skb_release_data() no longer has any users in other files.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch implements an API whereby an application can determine the
label of its peer's Unix datagram sockets via the auxiliary data mechanism of
recvmsg.
Patch purpose:
This patch enables a security-aware application to retrieve the
security context of the peer of a Unix datagram socket. The application
can then use this security context to determine the security context for
processing on behalf of the peer who sent the packet.
Patch design and implementation:
The design and implementation is very similar to the UDP case for INET
sockets. Basically we build upon the existing Unix domain socket API for
retrieving user credentials. Linux offers the API for obtaining user
credentials via ancillary messages (i.e., out of band/control messages
that are bundled together with a normal message). To retrieve the security
context, the application first indicates to the kernel such desire by
setting the SO_PASSSEC option via getsockopt. Then the application
retrieves the security context using the auxiliary data mechanism.
An example server application for Unix datagram socket should look like this:
toggle = 1;
toggle_len = sizeof(toggle);
setsockopt(sockfd, SOL_SOCKET, SO_PASSSEC, &toggle, &toggle_len);
recvmsg(sockfd, &msg_hdr, 0);
if (msg_hdr.msg_controllen > sizeof(struct cmsghdr)) {
cmsg_hdr = CMSG_FIRSTHDR(&msg_hdr);
if (cmsg_hdr->cmsg_len <= CMSG_LEN(sizeof(scontext)) &&
cmsg_hdr->cmsg_level == SOL_SOCKET &&
cmsg_hdr->cmsg_type == SCM_SECURITY) {
memcpy(&scontext, CMSG_DATA(cmsg_hdr), sizeof(scontext));
}
}
sock_setsockopt is enhanced with a new socket option SOCK_PASSSEC to allow
a server socket to receive security context of the peer.
Testing:
We have tested the patch by setting up Unix datagram client and server
applications. We verified that the server can retrieve the security context
using the auxiliary data mechanism of recvmsg.
Signed-off-by: Catherine Zhang <cxzhang@watson.ibm.com>
Acked-by: Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rather than having illegal_highdma as a macro when HIGHMEM is off, we
can turn it into an inline function that returns zero. This will catch
callers that give it bad arguments.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch encapsulates the usage of eff_cap (in netlink_skb_params) within
the security framework by extending security_netlink_recv to include a required
capability parameter and converting all direct usage of eff_caps outside
of the lsm modules to use the interface. It also updates the SELinux
implementation of the security_netlink_send and security_netlink_recv
hooks to take advantage of the sid in the netlink_skb_params struct.
This also enables SELinux to perform auditing of netlink capability checks.
Please apply, for 2.6.18 if possible.
Signed-off-by: Darrel Goeddel <dgoeddel@trustedcs.com>
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When GSO packets come from an untrusted source (e.g., a Xen guest domain),
we need to verify the header integrity before passing it to the hardware.
Since the first step in GSO is to verify the header, we can reuse that
code by adding a new bit to gso_type: SKB_GSO_DODGY. Packets with this
bit set can only be fed directly to devices with the corresponding bit
NETIF_F_GSO_ROBUST. If the device doesn't have that bit, then the skb
is fed to the GSO engine which will allow the packet to be sent to the
hardware if it passes the header check.
This patch changes the sg flag to a full features flag. The same method
can be used to implement TSO ECN support. We simply have to mark packets
with CWR set with SKB_GSO_ECN so that only hardware with a corresponding
NETIF_F_TSO_ECN can accept them. The GSO engine can either fully segment
the packet, or segment the first MTU and pass the rest to the hardware for
further segmentation.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netpoll system currently has a rx to tx path via:
netpoll_rx
__netpoll_rx
arp_reply
netpoll_send_skb
dev->hard_start_tx
This rx->tx loop places network drivers at risk of inadvertently causing a
deadlock or BUG halt by recursively trying to acquire a spinlock that is
used in both their rx and tx paths (this problem was origionally reported
to me in the 3c59x driver, which shares a spinlock between the
boomerang_interrupt and boomerang_start_xmit routines).
This patch breaks this loop, by queueing arp frames, so that they can be
responded to after all receive operations have been completed. Tested by
myself and the reported with successful results.
