[patch] add iocb to network protocols

[patch] add iocb to network protocols

[Benjamin LaHaise]

Hello Dave,

Below is a copy of the bk tree at:

	bk pull master.kernel.org:/home/bcrl/net-2.5

This tree adds the iocb parameter into the sendmsg and recvmsg 
operations within the network operations.  sock_read and sock_write 
are also replaced by sock_aio_read/sock_aio_write, so this requires 
the other aio changes to fs/read_write.c that are in fresh trees 
from Linus.  Comments?

This will update the following files:

 include/linux/net.h          |    9 ++
 include/net/inet_common.h    |    6 +
 include/net/sock.h           |   82 ++++++++++++++++---------
 include/net/tcp.h            |    5 -
 include/net/udp.h            |    3 
 net/atm/common.c             |    8 +-
 net/atm/common.h             |    8 +-
 net/ax25/af_ax25.c           |    7 +-
 net/bluetooth/af_bluetooth.c |    3 
 net/core/sock.c              |    8 +-
 net/decnet/af_decnet.c       |    8 +-
 net/econet/af_econet.c       |    9 +-
 net/ipv4/af_inet.c           |   17 ++---
 net/ipv4/raw.c               |    7 +-
 net/ipv4/tcp.c               |    5 -
 net/ipv4/udp.c               |    7 +-
 net/ipx/af_ipx.c             |    8 +-
 net/irda/af_irda.c           |    9 +-
 net/llc/af_llc.c             |    9 +-
 net/netlink/af_netlink.c     |    6 +
 net/netrom/af_netrom.c       |    8 +-
 net/packet/af_packet.c       |   10 +--
 net/rose/af_rose.c           |    9 +-
 net/socket.c                 |  141 ++++++++++++++++++++++++++-----------------
 net/unix/af_unix.c           |   12 ++-
 net/wanrouter/af_wanpipe.c   |    9 +-
 net/x25/af_x25.c             |    7 +-
 27 files changed, 255 insertions(+), 165 deletions(-)

through these ChangeSets:

<bcrl@redhat.com> (02/10/10 1.741)
   correct sock_aio_write prototype

<bcrl@redhat.com> (02/10/10 1.740)
   eliminate a compiler warning for aio_write in net/socket.c

<bcrl@bob.home.kvack.org> (02/10/10 1.733.4.2)
   clean up whitespace and patch import errors from net-kiocb patch

<bcrl@bob.home.kvack.org> (02/10/10 1.733.4.1)
   net-kiocb.diff


through these patches:

# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.740   -> 1.741  
#	        net/socket.c	1.31    -> 1.32   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 02/10/10	bcrl@redhat.com	1.741
# correct sock_aio_write prototype
# --------------------------------------------
#
diff -Nru a/net/socket.c b/net/socket.c
--- a/net/socket.c	Thu Oct 10 18:27:23 2002
+++ b/net/socket.c	Thu Oct 10 18:27:23 2002
@@ -92,7 +92,7 @@
 static int sock_no_open(struct inode *irrelevant, struct file *dontcare);
 static ssize_t sock_aio_read(struct kiocb *iocb, char *buf,
 			 size_t size, loff_t pos);
-static ssize_t sock_aio_write(struct kiocb *iocb, char *buf,
+static ssize_t sock_aio_write(struct kiocb *iocb, const char *buf,
 			  size_t size, loff_t pos);
 static int sock_mmap(struct file *file, struct vm_area_struct * vma);
 
# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.739   -> 1.740  
#	        net/socket.c	1.30    -> 1.31   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 02/10/10	bcrl@redhat.com	1.740
# eliminate a compiler warning for aio_write in net/socket.c
# --------------------------------------------
#
diff -Nru a/net/socket.c b/net/socket.c
--- a/net/socket.c	Thu Oct 10 18:27:24 2002
+++ b/net/socket.c	Thu Oct 10 18:27:24 2002
@@ -619,7 +619,7 @@
  *	is readable by the user process.
  */
 
