LinuxLists.cc - [PATCH v5 0/4] fs/ufs: Replace kmap() with kmap_local

2022-12-29 23:18:25

Subject: [PATCH v5 0/4] fs/ufs: Replace kmap() with kmap_local_page

2022-12-29 23:29:22

Subject: [PATCH v5 2/4] fs/ufs: Change the signature of ufs_get_page()

Change the signature of ufs_get_page() in order to prepare this function
to the conversion to the use of kmap_local_page(). Change also those call
sites which are required to conform its invocations to the new
signature.

Cc: Ira Weiny <[email protected]>
Suggested-by: Al Viro <[email protected]>
Reviewed-by: Ira Weiny <[email protected]>
Signed-off-by: Fabio M. De Francesco <[email protected]>
---
fs/ufs/dir.c | 49 +++++++++++++++++++++----------------------------
1 file changed, 21 insertions(+), 28 deletions(-)

diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c
index 69f78583c9c1..ae3b20354a28 100644
--- a/fs/ufs/dir.c
+++ b/fs/ufs/dir.c
@@ -185,7 +185,7 @@ static bool ufs_check_page(struct page *page)
return false;
}

-static struct page *ufs_get_page(struct inode *dir, unsigned long n)
+static void *ufs_get_page(struct inode *dir, unsigned long n, struct page **p)
{
struct address_space *mapping = dir->i_mapping;
struct page *page = read_mapping_page(mapping, n, NULL);
@@ -195,8 +195,10 @@ static struct page *ufs_get_page(struct inode *dir, unsigned long n)
if (!ufs_check_page(page))
goto fail;
}
+ *p = page;
+ return page_address(page);
}
- return page;
+ return ERR_CAST(page);

fail:
ufs_put_page(page);
@@ -227,15 +229,12 @@ ufs_next_entry(struct super_block *sb, struct ufs_dir_entry *p)

struct ufs_dir_entry *ufs_dotdot(struct inode *dir, struct page **p)
{
- struct page *page = ufs_get_page(dir, 0);
- struct ufs_dir_entry *de = NULL;
+ struct ufs_dir_entry *de = ufs_get_page(dir, 0, p);

- if (!IS_ERR(page)) {
- de = ufs_next_entry(dir->i_sb,
- (struct ufs_dir_entry *)page_address(page));
- *p = page;
- }
- return de;
+ if (!IS_ERR(de))
+ return ufs_next_entry(dir->i_sb, de);
+ else
+ return NULL;
}

/*
@@ -273,11 +272,10 @@ struct ufs_dir_entry *ufs_find_entry(struct inode *dir, const struct qstr *qstr,
start = 0;
n = start;
do {
- char *kaddr;
- page = ufs_get_page(dir, n);
- if (!IS_ERR(page)) {
- kaddr = page_address(page);
- de = (struct ufs_dir_entry *) kaddr;
+ char *kaddr = ufs_get_page(dir, n, &page);
+
+ if (!IS_ERR(kaddr)) {
+ de = (struct ufs_dir_entry *)kaddr;
kaddr += ufs_last_byte(dir, n) - reclen;
while ((char *) de <= kaddr) {
if (ufs_match(sb, namelen, name, de))
@@ -328,12 +326,10 @@ int ufs_add_link(struct dentry *dentry, struct inode *inode)
for (n = 0; n <= npages; n++) {
char *dir_end;

- page = ufs_get_page(dir, n);
- err = PTR_ERR(page);
- if (IS_ERR(page))
- goto out;
+ kaddr = ufs_get_page(dir, n, &page);
+ if (IS_ERR(kaddr))
+ return PTR_ERR(kaddr);
lock_page(page);
- kaddr = page_address(page);
dir_end = kaddr + ufs_last_byte(dir, n);
de = (struct ufs_dir_entry *)kaddr;
kaddr += PAGE_SIZE - reclen;
@@ -395,7 +391,6 @@ int ufs_add_link(struct dentry *dentry, struct inode *inode)
/* OFFSET_CACHE */
out_put:
ufs_put_page(page);
-out:
return err;
out_unlock:
unlock_page(page);
@@ -438,17 +433,16 @@ ufs_readdir(struct file *file, struct dir_context *ctx)
for ( ; n < npages; n++, offset = 0) {
char *kaddr, *limit;
struct ufs_dir_entry *de;
+ struct page *page;

- struct page *page = ufs_get_page(inode, n);
-
- if (IS_ERR(page)) {
+ kaddr = ufs_get_page(inode, n, &page);
+ if (IS_ERR(kaddr)) {
ufs_error(sb, __func__,
"bad page in #%lu",
inode->i_ino);
ctx->pos += PAGE_SIZE - offset;
return -EIO;
}
- kaddr = page_address(page);
if (unlikely(need_revalidate)) {
if (offset) {
offset = ufs_validate_entry(sb, kaddr, offset, chunk_mask);
@@ -595,12 +589,11 @@ int ufs_empty_dir(struct inode * inode)
for (i = 0; i < npages; i++) {
char *kaddr;
struct ufs_dir_entry *de;
- page = ufs_get_page(inode, i);

