pid: remove pidhash

pidhash is no longer required as all the information can be looked up
from idr tree.  nr_hashed represented the number of pids that had been
hashed.  Since, nr_hashed and PIDNS_HASH_ADDING are no longer relevant,
it has been renamed to pid_allocated and PIDNS_ADDING respectively.

[gs051095@gmail.com: v6]
  Link: http://lkml.kernel.org/r/1507760379-21662-3-git-send-email-gs051095@gmail.com
Link: http://lkml.kernel.org/r/1507583624-22146-3-git-send-email-gs051095@gmail.com
Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Tested-by: Tony Luck <tony.luck@intel.com>	[ia64]
Cc: Julia Lawall <julia.lawall@lip6.fr>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Pavel Tatashin <pasha.tatashin@oracle.com>
Cc: Kirill Tkhai <ktkhai@virtuozzo.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Gargi Sharma 2017-11-17 15:30:34 -08:00 committed by Linus Torvalds
parent 95846ecf9d
commit e8cfbc245e
8 changed files with 18 additions and 50 deletions

View File

@ -31,8 +31,8 @@ void foo(void)
DEFINE(SIGFRAME_SIZE, sizeof (struct sigframe)); DEFINE(SIGFRAME_SIZE, sizeof (struct sigframe));
DEFINE(UNW_FRAME_INFO_SIZE, sizeof (struct unw_frame_info)); DEFINE(UNW_FRAME_INFO_SIZE, sizeof (struct unw_frame_info));
BUILD_BUG_ON(sizeof(struct upid) != 32); BUILD_BUG_ON(sizeof(struct upid) != 16);
DEFINE(IA64_UPID_SHIFT, 5); DEFINE(IA64_UPID_SHIFT, 4);
BLANK(); BLANK();

View File

@ -105,7 +105,6 @@ extern struct group_info init_groups;
.numbers = { { \ .numbers = { { \
.nr = 0, \ .nr = 0, \
.ns = &init_pid_ns, \ .ns = &init_pid_ns, \
.pid_chain = { .next = NULL, .pprev = NULL }, \
}, } \ }, } \
} }

View File

@ -51,10 +51,8 @@ enum pid_type
*/ */
struct upid { struct upid {
/* Try to keep pid_chain in the same cacheline as nr for find_vpid */
int nr; int nr;
struct pid_namespace *ns; struct pid_namespace *ns;
struct hlist_node pid_chain;
}; };
struct pid struct pid

View File

@ -25,7 +25,7 @@ struct pid_namespace {
struct kref kref; struct kref kref;
struct idr idr; struct idr idr;
struct rcu_head rcu; struct rcu_head rcu;
unsigned int nr_hashed; unsigned int pid_allocated;
struct task_struct *child_reaper; struct task_struct *child_reaper;
struct kmem_cache *pid_cachep; struct kmem_cache *pid_cachep;
unsigned int level; unsigned int level;
@ -49,7 +49,7 @@ struct pid_namespace {
extern struct pid_namespace init_pid_ns; extern struct pid_namespace init_pid_ns;
#define PIDNS_HASH_ADDING (1U << 31) #define PIDNS_ADDING (1U << 31)
#ifdef CONFIG_PID_NS #ifdef CONFIG_PID_NS
static inline struct pid_namespace *get_pid_ns(struct pid_namespace *ns) static inline struct pid_namespace *get_pid_ns(struct pid_namespace *ns)

View File

@ -562,7 +562,6 @@ asmlinkage __visible void __init start_kernel(void)
* kmem_cache_init() * kmem_cache_init()
*/ */
setup_log_buf(0); setup_log_buf(0);
pidhash_init();
vfs_caches_init_early(); vfs_caches_init_early();
sort_main_extable(); sort_main_extable();
trap_init(); trap_init();

View File

@ -1871,7 +1871,7 @@ static __latent_entropy struct task_struct *copy_process(
retval = -ERESTARTNOINTR; retval = -ERESTARTNOINTR;
goto bad_fork_cancel_cgroup; goto bad_fork_cancel_cgroup;
} }
if (unlikely(!(ns_of_pid(pid)->nr_hashed & PIDNS_HASH_ADDING))) { if (unlikely(!(ns_of_pid(pid)->pid_allocated & PIDNS_ADDING))) {
retval = -ENOMEM; retval = -ENOMEM;
goto bad_fork_cancel_cgroup; goto bad_fork_cancel_cgroup;
} }

