[Devel] [PATCH RHEL9 COMMIT] ve/fs: support per-CT disable of filesystem sync operations
Konstantin Khorenko
khorenko at virtuozzo.com
Thu Dec 9 11:23:21 MSK 2021
The commit is pushed to "branch-rh9-5.14.vz9.1.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git
after rh9-5.14.0-4.vz9.10.34
------>
commit 5e0e63e3b426f44d261c76058855f9f740b3859c
Author: Nikita Yushchenko <nikita.yushchenko at virtuozzo.com>
Date: Thu Dec 9 11:23:21 2021 +0300
ve/fs: support per-CT disable of filesystem sync operations
This patch adds fs.fsync-enable sysctl (virtualized, can be changed from
Containers).
In ve0, the value of this sysctl is ignored.
In veX, setting this to zero will result in:
- sync(2), fsync(2), fdatasync(2), sync_file_range(2), msync(2)
silently do nothing,
- open(2) and fcntl(2) silently clear O_SYNC flag if it is passed.
Operation of AIO is not affected.
Changing this sysctl does not affect O_SYNC flag of already opened
file descriptors.
Default sysctl value: 1 (syncs are enabled in Container).
https://jira.sw.ru/browse/PSBM-44684
Feature: fs: per-CT sync behavior management
Signed-off-by: Nikita Yushchenko <nikita.yushchenko at virtuozzo.com>
---
fs/fcntl.c | 2 ++
fs/open.c | 3 +++
fs/sync.c | 26 ++++++++++++++++++++++++--
include/linux/fs.h | 12 ++++++++++++
include/linux/ve.h | 2 ++
kernel/ve/ve.c | 2 ++
kernel/ve/veowner.c | 8 ++++++++
mm/msync.c | 2 ++
8 files changed, 55 insertions(+), 2 deletions(-)
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 2e0c8515bd1a..8af146ea9231 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -68,6 +68,8 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
if (!may_use_odirect())
arg &= ~O_DIRECT;
+ if (ve_fsync_behavior() == FSYNC_NEVER)
+ arg &= ~O_SYNC;
/*
* O_APPEND cannot be cleared if the file is marked as append-only
* and the file is open for write.
diff --git a/fs/open.c b/fs/open.c
index 040df8bc6e76..65e60aa661a8 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -785,6 +785,9 @@ static int do_dentry_open(struct file *f,
if (!may_use_odirect())
f->f_flags &= ~O_DIRECT;
+ if (ve_fsync_behavior() == FSYNC_NEVER)
+ f->f_flags &= ~O_SYNC;
+
if (unlikely(f->f_flags & O_PATH)) {
f->f_mode = FMODE_PATH | FMODE_OPENED;
f->f_op = &empty_fops;
diff --git a/fs/sync.c b/fs/sync.c
index 1373a610dc78..cca10ec7a90f 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -16,6 +16,7 @@
#include <linux/pagemap.h>
#include <linux/quotaops.h>
#include <linux/backing-dev.h>
+#include <linux/ve.h>
#include "internal.h"
#define VALID_FLAGS (SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE| \
@@ -96,6 +97,17 @@ static void fdatawait_one_bdev(struct block_device *bdev, void *arg)
filemap_fdatawait_keep_errors(bdev->bd_inode->i_mapping);
}
+int ve_fsync_behavior(void)
+{
+ struct ve_struct *ve;
+
+ ve = get_exec_env();
+ if (ve_is_super(ve))
+ return FSYNC_ALWAYS;
+ else
+ return ve->fsync_enable;
+}
+
/*
* Sync everything. We start by waking flusher threads so that most of
* writeback runs on all devices in parallel. Then we sync all inodes reliably
@@ -110,6 +122,9 @@ void ksys_sync(void)
{
int nowait = 0, wait = 1;
+ if (ve_fsync_behavior() == FSYNC_NEVER)
+ return;
+
wakeup_flusher_threads(WB_REASON_SYNC);
iterate_supers(sync_inodes_one_sb, NULL);
