linux/fs/ioprio.c
Tejun Heo b2efa05265 block, cfq: unlink cfq_io_context's immediately
cic is association between io_context and request_queue.  A cic is
linked from both ioc and q and should be destroyed when either one
goes away.  As ioc and q both have their own locks, locking becomes a
bit complex - both orders work for removal from one but not from the
other.

Currently, cfq tries to circumvent this locking order issue with RCU.
ioc->lock nests inside queue_lock but the radix tree and cic's are
also protected by RCU allowing either side to walk their lists without
grabbing lock.

This rather unconventional use of RCU quickly devolves into extremely
fragile convolution.  e.g. The following is from cfqd going away too
soon after ioc and q exits raced.

 general protection fault: 0000 [#1] PREEMPT SMP
 CPU 2
 Modules linked in:
 [   88.503444]
 Pid: 599, comm: hexdump Not tainted 3.1.0-rc10-work+ #158 Bochs Bochs
 RIP: 0010:[<ffffffff81397628>]  [<ffffffff81397628>] cfq_exit_single_io_context+0x58/0xf0
 ...
 Call Trace:
  [<ffffffff81395a4a>] call_for_each_cic+0x5a/0x90
  [<ffffffff81395ab5>] cfq_exit_io_context+0x15/0x20
  [<ffffffff81389130>] exit_io_context+0x100/0x140
  [<ffffffff81098a29>] do_exit+0x579/0x850
  [<ffffffff81098d5b>] do_group_exit+0x5b/0xd0
  [<ffffffff81098de7>] sys_exit_group+0x17/0x20
  [<ffffffff81b02f2b>] system_call_fastpath+0x16/0x1b

The only real hot path here is cic lookup during request
initialization and avoiding extra locking requires very confined use
of RCU.  This patch makes cic removal from both ioc and request_queue
perform double-locking and unlink immediately.

* From q side, the change is almost trivial as ioc->lock nests inside
  queue_lock.  It just needs to grab each ioc->lock as it walks
  cic_list and unlink it.

* From ioc side, it's a bit more difficult because of inversed lock
  order.  ioc needs its lock to walk its cic_list but can't grab the
  matching queue_lock and needs to perform unlock-relock dancing.

  Unlinking is now wholly done from put_io_context() and fast path is
  optimized by using the queue_lock the caller already holds, which is
  by far the most common case.  If the ioc accessed multiple devices,
  it tries with trylock.  In unlikely cases of fast path failure, it
  falls back to full double-locking dance from workqueue.

Double-locking isn't the prettiest thing in the world but it's *far*
simpler and more understandable than RCU trick without adding any
meaningful overhead.

This still leaves a lot of now unnecessary RCU logics.  Future patches
will trim them.

-v2: Vivek pointed out that cic->q was being dereferenced after
     cic->release() was called.  Updated to use local variable @this_q
     instead.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2011-12-14 00:33:39 +01:00

