Skip to content

K
kernel_linux

OpenHarmony / kernel_linux
上一次同步 3 年多

通知 13
Star 8
Fork 2
K
kernel_linux
切换分支/标签
  1. 19 10月, 2005 3 次提交
    • T
      NFSv4: Fix a potential CLOSE race · 9512135d
      由 Trond Myklebust 提交于 
       Once the state_owner and lock_owner semaphores get removed, it will be
 possible for other OPEN requests to reopen the same file if they have
 lower sequence ids than our CLOSE call.
 This patch ensures that we recheck the file state once
 nfs_wait_on_sequence() has completed waiting.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
      9512135d
    • T
      NFSv4: Add functions to order RPC calls · cee54fc9
      由 Trond Myklebust 提交于 
       NFSv4 file state-changing functions such as OPEN, CLOSE, LOCK,... are all
 labelled with "sequence identifiers" in order to prevent the server from
 reordering RPC requests, as this could cause its file state to
 become out of sync with the client.

 Currently the NFS client code enforces this ordering locally using
 semaphores to restrict access to structures until the RPC call is done.
 This, of course, only works with synchronous RPC calls, since the
 user process must first grab the semaphore.
 By dropping semaphores, and instead teaching the RPC engine to hold
 the RPC calls until they are ready to be sent, we can extend this
 process to work nicely with asynchronous RPC calls too.

 This patch adds a new list called "rpc_sequence" that defines the order
 of the RPC calls to be sent. We add one such list for each state_owner.
 When an RPC call is ready to be sent, it checks if it is top of the
 rpc_sequence list. If so, it proceeds. If not, it goes back to sleep,
 and loops until it hits top of the list.
 Once the RPC call has completed, it can then bump the sequence id counter,
 and remove itself from the rpc_sequence list, and then wake up the next
 sleeper.

 Note that the state_owner sequence ids and lock_owner sequence ids are
 all indexed to the same rpc_sequence list, so OPEN, LOCK,... requests
 are all ordered w.r.t. each other.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
      cee54fc9
    • T
      RPC: allow call_encode() to delay transmission of an RPC call. · 5e5ce5be
      由 Trond Myklebust 提交于 
       Currently, call_encode will cause the entire RPC call to abort if it returns
 an error. This is unnecessarily rigid, and gets in the way of attempts
 to allow the NFSv4 layer to order RPC calls that carry sequence ids.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
      5e5ce5be
  3. 18 10月, 2005 2 次提交
  5. 17 10月, 2005 1 次提交
    • H
      [PATCH] list: add missing rcu_dereference on first element · b24d18aa
      由 Herbert Xu 提交于 
      It seems that all the list_*_rcu primitives are missing a memory barrier
on the very first dereference.  For example,

#define list_for_each_rcu(pos, head) \
	for (pos = (head)->next; prefetch(pos->next), pos != (head); \
		pos = rcu_dereference(pos->next))

It will go something like:

	pos = (head)->next

	prefetch(pos->next)

	pos != (head)

	do stuff

We're missing a barrier here.

	pos = rcu_dereference(pos->next)

		fetch pos->next

		barrier given by rcu_dereference(pos->next)

		store pos

Without the missing barrier, the pos->next value may turn out to be stale.
In fact, if "do stuff" were also dereferencing pos and relying on
list_for_each_rcu to provide the barrier then it may also break.

So here is a patch to make sure that we have a barrier for the first
element in the list.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Acked-by: N"Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      b24d18aa
  7. 16 10月, 2005 1 次提交
    • A
      [PATCH]: highest_possible_processor_id() has to be a macro · 688ce17b
      由 Al Viro 提交于 
      	... otherwise, things like alpha and sparc64 break and break
badly.  They define cpu_possible_map to something else in smp.h
*AFTER* having included cpumask.h.

	If that puppy is a macro, expansion will happen at the actual
caller, when we'd already seen #define cpu_possible_map ... and we will
get the right thing used.