- if (IS_ERR(page))
+ kaddr = ufs_get_page(inode, i, &page);
+ if (IS_ERR(kaddr))
continue;

- kaddr = page_address(page);
de = (struct ufs_dir_entry *)kaddr;
kaddr += ufs_last_byte(inode, i) - UFS_DIR_REC_LEN(1);

--
2.39.0

2022-12-29 23:35:51

by Fabio M. De Francesco

[permalink] [raw]

Subject: [PATCH v5 4/4] fs/ufs: Replace kmap() with kmap_local_page()

kmap() is being deprecated in favor of kmap_local_page().

There are two main problems with kmap(): (1) It comes with an overhead as
the mapping space is restricted and protected by a global lock for
synchronization and (2) it also requires global TLB invalidation when the
kmap’s pool wraps and it might block when the mapping space is fully
utilized until a slot becomes available.

With kmap_local_page() the mappings are per thread, CPU local, can take
page faults, and can be called from any context (including interrupts).
It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
the tasks can be preempted and, when they are scheduled to run again, the
kernel virtual addresses are restored and still valid.

The use of kmap_local_page() in fs/ufs is "safe" because (1) the kernel
virtual addresses are exclusively re-used by the thread which
established the mappings (i.e., thread locality is never violated) and (2)
the nestings of mappings and un-mappings are always stack based (LIFO).

Therefore, replace kmap() with kmap_local_page() in fs/ufs. kunmap_local()
requires the mapping address, so return that address from ufs_get_page()
and use it as parameter for the second argument of ufs_put_page().

Suggested-by: Al Viro <[email protected]>
Suggested-by: Ira Weiny <[email protected]>
Signed-off-by: Fabio M. De Francesco <[email protected]>
---
fs/ufs/dir.c | 72 +++++++++++++++++++++++++++++++++-----------------
fs/ufs/namei.c | 8 +++---
fs/ufs/ufs.h | 2 +-
3 files changed, 53 insertions(+), 29 deletions(-)

diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c
index 0bfd563ab0c2..8676a144e589 100644
--- a/fs/ufs/dir.c
+++ b/fs/ufs/dir.c
@@ -61,9 +61,9 @@ static int ufs_commit_chunk(struct page *page, loff_t pos, unsigned len)
return err;
}

-inline void ufs_put_page(struct page *page)
+inline void ufs_put_page(struct page *page, void *page_addr)
{
- kunmap(page);
+ kunmap_local(page_addr);
put_page(page);
}

@@ -76,7 +76,7 @@ ino_t ufs_inode_by_name(struct inode *dir, const struct qstr *qstr)
de = ufs_find_entry(dir, qstr, &page);
if (de) {
res = fs32_to_cpu(dir->i_sb, de->d_ino);
- ufs_put_page(page);
+ ufs_put_page(page, de);
}
return res;
}
@@ -99,18 +99,17 @@ void ufs_set_link(struct inode *dir, struct ufs_dir_entry *de,
ufs_set_de_type(dir->i_sb, de, inode->i_mode);

err = ufs_commit_chunk(page, pos, len);
- ufs_put_page(page);
+ ufs_put_page(page, de);
if (update_times)
dir->i_mtime = dir->i_ctime = current_time(dir);
mark_inode_dirty(dir);
}

-static bool ufs_check_page(struct page *page)
+static bool ufs_check_page(struct page *page, char *kaddr)
{
struct inode *dir = page->mapping->host;
struct super_block *sb = dir->i_sb;
- char *kaddr = page_address(page);
unsigned offs, rec_len;
unsigned limit = PAGE_SIZE;
const unsigned chunk_mask = UFS_SB(sb)->s_uspi->s_dirblksize - 1;
@@ -185,23 +184,32 @@ static bool ufs_check_page(struct page *page)
return false;
}

+/*
+ * Calls to ufs_get_page()/ufs_put_page() must be nested according to the
+ * rules documented in kmap_local_page()/kunmap_local().
+ *
+ * NOTE: ufs_find_entry() and ufs_dotdot() act as calls to ufs_get_page()
+ * and must be treated accordingly for nesting purposes.
+ */
static void *ufs_get_page(struct inode *dir, unsigned long n, struct page **p)
{
+ char *kaddr;
+
struct address_space *mapping = dir->i_mapping;
struct page *page = read_mapping_page(mapping, n, NULL);
if (!IS_ERR(page)) {
- kmap(page);
+ kaddr = kmap_local_page(page);
if (unlikely(!PageChecked(page))) {
- if (!ufs_check_page(page))
+ if (!ufs_check_page(page, kaddr))
goto fail;
}
*p = page;
- return page_address(page);
+ return kaddr;
}
return ERR_CAST(page);

fail:
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
return ERR_PTR(-EIO);
}