View File

@ -41,10 +41,6 @@
#include <linux/sched/task.h> #include <linux/sched/task.h>
#include <linux/idr.h> #include <linux/idr.h>
#define pid_hashfn(nr, ns) \
hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift)
static struct hlist_head *pid_hash;
static unsigned int pidhash_shift = 4;
struct pid init_struct_pid = INIT_STRUCT_PID; struct pid init_struct_pid = INIT_STRUCT_PID;
int pid_max = PID_MAX_DEFAULT; int pid_max = PID_MAX_DEFAULT;
@ -54,7 +50,6 @@ int pid_max = PID_MAX_DEFAULT;
int pid_max_min = RESERVED_PIDS + 1; int pid_max_min = RESERVED_PIDS + 1;
int pid_max_max = PID_MAX_LIMIT; int pid_max_max = PID_MAX_LIMIT;
/* /*
* PID-map pages start out as NULL, they get allocated upon * PID-map pages start out as NULL, they get allocated upon
* first use and are never deallocated. This way a low pid_max * first use and are never deallocated. This way a low pid_max
@ -64,7 +59,7 @@ int pid_max_max = PID_MAX_LIMIT;
struct pid_namespace init_pid_ns = { struct pid_namespace init_pid_ns = {
.kref = KREF_INIT(2), .kref = KREF_INIT(2),
.idr = IDR_INIT, .idr = IDR_INIT,
.nr_hashed = PIDNS_HASH_ADDING, .pid_allocated = PIDNS_ADDING,
.level = 0, .level = 0,
.child_reaper = &init_task, .child_reaper = &init_task,
.user_ns = &init_user_ns, .user_ns = &init_user_ns,
@ -123,8 +118,7 @@ void free_pid(struct pid *pid)
for (i = 0; i <= pid->level; i++) { for (i = 0; i <= pid->level; i++) {
struct upid *upid = pid->numbers + i; struct upid *upid = pid->numbers + i;
struct pid_namespace *ns = upid->ns; struct pid_namespace *ns = upid->ns;
hlist_del_rcu(&upid->pid_chain); switch (--ns->pid_allocated) {
switch (--ns->nr_hashed) {
case 2: case 2:
case 1: case 1:
/* When all that is left in the pid namespace /* When all that is left in the pid namespace
@ -133,10 +127,10 @@ void free_pid(struct pid *pid)
*/ */
wake_up_process(ns->child_reaper); wake_up_process(ns->child_reaper);
break; break;
case PIDNS_HASH_ADDING: case PIDNS_ADDING:
/* Handle a fork failure of the first process */ /* Handle a fork failure of the first process */
WARN_ON(ns->child_reaper); WARN_ON(ns->child_reaper);
ns->nr_hashed = 0; ns->pid_allocated = 0;
/* fall through */ /* fall through */
case 0: case 0:
schedule_work(&ns->proc_work); schedule_work(&ns->proc_work);
@ -212,14 +206,12 @@ struct pid *alloc_pid(struct pid_namespace *ns)
upid = pid->numbers + ns->level; upid = pid->numbers + ns->level;
spin_lock_irq(&pidmap_lock); spin_lock_irq(&pidmap_lock);
if (!(ns->nr_hashed & PIDNS_HASH_ADDING)) if (!(ns->pid_allocated & PIDNS_ADDING))
goto out_unlock; goto out_unlock;
for ( ; upid >= pid->numbers; --upid) { for ( ; upid >= pid->numbers; --upid) {
hlist_add_head_rcu(&upid->pid_chain,
&pid_hash[pid_hashfn(upid->nr, upid->ns)]);
/* Make the PID visible to find_pid_ns. */ /* Make the PID visible to find_pid_ns. */
idr_replace(&upid->ns->idr, pid, upid->nr); idr_replace(&upid->ns->idr, pid, upid->nr);
upid->ns->nr_hashed++; upid->ns->pid_allocated++;
} }
spin_unlock_irq(&pidmap_lock); spin_unlock_irq(&pidmap_lock);
@ -243,21 +235,13 @@ out_free:
void disable_pid_allocation(struct pid_namespace *ns) void disable_pid_allocation(struct pid_namespace *ns)
{ {
spin_lock_irq(&pidmap_lock); spin_lock_irq(&pidmap_lock);
ns->nr_hashed &= ~PIDNS_HASH_ADDING; ns->pid_allocated &= ~PIDNS_ADDING;
spin_unlock_irq(&pidmap_lock); spin_unlock_irq(&pidmap_lock);
} }
struct pid *find_pid_ns(int nr, struct pid_namespace *ns) struct pid *find_pid_ns(int nr, struct pid_namespace *ns)
{ {
struct upid *pnr; return idr_find(&ns->idr, nr);
hlist_for_each_entry_rcu(pnr,
&pid_hash[pid_hashfn(nr, ns)], pid_chain)
if (pnr->nr == nr && pnr->ns == ns)
return container_of(pnr, struct pid,
numbers[ns->level]);
return NULL;
} }
EXPORT_SYMBOL_GPL(find_pid_ns); EXPORT_SYMBOL_GPL(find_pid_ns);
@ -413,6 +397,7 @@ pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type,
if (type != PIDTYPE_PID) { if (type != PIDTYPE_PID) {
if (type == __PIDTYPE_TGID) if (type == __PIDTYPE_TGID)
type = PIDTYPE_PID; type = PIDTYPE_PID;
task = task->group_leader; task = task->group_leader;
} }
nr = pid_nr_ns(rcu_dereference(task->pids[type].pid), ns); nr = pid_nr_ns(rcu_dereference(task->pids[type].pid), ns);
@ -439,23 +424,10 @@ struct pid *find_ge_pid(int nr, struct pid_namespace *ns)
return idr_get_next(&ns->idr, &nr); return idr_get_next(&ns->idr, &nr);
} }
/*
* The pid hash table is scaled according to the amount of memory in the
* machine. From a minimum of 16 slots up to 4096 slots at one gigabyte or
* more.
*/
void __init pidhash_init(void)
{
pid_hash = alloc_large_system_hash("PID", sizeof(*pid_hash), 0, 18,
HASH_EARLY | HASH_SMALL | HASH_ZERO,
&pidhash_shift, NULL,
0, 4096);
}
void __init pid_idr_init(void) void __init pid_idr_init(void)
{ {
/* Verify no one has done anything silly: */ /* Verify no one has done anything silly: */
BUILD_BUG_ON(PID_MAX_LIMIT >= PIDNS_HASH_ADDING); BUILD_BUG_ON(PID_MAX_LIMIT >= PIDNS_ADDING);
/* bump default and minimum pid_max based on number of cpus */ /* bump default and minimum pid_max based on number of cpus */
pid_max = min(pid_max_max, max_t(int, pid_max, pid_max = min(pid_max_max, max_t(int, pid_max,