iterate_supers(sync_fs_one_sb, &nowait);
@@ -162,18 +177,22 @@ SYSCALL_DEFINE1(syncfs, int, fd)
{
struct fd f = fdget(fd);
struct super_block *sb;
- int ret, ret2;
+ int ret = 0, ret2 = 0;
if (!f.file)
return -EBADF;
sb = f.file->f_path.dentry->d_sb;
+ if (ve_fsync_behavior() == FSYNC_NEVER)
+ goto fdput;
+
down_read(&sb->s_umount);
ret = sync_filesystem(sb);
up_read(&sb->s_umount);
ret2 = errseq_check_and_advance(&sb->s_wb_err, &f.file->f_sb_err);
+fdput:
fdput(f);
return ret ? ret : ret2;
}
@@ -221,7 +240,10 @@ static int do_fsync(unsigned int fd, int datasync)
int ret = -EBADF;
if (f.file) {
- ret = vfs_fsync(f.file, datasync);
+ if (ve_fsync_behavior() != FSYNC_NEVER)
+ ret = vfs_fsync(f.file, datasync);
+ else
+ ret = 0;
fdput(f);
}
return ret;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 01419dbd864b..fb21d1a32cdb 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -3146,6 +3146,18 @@ extern bool path_is_under(const struct path *, const struct path *);
extern char *file_path(struct file *, char *, int);
+#define FSYNC_NEVER 0 /* ve syncs are ignored */
+#define FSYNC_ALWAYS 1 /* ve syncs work as ususal */
+
+#ifdef CONFIG_VE
+int ve_fsync_behavior(void);
+#else
+static inline int ve_fsync_behavior(void)
+{
+ return FSYNC_ALWAYS;
+}
+#endif
+
#include <linux/err.h>
/* needed for stackable file system support */
diff --git a/include/linux/ve.h b/include/linux/ve.h
index 1a66063d9ba8..4c8f7d308829 100644
--- a/include/linux/ve.h
+++ b/include/linux/ve.h
@@ -61,6 +61,8 @@ struct ve_struct {
struct kstat_lat_pcpu_struct sched_lat_ve;
int odirect_enable;
+ int fsync_enable;
+
#if IS_ENABLED(CONFIG_BINFMT_MISC)
struct binfmt_misc *binfmt_misc;
#endif
diff --git a/kernel/ve/ve.c b/kernel/ve/ve.c
index 48be49337a87..e94aa90aff25 100644
--- a/kernel/ve/ve.c
+++ b/kernel/ve/ve.c
@@ -70,6 +70,7 @@ struct ve_struct ve0 = {
.sched_lat_ve.cur = &ve0_lat_stats,
.netns_avail_nr = ATOMIC_INIT(INT_MAX),
.netns_max_nr = INT_MAX,
+ .fsync_enable = FSYNC_ALWAYS,
._randomize_va_space =
#ifdef CONFIG_COMPAT_BRK
1,
@@ -930,6 +931,7 @@ static struct cgroup_subsys_state *ve_create(struct cgroup_subsys_state *parent_
ve->meminfo_val = VE_MEMINFO_DEFAULT;
ve->odirect_enable = 2;
+ ve->fsync_enable = FSYNC_ALWAYS;
atomic_set(&ve->netns_avail_nr, NETNS_MAX_NR_DEFAULT);
ve->netns_max_nr = NETNS_MAX_NR_DEFAULT;
diff --git a/kernel/ve/veowner.c b/kernel/ve/veowner.c
index b0aba35b6be9..e255fe57d447 100644
--- a/kernel/ve/veowner.c
+++ b/kernel/ve/veowner.c
@@ -7,6 +7,7 @@
*
*/
+#include <linux/ve.h>
#include <linux/init.h>
#include <linux/module.h>
#include <linux/proc_fs.h>
@@ -66,6 +67,13 @@ static struct ctl_table vz_fs_table[] = {
.extra1 = &ve_mount_nr_min,
.extra2 = &ve_mount_nr_max,
},
+ {
+ .procname = "fsync-enable",
+ .data = &ve0.fsync_enable,
+ .maxlen = sizeof(int),
+ .mode = 0644 | S_ISVTX,
+ .proc_handler = &proc_dointvec_virtual,
+ },
{ }
};
diff --git a/mm/msync.c b/mm/msync.c
index 137d1c104f3e..20737eb4b76b 100644
--- a/mm/msync.c
+++ b/mm/msync.c
@@ -51,6 +51,8 @@ SYSCALL_DEFINE3(msync, unsigned long, start, size_t, len, int, flags)
if (end < start)
goto out;
error = 0;
+ if (ve_fsync_behavior() == FSYNC_NEVER)
+ goto out;
if (end == start)
goto out;
/*
More information about the Devel
mailing list