entry->rmap_item = rmap_item; + __entry->mm = mm; + __entry->err = err; + ), + + TP_printk("%spfn %lu rmap_item %p mm %p error %d", + (__entry->ksm_page ? "ksm " : ""), + __entry->pfn, __entry->rmap_item, __entry->mm, __entry->err) +); + +/** + * ksm_remove_ksm_page - called after a ksm page has been removed + * + * @pfn: page frame number of ksm page + * + * Allows to trace the removing of stable ksm pages. + */ +TRACE_EVENT(ksm_remove_ksm_page, + + TP_PROTO(unsigned long pfn), + + TP_ARGS(pfn), + + TP_STRUCT__entry( + __field(unsigned long, pfn) + ), + + TP_fast_assign( + __entry->pfn = pfn; + ), + + TP_printk("pfn %lu", __entry->pfn) +); + +/** + * ksm_remove_rmap_item - called after a rmap_item has been removed from the + * stable tree + * + * @pfn: page frame number of ksm page + * @rmap_item: address of rmap_item object + * @mm: address of the process mm struct + * + * Allows to trace the removal of pages from the stable tree list. + */ +TRACE_EVENT(ksm_remove_rmap_item, + + TP_PROTO(unsigned long pfn, void *rmap_item, void *mm), + + TP_ARGS(pfn, rmap_item, mm), + + TP_STRUCT__entry( + __field(unsigned long, pfn) + __field(void *, rmap_item) + __field(void *, mm) + ), + + TP_fast_assign( + __entry->pfn = pfn; + __entry->rmap_item = rmap_item; + __entry->mm = mm; + ), + + TP_printk("pfn %lu rmap_item %p mm %p", + __entry->pfn, __entry->rmap_item, __entry->mm) +); + +#endif /* _TRACE_KSM_H */ + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/mm/ksm.c b/mm/ksm.c index 5fa6b46dfa3b..88a969849955 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -45,6 +45,9 @@ #include "internal.h" #include "mm_slot.h"

+#define CREATE_TRACE_POINTS +#include <trace/events/ksm.h> + #ifdef CONFIG_NUMA #define NUMA(x) (x) #define DO_NUMA(x) do { (x); } while (0) @@ -633,10 +636,12 @@ static void remove_node_from_stable_tree(struct ksm_stable_node *stable_node) BUG_ON(stable_node->rmap_hlist_len < 0);

hlist_for_each_entry(rmap_item, &stable_node->hlist, hlist) { - if (rmap_item->hlist.next) + if (rmap_item->hlist.next) { + trace_ksm_remove_rmap_item(stable_node->kpfn, rmap_item, rmap_item->mm); ksm_pages_sharing--; - else + } else { ksm_pages_shared--; + }

rmap_item->mm->ksm_merging_pages--;

@@ -657,6 +662,7 @@ static void remove_node_from_stable_tree(struct ksm_stable_node *stable_node) BUILD_BUG_ON(STABLE_NODE_DUP_HEAD <= &migrate_nodes); BUILD_BUG_ON(STABLE_NODE_DUP_HEAD >= &migrate_nodes + 1);

+ trace_ksm_remove_ksm_page(stable_node->kpfn); if (stable_node->head == &migrate_nodes) list_del(&stable_node->list); else @@ -1319,6 +1325,8 @@ static int try_to_merge_with_ksm_page(struct ksm_rmap_item *rmap_item, get_anon_vma(vma->anon_vma); out: mmap_read_unlock(mm); + trace_ksm_merge_with_ksm_page(kpage, page_to_pfn(kpage ? kpage : page), + rmap_item, mm, err); return err; }