Specifically it was tested with netdump. Heres the BZ with details:
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=194055
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When transmitting a skb in netpoll_send_skb(), only retry a limited number
of times if the device queue is stopped.
Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Acked-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_find_text takes a "to" argument which is supposed to limit how
far into the skb it will search for the given text. At present,
it seems to ignore that argument on the first skb, and instead
return a match even if the text occurs beyond the limit.
Patch below fixes this, after adjusting for the "from" starting
point. This consequently fixes the netfilter string match's "--to"
handling, which currently is broken.
Signed-off-by: Phil Oester <kernel@linuxace.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix 2 problems in dev_hard_start_xmit():
1. nskb->next needs to link back to skb->next if hard_start_xmit()
returns non-zero.
2. Since the total number of GSO fragments may exceed MAX_SKB_FRAGS + 1,
it needs to stop transmitting if the netif_queue is stopped.
Signed-off-by: Michael Chan <mchan@broadcom.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
list_splice_init(list, head) does unneeded job if it is known that
list_empty(head) == 1. We can use list_replace_init() instead.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Warning(/var/linsrc/linux-2617-g4//include/linux/skbuff.h:304): No description found for parameter 'dma_cookie'
Warning(/var/linsrc/linux-2617-g4//include/net/sock.h:1274): No description found for parameter 'copied_early'
Warning(/var/linsrc/linux-2617-g4//net/core/dev.c:3309): No description found for parameter 'chan'
Warning(/var/linsrc/linux-2617-g4//net/core/dev.c:3309): No description found for parameter 'event'
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a generic segmentation offload toggle that can be turned
on/off for each net device. For now it only supports in TCPv4.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the GSO implementation for IPv4 TCP.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the infrastructure for generic segmentation offload.
The idea is to tap into the potential savings of TSO without hardware
support by postponing the allocation of segmented skb's until just
before the entry point into the NIC driver.
The same structure can be used to support software IPv6 TSO, as well as
UFO and segmentation offload for other relevant protocols, e.g., DCCP.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Having separate fields in sk_buff for TSO/UFO (tso_size/ufo_size) is not
going to scale if we add any more segmentation methods (e.g., DCCP). So
let's merge them.
They were used to tell the protocol of a packet. This function has been
subsumed by the new gso_type field. This is essentially a set of netdev
feature bits (shifted by 16 bits) that are required to process a specific
skb. As such it's easy to tell whether a given device can process a GSO
skb: you just have to and the gso_type field and the netdev's features
field.
I've made gso_type a conjunction. The idea is that you have a base type
(e.g., SKB_GSO_TCPV4) that can be modified further to support new features.
For example, if we add a hardware TSO type that supports ECN, they would
declare NETIF_F_TSO | NETIF_F_TSO_ECN. All TSO packets with CWR set would
have a gso_type of SKB_GSO_TCPV4 | SKB_GSO_TCPV4_ECN while all other TSO
packets would be SKB_GSO_TCPV4. This means that only the CWR packets need
to be emulated in software.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The dev_deactivate function has bit-rotted since the introduction of
lockless drivers. In particular, the spin_unlock_wait call at the end
has no effect on the xmit routine of lockless drivers.
With a little bit of work, we can make it much more useful by providing
the guarantee that when it returns, no more calls to the xmit routine
of the underlying driver will be made.
The idea is simple. There are two entry points in to the xmit routine.
The first comes from dev_queue_xmit. That one is easily stopped by
using synchronize_rcu. This works because we set the qdisc to noop_qdisc
before the synchronize_rcu call. That in turn causes all subsequent
packets sent to dev_queue_xmit to be dropped. The synchronize_rcu call
also ensures all outstanding calls leave their critical section.
The other entry point is from qdisc_run. Since we now have a bit that
indicates whether it's running, all we have to do is to wait until the
bit is off.
I've removed the loop to wait for __LINK_STATE_SCHED to clear. This is
useless because netif_wake_queue can cause it to be set again. It is
also harmless because we've disarmed qdisc_run.
I've also removed the spin_unlock_wait on xmit_lock because its only
purpose of making sure that all outstanding xmit_lock holders have
exited is also given by dev_watchdog_down.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
First of all it is unnecessary to allocate a new skb in skb_pad since
the existing one is not shared. More importantly, our hard_start_xmit
interface does not allow a new skb to be allocated since that breaks
requeueing.