@@ -227,6 +235,13 @@ ufs_next_entry(struct super_block *sb, struct ufs_dir_entry *p)
fs16_to_cpu(sb, p->d_reclen));
}

+/*
+ * Calls to ufs_get_page()/ufs_put_page() must be nested according to the
+ * rules documented in kmap_local_page()/kunmap_local().
+ *
+ * ufs_dotdot() acts as a call to ufs_get_page() and must be treated
+ * accordingly for nesting purposes.
+ */
struct ufs_dir_entry *ufs_dotdot(struct inode *dir, struct page **p)
{
struct ufs_dir_entry *de = ufs_get_page(dir, 0, p);
@@ -244,6 +259,11 @@ struct ufs_dir_entry *ufs_dotdot(struct inode *dir, struct page **p)
* returns the page in which the entry was found, and the entry itself
* (as a parameter - res_dir). Page is returned mapped and unlocked.
* Entry is guaranteed to be valid.
+ *
+ * On Success ufs_put_page() should be called on *res_page.
+ *
+ * ufs_find_entry() acts as a call to ufs_get_page() and must be treated
+ * accordingly for nesting purposes.
*/
struct ufs_dir_entry *ufs_find_entry(struct inode *dir, const struct qstr *qstr,
struct page **res_page)
@@ -282,7 +302,7 @@ struct ufs_dir_entry *ufs_find_entry(struct inode *dir, const struct qstr *qstr,
goto found;
de = ufs_next_entry(sb, de);
}
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
}
if (++n >= npages)
n = 0;
@@ -360,7 +380,7 @@ int ufs_add_link(struct dentry *dentry, struct inode *inode)
de = (struct ufs_dir_entry *) ((char *) de + rec_len);
}
unlock_page(page);
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
}
BUG();
return -EINVAL;
@@ -390,7 +410,7 @@ int ufs_add_link(struct dentry *dentry, struct inode *inode)
mark_inode_dirty(dir);
/* OFFSET_CACHE */
out_put:
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
return err;
out_unlock:
unlock_page(page);
@@ -468,13 +488,13 @@ ufs_readdir(struct file *file, struct dir_context *ctx)
ufs_get_de_namlen(sb, de),
fs32_to_cpu(sb, de->d_ino),
d_type)) {
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
return 0;
}
}
ctx->pos += fs16_to_cpu(sb, de->d_reclen);
}
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
}
return 0;
}
@@ -485,10 +505,15 @@ ufs_readdir(struct file *file, struct dir_context *ctx)
* previous entry.
*/
int ufs_delete_entry(struct inode *inode, struct ufs_dir_entry *dir,
- struct page * page)
+ struct page *page)
{
struct super_block *sb = inode->i_sb;
- char *kaddr = page_address(page);
+ /*
+ * The "dir" dentry points somewhere in the same page whose we need the
+ * address of; therefore, we can simply get the base address "kaddr" by
+ * masking the previous with PAGE_MASK.
+ */
+ char *kaddr = (char *)((unsigned long)dir & PAGE_MASK);
unsigned int from = offset_in_page(dir) & ~(UFS_SB(sb)->s_uspi->s_dirblksize - 1);
unsigned int to = offset_in_page(dir) + fs16_to_cpu(sb, dir->d_reclen);
loff_t pos;
@@ -527,7 +552,7 @@ int ufs_delete_entry(struct inode *inode, struct ufs_dir_entry *dir,
inode->i_ctime = inode->i_mtime = current_time(inode);
mark_inode_dirty(inode);
out:
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
UFSD("EXIT\n");
return err;
}
@@ -551,8 +576,7 @@ int ufs_make_empty(struct inode * inode, struct inode *dir)
goto fail;
}

- kmap(page);
- base = (char*)page_address(page);
+ base = kmap_local_page(page);
memset(base, 0, PAGE_SIZE);

de = (struct ufs_dir_entry *) base;
@@ -569,7 +593,7 @@ int ufs_make_empty(struct inode * inode, struct inode *dir)
de->d_reclen = cpu_to_fs16(sb, chunk_size - UFS_DIR_REC_LEN(1));
ufs_set_de_namlen(sb, de, 2);
strcpy (de->d_name, "..");
- kunmap(page);
+ kunmap_local(base);

err = ufs_commit_chunk(page, 0, chunk_size);
fail:
@@ -585,9 +609,9 @@ int ufs_empty_dir(struct inode * inode)
struct super_block *sb = inode->i_sb;
struct page *page = NULL;
unsigned long i, npages = dir_pages(inode);
+ char *kaddr;

for (i = 0; i < npages; i++) {
- char *kaddr;
struct ufs_dir_entry *de;