@@ -2137,6 +2145,8 @@ static void cmp_and_merge_page(struct page *page, struct ksm_rmap_item *rmap_ite if (vma) { err = try_to_merge_one_page(vma, page, ZERO_PAGE(rmap_item->address)); + trace_ksm_merge_one_page(page_to_pfn(ZERO_PAGE(rmap_item->address)), + rmap_item, mm, err); } else { /* * If the vma is out of date, we do not need to @@ -2270,6 +2280,8 @@ static struct ksm_rmap_item *scan_get_next_rmap_item(struct page **page)

mm_slot = ksm_scan.mm_slot; if (mm_slot == &ksm_mm_head) { + trace_ksm_start_scan(ksm_scan.seqnr, ksm_rmap_items); + /* * A number of pages can hang around indefinitely on per-cpu * pagevecs, raised page count preventing write_protect_page @@ -2433,6 +2445,7 @@ static struct ksm_rmap_item *scan_get_next_rmap_item(struct page **page) if (mm_slot != &ksm_mm_head) goto next_mm;

+ trace_ksm_stop_scan(ksm_scan.seqnr, ksm_rmap_items); ksm_scan.seqnr++; return NULL; } @@ -2585,6 +2598,7 @@ int __ksm_enter(struct mm_struct *mm, int flag) if (needs_wakeup) wake_up_interruptible(&ksm_thread_wait);

+ trace_ksm_enter(mm, flag); return 0; }

@@ -2653,6 +2667,8 @@ void __ksm_exit(struct mm_struct *mm, int flag) mmap_write_lock(mm); mmap_write_unlock(mm); } + + trace_ksm_exit(mm, flag); }

struct page *ksm_might_need_to_copy(struct page *page,

This splits off the pages_volatile function. The next patch will use this function.

Signed-off-by: Stefan Roesch shr@devkernel.io --- mm/ksm.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/mm/ksm.c b/mm/ksm.c index 88a969849955..19f434f97dc1 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -3182,8 +3182,7 @@ static ssize_t pages_unshared_show(struct kobject *kobj, } KSM_ATTR_RO(pages_unshared);

-static ssize_t pages_volatile_show(struct kobject *kobj, - struct kobj_attribute *attr, char *buf) +static long pages_volatile(void) { long ksm_pages_volatile;

@@ -3195,7 +3194,14 @@ static ssize_t pages_volatile_show(struct kobject *kobj, */ if (ksm_pages_volatile < 0) ksm_pages_volatile = 0; - return sysfs_emit(buf, "%ld\n", ksm_pages_volatile); + + return ksm_pages_volatile; +} + +static ssize_t pages_volatile_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return sysfs_emit(buf, "%ld\n", pages_volatile()); } KSM_ATTR_RO(pages_volatile);

This adds documentation for the new /sys/kernel/mm/ksm/general_profit knob.

Signed-off-by: Stefan Roesch shr@devkernel.io --- Documentation/ABI/testing/sysfs-kernel-mm-ksm | 8 ++++++++ Documentation/admin-guide/mm/ksm.rst | 2 ++ 2 files changed, 10 insertions(+)

diff --git a/Documentation/ABI/testing/sysfs-kernel-mm-ksm b/Documentation/ABI/testing/sysfs-kernel-mm-ksm index d244674a9480..7768e90f7a8f 100644 --- a/Documentation/ABI/testing/sysfs-kernel-mm-ksm +++ b/Documentation/ABI/testing/sysfs-kernel-mm-ksm @@ -51,3 +51,11 @@ Description: Control merging pages across different NUMA nodes.

When it is set to 0 only pages from the same node are merged, otherwise pages from all nodes can be merged together (default). + +What: /sys/kernel/mm/ksm/general_profit +Date: January 2023 +KernelVersion: 6.1 +Contact: Linux memory management mailing list linux-mm@kvack.org +Description: Measure how effective KSM is. + general_profit: how effective is KSM. The formula for the + calculation is in Documentation/admin-guide/mm/ksm.rst. diff --git a/Documentation/admin-guide/mm/ksm.rst b/Documentation/admin-guide/mm/ksm.rst index fb6ba2002a4b..72189bf1283e 100644 --- a/Documentation/admin-guide/mm/ksm.rst +++ b/Documentation/admin-guide/mm/ksm.rst @@ -159,6 +159,8 @@ stable_node_chains_prune_millisecs

The effectiveness of KSM and MADV_MERGEABLE is shown in ``/sys/kernel/mm/ksm/``:

+general_profit + how effective is KSM. The calculation is explained below. pages_shared how many shared pages are being used pages_sharing

The ksm documentation mentions the process profit metric and how to calculate it. This adds the calculation of the metric.