This patch uses pskb_expand_head to expand the existing skb and linearize
it if needed. Actually, someone should sift through every instance of
skb_pad on a non-linear skb as they do not fit the reasons why this was
originally created.
Incidentally, this fixes a minor bug when the skb is cloned (tcpdump,
TCP, etc.). As it is skb_pad will simply write over a cloned skb. Because
of the position of the write it is unlikely to cause problems but still
it's best if we don't do it.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function ethtool_get_ufo was referring to ETHTOOL_GTSO instead of
ETHTOOL_GUFO.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current stack treats NETIF_F_HW_CSUM and NETIF_F_NO_CSUM
identically so we test for them in quite a few places. For the sake
of brevity, I'm adding the macro NETIF_F_GEN_CSUM for these two. We
also test the disjunct of NETIF_F_IP_CSUM and the other two in various
places, for that purpose I've added NETIF_F_ALL_CSUM.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
It's better to warn and fail rather than rarely triggering BUG on paths
that incorrectly call skb_trim/__skb_trim on a non-linear skb.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The linearisation operation doesn't need to be super-optimised. So we can
replace __skb_linearize with __pskb_pull_tail which does the same thing but
is more general.
Also, most users of skb_linearize end up testing whether the skb is linear
or not so it helps to make skb_linearize do just that.
Some callers of skb_linearize also use it to copy cloned data, so it's
useful to have a new function skb_linearize_cow to copy the data if it's
either non-linear or cloned.
Last but not least, I've removed the gfp argument since nobody uses it
anymore. If it's ever needed we can easily add it back.
Misc bugs fixed by this patch:
* via-velocity error handling (also, no SG => no frags)
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Various drivers use xmit_lock internally to synchronise with their
transmission routines. They do so without setting xmit_lock_owner.
This is fine as long as netpoll is not in use.
With netpoll it is possible for deadlocks to occur if xmit_lock_owner
isn't set. This is because if a printk occurs while xmit_lock is held
and xmit_lock_owner is not set can cause netpoll to attempt to take
xmit_lock recursively.
While it is possible to resolve this by getting netpoll to use
trylock, it is suboptimal because netpoll's sole objective is to
maximise the chance of getting the printk out on the wire. So
delaying or dropping the message is to be avoided as much as possible.
So the only alternative is to always set xmit_lock_owner. The
following patch does this by introducing the netif_tx_lock family of
functions that take care of setting/unsetting xmit_lock_owner.
I renamed xmit_lock to _xmit_lock to indicate that it should not be
used directly. I didn't provide irq versions of the netif_tx_lock
functions since xmit_lock is meant to be a BH-disabling lock.
This is pretty much a straight text substitution except for a small
bug fix in winbond. It currently uses
netif_stop_queue/spin_unlock_wait to stop transmission. This is
unsafe as an IRQ can potentially wake up the queue. So it is safer to
use netif_tx_disable.
The hamradio bits used spin_lock_irq but it is unnecessary as
xmit_lock must never be taken in an IRQ handler.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a secmark field to the skbuff structure, to allow security subsystems to
place security markings on network packets. This is similar to the nfmark
field, except is intended for implementing security policy, rather than than
networking policy.
This patch was already acked in principle by Dave Miller.
Signed-off-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Any socket recv of less than this ammount will not be offloaded
Signed-off-by: Chris Leech <christopher.leech@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Adds an async_wait_queue and some additional fields to tcp_sock, and a
dma_cookie_t to sk_buff.
Signed-off-by: Chris Leech <christopher.leech@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Provides for pinning user space pages in memory, copying to iovecs,
and copying from sk_buffs including fragmented and chained sk_buffs.
Signed-off-by: Chris Leech <christopher.leech@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Attempts to allocate per-CPU DMA channels
Signed-off-by: Chris Leech <christopher.leech@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Noticed that dev_alloc_name() comment was incorrect, and more spellung
errors.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The classical IP over ATM code maintains its own IPv4 <-> <ATM stuff>
ARP table, using the standard neighbour-table code. The
neigh_table_init function adds this neighbour table to a linked list
of all neighbor tables which is used by the functions neigh_delete()
neigh_add() and neightbl_set(), all called by the netlink code.
