[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20250410000022.1901-2-sj@kernel.org>
Date: Wed, 9 Apr 2025 17:00:19 -0700
From: SeongJae Park <sj@...nel.org>
To: Andrew Morton <akpm@...ux-foundation.org>
Cc: SeongJae Park <sj@...nel.org>,
"Liam R.Howlett" <howlett@...il.com>,
David Hildenbrand <david@...hat.com>,
Lorenzo Stoakes <lorenzo.stoakes@...cle.com>,
Rik van Riel <riel@...riel.com>,
Shakeel Butt <shakeel.butt@...ux.dev>,
Vlastimil Babka <vbabka@...e.cz>,
kernel-team@...a.com,
linux-kernel@...r.kernel.org,
linux-mm@...ck.org
Subject: [PATCH v3 1/4] mm/madvise: define and use madvise_behavior struct for madvise_do_behavior()
To implement batched tlb flushes for MADV_DONTNEED[_LOCKED] and
MADV_FREE, an mmu_gather object in addition to the behavior integer need
to be passed to the internal logics. Using a struct can make it easy
without increasing the number of parameters of all code paths towards
the internal logic. Define a struct for the purpose and use it on the
code path that starts from madvise_do_behavior() and ends on
madvise_dontneed_free(). Note that this changes madvise_walk_vmas()
visitor type signature, too. Specifically, it changes its 'arg' type
from 'unsigned long' to the new struct pointer.
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
Signed-off-by: SeongJae Park <sj@...nel.org>
---
mm/madvise.c | 37 +++++++++++++++++++++++++------------
1 file changed, 25 insertions(+), 12 deletions(-)
diff --git a/mm/madvise.c b/mm/madvise.c
index b17f684322ad..26fa868b41af 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -48,6 +48,11 @@ struct madvise_walk_private {
bool pageout;
};
+struct madvise_behavior {
+ int behavior;
+ struct mmu_gather *tlb;
+};
+
/*
* Any behaviour which results in changes to the vma->vm_flags needs to
* take mmap_lock for writing. Others, which simply traverse vmas, need
@@ -893,8 +898,9 @@ static bool madvise_dontneed_free_valid_vma(struct vm_area_struct *vma,
static long madvise_dontneed_free(struct vm_area_struct *vma,
struct vm_area_struct **prev,
unsigned long start, unsigned long end,
- int behavior)
+ struct madvise_behavior *madv_behavior)
{
+ int behavior = madv_behavior->behavior;
struct mm_struct *mm = vma->vm_mm;
*prev = vma;
@@ -1249,8 +1255,10 @@ static long madvise_guard_remove(struct vm_area_struct *vma,
static int madvise_vma_behavior(struct vm_area_struct *vma,
struct vm_area_struct **prev,
unsigned long start, unsigned long end,
- unsigned long behavior)
+ void *behavior_arg)
{
+ struct madvise_behavior *arg = behavior_arg;
+ int behavior = arg->behavior;
int error;
struct anon_vma_name *anon_name;
unsigned long new_flags = vma->vm_flags;
@@ -1270,7 +1278,7 @@ static int madvise_vma_behavior(struct vm_area_struct *vma,
case MADV_FREE:
case MADV_DONTNEED:
case MADV_DONTNEED_LOCKED:
- return madvise_dontneed_free(vma, prev, start, end, behavior);
+ return madvise_dontneed_free(vma, prev, start, end, arg);
case MADV_NORMAL:
new_flags = new_flags & ~VM_RAND_READ & ~VM_SEQ_READ;
break;
@@ -1487,10 +1495,10 @@ static bool process_madvise_remote_valid(int behavior)
*/
static
int madvise_walk_vmas(struct mm_struct *mm, unsigned long start,
- unsigned long end, unsigned long arg,
+ unsigned long end, void *arg,
int (*visit)(struct vm_area_struct *vma,
struct vm_area_struct **prev, unsigned long start,
- unsigned long end, unsigned long arg))
+ unsigned long end, void *arg))
{
struct vm_area_struct *vma;
struct vm_area_struct *prev;
@@ -1548,7 +1556,7 @@ int madvise_walk_vmas(struct mm_struct *mm, unsigned long start,
static int madvise_vma_anon_name(struct vm_area_struct *vma,
struct vm_area_struct **prev,
unsigned long start, unsigned long end,
- unsigned long anon_name)
+ void *anon_name)
{
int error;
@@ -1557,7 +1565,7 @@ static int madvise_vma_anon_name(struct vm_area_struct *vma,
return -EBADF;
error = madvise_update_vma(vma, prev, start, end, vma->vm_flags,
- (struct anon_vma_name *)anon_name);
+ anon_name);
/*
* madvise() returns EAGAIN if kernel resources, such as
@@ -1589,7 +1597,7 @@ int madvise_set_anon_name(struct mm_struct *mm, unsigned long start,
if (end == start)
return 0;
- return madvise_walk_vmas(mm, start, end, (unsigned long)anon_name,
+ return madvise_walk_vmas(mm, start, end, anon_name,
madvise_vma_anon_name);
}
#endif /* CONFIG_ANON_VMA_NAME */
@@ -1677,8 +1685,10 @@ static bool is_madvise_populate(int behavior)
}
static int madvise_do_behavior(struct mm_struct *mm,
- unsigned long start, size_t len_in, int behavior)
+ unsigned long start, size_t len_in,
+ struct madvise_behavior *madv_behavior)
{
+ int behavior = madv_behavior->behavior;
struct blk_plug plug;
unsigned long end;
int error;
@@ -1692,7 +1702,7 @@ static int madvise_do_behavior(struct mm_struct *mm,
if (is_madvise_populate(behavior))
error = madvise_populate(mm, start, end, behavior);
else
- error = madvise_walk_vmas(mm, start, end, behavior,
+ error = madvise_walk_vmas(mm, start, end, madv_behavior,
madvise_vma_behavior);
blk_finish_plug(&plug);
return error;
@@ -1773,13 +1783,14 @@ static int madvise_do_behavior(struct mm_struct *mm,
int do_madvise(struct mm_struct *mm, unsigned long start, size_t len_in, int behavior)
{
int error;
+ struct madvise_behavior madv_behavior = {.behavior = behavior};
if (madvise_should_skip(start, len_in, behavior, &error))
return error;
error = madvise_lock(mm, behavior);
if (error)
return error;
- error = madvise_do_behavior(mm, start, len_in, behavior);
+ error = madvise_do_behavior(mm, start, len_in, &madv_behavior);
madvise_unlock(mm, behavior);
return error;
@@ -1796,6 +1807,7 @@ static ssize_t vector_madvise(struct mm_struct *mm, struct iov_iter *iter,
{
ssize_t ret = 0;
size_t total_len;
+ struct madvise_behavior madv_behavior = {.behavior = behavior};
total_len = iov_iter_count(iter);
@@ -1811,7 +1823,8 @@ static ssize_t vector_madvise(struct mm_struct *mm, struct iov_iter *iter,
if (madvise_should_skip(start, len_in, behavior, &error))
ret = error;
else
- ret = madvise_do_behavior(mm, start, len_in, behavior);
+ ret = madvise_do_behavior(mm, start, len_in,
+ &madv_behavior);
/*
* An madvise operation is attempting to restart the syscall,
* but we cannot proceed as it would not be correct to repeat
--
2.39.5
Powered by blists - more mailing lists