kaddr = ufs_get_page(inode, i, &page);
@@ -620,12 +644,12 @@ int ufs_empty_dir(struct inode * inode)
}
de = ufs_next_entry(sb, de);
}
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
}
return 1;

not_empty:
- ufs_put_page(page);
+ ufs_put_page(page, kaddr);
return 0;
}

diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
index 486b0f2e8b7a..7175d45e704c 100644
--- a/fs/ufs/namei.c
+++ b/fs/ufs/namei.c
@@ -250,7 +250,7 @@ static int ufs_rename(struct user_namespace *mnt_userns, struct inode *old_dir,
struct inode *old_inode = d_inode(old_dentry);
struct inode *new_inode = d_inode(new_dentry);
struct page *dir_page = NULL;
- struct ufs_dir_entry * dir_de = NULL;
+ struct ufs_dir_entry *dir_de = NULL;
struct page *old_page;
struct ufs_dir_entry *old_de;
int err = -ENOENT;
@@ -307,7 +307,7 @@ static int ufs_rename(struct user_namespace *mnt_userns, struct inode *old_dir,
if (old_dir != new_dir)
ufs_set_link(old_inode, dir_de, dir_page, new_dir, 0);
else {
- ufs_put_page(dir_page);
+ ufs_put_page(dir_page, dir_de);
}
inode_dec_link_count(old_dir);
}
@@ -316,10 +316,10 @@ static int ufs_rename(struct user_namespace *mnt_userns, struct inode *old_dir,

out_dir:
if (dir_de) {
- ufs_put_page(dir_page);
+ ufs_put_page(dir_page, dir_de);
}
out_old:
- ufs_put_page(old_page);
+ ufs_put_page(old_page, old_de);
out:
return err;
}
diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h
index f7ba8df25d03..942639e9a817 100644
--- a/fs/ufs/ufs.h
+++ b/fs/ufs/ufs.h
@@ -98,7 +98,7 @@ extern struct ufs_cg_private_info * ufs_load_cylinder (struct super_block *, uns
extern void ufs_put_cylinder (struct super_block *, unsigned);

/* dir.c */
-extern void ufs_put_page(struct page *page);
+extern void ufs_put_page(struct page *page, void *vaddr);
extern const struct inode_operations ufs_dir_inode_operations;
extern int ufs_add_link (struct dentry *, struct inode *);
extern ino_t ufs_inode_by_name(struct inode *, const struct qstr *);
--
2.39.0

2022-12-30 01:10:28

by Ira Weiny

[permalink] [raw]

Subject: Re: [PATCH v5 4/4] fs/ufs: Replace kmap() with kmap_local_page()

On Thu, Dec 29, 2022 at 11:51:00PM +0100, Fabio M. De Francesco wrote:
> kmap() is being deprecated in favor of kmap_local_page().
>
> There are two main problems with kmap(): (1) It comes with an overhead as
> the mapping space is restricted and protected by a global lock for
> synchronization and (2) it also requires global TLB invalidation when the
> kmap’s pool wraps and it might block when the mapping space is fully
> utilized until a slot becomes available.
>
> With kmap_local_page() the mappings are per thread, CPU local, can take
> page faults, and can be called from any context (including interrupts).
> It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
> the tasks can be preempted and, when they are scheduled to run again, the
> kernel virtual addresses are restored and still valid.
>
> The use of kmap_local_page() in fs/ufs is "safe" because (1) the kernel
> virtual addresses are exclusively re-used by the thread which
> established the mappings (i.e., thread locality is never violated) and (2)
> the nestings of mappings and un-mappings are always stack based (LIFO).
>
> Therefore, replace kmap() with kmap_local_page() in fs/ufs. kunmap_local()
> requires the mapping address, so return that address from ufs_get_page()
> and use it as parameter for the second argument of ufs_put_page().
>
> Suggested-by: Al Viro <[email protected]>
> Suggested-by: Ira Weiny <[email protected]>

