linux/ipc/namespace.c
Guillaume Knispel 0cfb6aee70 ipc: optimize semget/shmget/msgget for lots of keys
ipc_findkey() used to scan all objects to look for the wanted key.  This
is slow when using a high number of keys.  This change adds an rhashtable
of kern_ipc_perm objects in ipc_ids, so that one lookup cease to be O(n).

This change gives a 865% improvement of benchmark reaim.jobs_per_min on a
56 threads Intel(R) Xeon(R) CPU E5-2695 v3 @ 2.30GHz with 256G memory [1]

Other (more micro) benchmark results, by the author: On an i5 laptop, the
following loop executed right after a reboot took, without and with this
change:

    for (int i = 0, k=0x424242; i < KEYS; ++i)
        semget(k++, 1, IPC_CREAT | 0600);

                 total       total          max single  max single
   KEYS        without        with        call without   call with

      1            3.5         4.9   µs            3.5         4.9
     10            7.6         8.6   µs            3.7         4.7
     32           16.2        15.9   µs            4.3         5.3
    100           72.9        41.8   µs            3.7         4.7
   1000        5,630.0       502.0   µs             *           *
  10000    1,340,000.0     7,240.0   µs             *           *
  31900   17,600,000.0    22,200.0   µs             *           *

 *: unreliable measure: high variance

The duration for a lookup-only usage was obtained by the same loop once
the keys are present:

                 total       total          max single  max single
   KEYS        without        with        call without   call with

      1            2.1         2.5   µs            2.1         2.5
     10            4.5         4.8   µs            2.2         2.3
     32           13.0        10.8   µs            2.3         2.8
    100           82.9        25.1   µs             *          2.3
   1000        5,780.0       217.0   µs             *           *
  10000    1,470,000.0     2,520.0   µs             *           *
  31900   17,400,000.0     7,810.0   µs             *           *

Finally, executing each semget() in a new process gave, when still
summing only the durations of these syscalls:

creation:
                 total       total
   KEYS        without        with

      1            3.7         5.0   µs
     10           32.9        36.7   µs
     32          125.0       109.0   µs
    100          523.0       353.0   µs
   1000       20,300.0     3,280.0   µs
  10000    2,470,000.0    46,700.0   µs
  31900   27,800,000.0   219,000.0   µs

lookup-only:
                 total       total
   KEYS        without        with

      1            2.5         2.7   µs
     10           25.4        24.4   µs
     32          106.0        72.6   µs
    100          591.0       352.0   µs
   1000       22,400.0     2,250.0   µs
  10000    2,510,000.0    25,700.0   µs
  31900   28,200,000.0   115,000.0   µs

[1] http://lkml.kernel.org/r/20170814060507.GE23258@yexl-desktop

Link: http://lkml.kernel.org/r/20170815194954.ck32ta2z35yuzpwp@debix
Signed-off-by: Guillaume Knispel <guillaume.knispel@supersonicimagine.com>
Reviewed-by: Marc Pardo <marc.pardo@supersonicimagine.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Kees Cook <keescook@chromium.org>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: "Peter Zijlstra (Intel)" <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Andrey Vagin <avagin@openvz.org>
Cc: Guillaume Knispel <guillaume.knispel@supersonicimagine.com>
Cc: Marc Pardo <marc.pardo@supersonicimagine.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-09-08 18:26:51 -07:00