-static ssize_t sock_aio_write(struct kiocb *iocb, char *ubuf,
+static ssize_t sock_aio_write(struct kiocb *iocb, const char *ubuf,
 			  size_t size, loff_t pos)
 {
 	struct sock_iocb *x = kiocb_to_siocb(iocb);
# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.733.4.1 -> 1.733.4.2
#	        net/socket.c	1.29    -> 1.30   
#	  include/net/sock.h	1.21    -> 1.22   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 02/10/10	torvalds@penguin.transmeta.com	1.733.1.27
# Clean up after timers - move the "timers" Makefile info
# into the proper subdirectory (kernel) where it is used.
# 
# Drop unused variables.
# --------------------------------------------
# 02/10/10	axboe@suse.de	1.733.1.28
# [PATCH] excessive stack usage in cdrom
# 
# CD-ROM puts struct cdrom_changer_info on the stack in a few places, this
# is a bad idea since it's big (a bit over 1kb).  This makes us allocate
# it instead.
# 
# Noticed by Anton.
# --------------------------------------------
# 02/10/10	sam@ravnborg.org	1.733.1.29
# [PATCH] drivers/scsi - Makefile fix
# 
# Reference to .ver file incorrect after recent makefile changes.
# Grepped the kernel tree, and this is the only Makefile that
# uses $(MODVERDIR).
# --------------------------------------------
# 02/10/10	olaf.dietsche#list.linux-kernel@t-online.de	1.733.1.30
# [PATCH] 2.5.40: fix chmod/chown on procfs
# 
# This patch allows to change uid, gid and mode of files and directories
# located in procfs.
# 
# Without this patch you can change uid, gid and mode as long as the
# file is open. As soon as you close the file, it reverts back to its
# default, which is root:root and readonly usually.
# --------------------------------------------
# 02/10/10	paulus@samba.org	1.733.1.31
# [PATCH] add PCI device ID for Motorola MPC107
# 
# This patch adds the PCI device ID for the Motorola MPC107 host bridge.
# The entry is already in the list at pciids.sf.net but isn't in the
# kernel pci_ids.h file yet.  Please apply this to your tree.
# --------------------------------------------
# 02/10/10	paulus@samba.org	1.733.1.32
# [PATCH] adjust PPC sysctls
# 
# This patch takes out the unused KERN_PPC_ZEROPAGED sysctl, and
# restricts the KERN_PPC_POWERSAVE_NAP and KERN_PPC_L2CR sysctls to be
# present only on those PPC processors where they are useful.  This
# patch only affects PPC.
# --------------------------------------------
# 02/10/10	Andries.Brouwer@cwi.nl	1.733.1.33
# [PATCH] isofs fix
# 
# The patch below removes some dead code and nonsense code.
# The part that changes behaviour is
# 
#  -       if (sbi->s_cruft == 'n' &&
#  -           (volume_seq_no != 0) && (volume_seq_no != 1)) {
#  -               printk(KERN_WARNING "Warning: defective CD-ROM "
#  -                      "(volume sequence number %d). "
#  -                      "Enabling \"cruft\" mount option.\n", volume_seq_no);
#  -               sbi->s_cruft = 'y';
#  -       }
# 
# that has already bitten lots of people.
# 
# Nothing is wrong with a volume sequence number different from 0 or 1.
# (Cf. Ecma-119.pdf, Sections 4.17, 4.18, 6.6.)
# --------------------------------------------
# 02/10/10	torvalds@penguin.transmeta.com	1.733.1.34
# Merge http://linux-isdn.bkbits.net/linux-2.5.make
# into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.35
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# Instantiate a new file, include/linux/nfs4.h, which contains
# constants and typedef's for the NFSv4 protocol (by analogy with
# include/linux/nfs2.h and include/linux/nfs3.h).
# 
# Also #include this file in a few places where it will be needed
# later.
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.36
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# In a number of places in the NFS client, I had to change
# 
#   #ifdef CONFIG_NFS_V3
#      /* ... */
#   #endif
# 
# to
# 
#   #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
#      /* ... */
#   #endif
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.37
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# This patch changes the interface of the ->readdir() nfs_rpc_op
# so that its first argument is a dentry instead of an inode.
# 
# [Explanation: The dentry is required because in NFSv4, we need
#  to make use of the _parent_ directory's inode.  This is because
#  NFSv4 servers no longer return an entry for ".." in the READDIR
#  response, so the client kernel needs to fake this entry, inode
#  number and all.]
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.38
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# This patch changes the interface of the ->setattr() nfs_rpc_op
# so that its first argument is a dentry instead of an inode.
# 
# [Explanation: The dentry is required because in NFSv4, we may
#  need to OPEN the file before doing the SETATTR.  (This is
#  required if the file size is changed as part of the setattr.)
#  Opening the file requires making use of the containing
#  directory's inode.]
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.39
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# In NFSv4, there is no hard limit on the length of symlink text.
# This patch changes the -ENAMETOOLONG test in nfs_symlink() accordingly.
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.40
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# In NFSv4, an fsid is a 64-bit major number together with a 64-bit
# minor number.  In previous versions, an fsid is a single number.
# This patch changes 'struct nfs_fattr' accordingly.
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.41
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# This is a nontrivial change to the NFS client.
# 
# NFSv4 defines a new file attribute, change_attr.  This is a per-file
# opaque quantity returned by the server, whose value is required to
# change whenever the file is modified.  If it exists, we want to use
# it for all cache consistency checks in nfs_refresh_inode().  Some
# operations also return a "pre-operation" value of the change_attr;
# we want to take this into account too.
# 
# First, define flags
#   NFS_ATTR_FATTR_V4 - indicates that the 'struct nfs_fattr' is an
#                       NFSv4 fattr, so the change_attr field is valid
#   NFS_ATTR_PRE_CHANGE - indicates that the server returned a pre-operation
#                       change_attr, so the pre_change_attr field is valid
# 
# Second, change nfs_refresh_inode() so that the caches are invalidated
# if there is a change_attr mismatch.  Exception: If the pre_change_attr
# tells us that the mismatch was caused by our operation, then do not
# invalidate the caches.
# 
# This patch should leave the logic in nfs_refresh_inode() unchanged
# if neither of the new flags are set.
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.42
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# If the NFS_ATTR_FATTR_V4 flag is set, use the NFSv3 convention for
# the 'space_used' part of the fattr.
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.43
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# This is a nontrivial change to the NFS client.
# 
# Synchronous READ operations are currently done via the ->read() nfs_rpc_op.
# Therefore, the synchronous READ path can easily be adapted for NFSv4.  On
# the other hand, the asynchronous READ path contains several NFSv3-specific
# features, which make it difficult to adapt for NFSv4.
# 
# In this patch and the next, we modify the async READ path to be
# version-agnostic.  This patch just changes the 'struct nfs_read_data'
# so that the v2- and v3-specific parts are moved into a private area,
# with room for a v4-specific part in parallel.  None of the logic is
# changed.
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.44
# [PATCH] A basic NFSv4 client for 2.5.x
# 
# This is a nontrivial change to the NFS client.
# 
# In this patch, we finish modifying the async READ path so that it is
# version-agnostic.  We define a new nfs_rpc_op ->setup_read(), and move
# the v2- and v3-specific code in nfs_read_rpcsetup() there.  We also
# have to change nfs_readpage() result so that the 'count' of bytes
# read is a parameter.  The extra parameter means that it can no longer
# be ->tk_exit().  Instead, it is called from a version-specific ->tk_exit()
# routine which is set in ->read_setup().
# 
# The upshot of all this is that the version-specific part of the
# async READ path has been encapsulated in a new nfs_rpc_op
# ->read_setup(), and NFSv4 can share the logic for asynchronous
# READ's with NFSv2 and v3.
# --------------------------------------------
# 02/10/10	trond.myklebust@fys.uio.no	1.733.1.45
# [PATCH] Fix NFS locking over TCP
# 
# The 2.5.x RPC code is currently broken in that it demands that all
# tasks that call xprt_create_proto() in order to open a TCP socket must
# have CAP_NET_BIND_SERVICE capabilities, and must bind to a privileged
# port.
# 
# This breaks the NLM locking code and its use of the call_bind() RPC
# portmapper lookup feature.
# 
# This patch allows the built-in portmapper client to use unbound TCP
# sockets if the user does not have the necessary capabilities.
# --------------------------------------------
# 02/10/10	greg@kroah.com	1.733.1.46
# [PATCH] minor i386 timer changes for 2.5.41
# 
# Here's an additional patch that contains the cleanups I did to John's
# timer patches.  It does the following:
# 
# 	- uses C99 initializers
# 	- makes the timer list static
# 	- adds better documentation to the timer function structure
# 	- makes the timer init function return 0 on success
# 	- NULL terminates the list of timers to make further patches
# 	  easier.
# --------------------------------------------
# 02/10/10	dledford@redhat.com	1.733.1.47
# [PATCH] atp870 driver
# 
# This is a minimal patch to allow me to load/use the atp module so I can do
# further testing work on it.
# --------------------------------------------
# 02/10/10	bcrl@bob.home.kvack.org	1.733.4.2
# clean up whitespace and patch import errors from net-kiocb patch
# --------------------------------------------
#
diff -Nru a/include/net/sock.h b/include/net/sock.h
--- a/include/net/sock.h	Thu Oct 10 18:27:25 2002
+++ b/include/net/sock.h	Thu Oct 10 18:27:25 2002
@@ -318,30 +318,6 @@
 	return container_of((void *)si, struct kiocb, private);
 }
 
-/* sock_iocb: used to kick off async processing of socket ios */
-struct sock_iocb {
-	struct list_head	list;
-
-	int			flags;
-	int			size;
-	struct socket		*sock;
-	struct sock		*sk;
-	struct msghdr		*msg, async_msg;
-	struct iovec		async_iov;
-	struct scm_cookie	*scm, async_scm;
-};
-
-static inline struct sock_iocb *kiocb_to_siocb(struct kiocb *iocb)
-{
-	BUG_ON(sizeof(struct sock_iocb) > KIOCB_PRIVATE_SIZE);
-	return (struct sock_iocb *)iocb->private;
-}
-
-static inline struct kiocb *siocb_to_kiocb(struct sock_iocb *si)
-{
-	return container_of((void *)si, struct kiocb, private);
-}
-
 /* Used by processes to "lock" a socket state, so that
  * interrupts and bottom half handlers won't change it
  * from under us. It essentially blocks any incoming
diff -Nru a/net/socket.c b/net/socket.c
--- a/net/socket.c	Thu Oct 10 18:27:25 2002
+++ b/net/socket.c	Thu Oct 10 18:27:25 2002
@@ -117,7 +117,7 @@
 
 static struct file_operations socket_file_ops = {
 	.llseek =	no_llseek,
-	.aio_read =		sock_aio_read,
+	.aio_read =	sock_aio_read,
 	.aio_write =	sock_aio_write,
 	.poll =		sock_poll,
 	.ioctl =	sock_ioctl,
@@ -541,7 +541,7 @@
 	struct kiocb iocb;
 	int ret;
 