Reviewed-by: Ira Weiny <[email protected]>

> Signed-off-by: Fabio M. De Francesco <[email protected]>
> ---
> fs/ufs/dir.c | 72 +++++++++++++++++++++++++++++++++-----------------
> fs/ufs/namei.c | 8 +++---
> fs/ufs/ufs.h | 2 +-
> 3 files changed, 53 insertions(+), 29 deletions(-)
>
> diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c
> index 0bfd563ab0c2..8676a144e589 100644
> --- a/fs/ufs/dir.c
> +++ b/fs/ufs/dir.c
> @@ -61,9 +61,9 @@ static int ufs_commit_chunk(struct page *page, loff_t pos, unsigned len)
> return err;
> }
>
> -inline void ufs_put_page(struct page *page)
> +inline void ufs_put_page(struct page *page, void *page_addr)
> {
> - kunmap(page);
> + kunmap_local(page_addr);
> put_page(page);
> }
>
> @@ -76,7 +76,7 @@ ino_t ufs_inode_by_name(struct inode *dir, const struct qstr *qstr)
> de = ufs_find_entry(dir, qstr, &page);
> if (de) {
> res = fs32_to_cpu(dir->i_sb, de->d_ino);
> - ufs_put_page(page);
> + ufs_put_page(page, de);
> }
> return res;
> }
> @@ -99,18 +99,17 @@ void ufs_set_link(struct inode *dir, struct ufs_dir_entry *de,
> ufs_set_de_type(dir->i_sb, de, inode->i_mode);
>
> err = ufs_commit_chunk(page, pos, len);
> - ufs_put_page(page);
> + ufs_put_page(page, de);
> if (update_times)
> dir->i_mtime = dir->i_ctime = current_time(dir);
> mark_inode_dirty(dir);
> }
>
>
> -static bool ufs_check_page(struct page *page)
> +static bool ufs_check_page(struct page *page, char *kaddr)
> {
> struct inode *dir = page->mapping->host;
> struct super_block *sb = dir->i_sb;
> - char *kaddr = page_address(page);
> unsigned offs, rec_len;
> unsigned limit = PAGE_SIZE;
> const unsigned chunk_mask = UFS_SB(sb)->s_uspi->s_dirblksize - 1;
> @@ -185,23 +184,32 @@ static bool ufs_check_page(struct page *page)
> return false;
> }
>
> +/*
> + * Calls to ufs_get_page()/ufs_put_page() must be nested according to the
> + * rules documented in kmap_local_page()/kunmap_local().
> + *
> + * NOTE: ufs_find_entry() and ufs_dotdot() act as calls to ufs_get_page()
> + * and must be treated accordingly for nesting purposes.
> + */
> static void *ufs_get_page(struct inode *dir, unsigned long n, struct page **p)
> {
> + char *kaddr;
> +
> struct address_space *mapping = dir->i_mapping;
> struct page *page = read_mapping_page(mapping, n, NULL);
> if (!IS_ERR(page)) {
> - kmap(page);
> + kaddr = kmap_local_page(page);
> if (unlikely(!PageChecked(page))) {
> - if (!ufs_check_page(page))
> + if (!ufs_check_page(page, kaddr))
> goto fail;
> }
> *p = page;
> - return page_address(page);
> + return kaddr;
> }
> return ERR_CAST(page);
>
> fail:
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> return ERR_PTR(-EIO);
> }
>
> @@ -227,6 +235,13 @@ ufs_next_entry(struct super_block *sb, struct ufs_dir_entry *p)
> fs16_to_cpu(sb, p->d_reclen));
> }
>
> +/*
> + * Calls to ufs_get_page()/ufs_put_page() must be nested according to the
> + * rules documented in kmap_local_page()/kunmap_local().
> + *
> + * ufs_dotdot() acts as a call to ufs_get_page() and must be treated
> + * accordingly for nesting purposes.
> + */
> struct ufs_dir_entry *ufs_dotdot(struct inode *dir, struct page **p)
> {
> struct ufs_dir_entry *de = ufs_get_page(dir, 0, p);
> @@ -244,6 +259,11 @@ struct ufs_dir_entry *ufs_dotdot(struct inode *dir, struct page **p)
> * returns the page in which the entry was found, and the entry itself
> * (as a parameter - res_dir). Page is returned mapped and unlocked.
> * Entry is guaranteed to be valid.
> + *
> + * On Success ufs_put_page() should be called on *res_page.
> + *
> + * ufs_find_entry() acts as a call to ufs_get_page() and must be treated
> + * accordingly for nesting purposes.
> */
> struct ufs_dir_entry *ufs_find_entry(struct inode *dir, const struct qstr *qstr,
> struct page **res_page)
> @@ -282,7 +302,7 @@ struct ufs_dir_entry *ufs_find_entry(struct inode *dir, const struct qstr *qstr,
> goto found;
> de = ufs_next_entry(sb, de);
> }
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> }
> if (++n >= npages)
> n = 0;
> @@ -360,7 +380,7 @@ int ufs_add_link(struct dentry *dentry, struct inode *inode)
> de = (struct ufs_dir_entry *) ((char *) de + rec_len);
> }
> unlock_page(page);
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> }
> BUG();
> return -EINVAL;
> @@ -390,7 +410,7 @@ int ufs_add_link(struct dentry *dentry, struct inode *inode)
> mark_inode_dirty(dir);
> /* OFFSET_CACHE */
> out_put:
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> return err;
> out_unlock:
> unlock_page(page);
> @@ -468,13 +488,13 @@ ufs_readdir(struct file *file, struct dir_context *ctx)
> ufs_get_de_namlen(sb, de),
> fs32_to_cpu(sb, de->d_ino),
> d_type)) {
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> return 0;
> }
> }
> ctx->pos += fs16_to_cpu(sb, de->d_reclen);
> }
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> }
> return 0;
> }
> @@ -485,10 +505,15 @@ ufs_readdir(struct file *file, struct dir_context *ctx)
> * previous entry.
> */
> int ufs_delete_entry(struct inode *inode, struct ufs_dir_entry *dir,
> - struct page * page)
> + struct page *page)
> {
> struct super_block *sb = inode->i_sb;
> - char *kaddr = page_address(page);
> + /*
> + * The "dir" dentry points somewhere in the same page whose we need the
> + * address of; therefore, we can simply get the base address "kaddr" by
> + * masking the previous with PAGE_MASK.
> + */
> + char *kaddr = (char *)((unsigned long)dir & PAGE_MASK);
> unsigned int from = offset_in_page(dir) & ~(UFS_SB(sb)->s_uspi->s_dirblksize - 1);
> unsigned int to = offset_in_page(dir) + fs16_to_cpu(sb, dir->d_reclen);
> loff_t pos;
> @@ -527,7 +552,7 @@ int ufs_delete_entry(struct inode *inode, struct ufs_dir_entry *dir,
> inode->i_ctime = inode->i_mtime = current_time(inode);
> mark_inode_dirty(inode);
> out:
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> UFSD("EXIT\n");
> return err;
> }
> @@ -551,8 +576,7 @@ int ufs_make_empty(struct inode * inode, struct inode *dir)
> goto fail;
> }
>
> - kmap(page);
> - base = (char*)page_address(page);
> + base = kmap_local_page(page);
> memset(base, 0, PAGE_SIZE);
>
> de = (struct ufs_dir_entry *) base;
> @@ -569,7 +593,7 @@ int ufs_make_empty(struct inode * inode, struct inode *dir)
> de->d_reclen = cpu_to_fs16(sb, chunk_size - UFS_DIR_REC_LEN(1));
> ufs_set_de_namlen(sb, de, 2);
> strcpy (de->d_name, "..");
> - kunmap(page);
> + kunmap_local(base);
>
> err = ufs_commit_chunk(page, 0, chunk_size);
> fail:
> @@ -585,9 +609,9 @@ int ufs_empty_dir(struct inode * inode)
> struct super_block *sb = inode->i_sb;
> struct page *page = NULL;
> unsigned long i, npages = dir_pages(inode);
> + char *kaddr;
>
> for (i = 0; i < npages; i++) {
> - char *kaddr;
> struct ufs_dir_entry *de;
>
> kaddr = ufs_get_page(inode, i, &page);
> @@ -620,12 +644,12 @@ int ufs_empty_dir(struct inode * inode)
> }
> de = ufs_next_entry(sb, de);
> }
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> }
> return 1;
>
> not_empty:
> - ufs_put_page(page);
> + ufs_put_page(page, kaddr);
> return 0;
> }
>
> diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
> index 486b0f2e8b7a..7175d45e704c 100644
> --- a/fs/ufs/namei.c
> +++ b/fs/ufs/namei.c
> @@ -250,7 +250,7 @@ static int ufs_rename(struct user_namespace *mnt_userns, struct inode *old_dir,
> struct inode *old_inode = d_inode(old_dentry);
> struct inode *new_inode = d_inode(new_dentry);
> struct page *dir_page = NULL;
> - struct ufs_dir_entry * dir_de = NULL;
> + struct ufs_dir_entry *dir_de = NULL;
> struct page *old_page;
> struct ufs_dir_entry *old_de;
> int err = -ENOENT;
> @@ -307,7 +307,7 @@ static int ufs_rename(struct user_namespace *mnt_userns, struct inode *old_dir,
> if (old_dir != new_dir)
> ufs_set_link(old_inode, dir_de, dir_page, new_dir, 0);
> else {
> - ufs_put_page(dir_page);
> + ufs_put_page(dir_page, dir_de);
> }
> inode_dec_link_count(old_dir);
> }
> @@ -316,10 +316,10 @@ static int ufs_rename(struct user_namespace *mnt_userns, struct inode *old_dir,
>
> out_dir:
> if (dir_de) {
> - ufs_put_page(dir_page);
> + ufs_put_page(dir_page, dir_de);
> }
> out_old:
> - ufs_put_page(old_page);
> + ufs_put_page(old_page, old_de);
> out:
> return err;
> }
> diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h
> index f7ba8df25d03..942639e9a817 100644
> --- a/fs/ufs/ufs.h
> +++ b/fs/ufs/ufs.h
> @@ -98,7 +98,7 @@ extern struct ufs_cg_private_info * ufs_load_cylinder (struct super_block *, uns
> extern void ufs_put_cylinder (struct super_block *, unsigned);
>
> /* dir.c */
> -extern void ufs_put_page(struct page *page);
> +extern void ufs_put_page(struct page *page, void *vaddr);
> extern const struct inode_operations ufs_dir_inode_operations;
> extern int ufs_add_link (struct dentry *, struct inode *);
> extern ino_t ufs_inode_by_name(struct inode *, const struct qstr *);
> --
> 2.39.0
>