	As an inline helper it will be tokenized before we get to that
define and that's it; no matter what we define later, it won't affect
anything.  We get modules with dependency on cpu_possible_map instead
of the right symbol (phys_cpu_present_map in case of sparc64), or outright
link errors if they are built-in.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      688ce17b
  9. 15 10月, 2005 1 次提交
  11. 14 10月, 2005 1 次提交
  13. 13 10月, 2005 1 次提交
  15. 11 10月, 2005 7 次提交
  17. 10 10月, 2005 1 次提交
    • R
      [PATCH] x86_64: Set up safe page tables during resume · 3dd08325
      由 Rafael J. Wysocki 提交于 
      The following patch makes swsusp avoid the possible temporary corruption
of page translation tables during resume on x86-64.  This is achieved by
creating a copy of the relevant page tables that will not be modified by
swsusp and can be safely used by it on resume.

The problem is that during resume on x86-64 swsusp may temporarily
corrupt the page tables used for the direct mapping of RAM.  If that
happens, a page fault occurs and cannot be handled properly, which leads
to the solid hang of the affected system.  This leads to the loss of the
system's state from before suspend and may result in the loss of data or
the corruption of filesystems, so it is a serious issue.  Also, it
appears to happen quite often (for me, as often as 50% of the time).

The problem is related to the fact that (at least) one of the PMD
entries used in the direct memory mapping (starting at PAGE_OFFSET)
points to a page table the physical address of which is much greater
than the physical address of the PMD entry itself.  Moreover,
unfortunately, the physical address of the page table before suspend
(i.e.  the one stored in the suspend image) happens to be different to
the physical address of the corresponding page table used during resume
(i.e.  the one that is valid right before swsusp_arch_resume() in
arch/x86_64/kernel/suspend_asm.S is executed).  Thus while the image is
restored, the "offending" PMD entry gets overwritten, so it does not
point to the right physical address any more (i.e.  there's no page
table at the address pointed to by it, because it points to the address
the page table has been at during suspend).  Consequently, if the PMD
entry is used later on, and it _is_ used in the process of copying the
image pages, a page fault occurs, but it cannot be handled in the normal
way and the system hangs.

In principle we can call create_resume_mapping() from
swsusp_arch_resume() (ie.  from suspend_asm.S), but then the memory
allocations in create_resume_mapping(), resume_pud_mapping(), and
resume_pmd_mapping() must be made carefully so that we use _only_
NosaveFree pages in them (the other pages are overwritten by the loop in
swsusp_arch_resume()).  Additionally, we are in atomic context at that
time, so we cannot use GFP_KERNEL.  Moreover, if one of the allocations
fails, we should free all of the allocated pages, so we need to trace
them somehow.

All of this is done in the appended patch, except that the functions
populating the page tables are located in arch/x86_64/kernel/suspend.c
rather than in init.c.  It may be done in a more elegan way in the
future, with the help of some swsusp patches that are in the works now.

[AK: move some externs into headers, renamed a function]
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NAndi Kleen <ak@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      3dd08325
  19. 09 10月, 2005 2 次提交
  21. 07 10月, 2005 1 次提交
  23. 05 10月, 2005 4 次提交
  25. 04 10月, 2005 3 次提交
    • H
      [IPV4]: Replace __in_dev_get with __in_dev_get_rcu/rtnl · e5ed6399
      由 Herbert Xu 提交于 
      The following patch renames __in_dev_get() to __in_dev_get_rtnl() and
introduces __in_dev_get_rcu() to cover the second case.

1) RCU with refcnt should use in_dev_get().
2) RCU without refcnt should use __in_dev_get_rcu().
3) All others must hold RTNL and use __in_dev_get_rtnl().

There is one exception in net/ipv4/route.c which is in fact a pre-existing
race condition.  I've marked it as such so that we remember to fix it.

This patch is based on suggestions and prior work by Suzanne Wood and
Paul McKenney.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      e5ed6399
    • E
      [INET]: speedup inet (tcp/dccp) lookups · 81c3d547
      由 Eric Dumazet 提交于 
      Arnaldo and I agreed it could be applied now, because I have other
pending patches depending on this one (Thank you Arnaldo)

(The other important patch moves skc_refcnt in a separate cache line,
so that the SMP/NUMA performance doesnt suffer from cache line ping pongs)

1) First some performance data :
--------------------------------

tcp_v4_rcv() wastes a *lot* of time in __inet_lookup_established()

The most time critical code is :

sk_for_each(sk, node, &head->chain) {
     if (INET_MATCH(sk, acookie, saddr, daddr, ports, dif))
         goto hit; /* You sunk my battleship! */
}

The sk_for_each() does use prefetch() hints but only the begining of
"struct sock" is prefetched.