Once the ATM neighbour table is added to the list, there are two
tables with family == AF_INET there, and ARP entries sent via netlink
go into the first table with matching family. This is indeterminate
and often wrong.
To see the bug, on a kernel with CLIP enabled, create a standard IPv4
ARP entry by pinging an unused address on a local subnet. Then attempt
to complete that entry by doing
ip neigh replace <ip address> lladdr <some mac address> nud reachable
Looking at the ARP tables by using
ip neigh show
will reveal two ARP entries for the same address. One of these can be
found in /proc/net/arp, and the other in /proc/net/atm/arp.
This patch adds a new function, neigh_table_init_no_netlink() which
does everything the neigh_table_init() does, except add the table to
the netlink all-arp-tables chain. In addition neigh_table_init() has a
check that all tables on the chain have a distinct address family.
The init call in clip.c is changed to call
neigh_table_init_no_netlink().
Since ATM ARP tables are rather more complicated than can currently be
handled by the available rtattrs in the netlink protocol, no
functionality is lost by this patch, and non-ATM ARP manipulation via
netlink is rescued. A more complete solution would involve a rtattr
for ATM ARP entries and some way for the netlink code to give
neigh_add and friends more information than just address family with
which to find the correct ARP table.
[ I've changed the assertion checking in neigh_table_init() to not
use BUG_ON() while holding neigh_tbl_lock. Instead we remember that
we found an existing tbl with the same family, and after dropping
the lock we'll give a diagnostic kernel log message and a stack dump.
-DaveM ]
Signed-off-by: Simon Kelley <simon@thekelleys.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The last step of netdevice registration was being done by a delayed
call, but because it was delayed, it was impossible to return any error
code if the class_device registration failed.
Side effects:
* one state in registration process is unnecessary.
* register_netdevice can sleep inside class_device registration/hotplug
* code in netdev_run_todo only does unregistration so it is simpler.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The test used in the linkwatch does not handle wrap-arounds correctly.
Since the intention of the code is to eliminate bursts of messages we
can afford to delay things up to a second. Using that fact we can
easily handle wrap-arounds by making sure that we don't delay things
by more than one second.
This is based on diagnosis and a patch by Stefan Rompf.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Stefan Rompf <stefan@loplof.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
From: Alan Stern <stern@rowland.harvard.edu>
This chain does it's own locking via the RTNL semaphore, and
can also run recursively so adding a new mutex here was causing
deadlocks.
Signed-off-by: David S. Miller <davem@davemloft.net>
Atomically create attributes when class device is added. This avoids
the race between registering class_device (which generates hotplug
event), and the creation of attribute groups.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add some sanity checking. truesize should be at least sizeof(struct
sk_buff) plus the current packet length. If not, then truesize is
seriously mangled and deserves a kernel log message.
Currently we'll do the check for release of stream socket buffers.
But we can add checks to more spots over time.
Incorporating ideas from Herbert Xu.
Signed-off-by: David S. Miller <davem@davemloft.net>
I've just realised that the RtNetlink code does not check the
permission for SIOCGIWENCODE and SIOCGIWENCODEEXT, which means that
any user can read the encryption keys. The fix is trivial and should
go in 2.6.17 alonside the two other patch I sent you last week.
Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Check the permissions when user-space try to read the
encryption parameters via SIOCGIWENCODEEXT. This is trivial and
probably should go in 2.6.17...
Bug was found by Brian Eaton <eaton.lists@gmail.com>, thanks !
Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This patch fixes unaligned access warnings noticed on IA64
in sk_run_filter(). 'ptr' can be unaligned.
Signed-off-By: Dmitry Mishin <dim@openvz.org>
Signed-off-By: Kirill Korotaev <dev@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
for_each_cpu() actually iterates across all possible CPUs. We've had mistakes
in the past where people were using for_each_cpu() where they should have been
iterating across only online or present CPUs. This is inefficient and
possibly buggy.
We're renaming for_each_cpu() to for_each_possible_cpu() to avoid this in the
future.