2023-03-27 10:24:07

by Fabio M. De Francesco

[permalink] [raw]

Subject: Re: [PATCH v5 0/4] fs/ufs: Replace kmap() with kmap_local_page

On giovedì 29 dicembre 2022 23:50:56 CEST Fabio M. De Francesco wrote:
> kmap() is being deprecated in favor of kmap_local_page().
>
> There are two main problems with kmap(): (1) It comes with an overhead as
> the mapping space is restricted and protected by a global lock for
> synchronization and (2) it also requires global TLB invalidation when the
> kmap’s pool wraps and it might block when the mapping space is fully
> utilized until a slot becomes available.
>
> With kmap_local_page() the mappings are per thread, CPU local, can take
> page faults, and can be called from any context (including interrupts).
> It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
> the tasks can be preempted and, when they are scheduled to run again, the
> kernel virtual addresses are restored and still valid.
>
> Since its use in fs/ufs is safe everywhere, it should be preferred.
>
> Therefore, replace kmap() with kmap_local_page() in fs/ufs. kunmap_local()
> requires the mapping address, so return that address from ufs_get_page()
> to be used in ufs_put_page().

Hi Al,

I see that this series is here since Dec 29, 2022.
Is there anything that prevents its merging?
Can you please its four patches in your tree?

Thanks,

Fabio

>
> This series could have not been ever made because nothing prevented the
> previous patch from working properly but Al Viro made a long series of
> very appreciated comments about how many unnecessary and redundant lines
> of code I could have removed. He could see things I was entirely unable
> to notice. Furthermore, he also provided solutions and details about how
> I could decompose a single patch into a small series of three
> independent units.[1][2][3]
>
> I want to thank him so much for the patience, kindness and the time he
> decided to spend to provide those analysis and write three messages full
> of interesting insights.[1][2][3]
>
> Cc: Al Viro <[email protected]>
> Cc: Ira Weiny <[email protected]>
> Signed-off-by: Fabio M. De Francesco <[email protected]>
>
> Changes from v1:
> 1/3: No changes.
> 2/3: Restore the return of "err" that was mistakenly deleted
> together with the removal of the "out" label in
> ufs_add_link(). Thanks to Al Viro.[4]
> Return the address of the kmap()'ed page instead of a
> pointer to a pointer to the mapped page; a page_address()
> had been overlooked in ufs_get_page(). Thanks to Al
> Viro.[5]
> 3/3: Return the kernel virtual address got from the call to
> kmap_local_page() after conversion from kmap(). Again
> thanks to Al Viro.[6]
>
> Changes from v2:
> 1/3: No changes.
> 2/3: Rework ufs_get_page() because the previous version had two
> errors: (1) It could return an invalid pages with the out
> argument "page" and (2) it could return "page_address(page)"
> also in cases where read_mapping_page() returned an error
> and the page is never kmap()'ed. Thanks to Al Viro.[7]
> 3/3: Rework ufs_get_page() after conversion to
> kmap_local_page(), in accordance to the last changes in 2/3.
>
> Changes from v3:
> 1/3: No changes.
> 2/3: No changes.
> 3/3: Replace kunmap() with kunmap_local().
>
> Changes from v4:
> 1/4: It was 1/3.
> 2/4: Move the declaration of a page into an inner loop. Add Ira
> Weiny's "Reviewed-by" tag (thanks!).
> 3/4: Add this patch to use ufs_put_page() to replace three kunmap()
> and put_page() in namei.c. Thanks to Ira Weiny who noticed that
> I had overlooked their presence.
> 4/4: Remove an unnecessary masking that is already carried out by
> kunmap_local() via kunmap_local_indexed(). Add a comment to
> clarify that a ufs_dir_entry passed to ufs_delete_entry()
> points in the same page we need the address of. Suggested by
> Ira Weiny.
>
> [1] https://lore.kernel.org/lkml/Y4E++JERgUMoqfjG@ZenIV/#t
> [2] https://lore.kernel.org/lkml/Y4FG0O7VWTTng5yh@ZenIV/#t
> [3] https://lore.kernel.org/lkml/Y4ONIFJatIGsVNpf@ZenIV/#t
> [4] https://lore.kernel.org/lkml/Y5Zc0qZ3+zsI74OZ@ZenIV/#t
> [5] https://lore.kernel.org/lkml/Y5ZZy23FFAnQDR3C@ZenIV/#t
> [6] https://lore.kernel.org/lkml/Y5ZcMPzPG9h6C9eh@ZenIV/#t
> [7] https://lore.kernel.org/lkml/Y5glgpD7fFifC4Fi@ZenIV/#t
>
> The cover letter of the v1 series is at
> https://lore.kernel.org/lkml/[email protected]/
> The cover letter of the v2 series is at
> https://lore.kernel.org/lkml/[email protected]/
> The cover letter of the v3 series is at
> https://lore.kernel.org/lkml/[email protected]/
> The cover letter of the v4 series is at
> https://lore.kernel.org/lkml/[email protected]/
>
> Fabio M. De Francesco (4):
> fs/ufs: Use the offset_in_page() helper
> fs/ufs: Change the signature of ufs_get_page()
> fs/ufs: Use ufs_put_page() in ufs_rename()
> fs/ufs: Replace kmap() with kmap_local_page()
>
> fs/ufs/dir.c | 131 +++++++++++++++++++++++++++----------------------
> fs/ufs/namei.c | 11 ++---
> fs/ufs/ufs.h | 1 +
> 3 files changed, 78 insertions(+), 65 deletions(-)
>
> --
> 2.39.0