As INET_MATCH first comparison uses inet_sk(__sk)->daddr, wich is far
away from the begining of "struct sock", it has to bring into CPU
cache cold cache line. Each iteration has to use at least 2 cache
lines.

This can be problematic if some chains are very long.

2) The goal
-----------

The idea I had is to change things so that INET_MATCH() may return
FALSE in 99% of cases only using the data already in the CPU cache,
using one cache line per iteration.

3) Description of the patch
---------------------------

Adds a new 'unsigned int skc_hash' field in 'struct sock_common',
filling a 32 bits hole on 64 bits platform.

struct sock_common {
	unsigned short		skc_family;
	volatile unsigned char	skc_state;
	unsigned char		skc_reuse;
	int			skc_bound_dev_if;
	struct hlist_node	skc_node;
	struct hlist_node	skc_bind_node;
	atomic_t		skc_refcnt;
+	unsigned int		skc_hash;
	struct proto		*skc_prot;
};

Store in this 32 bits field the full hash, not masked by (ehash_size -
1) Using this full hash as the first comparison done in INET_MATCH
permits us immediatly skip the element without touching a second cache
line in case of a miss.

Suppress the sk_hashent/tw_hashent fields since skc_hash (aliased to
sk_hash and tw_hash) already contains the slot number if we mask with
(ehash_size - 1)

File include/net/inet_hashtables.h

64 bits platforms :
#define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
     (((__sk)->sk_hash == (__hash))
     ((*((__u64 *)&(inet_sk(__sk)->daddr)))== (__cookie))   &&  \
     ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports))   &&  \
     (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))

32bits platforms:
#define TCP_IPV4_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
     (((__sk)->sk_hash == (__hash))                 &&  \
     (inet_sk(__sk)->daddr          == (__saddr))   &&  \
     (inet_sk(__sk)->rcv_saddr      == (__daddr))   &&  \
     (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))


- Adds a prefetch(head->chain.first) in 
__inet_lookup_established()/__tcp_v4_check_established() and 
__inet6_lookup_established()/__tcp_v6_check_established() and 
__dccp_v4_check_established() to bring into cache the first element of the 
list, before the {read|write}_lock(&head->lock);
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Acked-by: NArnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      81c3d547
    • H
      [NET]: Fix packet timestamping. · 325ed823
      由 Herbert Xu 提交于 
      I've found the problem in general.  It affects any 64-bit
architecture.  The problem occurs when you change the system time.

Suppose that when you boot your system clock is forward by a day.
This gets recorded down in skb_tv_base.  You then wind the clock back
by a day.  From that point onwards the offset will be negative which
essentially overflows the 32-bit variables they're stored in.

In fact, why don't we just store the real time stamp in those 32-bit
variables? After all, we're not going to overflow for quite a while
yet.

When we do overflow, we'll need a better solution of course.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      325ed823
  27. 02 10月, 2005 1 次提交
  29. 01 10月, 2005 1 次提交
    • Z
      [PATCH] aio: remove unlocked task_list test and resulting race · 897f15fb
      由 Zach Brown 提交于 
      Only one of the run or kick path is supposed to put an iocb on the run
list.  If both of them do it than one of them can end up referencing a
freed iocb.  The kick path could delete the task_list item from the wait
queue before getting the ctx_lock and putting the iocb on the run list.
The run path was testing the task_list item outside the lock so that it
could catch ki_retry methods that return -EIOCBRETRY *without* putting the
iocb on a wait queue and promising to call kick_iocb.  This unlocked check
could then race with the kick path to cause both to try and put the iocb on
the run list.

The patch stops the run path from testing task_list by requring that any
ki_retry that returns -EIOCBRETRY *must* guarantee that kick_iocb() will be
called in the future.  aio_p{read,write}, the only in-tree -EIOCBRETRY
users, are updated.
Signed-off-by: NZach Brown <zach.brown@oracle.com>
Signed-off-by: NBenjamin LaHaise <bcrl@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      897f15fb
  31. 30 9月, 2005 2 次提交
    • L
      Revert task flag re-ordering, add comments · 4a8342d2
      由 Linus Torvalds 提交于 
      Roland points out that the flags end up having non-obvious dependencies
elsewhere, so revert aa55a086 and add
some comments about why things are as they are.

