2022-08-22 16:02:16

by Liam R. Howlett

Subject: [PATCH v13 47/70] perf: use VMA iterator

From: "Matthew Wilcox (Oracle)" <[email protected]>

The VMA iterator is faster than the linked list and removing the linked
list will shrink the vm_area_struct.

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
Signed-off-by: Liam R. Howlett <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
---
kernel/events/core.c | 3 ++-
kernel/events/uprobes.c | 9 ++++++---
2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 2621fd24ad26..101c5912c3fc 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -10229,8 +10229,9 @@ static void perf_addr_filter_apply(struct perf_addr_filter *filter,
struct perf_addr_filter_range *fr)
{
struct vm_area_struct *vma;
+ VMA_ITERATOR(vmi, mm, 0);

- for (vma = mm->mmap; vma; vma = vma->vm_next) {
+ for_each_vma(vmi, vma) {
if (!vma->vm_file)
continue;

diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 2eaa327f8158..401bc2d24ce0 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -349,9 +349,10 @@ static bool valid_ref_ctr_vma(struct uprobe *uprobe,
static struct vm_area_struct *
find_ref_ctr_vma(struct uprobe *uprobe, struct mm_struct *mm)
{
+ VMA_ITERATOR(vmi, mm, 0);
struct vm_area_struct *tmp;

- for (tmp = mm->mmap; tmp; tmp = tmp->vm_next)
+ for_each_vma(vmi, tmp)
if (valid_ref_ctr_vma(uprobe, tmp))
return tmp;

@@ -1231,11 +1232,12 @@ int uprobe_apply(struct inode *inode, loff_t offset,

static int unapply_uprobe(struct uprobe *uprobe, struct mm_struct *mm)
{
+ VMA_ITERATOR(vmi, mm, 0);
struct vm_area_struct *vma;
int err = 0;

mmap_read_lock(mm);
- for (vma = mm->mmap; vma; vma = vma->vm_next) {
+ for_each_vma(vmi, vma) {
unsigned long vaddr;
loff_t offset;

@@ -1983,9 +1985,10 @@ bool uprobe_deny_signal(void)

static void mmf_recalc_uprobes(struct mm_struct *mm)
{
+ VMA_ITERATOR(vmi, mm, 0);
struct vm_area_struct *vma;

- for (vma = mm->mmap; vma; vma = vma->vm_next) {
+ for_each_vma(vmi, vma) {
if (!valid_vma(vma, false))
continue;
/*
--
2.35.1

2022-08-23 19:34:19

by Davidlohr Bueso

Subject: Re: [PATCH v13 47/70] perf: use VMA iterator

On Mon, 22 Aug 2022, Liam Howlett wrote:

>From: "Matthew Wilcox (Oracle)" <[email protected]>
>
>The VMA iterator is faster than the linked list and removing the linked
>list will shrink the vm_area_struct.
>
>Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
>Signed-off-by: Liam R. Howlett <[email protected]>
>Acked-by: Vlastimil Babka <[email protected]>

Reviewed-by: Davidlohr Bueso <[email protected]>