routine meant, so I thought I'd fool with the argument names in an effort to
make them more helpful.

  1. Currently rpc_queue_upcall returns -EPIPE if we make an upcall on a pipe
     that userland hasn't opened yet, and we timeout and retry later.  This
     can lead to an unnecessary delay on mount, because rpc.gssd is racing
     to open the newly created pipe while the nfs code is making the first
     upcall.  If rpc.gssd loses, then we end up with a delay equal to the
     length of the timeout.  So instead we allow rpc_queue_upcall to queue
     upcalls on pipes that aren't opened yet.  To deal with the case of
     other upcall-users (e.g., the name<->uid mapping upcall code) who
     do want to know if the pipe isn't open (in the name<->uid case you can
     choose just to map everyone to nobody if the user doesn't want to run
     idmapd), we add a flag parameter to rpc_mkpipe that allows us to choose
     the kind of behavior we want at the time we create the pipe.

  2. Currently gss_msg's are destroyed the moment they have been completely
     read (by the call to destroy_msg in rpc_pipe_read).  This means an
     rpc_wake_up is done then, and can't be done later (because the gss_msg is
     gone, along with gss_msg->waitq).  It will typically be some time yet
     before the downcall comes, so the woken-up processes will have to wait and
     retry later; as above this leads to unnecessary delays.  Also, since the
     gss_msg is deleted from the list of gss_msgs's, we forget that an upcall
     to get creds for the user in question is still pending, so multiple
     unnecessary upcalls will be made.  This patch changes gss_pipe_upcall to
     never update msg->copied so that rpc_pipe_read never destroys the message.
     Instead, we wait till a downcall arrives to remove the upcall, using the
     new function __rpc_purge_one_upcall, which searches the list of pending
     rpc_pipe_msg's on the inode as well as checking the current upcall, to
     deal with the case where rpc.gssd might preemptively create a context for
     a user that there's already a pending upcall for.  Note also that this
     means that repeated reads by rpc.gssd will return the same data until
     rpc.gssd does a downcall.  This also gives us a better chance of
     recovering from rpc.gssd crashes.

the kernel can return -EACCES when a user lacks credentials, instead of just
hanging until they kinit.

null calls that was specified in the mount options.

but gss_refresh depends on gss_cred_get_ctx returning NULL whenever the cred is
not up to date.  So, I replaced the single gss_cred_get_ctx by a
gss_cred_get_ctx and a gss_cred_get_uptodate_ctx.

the XID. Following the introduction of zero-copy under UDP, the data
may be entirely located in pages under skb_shinfo(skb)->frags[].

into home.transmeta.com:/home/torvalds/v2.5/linux

into osdl.org:/home/mochel/src/kernel/devel/linux-2.5-core

into home.transmeta.com:/home/torvalds/v2.5/linux

Patch from Nicolas Pitre

> Patch #1540/1 has had the following note added:
> 
> Ok, it seems that we've had this for years and years, and it hasn't been
> causing a problem.  We've now got -Wa,-mno-fpu which should catch anything
> which shouldn't be used anyway.
> 
> Could you send a new patch for this change please?
> 
> Update by: Russell King

Here it is.

Patch from Peter Milne

Peripheral Memory Mapped Register defs for Messaging, PBI, DMA, I2C

Patch from Nicolas Pitre

The -march=armv5te issue looks to be fixed in gcc-3.3.
Since gcc-3.3 is the only gcc version >= 3 that seems to correctly
compile kernels out of the box I think it would be reasonable to
apply this patch and no bother with older gcc versions.

Yet gcc-3.3 emits V5TE instructions only when -mcpu=xscale is used,
but that seems to have been fixed in current CVS so next gcc release
will use all available instructions properly with -march=armv5te.

Patch from Deepak Saxena

This patch adds some hooks to arch/arm/Makefile to enable proper building
of big-endian kernels.  It also adds an option to the the top level
arm KConfig that only allows a big-endian kernel to be built if the sub-arch
KConfig sets the CONFIG_ARCH_SUPPORTS_BIG_ENDIAN boolean to true. This is
mainly there to keep people from attemtpting to build a big-endian kernel
for a system that is not capable of running in big-endian mode.

Patch from Deepak Saxena

into home.transmeta.com:/home/torvalds/v2.5/linux

Rename the socket_info_t inside tcic.c to struct tcic_socket. At one
time there were at least five different "socket_info_t" within cardmgr
/ linux/drivers/pcmcia. Let's get rid of all of them to avoid
confusion in future.

Rename the socket_info_t inside tcic.c to struct i82365_socket. At one
time there were at least five different "socket_info_t" within cardmgr
/ linux/drivers/pcmcia. Let's get rid of all of them to avoid
confusion in future.

The i82365 doesn't work unless CONFIG_ISA is set -- see this from
i82365_init:

...

    sockets = 0;

#ifdef CONFIG_ISA
    isa_probe();
#endif

    if (sockets == 0) {
        printk("not found.\n");
        driver_unregister(&i82365_driver);
        return -ENODEV;
    }

...