-        init_sync_kiocb(&iocb, NULL);
+	init_sync_kiocb(&iocb, NULL);
 	ret = __sock_sendmsg(&iocb, sock, msg, size);
 	if (-EIOCBQUEUED == ret)
 		ret = wait_on_sync_kiocb(&iocb);
# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.733.1.15 -> 1.733.4.1
#	    net/llc/af_llc.c	1.31    -> 1.32   
#	      net/ipv4/raw.c	1.11    -> 1.12   
#	   include/net/tcp.h	1.19    -> 1.20   
#	  net/ipv4/af_inet.c	1.19    -> 1.20   
#	    net/ipx/af_ipx.c	1.21    -> 1.22   
#	      net/ipv4/udp.c	1.11    -> 1.12   
#	net/bluetooth/af_bluetooth.c	1.6     -> 1.7    
#	      net/ipv4/tcp.c	1.27    -> 1.28   
#	  net/irda/af_irda.c	1.31    -> 1.32   
#	        net/socket.c	1.28    -> 1.29   
#	 include/linux/net.h	1.4     -> 1.5    
#	net/econet/af_econet.c	1.10    -> 1.11   
#	  net/rose/af_rose.c	1.17    -> 1.18   
#	net/netrom/af_netrom.c	1.20    -> 1.21   
#	net/netlink/af_netlink.c	1.11    -> 1.12   
#	net/decnet/af_decnet.c	1.17    -> 1.18   
#	  net/unix/af_unix.c	1.28    -> 1.29   
#	net/packet/af_packet.c	1.15    -> 1.16   
#	  include/net/sock.h	1.19    -> 1.21   
#	     net/core/sock.c	1.12    -> 1.13   
#	include/net/inet_common.h	1.3     -> 1.4    
#	    net/atm/common.h	1.1     -> 1.2    
#	   include/net/udp.h	1.4     -> 1.5    
#	  net/ax25/af_ax25.c	1.15    -> 1.16   
#	    net/x25/af_x25.c	1.20    -> 1.21   
#	net/wanrouter/af_wanpipe.c	1.9     -> 1.10   
#	    net/atm/common.c	1.9     -> 1.10   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 02/10/10	bcrl@redhat.com	1.736
# Merge redhat.com:/md0/linus-2.5 into redhat.com:/md0/aio-2.5
# --------------------------------------------
# 02/10/10	mingo@elte.hu	1.733.1.16
# [PATCH] timer cleanups
# 
# This is my latest timer patchset, it makes del_timer_sync() a bit more
# robust wrt.  code that re-adds timers from the timer handler.
# 
# Other changes in the patch:
# 
#  - clean up cascading a bit.
# 
#  - do not save flags in __run_timer_list - we enter from an irqs-enabled
#    tasklet.
# --------------------------------------------
# 02/10/10	akpm@digeo.com	1.733.1.17
# [PATCH] mremap use-after-free bugfix
# 
# I have invented a new software development methodology! You send an
# email to Hugh saying "I don't have the foggiest idea why this guy's
# kernel is oopsing" and next morning, you get a patch! I shall patent
# this.
# 
# Since 2.5.3, move_vma() has been passing a freed vma into
# move_page_tables().  Fix it to move back to the previous vma in the
# list if we're about to delete this one.
# 
# Thanks to Morten Helgesen for patient reporting, diagnosis and testing.
# --------------------------------------------
# 02/10/10	akpm@digeo.com	1.733.1.18
# [PATCH] move_one_page atomicity fix
# 
# The atomicicty fix for move_one_page() was not quite right.
# 
# We only do the page_table_present() test if CONFIG_HIGHPTE=y.  Which is
# fine, but even with CONFIG_HIGHPTE=n, the pte mapping functions still
# do an inc_preempt_count() due to their unconditional kmap_atomic().  So
# we get a might_sleep() warning.
# 
# The warning is actually bogus, because those pte's are always in
# direct-mapped memory.
# 
# So hm.  Three fixes suggest themselves:
# 
# 1: Run the page_table_present() test if CONFIG_HIGHMEM.
# 
#    Rejected: penalises non-pte_highmem setups
# 
# 2: Make kmap_atomic() not do inc_preempt_count() is the page was
#    direct mapped.
# 
#    Rejected: I don't think we want kmap_atomic side effects to be
#    varying according to the page which was passed.
# 
# 3: Change the pte mapping functions so they don't run kmap_atomic at
#    all if CONFIG_HIGHPTE=n
# 
#    This is what I did.  And guess what?  For CONFIG_HIGHMEM=y,
#    CONFIG_HIGHPTE=n this patch shrinks the kernel by 5 kbytes.  Because
#    kmap_atomic is inlined.
# 
#    The lesson: we do way too much damn inlining.
# --------------------------------------------
# 02/10/10	akpm@digeo.com	1.733.1.19
# [PATCH] fix the raw driver
# 
# Fix the raw driver by tricking it into performing O_DIRECT IO against
# the bound blockdev.
# 
# - rewrite the i_mapping for /dev/raw/raw0 to point at the same thing
#   as bdev->bd_inode->i_mapping.  We've performed a bdget() against the
#   blockdev, which should pin it for the correct lifetime.
# 
# - set the O_DIRECT bit on the caller's file->flags.
# --------------------------------------------
# 02/10/10	akpm@digeo.com	1.733.1.20
# [PATCH] remove radix_tree_reserve()
# 
# From Hugh Dickins.
# 
# radix_tree_reserve() exists solely for the tmpfs move_to_swap_cache()
# and move_from_swap_cache() functions, and yet they don't need it: there
# is no problem in the one page being simultaneously listed in two radix
# trees (while both locks are held).  Use radix_tree_insert(), and remove
# radix_tree_reserve(); also removed a few blank lines.
# --------------------------------------------
# 02/10/10	akpm@digeo.com	1.733.1.21
# [PATCH] remove the sched_yield from the ext3 fsync path
# 
# The changed sched_yield() semantics have made ext3's transaction
# batching terribly slow.
# 
# Apparently a schedule() fixes that, although it probably breaks
# transaction batching.
# 
# This patch largely fixes my complaints about the new scheduler being
# extremely sluggish to interactive applications.  Evidently those
# applications were calling fsync() and were spending extremely long
# periods in sched_yield().
# --------------------------------------------
# 02/10/10	akpm@digeo.com	1.733.1.22
# [PATCH] make readv/writev return 0 for 0 segments
# 
# Should resolve an ongoing fiasco concerning what we should return to
# userspace if they do a readv or writev of zero segments.
# 
# SuS is ambiguous, but implies EINVAL.  We're currently returning
# EINVAL, but 2.4 returns zero.
# 
# I think zero makes more sense, and it is what 2.4 does.
# --------------------------------------------
# 02/10/10	bcrl@redhat.com	1.737
# fix symbol export in fs/read_write.c
# --------------------------------------------
# 02/10/10	johnstul@us.ibm.com	1.733.1.23
# [PATCH] linux-2.5.41_timer-changes_A4 (1/3 - infrastructure)
# 
#         The i386 time.c code is turning into a mess. We've got multiple
# functions that do the same thing, only with different hardware, all
# surrounded #ifdefs and even more difficult to follow #ifndefs. George
# Anzinger is introducing a new ACPIpm time source, I'm going to attempt
# to add the cyclone counter as a time source, and in the future there
# will be HPET to deal with. These will not go in cleanly together as
# things are now.
# 
#         Inspired by suggestions from Alan, this collection of patches
# tries to clean up time.c by breaking out the PIT and TSC specific parts
# into their own files. Additionally the patch creates an abstract
# interface to use these existing time soruces, as well as make it easier
# to add future time sources.
# 
#         It introduces "struct timer_ops" which gives the time code a
# clear interface to use these different time sources. It also allows for
# clearer conditional compilation of these various time sources.
# 
#         This first patch (part 1 of 3) provides the infrastructure
# needed via the timer_ops structure, as well as the select_timer()
# function for choosing the best available timer.
# --------------------------------------------
# 02/10/10	johnstul@us.ibm.com	1.733.1.24
# [PATCH] linux-2.5.41_timer-changes_A4 (2/3 - bulk move)
# 
#         This is part 2 of 3 of my timer-change patch. Part 2 is just a
# bulk move of code out of time.c and into timer_pit.c and timer_tsc.c. No
# code is changed, only moved.
# 
#         Please note, this code will not compile without the final third
# part of this patch collection. This was done for readability alone.
# --------------------------------------------
# 02/10/10	johnstul@us.ibm.com	1.733.1.25
# [PATCH] linux-2.5.41_timer-changes_A4 (3/3 - integration)
# 
#         This is the final part 3 of 3 of my timer-change patch. Part 3
# integrates the moved code (from part 2) into the new infrastructure
# (from part 1).
# --------------------------------------------
# 02/10/10	johnstul@us.ibm.com	1.733.1.26
# [PATCH] linux-2.5.41_cyclone-timer_B2
# 
#         In order to demonstrate how new time-sources are added to my
# timer-changes patch. Here is my current version of my cyclone-timer
# patch for 2.5.41. This uses the infrastructure set up in the
# timer-changes_A4 patch set to add the cyclone counter (found on IBM
# Summit Based hardware) as a time-source.
# 
#         The current code is not enabled as it also depends on James
# Cleverdon's 2.5 summit patch, however it illustrates how cleanly new
# time-sources can be added.
# --------------------------------------------
# 02/10/10	stevef@smfhome1.austin.rr.com	1.733.3.1
# Initial check in of cifs filesystem version 0.54 for Linux 2.5 (to clean tree as one changeset)
# --------------------------------------------
# 02/10/10	bcrl@bob.home.kvack.org	1.733.4.1
# net-kiocb.diff
# --------------------------------------------
#
diff -Nru a/include/linux/net.h b/include/linux/net.h
--- a/include/linux/net.h	Thu Oct 10 18:27:27 2002
+++ b/include/linux/net.h	Thu Oct 10 18:27:27 2002
@@ -81,6 +81,7 @@
 struct scm_cookie;
 struct vm_area_struct;
 struct page;
+struct kiocb;
 