Signed-off-by: Stefan Roesch shr@devkernel.io --- include/linux/ksm.h | 4 ++++ mm/ksm.c | 8 ++++++++ 2 files changed, 12 insertions(+)

diff --git a/include/linux/ksm.h b/include/linux/ksm.h index d38a05a36298..a18cd03efcfb 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -55,6 +55,10 @@ struct page *ksm_might_need_to_copy(struct page *page, void rmap_walk_ksm(struct folio *folio, struct rmap_walk_control *rwc); void folio_migrate_ksm(struct folio *newfolio, struct folio *folio);

+#ifdef CONFIG_PROC_FS +long ksm_process_profit(struct mm_struct *); +#endif /* CONFIG_PROC_FS */ + #else /* !CONFIG_KSM */

static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) diff --git a/mm/ksm.c b/mm/ksm.c index d29454a802a0..288689b59527 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -2935,6 +2935,14 @@ static void wait_while_offlining(void) } #endif /* CONFIG_MEMORY_HOTREMOVE */

+#ifdef CONFIG_PROC_FS +long ksm_process_profit(struct mm_struct *mm) +{ + return (long)mm->ksm_merging_pages * PAGE_SIZE - + mm->ksm_rmap_items * sizeof(struct ksm_rmap_item); +} +#endif /* CONFIG_PROC_FS */ + #ifdef CONFIG_SYSFS /* * This all compiles without CONFIG_SYSFS, but is a waste of space.

This exposes the ksm process profit metric in /proc/<pid>/ksm_stat. The documentation mentions the formula for the ksm process profit metric, however it does not calculate it. In addition the formula depends on the size of internal structures. So it makes sense to expose it.

Signed-off-by: Stefan Roesch shr@devkernel.io --- fs/proc/base.c | 3 +++ 1 file changed, 3 insertions(+)

diff --git a/fs/proc/base.c b/fs/proc/base.c index 9e479d7d202b..a2b7779f76f6 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -96,6 +96,7 @@ #include <linux/time_namespace.h> #include <linux/resctrl.h> #include <linux/cn_proc.h> +#include <linux/ksm.h> #include <trace/events/oom.h> #include "internal.h" #include "fd.h" @@ -3207,6 +3208,8 @@ static int proc_pid_ksm_stat(struct seq_file *m, struct pid_namespace *ns, mm = get_task_mm(task); if (mm) { seq_printf(m, "ksm_rmap_items %lu\n", mm->ksm_rmap_items); + seq_printf(m, "ksm_merging_pages %lu\n", mm->ksm_merging_pages); + seq_printf(m, "ksm_process_profit %ld\n", ksm_process_profit(mm)); mmput(mm); }

This documents the ksm_process_profit and ksm_merge_type settings in ksm_stat.

Signed-off-by: Stefan Roesch shr@devkernel.io --- Documentation/admin-guide/mm/ksm.rst | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/Documentation/admin-guide/mm/ksm.rst b/Documentation/admin-guide/mm/ksm.rst index 72189bf1283e..70896f01d641 100644 --- a/Documentation/admin-guide/mm/ksm.rst +++ b/Documentation/admin-guide/mm/ksm.rst @@ -211,7 +211,8 @@ several times, which are unprofitable memory consumed. ksm_rmap_items * sizeof(rmap_item).

where ksm_merging_pages is shown under the directory ``/proc/<pid>/``, - and ksm_rmap_items is shown in ``/proc/<pid>/ksm_stat``. + and ksm_rmap_items is shown in ``/proc/<pid>/ksm_stat``. The process profit + is also shown in ``/proc/<pid>/ksm_stat`` as ksm_process_profit.

From the perspective of application, a high ratio of ``ksm_rmap_items`` to ``ksm_merging_pages`` means a bad madvise-applied policy, so developers or @@ -222,6 +223,9 @@ so if the ``ksm_rmap_items/ksm_merging_pages`` ratio exceeds 64 on 64-bit CPU or exceeds 128 on 32-bit CPU, then the app's madvise policy should be dropped, because the ksm profit is approximately zero or negative.