This patch replaces for_each_cpu with for_each_possible_cpu under /net
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
From: Thomas de Grenier de Latour <degrenier@easyconnect.fr>
On Sun, 9 Apr 2006 21:56:59 +0400,
Sergey Vlasov <vsu@altlinux.ru> wrote:
> However, show_address() does not output anything unless
> dev->reg_state == NETREG_REGISTERED - and this state is set by
> netdev_run_todo() only after netdev_register_sysfs() returns, so in
> the meantime (while netdev_register_sysfs() is busy adding the
> "statistics" attribute group) some process may see an empty "address"
> attribute.
I've tried the attached patch, suggested by Sergey Vlasov on
hotplug-devel@, and as far as i can test it works just fine.
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes an off-by-21-or-49 error ;-) spotted by the Coverity
checker.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Put a comment in there explaining why we double the setsockopt()
caller's SO_RCVBUF. People keep wondering.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
On a allyesconfig'ured kernel:
Size Uses Wasted Name and definition
===== ==== ====== ================================================
95 162 12075 netif_wake_queue include/linux/netdevice.h
129 86 9265 dev_kfree_skb_any include/linux/netdevice.h
127 56 5885 netif_device_attach include/linux/netdevice.h
73 86 4505 dev_kfree_skb_irq include/linux/netdevice.h
46 60 1534 netif_device_detach include/linux/netdevice.h
119 16 1485 __netif_rx_schedule include/linux/netdevice.h
143 5 492 netif_rx_schedule include/linux/netdevice.h
81 7 366 netif_schedule include/linux/netdevice.h
netif_wake_queue is big because __netif_schedule is a big inline:
static inline void __netif_schedule(struct net_device *dev)
{
if (!test_and_set_bit(__LINK_STATE_SCHED, &dev->state)) {
unsigned long flags;
struct softnet_data *sd;
local_irq_save(flags);
sd = &__get_cpu_var(softnet_data);
dev->next_sched = sd->output_queue;
sd->output_queue = dev;
raise_softirq_irqoff(NET_TX_SOFTIRQ);
local_irq_restore(flags);
}
}
static inline void netif_wake_queue(struct net_device *dev)
{
#ifdef CONFIG_NETPOLL_TRAP
if (netpoll_trap())
return;
#endif
if (test_and_clear_bit(__LINK_STATE_XOFF, &dev->state))
__netif_schedule(dev);
}
By de-inlining __netif_schedule we are saving a lot of text
at each callsite of netif_wake_queue and netif_schedule.
__netif_rx_schedule is also big, and it makes more sense to keep
both of them out of line.
Patch also deinlines dev_kfree_skb_any. We can deinline dev_kfree_skb_irq
instead... oh well.
netif_device_attach/detach are not hot paths, we can deinline them too.
Signed-off-by: Denis Vlasenko <vda@ilport.com.ua>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The kernel's implementation of notifier chains is unsafe. There is no
protection against entries being added to or removed from a chain while the
chain is in use. The issues were discussed in this thread:
http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2
We noticed that notifier chains in the kernel fall into two basic usage
classes:
"Blocking" chains are always called from a process context
and the callout routines are allowed to sleep;
"Atomic" chains can be called from an atomic context and
the callout routines are not allowed to sleep.
We decided to codify this distinction and make it part of the API. Therefore
this set of patches introduces three new, parallel APIs: one for blocking
notifiers, one for atomic notifiers, and one for "raw" notifiers (which is
really just the old API under a new name). New kinds of data structures are
used for the heads of the chains, and new routines are defined for
registration, unregistration, and calling a chain. The three APIs are
explained in include/linux/notifier.h and their implementation is in
kernel/sys.c.
With atomic and blocking chains, the implementation guarantees that the chain
links will not be corrupted and that chain callers will not get messed up by
entries being added or removed. For raw chains the implementation provides no
guarantees at all; users of this API must provide their own protections. (The
idea was that situations may come up where the assumptions of the atomic and
blocking APIs are not appropriate, so it should be possible for users to
handle these things in their own way.)