 struct proto_ops {
   int	family;
@@ -104,8 +105,12 @@
 			 char *optval, int optlen);
   int	(*getsockopt)	(struct socket *sock, int level, int optname,
 			 char *optval, int *optlen);
-  int   (*sendmsg)	(struct socket *sock, struct msghdr *m, int total_len, struct scm_cookie *scm);
-  int   (*recvmsg)	(struct socket *sock, struct msghdr *m, int total_len, int flags, struct scm_cookie *scm);
+  int   (*sendmsg)	(struct kiocb *iocb, struct socket *sock,
+			 struct msghdr *m, int total_len,
+			 struct scm_cookie *scm);
+  int   (*recvmsg)	(struct kiocb *iocb, struct socket *sock,
+			 struct msghdr *m, int total_len, int flags,
+			 struct scm_cookie *scm);
   int	(*mmap)		(struct file *file, struct socket *sock, struct vm_area_struct * vma);
   ssize_t (*sendpage)	(struct socket *sock, struct page *page, int offset, size_t size, int flags);
 };
diff -Nru a/include/net/inet_common.h b/include/net/inet_common.h
--- a/include/net/inet_common.h	Thu Oct 10 18:27:27 2002
+++ b/include/net/inet_common.h	Thu Oct 10 18:27:27 2002
@@ -20,10 +20,12 @@
 						   int addr_len, int flags);
 extern int			inet_accept(struct socket *sock, 
 					    struct socket *newsock, int flags);
-extern int			inet_recvmsg(struct socket *sock, 
+extern int			inet_recvmsg(struct kiocb *iocb,
+					     struct socket *sock, 
 					     struct msghdr *ubuf, 
 					     int size, int flags, struct scm_cookie *scm);
-extern int			inet_sendmsg(struct socket *sock, 
+extern int			inet_sendmsg(struct kiocb *iocb,
+					     struct socket *sock, 
 					     struct msghdr *msg, 
 					     int size, struct scm_cookie *scm);
 extern int			inet_shutdown(struct socket *sock, int how);
diff -Nru a/include/net/sock.h b/include/net/sock.h
--- a/include/net/sock.h	Thu Oct 10 18:27:27 2002
+++ b/include/net/sock.h	Thu Oct 10 18:27:27 2002
@@ -51,6 +51,7 @@
 
 #include <asm/atomic.h>
 #include <net/dst.h>
+#include <net/scm.h>	/* for sock_iocb */
 
 /*
  * This structure really needs to be cleaned up.
@@ -242,9 +243,10 @@
 	int			(*getsockopt)(struct sock *sk, int level, 
 					int optname, char *optval, 
 					int *option);  	 
-	int			(*sendmsg)(struct sock *sk, struct msghdr *msg,
-					   int len);
-	int			(*recvmsg)(struct sock *sk, struct msghdr *msg,
+	int			(*sendmsg)(struct kiocb *iocb, struct sock *sk,
+					   struct msghdr *msg, int len);
+	int			(*recvmsg)(struct kiocb *iocb, struct sock *sk,
+					   struct msghdr *msg,
 					int len, int noblock, int flags, 
 					int *addr_len);
 	int			(*bind)(struct sock *sk, 
@@ -292,7 +294,53 @@
 #define SOCK_BINDADDR_LOCK	4
 #define SOCK_BINDPORT_LOCK	8
 
+/* sock_iocb: used to kick off async processing of socket ios */
+struct sock_iocb {
+	struct list_head	list;
+
+	int			flags;
+	int			size;
+	struct socket		*sock;
+	struct sock		*sk;
+	struct msghdr		*msg, async_msg;
+	struct iovec		async_iov;
+	struct scm_cookie	*scm, async_scm;
+};
+
+static inline struct sock_iocb *kiocb_to_siocb(struct kiocb *iocb)
+{
+	BUG_ON(sizeof(struct sock_iocb) > KIOCB_PRIVATE_SIZE);
+	return (struct sock_iocb *)iocb->private;
+}
+
+static inline struct kiocb *siocb_to_kiocb(struct sock_iocb *si)
+{
+	return container_of((void *)si, struct kiocb, private);
+}
 
+/* sock_iocb: used to kick off async processing of socket ios */
+struct sock_iocb {
+	struct list_head	list;
+
+	int			flags;
+	int			size;
+	struct socket		*sock;
+	struct sock		*sk;
+	struct msghdr		*msg, async_msg;
+	struct iovec		async_iov;
+	struct scm_cookie	*scm, async_scm;
+};
+
+static inline struct sock_iocb *kiocb_to_siocb(struct kiocb *iocb)
+{
+	BUG_ON(sizeof(struct sock_iocb) > KIOCB_PRIVATE_SIZE);
+	return (struct sock_iocb *)iocb->private;
+}
+
+static inline struct kiocb *siocb_to_kiocb(struct sock_iocb *si)
+{
+	return container_of((void *)si, struct kiocb, private);
+}
 
 /* Used by processes to "lock" a socket state, so that
  * interrupts and bottom half handlers won't change it
@@ -390,10 +438,10 @@
 						   char *, int *);
 extern int			sock_no_setsockopt(struct socket *, int, int,
 						   char *, int);
-extern int                      sock_no_sendmsg(struct socket *,
+extern int                      sock_no_sendmsg(struct kiocb *, struct socket *,
 						struct msghdr *, int,
 						struct scm_cookie *);
-extern int                      sock_no_recvmsg(struct socket *,
+extern int                      sock_no_recvmsg(struct kiocb *, struct socket *,
 						struct msghdr *, int, int,
 						struct scm_cookie *);
 extern int			sock_no_mmap(struct file *file,
diff -Nru a/include/net/tcp.h b/include/net/tcp.h
--- a/include/net/tcp.h	Thu Oct 10 18:27:27 2002
+++ b/include/net/tcp.h	Thu Oct 10 18:27:27 2002
@@ -648,7 +648,8 @@
 
 extern int		    	tcp_v4_tw_remember_stamp(struct tcp_tw_bucket *tw);
 
-extern int			tcp_sendmsg(struct sock *sk, struct msghdr *msg, int size);
+extern int			tcp_sendmsg(struct kiocb *iocb, struct sock *sk,
+					    struct msghdr *msg, int size);
 extern ssize_t			tcp_sendpage(struct socket *sock, struct page *page, int offset, size_t size, int flags);
 
 extern int			tcp_ioctl(struct sock *sk, 
@@ -739,7 +740,7 @@
 					       int optname, char *optval, 
 					       int optlen);
 extern void			tcp_set_keepalive(struct sock *sk, int val);
-extern int			tcp_recvmsg(struct sock *sk, 
+extern int			tcp_recvmsg(struct kiocb *iocb, struct sock *sk,
 					    struct msghdr *msg,
 					    int len, int nonblock, 
 					    int flags, int *addr_len);
diff -Nru a/include/net/udp.h b/include/net/udp.h
--- a/include/net/udp.h	Thu Oct 10 18:27:27 2002
+++ b/include/net/udp.h	Thu Oct 10 18:27:27 2002
@@ -64,7 +64,8 @@
 extern int	udp_connect(struct sock *sk,
 			    struct sockaddr *usin, int addr_len);
 
-extern int	udp_sendmsg(struct sock *sk, struct msghdr *msg, int len);
+extern int	udp_sendmsg(struct kiocb *iocb, struct sock *sk,
+			    struct msghdr *msg, int len);
 
 extern int	udp_rcv(struct sk_buff *skb);
 extern int	udp_ioctl(struct sock *sk, int cmd, unsigned long arg);
diff -Nru a/net/atm/common.c b/net/atm/common.c
--- a/net/atm/common.c	Thu Oct 10 18:27:27 2002
+++ b/net/atm/common.c	Thu Oct 10 18:27:27 2002
@@ -336,8 +336,8 @@
 }
 
 
-int atm_recvmsg(struct socket *sock,struct msghdr *m,int total_len,
-    int flags,struct scm_cookie *scm)
+int atm_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m,
+		int total_len, int flags, struct scm_cookie *scm)
 {
 	DECLARE_WAITQUEUE(wait,current);
 	struct atm_vcc *vcc;
@@ -417,8 +417,8 @@
 }
 