+The ksm_merge_type in ``/proc/<pid>/ksm_stat`` shows the merge type of the +process. Valid values are ``none``, ``madvise`` and ``process``. + Monitoring KSM events =====================

This adds the new prctl flags to the include file prct.h in the tools directory. This makes sure they are available for testing.

Signed-off-by: Stefan Roesch shr@devkernel.io --- tools/include/uapi/linux/prctl.h | 2 ++ 1 file changed, 2 insertions(+)

diff --git a/tools/include/uapi/linux/prctl.h b/tools/include/uapi/linux/prctl.h index a5e06dcbba13..b8ba85f18e43 100644 --- a/tools/include/uapi/linux/prctl.h +++ b/tools/include/uapi/linux/prctl.h @@ -284,4 +284,6 @@ struct prctl_mm_map { #define PR_SET_VMA 0x53564d41 # define PR_SET_VMA_ANON_NAME 0

+#define PR_SET_MEMORY_MERGE 65 +#define PR_GET_MEMORY_MERGE 66 #endif /* _LINUX_PRCTL_H */

This adds the -G flag to the ksm_tests to query if prctl has been used to enable ksm merging.

Signed-off-by: Stefan Roesch shr@devkernel.io --- tools/testing/selftests/vm/ksm_tests.c | 37 ++++++++++++++++++++++++-- 1 file changed, 35 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/vm/ksm_tests.c b/tools/testing/selftests/vm/ksm_tests.c index 386a0929c8a3..9667cb3b8c6a 100644 --- a/tools/testing/selftests/vm/ksm_tests.c +++ b/tools/testing/selftests/vm/ksm_tests.c @@ -45,6 +45,7 @@ enum ksm_merge_type { enum ksm_test_name { CHECK_KSM_MERGE, CHECK_KSM_UNMERGE, + CHECK_KSM_GET_MERGE_TYPE, CHECK_KSM_ZERO_PAGE_MERGE, CHECK_KSM_NUMA_MERGE, KSM_MERGE_TIME, @@ -124,7 +125,8 @@ static void print_help(void) " -D evaluate unmerging time and speed when disabling KSM.\n" " For this test, the size of duplicated memory area (in MiB)\n" " must be provided using -s option\n" - " -C evaluate the time required to break COW of merged pages.\n\n"); + " -C evaluate the time required to break COW of merged pages.\n" + " -G query merge mode\n\n");

printf(" -a: specify the access protections of pages.\n" " <prot> must be of the form [rwx].\n" @@ -323,6 +325,31 @@ static int check_ksm_merge(int merge_type, int mapping, int prot, return KSFT_FAIL; }

+static int check_ksm_get_merge_type(void) +{ + if (prctl(PR_SET_MEMORY_MERGE, 1)) { + perror("prctl set"); + return 1; + } + + int is_on = prctl(PR_GET_MEMORY_MERGE, 0); + + if (prctl(PR_SET_MEMORY_MERGE, 0)) { + perror("prctl set"); + return 1; + } + + int is_off = prctl(PR_GET_MEMORY_MERGE, 0); + + if (is_on && is_off) { + printf("OK\n"); + return KSFT_PASS; + } + + printf("Not OK\n"); + return KSFT_FAIL; +} + static int check_ksm_unmerge(int merge_type, int mapping, int prot, int timeout, size_t page_size) { void *map_ptr; @@ -733,7 +760,7 @@ int main(int argc, char *argv[]) bool merge_across_nodes = KSM_MERGE_ACROSS_NODES_DEFAULT; long size_MB = 0;