So, remove all instances of CONFIG_ISA from i82365.c and add a
dependency to the Kconfig file.

Get rid of the

typedef struct pcmcia_socket socket_info_t;

by replacing all remaining usages of cs_internal's socket_info_t with
struct pcmcia_socket.

Linus said [on April 25th of this year]:
> There are two reasons I did pci_socket, and one of them is stale
> and the other one is a matter of taste.

As the second reason is stale now, too, unify yenta.c and
pci_socket.c.

Remove socket_no from struct pcmcia_bus_socket - it's only used in
printks, and the one additional level of indirection doesn't hurt there.

 ds.c |   10 ++++------
 1 files changed, 4 insertions(+), 6 deletions(-)

Replace the socket number with struct pcmcia_socket from mtd_bind_t
and bind_req_t.

 drivers/pcmcia/cs.c |    4 ++--
 drivers/pcmcia/ds.c |    6 +++---
 include/pcmcia/cs.h |    4 ++--
 3 files changed, 7 insertions(+), 7 deletions(-)

As "socket_no" is deprecated, replace it with struct pcmcia_socket in 
client_handle_t.

 bulkmem.c     |    4 ++--
 cs.c          |   22 ++++++++++------------
 cs_internal.h |    4 ++--
 3 files changed, 14 insertions(+), 16 deletions(-)

This patch updates the callbacks to the socket drivers to take the
corresponding struct pcmcia_socket as argument instead of the "socket
number".

 drivers/pcmcia/bulkmem.c    |    6 +++---
 drivers/pcmcia/cistpl.c     |    4 ++--
 drivers/pcmcia/cs.c         |   18 +++++++++---------
 drivers/pcmcia/i82092.c     |   36 +++++++++++++++++++++++-------------
 drivers/pcmcia/i82092aa.h   |   20 ++++++++++----------
 drivers/pcmcia/i82365.c     |   34 +++++++++++++++++++++++-----------
 drivers/pcmcia/pci_socket.c |   40 ++++++++++++++++++++--------------------
 drivers/pcmcia/rsrc_mgr.c   |    2 +-
 drivers/pcmcia/tcic.c       |   35 ++++++++++++++++++-----------------
 include/pcmcia/ss.h         |   22 ++++++++++++----------
 10 files changed, 121 insertions(+), 96 deletions(-)

Re-enable the suspend and resume methods for SA11xx PCMCIA devices,
and update the initialisation/cleanup code for Dominik's recent
changes.

Add a more sane socket registration interface.

Previously, it was messed up because Greg's struct class hadn't been
invented when I wrote the code: there may be multiple sockets per
"struct device", and there is the need for one "struct class_device"
for each socket.

 drivers/pcmcia/cs.c         |  217 +++++++++++++++++++++----------------------- drivers/pcmcia/ds.c         |   56 ++---------
 drivers/pcmcia/i82092.c     |   48 ++++-----
 drivers/pcmcia/i82365.c     |   41 ++++----
 drivers/pcmcia/pci_socket.c |   23 +---
 drivers/pcmcia/pci_socket.h |    2
 drivers/pcmcia/tcic.c       |   38 ++++---
 include/pcmcia/ss.h         |   25 +++--
 8 files changed, 210 insertions(+), 240 deletions(-)

Split up the socket initialization in cs.c into two pieces:

        pcmcia_register_socket / pcmcia_unregister_socket 
will focus on registering sockets with the device core in future, set
all necessary fields etc.

        pcmcia_add_socket / pcmcia_remove_socket 
will use an interface to the device class "pcmcia_socket_class" in
future, and mainly take care of what initialization cs.c needs to work
properly.

This patch is almost completely limited to shuffling code around

 cs.c |  157 +++++++++++++++++++++++++++++++++++-------------------------------- 1 files changed, 83 insertions(+), 74 deletions(-)

From Jeremy Fitzhardinge:

With the current system device changes (I picked them up in 2.5.70-mm8),
the system device class assumes that all system device drivers are
registered before any system devices are registered.

Unfortunately, this is often not the case.  CPU devices are registered
very early, but cpufreq registers drivers for them; since cpufreq
drivers can be loaded as modules, they clearly can't be registered
before the device is.

This patch keeps a list of all registered devices hanging off the system
device class.  When a new driver is registered, it calls the driver's
add() function with all existing devices.

Conversely, when a driver is unregistered, it calls the driver's
remove() function for all existing devices so the driver can clean up.

Note: the list in the class's embedded kset is used, rather than creating
a new field.

From Greg: 

I took out the other put_device() in the -bk tree in class_device_del() 
but forgot to remove this one. 

From Greg Kroah-Hartman

into home.transmeta.com:/home/torvalds/v2.5/linux

into kroah.com:/home/greg/linux/BK/i2c-2.5

into home.transmeta.com:/home/torvalds/v2.5/linux