 
-int atm_sendmsg(struct socket *sock,struct msghdr *m,int total_len,
-    struct scm_cookie *scm)
+int atm_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m,
+		int total_len, struct scm_cookie *scm)
 {
 	DECLARE_WAITQUEUE(wait,current);
 	struct atm_vcc *vcc;
diff -Nru a/net/atm/common.h b/net/atm/common.h
--- a/net/atm/common.h	Thu Oct 10 18:27:27 2002
+++ b/net/atm/common.h	Thu Oct 10 18:27:27 2002
@@ -13,10 +13,10 @@
 int atm_create(struct socket *sock,int protocol,int family);
 int atm_release(struct socket *sock);
 int atm_connect(struct socket *sock,int itf,short vpi,int vci);
-int atm_recvmsg(struct socket *sock,struct msghdr *m,int total_len,
-    int flags,struct scm_cookie *scm);
-int atm_sendmsg(struct socket *sock,struct msghdr *m,int total_len,
-  struct scm_cookie *scm);
+int atm_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m,
+		int total_len, int flags, struct scm_cookie *scm);
+int atm_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m,
+		int total_len, struct scm_cookie *scm);
 unsigned int atm_poll(struct file *file,struct socket *sock,poll_table *wait);
 int atm_ioctl(struct socket *sock,unsigned int cmd,unsigned long arg);
 int atm_setsockopt(struct socket *sock,int level,int optname,char *optval,
diff -Nru a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c
--- a/net/ax25/af_ax25.c	Thu Oct 10 18:27:27 2002
+++ b/net/ax25/af_ax25.c	Thu Oct 10 18:27:27 2002
@@ -1410,7 +1410,8 @@
 	return err;
 }
 
-static int ax25_sendmsg(struct socket *sock, struct msghdr *msg, int len,
+static int ax25_sendmsg(struct kiocb *iocb, struct socket *sock,
+			struct msghdr *msg, int len,
 	struct scm_cookie *scm)
 {
 	struct sockaddr_ax25 *usax = (struct sockaddr_ax25 *)msg->msg_name;
@@ -1588,8 +1589,8 @@
 	return err;
 }
 
-static int ax25_recvmsg(struct socket *sock, struct msghdr *msg, int size,
-	int flags, struct scm_cookie *scm)
+static int ax25_recvmsg(struct kiocb *iocb, struct socket *sock,
+	struct msghdr *msg, int size, int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct sk_buff *skb;
diff -Nru a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c
--- a/net/bluetooth/af_bluetooth.c	Thu Oct 10 18:27:27 2002
+++ b/net/bluetooth/af_bluetooth.c	Thu Oct 10 18:27:27 2002
@@ -207,7 +207,8 @@
 	return NULL;
 }
 
-int bluez_sock_recvmsg(struct socket *sock, struct msghdr *msg, int len, int flags, struct scm_cookie *scm)
+int bluez_sock_recvmsg(struct kiocb *iocb, struct socket *sock,
+	struct msghdr *msg, int len, int flags, struct scm_cookie *scm)
 {
 	int noblock = flags & MSG_DONTWAIT;
 	struct sock *sk = sock->sk;
diff -Nru a/net/core/sock.c b/net/core/sock.c
--- a/net/core/sock.c	Thu Oct 10 18:27:27 2002
+++ b/net/core/sock.c	Thu Oct 10 18:27:27 2002
@@ -1047,14 +1047,14 @@
 	return -EOPNOTSUPP;
 }
 
-int sock_no_sendmsg(struct socket *sock, struct msghdr *m, int flags,
-		    struct scm_cookie *scm)
+int sock_no_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m,
+		    int flags, struct scm_cookie *scm)
 {
 	return -EOPNOTSUPP;
 }
 
-int sock_no_recvmsg(struct socket *sock, struct msghdr *m, int len, int flags,
-		    struct scm_cookie *scm)
+int sock_no_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m,
+		    int len, int flags, struct scm_cookie *scm)
 {
 	return -EOPNOTSUPP;
 }
diff -Nru a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
--- a/net/decnet/af_decnet.c	Thu Oct 10 18:27:27 2002
+++ b/net/decnet/af_decnet.c	Thu Oct 10 18:27:27 2002
@@ -1733,8 +1733,8 @@
 }
 
 
-static int dn_recvmsg(struct socket *sock, struct msghdr *msg, int size,
-	int flags, struct scm_cookie *scm)
+static int dn_recvmsg(struct kiocb *iocb, struct socket *sock,
+	struct msghdr *msg, int size, int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct dn_scp *scp = DN_SK(sk);
@@ -1901,8 +1901,8 @@
 	return 0;
 }
 
-static int dn_sendmsg(struct socket *sock, struct msghdr *msg, int size, 
-	   struct scm_cookie *scm)
+static int dn_sendmsg(struct kiocb *iocb, struct socket *sock,
+	   struct msghdr *msg, int size, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct dn_scp *scp = DN_SK(sk);
diff -Nru a/net/econet/af_econet.c b/net/econet/af_econet.c
--- a/net/econet/af_econet.c	Thu Oct 10 18:27:27 2002
+++ b/net/econet/af_econet.c	Thu Oct 10 18:27:27 2002
@@ -97,8 +97,9 @@
  *	If necessary we block.
  */
 
-static int econet_recvmsg(struct socket *sock, struct msghdr *msg, int len,
-			  int flags, struct scm_cookie *scm)
+static int econet_recvmsg(struct kiocb *iocb, struct socket *sock,
+			  struct msghdr *msg, int len, int flags,
+			  struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct sk_buff *skb;
@@ -230,8 +231,8 @@
  *	and hence whether to use real Econet or the UDP emulation.
  */
 
-static int econet_sendmsg(struct socket *sock, struct msghdr *msg, int len,
-			  struct scm_cookie *scm)
+static int econet_sendmsg(struct kiocb *iocb, struct socket *sock,
+			  struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct sockaddr_ec *saddr=(struct sockaddr_ec *)msg->msg_name;
diff -Nru a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
--- a/net/ipv4/af_inet.c	Thu Oct 10 18:27:27 2002
+++ b/net/ipv4/af_inet.c	Thu Oct 10 18:27:27 2002
@@ -753,22 +753,23 @@
 }
 
 
-
-int inet_recvmsg(struct socket *sock, struct msghdr *msg, int size,
-		 int flags, struct scm_cookie *scm)
+int inet_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
+		 int size, int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	int addr_len = 0;
-	int err = sk->prot->recvmsg(sk, msg, size, flags & MSG_DONTWAIT,
-				    flags & ~MSG_DONTWAIT, &addr_len);
+	int err;
+
+	err = sk->prot->recvmsg(iocb, sk, msg, size, flags & MSG_DONTWAIT,
+				flags & ~MSG_DONTWAIT, &addr_len);
 	if (err >= 0)
 		msg->msg_namelen = addr_len;
 	return err;
 }
 
 
-int inet_sendmsg(struct socket *sock, struct msghdr *msg, int size,
-		 struct scm_cookie *scm)
+int inet_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
+		 int size, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 
@@ -776,7 +777,7 @@
 	if (!inet_sk(sk)->num && inet_autobind(sk))
 		return -EAGAIN;
 
-	return sk->prot->sendmsg(sk, msg, size);
+	return sk->prot->sendmsg(iocb, sk, msg, size);
 }
 
 int inet_shutdown(struct socket *sock, int how)
diff -Nru a/net/ipv4/raw.c b/net/ipv4/raw.c
--- a/net/ipv4/raw.c	Thu Oct 10 18:27:27 2002
+++ b/net/ipv4/raw.c	Thu Oct 10 18:27:27 2002
@@ -295,7 +295,8 @@
 	return 0;
 }
 
-static int raw_sendmsg(struct sock *sk, struct msghdr *msg, int len)
+static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+		       int len)
 {
 	struct inet_opt *inet = inet_sk(sk);
 	struct ipcm_cookie ipc;
@@ -476,8 +477,8 @@
  *	we return it, otherwise we block.
  */
 
-int raw_recvmsg(struct sock *sk, struct msghdr *msg, int len,
-		int noblock, int flags, int *addr_len)
+int raw_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+		int len, int noblock, int flags, int *addr_len)
 {
 	struct inet_opt *inet = inet_sk(sk);
 	int copied = 0;
diff -Nru a/net/ipv4/tcp.c b/net/ipv4/tcp.c
--- a/net/ipv4/tcp.c	Thu Oct 10 18:27:27 2002
+++ b/net/ipv4/tcp.c	Thu Oct 10 18:27:27 2002
@@ -1013,7 +1013,8 @@
 	return tmp;
 }
 
-int tcp_sendmsg(struct sock *sk, struct msghdr *msg, int size)
+int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+		int size)
 {
 	struct iovec *iov;
 	struct tcp_opt *tp = tcp_sk(sk);
@@ -1475,7 +1476,7 @@
  *	Probably, code can be easily improved even more.
  */
 