218 lines
4.8 KiB
C

/*
* linux/ipc/namespace.c
* Copyright (C) 2006 Pavel Emelyanov <xemul@openvz.org> OpenVZ, SWsoft Inc.
*/
#include <linux/ipc.h>
#include <linux/msg.h>
#include <linux/ipc_namespace.h>
#include <linux/rcupdate.h>
#include <linux/nsproxy.h>
#include <linux/slab.h>
#include <linux/cred.h>
#include <linux/fs.h>
#include <linux/mount.h>
#include <linux/user_namespace.h>
#include <linux/proc_ns.h>
#include <linux/sched/task.h>
#include "util.h"
static struct ucounts *inc_ipc_namespaces(struct user_namespace *ns)
{
return inc_ucount(ns, current_euid(), UCOUNT_IPC_NAMESPACES);
}
static void dec_ipc_namespaces(struct ucounts *ucounts)
{
dec_ucount(ucounts, UCOUNT_IPC_NAMESPACES);
}
static struct ipc_namespace *create_ipc_ns(struct user_namespace *user_ns,
struct ipc_namespace *old_ns)
{
struct ipc_namespace *ns;
struct ucounts *ucounts;
int err;
err = -ENOSPC;
ucounts = inc_ipc_namespaces(user_ns);
if (!ucounts)
goto fail;
err = -ENOMEM;
ns = kmalloc(sizeof(struct ipc_namespace), GFP_KERNEL);
if (ns == NULL)
goto fail_dec;
err = ns_alloc_inum(&ns->ns);
if (err)
goto fail_free;
ns->ns.ops = &ipcns_operations;
refcount_set(&ns->count, 1);
ns->user_ns = get_user_ns(user_ns);
ns->ucounts = ucounts;
err = sem_init_ns(ns);
if (err)
goto fail_put;
err = msg_init_ns(ns);
if (err)
goto fail_destroy_sem;
err = shm_init_ns(ns);
if (err)
goto fail_destroy_msg;
err = mq_init_ns(ns);
if (err)
goto fail_destroy_shm;
return ns;
fail_destroy_shm:
shm_exit_ns(ns);
fail_destroy_msg:
msg_exit_ns(ns);
fail_destroy_sem:
sem_exit_ns(ns);
fail_put:
put_user_ns(ns->user_ns);
ns_free_inum(&ns->ns);
fail_free:
kfree(ns);
fail_dec:
dec_ipc_namespaces(ucounts);
fail:
return ERR_PTR(err);
}
struct ipc_namespace *copy_ipcs(unsigned long flags,
struct user_namespace *user_ns, struct ipc_namespace *ns)
{
if (!(flags & CLONE_NEWIPC))
return get_ipc_ns(ns);
return create_ipc_ns(user_ns, ns);
}
/*
* free_ipcs - free all ipcs of one type
* @ns: the namespace to remove the ipcs from
* @ids: the table of ipcs to free
* @free: the function called to free each individual ipc
*
* Called for each kind of ipc when an ipc_namespace exits.
*/
void free_ipcs(struct ipc_namespace *ns, struct ipc_ids *ids,
void (*free)(struct ipc_namespace *, struct kern_ipc_perm *))
{
struct kern_ipc_perm *perm;
int next_id;
int total, in_use;
down_write(&ids->rwsem);
in_use = ids->in_use;
for (total = 0, next_id = 0; total < in_use; next_id++) {
perm = idr_find(&ids->ipcs_idr, next_id);
if (perm == NULL)
continue;
rcu_read_lock();
ipc_lock_object(perm);
free(ns, perm);
total++;
}
up_write(&ids->rwsem);
}
static void free_ipc_ns(struct ipc_namespace *ns)
{
sem_exit_ns(ns);
msg_exit_ns(ns);
shm_exit_ns(ns);
dec_ipc_namespaces(ns->ucounts);
put_user_ns(ns->user_ns);
ns_free_inum(&ns->ns);
kfree(ns);
}
/*
* put_ipc_ns - drop a reference to an ipc namespace.
* @ns: the namespace to put
*
* If this is the last task in the namespace exiting, and
* it is dropping the refcount to 0, then it can race with
* a task in another ipc namespace but in a mounts namespace
* which has this ipcns's mqueuefs mounted, doing some action
* with one of the mqueuefs files. That can raise the refcount.
* So dropping the refcount, and raising the refcount when
* accessing it through the VFS, are protected with mq_lock.
*
* (Clearly, a task raising the refcount on its own ipc_ns
* needn't take mq_lock since it can't race with the last task
* in the ipcns exiting).
*/
void put_ipc_ns(struct ipc_namespace *ns)
{
if (refcount_dec_and_lock(&ns->count, &mq_lock)) {
mq_clear_sbinfo(ns);
spin_unlock(&mq_lock);
mq_put_mnt(ns);
free_ipc_ns(ns);
}
}
static inline struct ipc_namespace *to_ipc_ns(struct ns_common *ns)
{
return container_of(ns, struct ipc_namespace, ns);
}
static struct ns_common *ipcns_get(struct task_struct *task)
{
struct ipc_namespace *ns = NULL;
struct nsproxy *nsproxy;
task_lock(task);
nsproxy = task->nsproxy;
if (nsproxy)
ns = get_ipc_ns(nsproxy->ipc_ns);
task_unlock(task);
return ns ? &ns->ns : NULL;
}
static void ipcns_put(struct ns_common *ns)
{
return put_ipc_ns(to_ipc_ns(ns));
}
static int ipcns_install(struct nsproxy *nsproxy, struct ns_common *new)
{
struct ipc_namespace *ns = to_ipc_ns(new);
if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN) ||
!ns_capable(current_user_ns(), CAP_SYS_ADMIN))
return -EPERM;
/* Ditch state from the old ipc namespace */
exit_sem(current);
put_ipc_ns(nsproxy->ipc_ns);
nsproxy->ipc_ns = get_ipc_ns(ns);
return 0;
}
static struct user_namespace *ipcns_owner(struct ns_common *ns)
{
return to_ipc_ns(ns)->user_ns;
}
const struct proc_ns_operations ipcns_operations = {
.name = "ipc",
.type = CLONE_NEWIPC,
.get = ipcns_get,
.put = ipcns_put,
.install = ipcns_install,
.owner = ipcns_owner,
};