2023-03-27 13:27:43

by Matthew Wilcox

[permalink] [raw]

Subject: Re: [PATCH v5 0/4] fs/ufs: Replace kmap() with kmap_local_page

On Mon, Mar 27, 2023 at 12:13:08PM +0200, Fabio M. De Francesco wrote:
> On giovedì 29 dicembre 2022 23:50:56 CEST Fabio M. De Francesco wrote:
> > kmap() is being deprecated in favor of kmap_local_page().
> >
> > There are two main problems with kmap(): (1) It comes with an overhead as
> > the mapping space is restricted and protected by a global lock for
> > synchronization and (2) it also requires global TLB invalidation when the
> > kmap’s pool wraps and it might block when the mapping space is fully
> > utilized until a slot becomes available.
> >
> > With kmap_local_page() the mappings are per thread, CPU local, can take
> > page faults, and can be called from any context (including interrupts).
> > It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
> > the tasks can be preempted and, when they are scheduled to run again, the
> > kernel virtual addresses are restored and still valid.
> >
> > Since its use in fs/ufs is safe everywhere, it should be preferred.
> >
> > Therefore, replace kmap() with kmap_local_page() in fs/ufs. kunmap_local()
> > requires the mapping address, so return that address from ufs_get_page()
> > to be used in ufs_put_page().
>
> Hi Al,
>
> I see that this series is here since Dec 29, 2022.
> Is there anything that prevents its merging?
> Can you please its four patches in your tree?

I'm pretty sure UFS directories should simply be allocated from lowmem.
There's really no reason to put them in highmem these days.

2023-04-03 05:33:31

by Fabio M. De Francesco

[permalink] [raw]

Subject: Re: [PATCH v5 0/4] fs/ufs: Replace kmap() with kmap_local_page

On lunedì 27 marzo 2023 15:23:41 CEST Matthew Wilcox wrote:
> On Mon, Mar 27, 2023 at 12:13:08PM +0200, Fabio M. De Francesco wrote:
> > On giovedì 29 dicembre 2022 23:50:56 CEST Fabio M. De Francesco wrote:
> > > kmap() is being deprecated in favor of kmap_local_page().
> > >
> > > There are two main problems with kmap(): (1) It comes with an overhead
as
> > > the mapping space is restricted and protected by a global lock for
> > > synchronization and (2) it also requires global TLB invalidation when
the
> > > kmap’s pool wraps and it might block when the mapping space is fully
> > > utilized until a slot becomes available.
> > >
> > > With kmap_local_page() the mappings are per thread, CPU local, can take
> > > page faults, and can be called from any context (including interrupts).
> > > It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
> > > the tasks can be preempted and, when they are scheduled to run again,
the
> > > kernel virtual addresses are restored and still valid.
> > >
> > > Since its use in fs/ufs is safe everywhere, it should be preferred.
> > >
> > > Therefore, replace kmap() with kmap_local_page() in fs/ufs.
kunmap_local()
> > > requires the mapping address, so return that address from ufs_get_page()
> > > to be used in ufs_put_page().
> >
> > Hi Al,
> >
> > I see that this series is here since Dec 29, 2022.
> > Is there anything that prevents its merging?
> > Can you please its four patches in your tree?
>
> I'm pretty sure UFS directories should simply be allocated from lowmem.
> There's really no reason to put them in highmem these days.

Matthew,

It's been a few days since you wrote but I still haven't done anything or
replied. For now, I just want you to know that it's not my intention to ignore
your comment.

The only reasons I hesitate to follow your directions are that I built this
series based on specific suggestions from Al.

While I understand and, for sure, heed what you said, I also think I should
wait for Al to chime in with what he thinks about all this before taking a
different route for removing kmap() from fs/ufs.

Thanks for looking at this series.

Fabio