-int tcp_recvmsg(struct sock *sk, struct msghdr *msg,
+int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		int len, int nonblock, int flags, int *addr_len)
 {
 	struct tcp_opt *tp = tcp_sk(sk);
diff -Nru a/net/ipv4/udp.c b/net/ipv4/udp.c
--- a/net/ipv4/udp.c	Thu Oct 10 18:27:27 2002
+++ b/net/ipv4/udp.c	Thu Oct 10 18:27:27 2002
@@ -423,7 +423,8 @@
 				   fraglen);
 }
 
-int udp_sendmsg(struct sock *sk, struct msghdr *msg, int len)
+int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+		int len)
 {
 	struct inet_opt *inet = inet_sk(sk);
 	int ulen = len + sizeof(struct udphdr);
@@ -635,8 +636,8 @@
  * 	return it, otherwise we block.
  */
 
-int udp_recvmsg(struct sock *sk, struct msghdr *msg, int len,
-		int noblock, int flags, int *addr_len)
+int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+		int len, int noblock, int flags, int *addr_len)
 {
 	struct inet_opt *inet = inet_sk(sk);
   	struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
diff -Nru a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c
--- a/net/ipx/af_ipx.c	Thu Oct 10 18:27:27 2002
+++ b/net/ipx/af_ipx.c	Thu Oct 10 18:27:27 2002
@@ -2009,8 +2009,8 @@
 out:	return ret;
 }
 
-static int ipx_sendmsg(struct socket *sock, struct msghdr *msg, int len,
-	struct scm_cookie *scm)
+static int ipx_sendmsg(struct kiocb *iocb, struct socket *sock,
+	struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct ipx_opt *ipxs = ipx_sk(sk);
@@ -2069,8 +2069,8 @@
 }
 
 
-static int ipx_recvmsg(struct socket *sock, struct msghdr *msg, int size,
-		int flags, struct scm_cookie *scm)
+static int ipx_recvmsg(struct kiocb *iocb, struct socket *sock,
+		struct msghdr *msg, int size, int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct ipx_opt *ipxs = ipx_sk(sk);
diff -Nru a/net/irda/af_irda.c b/net/irda/af_irda.c
--- a/net/irda/af_irda.c	Thu Oct 10 18:27:27 2002
+++ b/net/irda/af_irda.c	Thu Oct 10 18:27:27 2002
@@ -1259,8 +1259,8 @@
  *    SEQPACK services. This is possible since it forces the client to
  *    fragment the message if necessary
  */
-static int irda_sendmsg(struct socket *sock, struct msghdr *msg, int len,
-			struct scm_cookie *scm)
+static int irda_sendmsg(struct kiocb *iocb, struct socket *sock,
+			struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct irda_sock *self;
@@ -1331,8 +1331,9 @@
  *    Try to receive message and copy it to user. The frame is discarded
  *    after being read, regardless of how much the user actually read
  */
-static int irda_recvmsg_dgram(struct socket *sock, struct msghdr *msg,
-			      int size, int flags, struct scm_cookie *scm)
+static int irda_recvmsg_dgram(struct kiocb *iocb, struct socket *sock,
+			      struct msghdr *msg, int size, int flags,
+			      struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct irda_sock *self = irda_sk(sk);
diff -Nru a/net/llc/af_llc.c b/net/llc/af_llc.c
--- a/net/llc/af_llc.c	Thu Oct 10 18:27:27 2002
+++ b/net/llc/af_llc.c	Thu Oct 10 18:27:27 2002
@@ -677,8 +677,9 @@
  *	Copy received data to the socket user.
  *	Returns non-negative upon success, negative otherwise.
  */
-static int llc_ui_recvmsg(struct socket *sock, struct msghdr *msg, int size,
-			  int flags, struct scm_cookie *scm)
+static int llc_ui_recvmsg(struct kiocb *iocb, struct socket *sock,
+			  struct msghdr *msg, int size, int flags,
+			  struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct sockaddr_llc *uaddr = (struct sockaddr_llc *)msg->msg_name;
@@ -731,8 +732,8 @@
  *	Transmit data provided by the socket user.
  *	Returns non-negative upon success, negative otherwise.
  */
-static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, int len,
-			  struct scm_cookie *scm)
+static int llc_ui_sendmsg(struct kiocb *iocb, struct socket *sock,
+			  struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct llc_opt *llc = llc_sk(sk);
diff -Nru a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
--- a/net/netlink/af_netlink.c	Thu Oct 10 18:27:27 2002
+++ b/net/netlink/af_netlink.c	Thu Oct 10 18:27:27 2002
@@ -570,7 +570,8 @@
 	read_unlock(&nl_table_lock);
 }
 
-static int netlink_sendmsg(struct socket *sock, struct msghdr *msg, int len,
+static int netlink_sendmsg(struct kiocb *iocb, struct socket *sock,
+			   struct msghdr *msg, int len,
 			   struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
@@ -639,7 +640,8 @@
 	return err;
 }
 
-static int netlink_recvmsg(struct socket *sock, struct msghdr *msg, int len,
+static int netlink_recvmsg(struct kiocb *iocb, struct socket *sock,
+			   struct msghdr *msg, int len,
 			   int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
diff -Nru a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
--- a/net/netrom/af_netrom.c	Thu Oct 10 18:27:27 2002
+++ b/net/netrom/af_netrom.c	Thu Oct 10 18:27:27 2002
@@ -966,7 +966,8 @@
 	return 1;
 }
 
-static int nr_sendmsg(struct socket *sock, struct msghdr *msg, int len, struct scm_cookie *scm)
+static int nr_sendmsg(struct kiocb *iocb, struct socket *sock,
+		      struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	nr_cb *nr = nr_sk(sk);
@@ -1056,8 +1057,9 @@
 	return len;
 }
 
-static int nr_recvmsg(struct socket *sock, struct msghdr *msg, int size, 
-	int flags, struct scm_cookie *scm)
+static int nr_recvmsg(struct kiocb *iocb, struct socket *sock,
+		      struct msghdr *msg, int size, int flags,
+		      struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct sockaddr_ax25 *sax = (struct sockaddr_ax25 *)msg->msg_name;
diff -Nru a/net/packet/af_packet.c b/net/packet/af_packet.c
--- a/net/packet/af_packet.c	Thu Oct 10 18:27:27 2002
+++ b/net/packet/af_packet.c	Thu Oct 10 18:27:27 2002
@@ -288,7 +288,8 @@
  *	protocol layers and you must therefore supply it with a complete frame
  */
  
-static int packet_sendmsg_spkt(struct socket *sock, struct msghdr *msg, int len,
+static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock,
+			       struct msghdr *msg, int len,
 			       struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
@@ -665,8 +666,8 @@
 #endif
 
 
-static int packet_sendmsg(struct socket *sock, struct msghdr *msg, int len,
-			  struct scm_cookie *scm)
+static int packet_sendmsg(struct kiocb *iocb, struct socket *sock,
+			  struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct sockaddr_ll *saddr=(struct sockaddr_ll *)msg->msg_name;
@@ -1020,7 +1021,8 @@
  *	If necessary we block.
  */
 
-static int packet_recvmsg(struct socket *sock, struct msghdr *msg, int len,
+static int packet_recvmsg(struct kiocb *iocb, struct socket *sock,
+			  struct msghdr *msg, int len,
 			  int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
diff -Nru a/net/rose/af_rose.c b/net/rose/af_rose.c
--- a/net/rose/af_rose.c	Thu Oct 10 18:27:27 2002
+++ b/net/rose/af_rose.c	Thu Oct 10 18:27:27 2002
@@ -1025,8 +1025,8 @@
 	return 1;
 }
 
-static int rose_sendmsg(struct socket *sock, struct msghdr *msg, int len,
-				struct scm_cookie *scm)
+static int rose_sendmsg(struct kiocb *iocb, struct socket *sock,
+			struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	rose_cb *rose = rose_sk(sk);
@@ -1189,8 +1189,9 @@
 }
 