- while ((opt = getopt(argc, argv, "ha:p:l:z:m:s:t:MUZNPCHD")) != -1) { + while ((opt = getopt(argc, argv, "ha:p:l:z:m:s:t:GMUZNPCHD")) != -1) { switch (opt) { case 'a': prot = str_to_prot(optarg); @@ -792,6 +819,9 @@ int main(int argc, char *argv[]) case 'Z': test_name = CHECK_KSM_ZERO_PAGE_MERGE; break; + case 'G': + test_name = CHECK_KSM_GET_MERGE_TYPE; + break; case 'N': test_name = CHECK_KSM_NUMA_MERGE; break; @@ -841,6 +871,9 @@ int main(int argc, char *argv[]) ret = check_ksm_unmerge(merge_type, MAP_PRIVATE | MAP_ANONYMOUS, prot, ksm_scan_limit_sec, page_size); break; + case CHECK_KSM_GET_MERGE_TYPE: + ret = check_ksm_get_merge_type(); + break; case CHECK_KSM_ZERO_PAGE_MERGE: ret = check_ksm_zero_page_merge(merge_type, MAP_PRIVATE | MAP_ANONYMOUS, prot, page_count, ksm_scan_limit_sec, use_zero_pages,

This adds two functions to report the metrics in /proc/self/ksm_stat and /sys/kernel/debug/mm/ksm.

The debugging can be enabled with the following command line: make -C tools/testing/selftests TARGETS="vm" --keep-going \ EXTRA_CFLAGS=-DDEBUG=1

Signed-off-by: Stefan Roesch shr@devkernel.io --- tools/testing/selftests/vm/ksm_tests.c | 54 ++++++++++++++++++++++++++ 1 file changed, 54 insertions(+)

diff --git a/tools/testing/selftests/vm/ksm_tests.c b/tools/testing/selftests/vm/ksm_tests.c index a0a48ac43b29..9fb21b982dc9 100644 --- a/tools/testing/selftests/vm/ksm_tests.c +++ b/tools/testing/selftests/vm/ksm_tests.c @@ -93,6 +93,55 @@ static int ksm_read_sysfs(const char *file_path, unsigned long *val) return 0; }

+#ifdef DEBUG +static void ksm_print_sysfs(void) +{ + unsigned long max_page_sharing, pages_sharing, pages_shared; + unsigned long full_scans, pages_unshared, pages_volatile; + unsigned long stable_node_chains, stable_node_dups; + long general_profit; + + if (ksm_read_sysfs(KSM_FP("pages_shared"), &pages_shared) || + ksm_read_sysfs(KSM_FP("pages_sharing"), &pages_sharing) || + ksm_read_sysfs(KSM_FP("max_page_sharing"), &max_page_sharing) || + ksm_read_sysfs(KSM_FP("full_scans"), &full_scans) || + ksm_read_sysfs(KSM_FP("pages_unshared"), &pages_unshared) || + ksm_read_sysfs(KSM_FP("pages_volatile"), &pages_volatile) || + ksm_read_sysfs(KSM_FP("stable_node_chains"), &stable_node_chains) || + ksm_read_sysfs(KSM_FP("stable_node_dups"), &stable_node_dups) || + ksm_read_sysfs(KSM_FP("general_profit"), (unsigned long *)&general_profit)) + return; + + printf("pages_shared : %lu\n", pages_shared); + printf("pages_sharing : %lu\n", pages_sharing); + printf("max_page_sharing : %lu\n", max_page_sharing); + printf("full_scans : %lu\n", full_scans); + printf("pages_unshared : %lu\n", pages_unshared); + printf("pages_volatile : %lu\n", pages_volatile); + printf("stable_node_chains: %lu\n", stable_node_chains); + printf("stable_node_dups : %lu\n", stable_node_dups); + printf("general_profit : %ld\n", general_profit); +} + +static void ksm_print_procfs(void) +{ + const char *file_name = "/proc/self/ksm_stat"; + char buffer[512]; + FILE *f = fopen(file_name, "r"); + + if (!f) { + fprintf(stderr, "f %s\n", file_name); + perror("fopen"); + return; + } + + while (fgets(buffer, sizeof(buffer), f)) + printf("%s", buffer); + + fclose(f); +} +#endif + static int str_to_prot(char *prot_str) { int prot = 0; @@ -237,6 +286,11 @@ static bool assert_ksm_pages_count(long dupl_page_count) ksm_read_sysfs(KSM_FP("max_page_sharing"), &max_page_sharing)) return false;

+#ifdef DEBUG + ksm_print_sysfs(); + ksm_print_procfs(); +#endif + /* * Since there must be at least 2 pages for merging and 1 page can be * shared with the limited number of pages (max_page_sharing), sometimes