There are some limitations, which should not be too hard to live with. For
atomic/blocking chains, registration and unregistration must always be done in
a process context since the chain is protected by a mutex/rwsem. Also, a
callout routine for a non-raw chain must not try to register or unregister
entries on its own chain. (This did happen in a couple of places and the code
had to be changed to avoid it.)
Since atomic chains may be called from within an NMI handler, they cannot use
spinlocks for synchronization. Instead we use RCU. The overhead falls almost
entirely in the unregister routine, which is okay since unregistration is much
less frequent that calling a chain.
Here is the list of chains that we adjusted and their classifications. None
of them use the raw API, so for the moment it is only a placeholder.
ATOMIC CHAINS
-------------
arch/i386/kernel/traps.c: i386die_chain
arch/ia64/kernel/traps.c: ia64die_chain
arch/powerpc/kernel/traps.c: powerpc_die_chain
arch/sparc64/kernel/traps.c: sparc64die_chain
arch/x86_64/kernel/traps.c: die_chain
drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list
kernel/panic.c: panic_notifier_list
kernel/profile.c: task_free_notifier
net/bluetooth/hci_core.c: hci_notifier
net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain
net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain
net/ipv6/addrconf.c: inet6addr_chain
net/netfilter/nf_conntrack_core.c: nf_conntrack_chain
net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain
net/netlink/af_netlink.c: netlink_chain
BLOCKING CHAINS
---------------
arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain
arch/s390/kernel/process.c: idle_chain
arch/x86_64/kernel/process.c idle_notifier
drivers/base/memory.c: memory_chain
drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list
drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list
drivers/macintosh/adb.c: adb_client_list
drivers/macintosh/via-pmu.c sleep_notifier_list
drivers/macintosh/via-pmu68k.c sleep_notifier_list
drivers/macintosh/windfarm_core.c wf_client_list
drivers/usb/core/notify.c usb_notifier_list
drivers/video/fbmem.c fb_notifier_list
kernel/cpu.c cpu_chain
kernel/module.c module_notify_list
kernel/profile.c munmap_notifier
kernel/profile.c task_exit_notifier
kernel/sys.c reboot_notifier_list
net/core/dev.c netdev_chain
net/decnet/dn_dev.c: dnaddr_chain
net/ipv4/devinet.c: inetaddr_chain
It's possible that some of these classifications are wrong. If they are,
please let us know or submit a patch to fix them. Note that any chain that
gets called very frequently should be atomic, because the rwsem read-locking
used for blocking chains is very likely to incur cache misses on SMP systems.
(However, if the chain's callout routines may sleep then the chain cannot be
atomic.)
The patch set was written by Alan Stern and Chandra Seetharaman, incorporating
material written by Keith Owens and suggestions from Paul McKenney and Andrew
Morton.
[jes@sgi.com: restructure the notifier chain initialization macros]
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com>
Signed-off-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Just noticed that request_sock.[ch] contain a useless assignment of
rskq_accept_head to itself. I assume this is a typo and the 2nd one
was supposed to be _tail. However, setting _tail to NULL is not
needed, so the patch below just drops the 2nd assignment.
Signed-off-By: Norbert Kiesel <nkiesel@tbdnetworks.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement the half-closed devices notifiation, by adding a new POLLRDHUP
(and its alias EPOLLRDHUP) bit to the existing poll/select sets. Since the
existing POLLHUP handling, that does not report correctly half-closed
devices, was feared to be changed, this implementation leaves the current
POLLHUP reporting unchanged and simply add a new bit that is set in the few
places where it makes sense. The same thing was discussed and conceptually
agreed quite some time ago:
http://lkml.org/lkml/2003/7/12/116
Since this new event bit is added to the existing Linux poll infrastruture,
even the existing poll/select system calls will be able to use it. As far
as the existing POLLHUP handling, the patch leaves it as is. The
pollrdhup-2.6.16.rc5-0.10.diff defines the POLLRDHUP for all the existing
archs and sets the bit in the six relevant files. The other attached diff
is the simple change required to sys/epoll.h to add the EPOLLRDHUP
definition.
There is "a stupid program" to test POLLRDHUP delivery here:
http://www.xmailserver.org/pollrdhup-test.c
It tests poll(2), but since the delivery is same epoll(2) will work equally.
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jörn Engel