 
-static int rose_recvmsg(struct socket *sock, struct msghdr *msg, int size,
-		   int flags, struct scm_cookie *scm)
+static int rose_recvmsg(struct kiocb *iocb, struct socket *sock,
+			struct msghdr *msg, int size, int flags,
+			struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	rose_cb *rose = rose_sk(sk);
diff -Nru a/net/socket.c b/net/socket.c
--- a/net/socket.c	Thu Oct 10 18:27:27 2002
+++ b/net/socket.c	Thu Oct 10 18:27:27 2002
@@ -90,10 +90,10 @@
 #include <linux/netfilter.h>
 
 static int sock_no_open(struct inode *irrelevant, struct file *dontcare);
-static ssize_t sock_read(struct file *file, char *buf,
-			 size_t size, loff_t *ppos);
-static ssize_t sock_write(struct file *file, const char *buf,
-			  size_t size, loff_t *ppos);
+static ssize_t sock_aio_read(struct kiocb *iocb, char *buf,
+			 size_t size, loff_t pos);
+static ssize_t sock_aio_write(struct kiocb *iocb, char *buf,
+			  size_t size, loff_t pos);
 static int sock_mmap(struct file *file, struct vm_area_struct * vma);
 
 static int sock_close(struct inode *inode, struct file *file);
@@ -117,8 +117,8 @@
 
 static struct file_operations socket_file_ops = {
 	.llseek =	no_llseek,
-	.read =		sock_read,
-	.write =	sock_write,
+	.aio_read =		sock_aio_read,
+	.aio_write =	sock_aio_write,
 	.poll =		sock_poll,
 	.ioctl =	sock_ioctl,
 	.mmap =		sock_mmap,
@@ -517,64 +517,100 @@
 	sock->file=NULL;
 }
 
-int sock_sendmsg(struct socket *sock, struct msghdr *msg, int size)
+static int __sock_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, int size)
 {
+	struct sock_iocb *si = kiocb_to_siocb(iocb);
 	int err;
-	struct scm_cookie scm;
 
-	err = scm_send(sock, msg, &scm);
+	si->scm = &si->async_scm;
+	si->sock = sock;
+	si->msg = msg;
+	si->size = size;
+
+	err = scm_send(sock, msg, si->scm);
 	if (err >= 0) {
-		err = sock->ops->sendmsg(sock, msg, size, &scm);
-		scm_destroy(&scm);
+		err = sock->ops->sendmsg(iocb, sock, msg, size, si->scm);
+		if (-EIOCBQUEUED != err)
+			scm_destroy(si->scm);
 	}
 	return err;
 }
 
-int sock_recvmsg(struct socket *sock, struct msghdr *msg, int size, int flags)
+int sock_sendmsg(struct socket *sock, struct msghdr *msg, int size)
+{
+	struct kiocb iocb;
+	int ret;
+
+        init_sync_kiocb(&iocb, NULL);
+	ret = __sock_sendmsg(&iocb, sock, msg, size);
+	if (-EIOCBQUEUED == ret)
+		ret = wait_on_sync_kiocb(&iocb);
+	return ret;
+}
+
+
+int __sock_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, int size, int flags)
 {
-	struct scm_cookie scm;
+	struct sock_iocb *si = kiocb_to_siocb(iocb);
 
-	memset(&scm, 0, sizeof(scm));
+	si->sock = sock;
+	si->scm = &si->async_scm;
+	si->sock = sock;
+	si->msg = msg;
+	si->size = size;
+	si->flags = flags;
 
-	size = sock->ops->recvmsg(sock, msg, size, flags, &scm);
+	memset(si->scm, 0, sizeof(*si->scm));
+
+	size = sock->ops->recvmsg(iocb, sock, msg, size, flags, si->scm);
 	if (size >= 0)
-		scm_recv(sock, msg, &scm, flags);
+		scm_recv(sock, msg, si->scm, flags);
 
 	return size;
 }
 
+int sock_recvmsg(struct socket *sock, struct msghdr *msg, int size, int flags)
+{
+	struct kiocb iocb;
+	int ret;
+
+        init_sync_kiocb(&iocb, NULL);
+	ret = __sock_recvmsg(&iocb, sock, msg, size, flags);
+	if (-EIOCBQUEUED == ret)
+		ret = wait_on_sync_kiocb(&iocb);
+	return ret;
+}
 
 /*
  *	Read data from a socket. ubuf is a user mode pointer. We make sure the user
  *	area ubuf...ubuf+size-1 is writable before asking the protocol.
  */
 
-static ssize_t sock_read(struct file *file, char *ubuf,
-			 size_t size, loff_t *ppos)
+static ssize_t sock_aio_read(struct kiocb *iocb, char *ubuf,
+			 size_t size, loff_t pos)
 {
+	struct sock_iocb *x = kiocb_to_siocb(iocb);
 	struct socket *sock;
-	struct iovec iov;
-	struct msghdr msg;
 	int flags;
 
-	if (ppos != &file->f_pos)
+	if (pos != 0)
 		return -ESPIPE;
 	if (size==0)		/* Match SYS5 behaviour */
 		return 0;
 
-	sock = SOCKET_I(file->f_dentry->d_inode); 
+	sock = SOCKET_I(iocb->ki_filp->f_dentry->d_inode); 
 
-	msg.msg_name=NULL;
-	msg.msg_namelen=0;
-	msg.msg_iov=&iov;
-	msg.msg_iovlen=1;
-	msg.msg_control=NULL;
-	msg.msg_controllen=0;
-	iov.iov_base=ubuf;
-	iov.iov_len=size;
-	flags = !(file->f_flags & O_NONBLOCK) ? 0 : MSG_DONTWAIT;
+	x->async_msg.msg_name = NULL;
+	x->async_msg.msg_namelen = 0;
+	x->async_msg.msg_iov = &x->async_iov;
+	x->async_msg.msg_iovlen = 1;
+	x->async_msg.msg_control = NULL;
+	x->async_msg.msg_controllen = 0;
+	x->async_iov.iov_base = ubuf;
+	x->async_iov.iov_len = size;
+	flags = !(iocb->ki_filp->f_flags & O_NONBLOCK) ? 0 : MSG_DONTWAIT;
 
-	return sock_recvmsg(sock, &msg, size, flags);
+	return __sock_recvmsg(iocb, sock, &x->async_msg, size, flags);
 }
 
 
@@ -583,33 +619,32 @@
  *	is readable by the user process.
  */
 
-static ssize_t sock_write(struct file *file, const char *ubuf,
-			  size_t size, loff_t *ppos)
+static ssize_t sock_aio_write(struct kiocb *iocb, char *ubuf,
+			  size_t size, loff_t pos)
 {
+	struct sock_iocb *x = kiocb_to_siocb(iocb);
 	struct socket *sock;
-	struct msghdr msg;
-	struct iovec iov;
 	
-	if (ppos != &file->f_pos)
+	if (pos != 0)
 		return -ESPIPE;
 	if(size==0)		/* Match SYS5 behaviour */
 		return 0;
 
-	sock = SOCKET_I(file->f_dentry->d_inode); 
+	sock = SOCKET_I(iocb->ki_filp->f_dentry->d_inode); 
 
-	msg.msg_name=NULL;
-	msg.msg_namelen=0;
-	msg.msg_iov=&iov;
-	msg.msg_iovlen=1;
-	msg.msg_control=NULL;
-	msg.msg_controllen=0;
-	msg.msg_flags=!(file->f_flags & O_NONBLOCK) ? 0 : MSG_DONTWAIT;
+	x->async_msg.msg_name = NULL;
+	x->async_msg.msg_namelen = 0;
+	x->async_msg.msg_iov = &x->async_iov;
+	x->async_msg.msg_iovlen = 1;
+	x->async_msg.msg_control = NULL;
+	x->async_msg.msg_controllen = 0;
+	x->async_msg.msg_flags = !(iocb->ki_filp->f_flags & O_NONBLOCK) ? 0 : MSG_DONTWAIT;
 	if (sock->type == SOCK_SEQPACKET)
-		msg.msg_flags |= MSG_EOR;
-	iov.iov_base=(void *)ubuf;
-	iov.iov_len=size;
+		x->async_msg.msg_flags |= MSG_EOR;
+	x->async_iov.iov_base = (void *)ubuf;
+	x->async_iov.iov_len = size;
 	
-	return sock_sendmsg(sock, &msg, size);
+	return __sock_sendmsg(iocb, sock, &x->async_msg, size);
 }
 
 ssize_t sock_sendpage(struct file *file, struct page *page,
diff -Nru a/net/unix/af_unix.c b/net/unix/af_unix.c
--- a/net/unix/af_unix.c	Thu Oct 10 18:27:27 2002
+++ b/net/unix/af_unix.c	Thu Oct 10 18:27:27 2002
@@ -1175,7 +1175,8 @@
  *	Send AF_UNIX data.
  */
 