We'll just have to fix up the broken comparisons. Roland has a patch.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      4a8342d2
    • O
      [PATCH] fix TASK_STOPPED vs TASK_NONINTERACTIVE interaction · aa55a086
      由 Oleg Nesterov 提交于 
      do_signal_stop:

	for_each_thread(t) {
		if (t->state < TASK_STOPPED)
			++sig->group_stop_count;
	}

However, TASK_NONINTERACTIVE > TASK_STOPPED, so this loop will not
count TASK_INTERRUPTIBLE | TASK_NONINTERACTIVE threads.

See also wait_task_stopped(), which checks ->state > TASK_STOPPED.
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>

[ We really probably should always use the appropriate bitmasks to test
  task states, not do it like this. Using something like

	#define TASK_RUNNABLE (TASK_RUNNING | TASK_INTERRUPTIBLE | \
				TASK_UNINTERRUPTIBLE | TASK_NONINTERACTIVE)

  and then doing "if (task->state & TASK_RUNNABLE)" or similar. But the
  ordering of the task states is historical, and keeping the ordering
  does make sense regardless. ]
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      aa55a086
  33. 29 9月, 2005 1 次提交
    • D
      [PATCH] Keys: Add possessor permissions to keys [try #3] · 664cceb0
      由 David Howells 提交于 
      The attached patch adds extra permission grants to keys for the possessor of a
key in addition to the owner, group and other permissions bits. This makes
SUID binaries easier to support without going as far as labelling keys and key
targets using the LSM facilities.

This patch adds a second "pointer type" to key structures (struct key_ref *)
that can have the bottom bit of the address set to indicate the possession of
a key. This is propagated through searches from the keyring to the discovered
key. It has been made a separate type so that the compiler can spot attempts
to dereference a potentially incorrect pointer.

The "possession" attribute can't be attached to a key structure directly as
it's not an intrinsic property of a key.

Pointers to keys have been replaced with struct key_ref *'s wherever
possession information needs to be passed through.

This does assume that the bottom bit of the pointer will always be zero on
return from kmem_cache_alloc().

The key reference type has been made into a typedef so that at least it can be
located in the sources, even though it's basically a pointer to an undefined
type. I've also renamed the accessor functions to be more useful, and all
reference variables should now end in "_ref".
Signed-Off-By: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      664cceb0
  35. 28 9月, 2005 3 次提交
    • B
      [NET]: Fix GCC4 compile error: sysctl in linux/if_ether.h · 2fab35d7
      由 Ben Dooks 提交于 
      The following is generated when compiling a
recent (2.6.14-rc2-git5) kernel configured for
ARM, with GCC4. 

  CC      init/main.o
In file included from include/linux/netdevice.h:29,
                 from include/net/sock.h:48,
                 from init/main.c:50:
include/linux/if_ether.h:114: error: array type has incomplete element type

It seems that if CONFIG_SYSCTL is not set, then
the compiler will throw an error due to the definition
of the ether_table[] array

Attached is a solution to the problem
Signed-off-by: NBen Dooks <ben-linux@fluff.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      2fab35d7
    • D
      [NET]: Add Sun Cassini driver. · 1f26dac3
      由 David S. Miller 提交于 
      Written by Adrian Sun (asun@darksunrising.com).
Ported to 2.6.x by Tom 'spot' Callaway <tcallawa@redhat.com>.
Further cleaned up and integrated by David S. Miller
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      1f26dac3
    • E
      [NET]: Reorder some hot fields of struct net_device · 9356b8fc
      由 Eric Dumazet 提交于 
      Place them on separate cache lines in SMP to lower memory bouncing
between multiple CPU accessing the device.

     - One part is mostly used on receive path (including
       eth_type_trans()) (poll_list, poll, quota, weight, last_rx,
       dev_addr, broadcast)

     - One part is mostly used on queue transmit path (qdisc)
      (queue_lock, qdisc, qdisc_sleeping, qdisc_list, tx_queue_len)

     - One part is mostly used on xmit path (device)
      (xmit_lock, xmit_lock_owner, priv, hard_start_xmit, trans_start)

'features' is placed outside of these hot points, in a location that
may be shared by all cpus (because mostly read)

name_hlist is moved close to name[IFNAMSIZ] to speedup __dev_get_by_name()
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      9356b8fc
  37. 27 9月, 2005 3 次提交
  39. 26 9月, 2005 1 次提交