2
0
mirror of https://github.com/checkpoint-restore/criu synced 2025-08-22 18:07:57 +00:00
criu/file-ids.c

120 lines
2.0 KiB
C
Raw Permalink Normal View History

files: Use sys_kcmp to find file descriptor duplicates v4 We switch generic-object-id concept with sys_kcmp approach, which implies changes of image format a bit (and since it's early time for project overall, we're allowed to). In short -- previously every file descriptor had an ID generated by a kernel and exported via procfs. If the appropriate file descriptors were the same objects in kernel memory -- the IDs did match up to bit. It allows us to figure out which files were actually the identical ones and should be restored in a special way. Once sys_kcmp system call was merged into the kernel, we've got a new opprotunity -- to use this syscall instead. The syscall basically compares kernel objects and returns ordered results suitable for objects sorting in a userspace. For us it means -- we treat every file descriptor as a combination of 'genid' and 'subid'. While 'genid' serves for fast comparison between fds, the 'subid' is kind of a second key, which guarantees uniqueness of genid+subid tuple over all file descritors found in a process (or group of processes). To be able to find and dump file descriptors in a single pass we collect every fd into a global rbtree, where (!) each node might become a root for a subtree as well. The main tree carries only non-equal genid. If we find genid which is already in tree, we need to make sure that it's either indeed a duplicate or not. For this we use sys_kcmp syscall and if we find that file descriptors are different -- we simply put new fd into a subtree. Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Acked-by: Pavel Emelyanov <xemul@parallels.com>
2012-02-28 18:27:28 +04:00
#include <stdio.h>
#include <stdlib.h>
#include <stdarg.h>
#include <signal.h>
#include <limits.h>
#include <unistd.h>
#include <errno.h>
#include <string.h>
#include <sys/stat.h>
files: Use sys_kcmp to find file descriptor duplicates v4 We switch generic-object-id concept with sys_kcmp approach, which implies changes of image format a bit (and since it's early time for project overall, we're allowed to). In short -- previously every file descriptor had an ID generated by a kernel and exported via procfs. If the appropriate file descriptors were the same objects in kernel memory -- the IDs did match up to bit. It allows us to figure out which files were actually the identical ones and should be restored in a special way. Once sys_kcmp system call was merged into the kernel, we've got a new opprotunity -- to use this syscall instead. The syscall basically compares kernel objects and returns ordered results suitable for objects sorting in a userspace. For us it means -- we treat every file descriptor as a combination of 'genid' and 'subid'. While 'genid' serves for fast comparison between fds, the 'subid' is kind of a second key, which guarantees uniqueness of genid+subid tuple over all file descritors found in a process (or group of processes). To be able to find and dump file descriptors in a single pass we collect every fd into a global rbtree, where (!) each node might become a root for a subtree as well. The main tree carries only non-equal genid. If we find genid which is already in tree, we need to make sure that it's either indeed a duplicate or not. For this we use sys_kcmp syscall and if we find that file descriptors are different -- we simply put new fd into a subtree. Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Acked-by: Pavel Emelyanov <xemul@parallels.com>
2012-02-28 18:27:28 +04:00
#include <sys/types.h>
#include "asm/types.h"
files: Use sys_kcmp to find file descriptor duplicates v4 We switch generic-object-id concept with sys_kcmp approach, which implies changes of image format a bit (and since it's early time for project overall, we're allowed to). In short -- previously every file descriptor had an ID generated by a kernel and exported via procfs. If the appropriate file descriptors were the same objects in kernel memory -- the IDs did match up to bit. It allows us to figure out which files were actually the identical ones and should be restored in a special way. Once sys_kcmp system call was merged into the kernel, we've got a new opprotunity -- to use this syscall instead. The syscall basically compares kernel objects and returns ordered results suitable for objects sorting in a userspace. For us it means -- we treat every file descriptor as a combination of 'genid' and 'subid'. While 'genid' serves for fast comparison between fds, the 'subid' is kind of a second key, which guarantees uniqueness of genid+subid tuple over all file descritors found in a process (or group of processes). To be able to find and dump file descriptors in a single pass we collect every fd into a global rbtree, where (!) each node might become a root for a subtree as well. The main tree carries only non-equal genid. If we find genid which is already in tree, we need to make sure that it's either indeed a duplicate or not. For this we use sys_kcmp syscall and if we find that file descriptors are different -- we simply put new fd into a subtree. Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Acked-by: Pavel Emelyanov <xemul@parallels.com>
2012-02-28 18:27:28 +04:00
#include "file-ids.h"
#include "rbtree.h"
#include "kcmp-ids.h"