-static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, int len,
+static int unix_dgram_sendmsg(struct kiocb *iocb, struct socket *sock,
+			      struct msghdr *msg, int len,
 			      struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
@@ -1307,7 +1308,8 @@
 }
 
 		
-static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, int len,
+static int unix_stream_sendmsg(struct kiocb *iocb, struct socket *sock,
+			       struct msghdr *msg, int len,
 			       struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
@@ -1415,7 +1417,8 @@
 	}
 }
 
-static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, int size,
+static int unix_dgram_recvmsg(struct kiocb *iocb, struct socket *sock,
+			      struct msghdr *msg, int size,
 			      int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
@@ -1517,7 +1520,8 @@
 
 
 
-static int unix_stream_recvmsg(struct socket *sock, struct msghdr *msg, int size,
+static int unix_stream_recvmsg(struct kiocb *iocb, struct socket *sock,
+			       struct msghdr *msg, int size,
 			       int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
diff -Nru a/net/wanrouter/af_wanpipe.c b/net/wanrouter/af_wanpipe.c
--- a/net/wanrouter/af_wanpipe.c	Thu Oct 10 18:27:27 2002
+++ b/net/wanrouter/af_wanpipe.c	Thu Oct 10 18:27:27 2002
@@ -540,8 +540,8 @@
  *      a packet is queued into sk->write_queue.
  *===========================================================*/
 
-static int wanpipe_sendmsg(struct socket *sock, struct msghdr *msg, int len,
-			  struct scm_cookie *scm)
+static int wanpipe_sendmsg(struct kiocb *iocb, struct socket *sock,
+			   struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	wanpipe_opt *wp;
 	struct sock *sk = sock->sk;
@@ -1647,8 +1647,9 @@
  *      to the user. If necessary we block.
  *===========================================================*/
 
-static int wanpipe_recvmsg(struct socket *sock, struct msghdr *msg, int len,
-			  int flags, struct scm_cookie *scm)
+static int wanpipe_recvmsg(struct kiocb *iocb, struct socket *sock,
+			   struct msghdr *msg, int len, int flags,
+			   struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct sk_buff *skb;
diff -Nru a/net/x25/af_x25.c b/net/x25/af_x25.c
--- a/net/x25/af_x25.c	Thu Oct 10 18:27:27 2002
+++ b/net/x25/af_x25.c	Thu Oct 10 18:27:27 2002
@@ -916,8 +916,8 @@
 	goto out;
 }
 
-static int x25_sendmsg(struct socket *sock, struct msghdr *msg, int len,
-		       struct scm_cookie *scm)
+static int x25_sendmsg(struct kiocb *iocb, struct socket *sock,
+		       struct msghdr *msg, int len, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;
 	struct x25_opt *x25 = x25_sk(sk);
@@ -1091,7 +1091,8 @@
 }
 
 
-static int x25_recvmsg(struct socket *sock, struct msghdr *msg, int size,
+static int x25_recvmsg(struct kiocb *iocb, struct socket *sock,
+		       struct msghdr *msg, int size,
 		       int flags, struct scm_cookie *scm)
 {
 	struct sock *sk = sock->sk;

Re: [patch] add iocb to network protocols

[David S. Miller]

+#include <net/scm.h>	/* for sock_iocb */

What is this?

Re: [patch] add iocb to network protocols

[David S. Miller]

   From: Benjamin LaHaise <bcrl@redhat.com>
   Date: Thu, 10 Oct 2002 19:40:25 -0400

   On Thu, Oct 10, 2002 at 04:22:40PM -0700, David S. Miller wrote:
   > 
   > +#include <net/scm.h>	/* for sock_iocb */
   > 
   > What is this?
   
   The private area of an iocb for network operations contains a struct 
   scm_cookie.

I know this.  But the comment is wrong, when you say "for FOO"
next to an include that means "I'm including this to get the
definition of FOO".

Just delete the comment entirely, it really isn't needed.

Otherwise looks ok at first glance.

Re: [patch] add iocb to network protocols

[Benjamin LaHaise]

On Thu, Oct 10, 2002 at 04:22:40PM -0700, David S. Miller wrote:
> 
> +#include <net/scm.h>	/* for sock_iocb */
> 
> What is this?

The private area of an iocb for network operations contains a struct 
scm_cookie.  The alternative is to allocate it via kmalloc when needed, 
but there isn't currently a way of identifying which network protocols 
require the scm_cookie.  Since most network protocols do not use it, 
would you rather remove the scm_cookie out of the default path and 
allow the protocol to choose to create it?  That would remove the 
dependancy on scm_cookie from sock_iocb entirely.

		-ben
-- 
"Do you seek knowledge in time travel?"

Re: [patch] add iocb to network protocols

[Steven Whitehouse]

Hi,

> 
> Hello Dave,
> 
> Below is a copy of the bk tree at:
> 
> 	bk pull master.kernel.org:/home/bcrl/net-2.5
> 
> This tree adds the iocb parameter into the sendmsg and recvmsg 
> operations within the network operations.  sock_read and sock_write 
> are also replaced by sock_aio_read/sock_aio_write, so this requires 
> the other aio changes to fs/read_write.c that are in fresh trees 
> from Linus.  Comments?
>
[patch etc. cut]

I have a few questions about the way aio interacts with the socket locking
and the consequences for atomicity of requests:

Given two aio requests, A and B both for the same socket, can I be sure
that they will complete in the order that I submit them ? Can I also
be sure that parts of the I/O request for A will not be mixed up with
B ? Will that still be true if one of the requests is aio and one a "normal"
send/recvmsg() for example ?

Currently with multiple sendmsg() calls (for example) because the socket
lock is dropped to allow packet delivery to the socket during periods of
waiting for events, it would be possible for our two I/O requests A and B
to be interleaved so that B could be sent out sandwiched between two
parts of request A. Of course you have to submit the I/O from separate
threads for this to happen, but I'm wondering what the implications are 
for aio in this area. Do the standards have anything to say in this area ?

Steve.

Re: [patch] add iocb to network protocols

[Benjamin LaHaise]

On Fri, Oct 11, 2002 at 10:31:23AM +0100, Steven Whitehouse wrote:
> Given two aio requests, A and B both for the same socket, can I be sure
> that they will complete in the order that I submit them ? Can I also
> be sure that parts of the I/O request for A will not be mixed up with
> B ? Will that still be true if one of the requests is aio and one a "normal"
> send/recvmsg() for example ?

The POSIX standard does not seem to require any ordering between requests, 
and some implementations take advantage of this by using threads to execute 
requests.  That said, providing intra request ordering for sockets is easy 
to do, and is one of the guarantees I'm trying to make as it allows the 
implementation to provide the same semantics as are required for things 
like zero copy tx.

		-ben
-- 
"Do you seek knowledge in time travel?"

Re: [patch] add iocb to network protocols

[kuznet]

Hello!

> The private area of an iocb for network operations contains a struct 
> scm_cookie.  The alternative is to allocate it via kmalloc when needed, 
> but there isn't currently a way of identifying which network protocols 
> require the scm_cookie.

Kill it. It was not killed in 2.4 by plain luck, sorry, misfortune.

However, I do not understand this. You need to pass forth and back
SCM_RIGHTS is some way. It is quite dubious to make this asynchronously.
The same is surely true about credentials, they should be latched
on entry, not at the moment of real execution.

Kill anyway, at least in synchronous calls this was my mistake.

Alexey

Re: [patch] add iocb to network protocols

[Benjamin LaHaise]

Hello,

On Sat, Oct 12, 2002 at 12:15:48AM +0400, kuznet@ms2.inr.ac.ru wrote:
> However, I do not understand this. You need to pass forth and back
> SCM_RIGHTS is some way. It is quite dubious to make this asynchronously.
> The same is surely true about credentials, they should be latched
> on entry, not at the moment of real execution.

You actually do get to execute synchronously in the user's context the 
first time the function is entered to queue the operation, so collecting 
the credentials on entry and storing them for future use will work.  I'll 
submit a cleanup that does this.

		-ben
-- 
"Do you seek knowledge in time travel?"

Re: [patch] add iocb to network protocols

[David S. Miller]

   From: Benjamin LaHaise <bcrl@redhat.com>
   Date: Thu, 10 Oct 2002 18:35:29 -0400

   Below is a copy of the bk tree at:
   
   	bk pull master.kernel.org:/home/bcrl/net-2.5

Pulled, thanks.

We can work out the issues Steve brought up independantly.