View File

@ -133,7 +133,7 @@ static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns
ns->parent = get_pid_ns(parent_pid_ns); ns->parent = get_pid_ns(parent_pid_ns);
ns->user_ns = get_user_ns(user_ns); ns->user_ns = get_user_ns(user_ns);
ns->ucounts = ucounts; ns->ucounts = ucounts;
ns->nr_hashed = PIDNS_HASH_ADDING; ns->pid_allocated = PIDNS_ADDING;
INIT_WORK(&ns->proc_work, proc_cleanup_work); INIT_WORK(&ns->proc_work, proc_cleanup_work);
return ns; return ns;
@ -254,7 +254,7 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns)
* sys_wait4() above can't reap the EXIT_DEAD children but we do not * sys_wait4() above can't reap the EXIT_DEAD children but we do not
* really care, we could reparent them to the global init. We could * really care, we could reparent them to the global init. We could
* exit and reap ->child_reaper even if it is not the last thread in * exit and reap ->child_reaper even if it is not the last thread in
* this pid_ns, free_pid(nr_hashed == 0) calls proc_cleanup_work(), * this pid_ns, free_pid(pid_allocated == 0) calls proc_cleanup_work(),
* pid_ns can not go away until proc_kill_sb() drops the reference. * pid_ns can not go away until proc_kill_sb() drops the reference.
* *
* But this ns can also have other tasks injected by setns()+fork(). * But this ns can also have other tasks injected by setns()+fork().
@ -268,7 +268,7 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns)
*/ */
for (;;) { for (;;) {
set_current_state(TASK_INTERRUPTIBLE); set_current_state(TASK_INTERRUPTIBLE);
if (pid_ns->nr_hashed == init_pids) if (pid_ns->pid_allocated == init_pids)
break; break;
schedule(); schedule();
} }