files: Use sys_kcmp to find file descriptor duplicates v4 We switch generic-object-id concept with sys_kcmp approach, which implies changes of image format a bit (and since it's early time for project overall, we're allowed to). In short -- previously every file descriptor had an ID generated by a kernel and exported via procfs. If the appropriate file descriptors were the same objects in kernel memory -- the IDs did match up to bit. It allows us to figure out which files were actually the identical ones and should be restored in a special way. Once sys_kcmp system call was merged into the kernel, we've got a new opprotunity -- to use this syscall instead. The syscall basically compares kernel objects and returns ordered results suitable for objects sorting in a userspace. For us it means -- we treat every file descriptor as a combination of 'genid' and 'subid'. While 'genid' serves for fast comparison between fds, the 'subid' is kind of a second key, which guarantees uniqueness of genid+subid tuple over all file descritors found in a process (or group of processes). To be able to find and dump file descriptors in a single pass we collect every fd into a global rbtree, where (!) each node might become a root for a subtree as well. The main tree carries only non-equal genid. If we find genid which is already in tree, we need to make sure that it's either indeed a duplicate or not. For this we use sys_kcmp syscall and if we find that file descriptors are different -- we simply put new fd into a subtree. Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Acked-by: Pavel Emelyanov <xemul@parallels.com>
2012-02-28 18:27:28 +04:00
#include "compiler.h"
#include "syscall.h"
#include "image.h"
#include "util.h"
#include "irmap.h"
#include "files.h"
files: Use sys_kcmp to find file descriptor duplicates v4 We switch generic-object-id concept with sys_kcmp approach, which implies changes of image format a bit (and since it's early time for project overall, we're allowed to). In short -- previously every file descriptor had an ID generated by a kernel and exported via procfs. If the appropriate file descriptors were the same objects in kernel memory -- the IDs did match up to bit. It allows us to figure out which files were actually the identical ones and should be restored in a special way. Once sys_kcmp system call was merged into the kernel, we've got a new opprotunity -- to use this syscall instead. The syscall basically compares kernel objects and returns ordered results suitable for objects sorting in a userspace. For us it means -- we treat every file descriptor as a combination of 'genid' and 'subid'. While 'genid' serves for fast comparison between fds, the 'subid' is kind of a second key, which guarantees uniqueness of genid+subid tuple over all file descritors found in a process (or group of processes). To be able to find and dump file descriptors in a single pass we collect every fd into a global rbtree, where (!) each node might become a root for a subtree as well. The main tree carries only non-equal genid. If we find genid which is already in tree, we need to make sure that it's either indeed a duplicate or not. For this we use sys_kcmp syscall and if we find that file descriptors are different -- we simply put new fd into a subtree. Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Acked-by: Pavel Emelyanov <xemul@parallels.com>
2012-02-28 18:27:28 +04:00
static DECLARE_KCMP_TREE(fd_tree, KCMP_FILE);
void fd_id_show_tree(void)
{
kid_show_tree(&fd_tree);
}
#define FDID_BITS 5
#define FDID_SIZE (1 << FDID_BITS)
#define FDID_MASK (FDID_SIZE - 1)
static inline int fdid_hashfn(unsigned int s_dev, unsigned long i_ino)
{
return (s_dev + i_ino) & FDID_MASK;
}
struct fd_id {
int mnt_id;
unsigned int dev;
unsigned long ino;
u32 id;
struct fd_id *n;
};
static struct fd_id *fd_id_cache[FDID_SIZE];
static void fd_id_cache_one(u32 id, struct fd_parms *p)
{
struct fd_id *fi;
unsigned hv;
fi = xmalloc(sizeof(*fi));
if (fi) {
fi->dev = p->stat.st_dev;
fi->ino = p->stat.st_ino;
fi->mnt_id = p->mnt_id;
fi->id = id;
hv = fdid_hashfn(p->stat.st_dev, p->stat.st_ino);
fi->n = fd_id_cache[hv];
fd_id_cache[hv] = fi;
}
}
static struct fd_id *fd_id_cache_lookup(struct fd_parms *p)
{
struct stat *st = &p->stat;
struct fd_id *fi;
for (fi = fd_id_cache[fdid_hashfn(st->st_dev, st->st_ino)];
fi; fi = fi->n)
if (fi->dev == st->st_dev &&
fi->ino == st->st_ino &&
fi->mnt_id == p->mnt_id)
return fi;
return NULL;
}
int fd_id_generate_special(struct fd_parms *p, u32 *id)
{
if (p) {
struct fd_id *fi;
fi = fd_id_cache_lookup(p);
if (fi) {
*id = fi->id;
return 0;
}
}
*id = fd_tree.subid++;
if (p)
fd_id_cache_one(*id, p);
return 1;
}
int fd_id_generate(pid_t pid, FdinfoEntry *fe, struct fd_parms *p)
{
u32 id;
struct kid_elem e;
int new_id = 0;
e.pid = pid;
e.genid = fe->id;
e.idx = fe->fd;
id = kid_generate_gen(&fd_tree, &e, &new_id);
if (!id)
return -ENOMEM;
if (new_id)
fd_id_cache_one(id, p);
fe->id = id;
return new_id;
files: Use sys_kcmp to find file descriptor duplicates v4 We switch generic-object-id concept with sys_kcmp approach, which implies changes of image format a bit (and since it's early time for project overall, we're allowed to). In short -- previously every file descriptor had an ID generated by a kernel and exported via procfs. If the appropriate file descriptors were the same objects in kernel memory -- the IDs did match up to bit. It allows us to figure out which files were actually the identical ones and should be restored in a special way. Once sys_kcmp system call was merged into the kernel, we've got a new opprotunity -- to use this syscall instead. The syscall basically compares kernel objects and returns ordered results suitable for objects sorting in a userspace. For us it means -- we treat every file descriptor as a combination of 'genid' and 'subid'. While 'genid' serves for fast comparison between fds, the 'subid' is kind of a second key, which guarantees uniqueness of genid+subid tuple over all file descritors found in a process (or group of processes). To be able to find and dump file descriptors in a single pass we collect every fd into a global rbtree, where (!) each node might become a root for a subtree as well. The main tree carries only non-equal genid. If we find genid which is already in tree, we need to make sure that it's either indeed a duplicate or not. For this we use sys_kcmp syscall and if we find that file descriptors are different -- we simply put new fd into a subtree. Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Acked-by: Pavel Emelyanov <xemul@parallels.com>
2012-02-28 18:27:28 +04:00
}