236 lines
4.8 KiB
C

/*
* fs/ioprio.c
*
* Copyright (C) 2004 Jens Axboe <axboe@kernel.dk>
*
* Helper functions for setting/querying io priorities of processes. The
* system calls closely mimmick getpriority/setpriority, see the man page for
* those. The prio argument is a composite of prio class and prio data, where
* the data argument has meaning within that class. The standard scheduling
* classes have 8 distinct prio levels, with 0 being the highest prio and 7
* being the lowest.
*
* IOW, setting BE scheduling class with prio 2 is done ala:
*
* unsigned int prio = (IOPRIO_CLASS_BE << IOPRIO_CLASS_SHIFT) | 2;
*
* ioprio_set(PRIO_PROCESS, pid, prio);
*
* See also Documentation/block/ioprio.txt
*
*/
#include <linux/gfp.h>
#include <linux/kernel.h>
#include <linux/export.h>
#include <linux/ioprio.h>
#include <linux/blkdev.h>
#include <linux/capability.h>
#include <linux/syscalls.h>
#include <linux/security.h>
#include <linux/pid_namespace.h>
int set_task_ioprio(struct task_struct *task, int ioprio)
{
int err;
struct io_context *ioc;
const struct cred *cred = current_cred(), *tcred;
rcu_read_lock();
tcred = __task_cred(task);
if (tcred->uid != cred->euid &&
tcred->uid != cred->uid && !capable(CAP_SYS_NICE)) {
rcu_read_unlock();
return -EPERM;
}
rcu_read_unlock();
err = security_task_setioprio(task, ioprio);
if (err)
return err;
ioc = get_task_io_context(task, GFP_ATOMIC, NUMA_NO_NODE);
if (ioc) {
ioc_ioprio_changed(ioc, ioprio);
put_io_context(ioc, NULL);
}
return err;
}
EXPORT_SYMBOL_GPL(set_task_ioprio);
SYSCALL_DEFINE3(ioprio_set, int, which, int, who, int, ioprio)
{
int class = IOPRIO_PRIO_CLASS(ioprio);
int data = IOPRIO_PRIO_DATA(ioprio);
struct task_struct *p, *g;
struct user_struct *user;
struct pid *pgrp;
int ret;
switch (class) {
case IOPRIO_CLASS_RT:
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
/* fall through, rt has prio field too */
case IOPRIO_CLASS_BE:
if (data >= IOPRIO_BE_NR || data < 0)
return -EINVAL;
break;
case IOPRIO_CLASS_IDLE:
break;
case IOPRIO_CLASS_NONE:
if (data)
return -EINVAL;
break;
default:
return -EINVAL;
}
ret = -ESRCH;
rcu_read_lock();
switch (which) {
case IOPRIO_WHO_PROCESS:
if (!who)
p = current;
else
p = find_task_by_vpid(who);
if (p)
ret = set_task_ioprio(p, ioprio);
break;
case IOPRIO_WHO_PGRP:
if (!who)
pgrp = task_pgrp(current);
else
pgrp = find_vpid(who);
do_each_pid_thread(pgrp, PIDTYPE_PGID, p) {
ret = set_task_ioprio(p, ioprio);
if (ret)
break;
} while_each_pid_thread(pgrp, PIDTYPE_PGID, p);
break;
case IOPRIO_WHO_USER:
if (!who)
user = current_user();
else
user = find_user(who);
if (!user)
break;
do_each_thread(g, p) {
if (__task_cred(p)->uid != who)
continue;
ret = set_task_ioprio(p, ioprio);
if (ret)
goto free_uid;
} while_each_thread(g, p);
free_uid:
if (who)
free_uid(user);
break;
default:
ret = -EINVAL;
}
rcu_read_unlock();
return ret;
}
static int get_task_ioprio(struct task_struct *p)
{
int ret;
ret = security_task_getioprio(p);
if (ret)
goto out;
ret = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_NONE, IOPRIO_NORM);
if (p->io_context)
ret = p->io_context->ioprio;
out:
return ret;
}
int ioprio_best(unsigned short aprio, unsigned short bprio)
{
unsigned short aclass = IOPRIO_PRIO_CLASS(aprio);
unsigned short bclass = IOPRIO_PRIO_CLASS(bprio);
if (aclass == IOPRIO_CLASS_NONE)
aclass = IOPRIO_CLASS_BE;
if (bclass == IOPRIO_CLASS_NONE)
bclass = IOPRIO_CLASS_BE;
if (aclass == bclass)
return min(aprio, bprio);
if (aclass > bclass)
return bprio;
else
return aprio;
}
SYSCALL_DEFINE2(ioprio_get, int, which, int, who)
{
struct task_struct *g, *p;
struct user_struct *user;
struct pid *pgrp;
int ret = -ESRCH;
int tmpio;
rcu_read_lock();
switch (which) {
case IOPRIO_WHO_PROCESS:
if (!who)
p = current;
else
p = find_task_by_vpid(who);
if (p)
ret = get_task_ioprio(p);
break;
case IOPRIO_WHO_PGRP:
if (!who)
pgrp = task_pgrp(current);
else
pgrp = find_vpid(who);
do_each_pid_thread(pgrp, PIDTYPE_PGID, p) {
tmpio = get_task_ioprio(p);
if (tmpio < 0)
continue;
if (ret == -ESRCH)
ret = tmpio;
else
ret = ioprio_best(ret, tmpio);
} while_each_pid_thread(pgrp, PIDTYPE_PGID, p);
break;
case IOPRIO_WHO_USER:
if (!who)
user = current_user();
else
user = find_user(who);
if (!user)
break;
do_each_thread(g, p) {
if (__task_cred(p)->uid != user->uid)
continue;
tmpio = get_task_ioprio(p);
if (tmpio < 0)
continue;
if (ret == -ESRCH)
ret = tmpio;
else
ret = ioprio_best(ret, tmpio);
} while_each_thread(g, p);
if (who)
free_uid(user);
break;
default:
ret = -EINVAL;
}
rcu_read_unlock();
return ret;
}