IrDA patches on the way...

IrDA patches on the way...

[Jean Tourrilhes]

	Hi Linus,

	It's now a tradition : here a few patches for you to merge in
the 2.5.6 kernel. A few cleanup that have been waiting for a while,
and a few critical fixes that recently showed up.
	I've asked Jeff if he wanted to be my "Patch Penguin (TM)" for
Linux-IrDA patches, but he was not very enthousiastic about IrDA (I
don't understand why ;-) I'll make seperate patches for Marcello.
	Patches have been tested in 2.5.6-pre2 (SMP). All non critical
patches have been on the IrDA list for more than one month (in
particular all the changes to LAP Tx queue).
        Have fun...

        Jean


[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir256_bus_to_virt.diff :
----------------------
	o [CRITICA] Fix ISA FIR drivers for new DMA API
	<PCI FIR drivers are still broken and need fixing>

ir256_sock_connect_cli.diff :
---------------------------
	o [CRITICA] Fix socket connect to remove dangerous cli()
	<Tested on SMP>

ir256_irnet_disc_ind.diff :
-------------------------
	o [CORRECT] Fix IrNET disconnection to not reconnect but
	  instead to hangup pppd

ir256_lap_icmd_fix-4.diff :
-------------------------
	o [CORRECT] Fix Tx queue handling (remove race, keep packets in order)
	o [CORRECT] Synchronise window_size & line_capacity and make sure
	  we never forget to increase them (would stall Tx queue)
	o [FEATURE] Group common code out of if-then-else
	o [FEATURE] Don't harcode LAP header size, use proper constant
	o [FEATURE] Inline irlap_next_state() to decrease bloat

ir256_usb_cow_urballoc.diff :
---------------------------
	o [FEATURE] Don't use skb_cow() unless we really need to
	o [CORRECT] Reorder URB init to avoid races
	o [CORRECT] USB dealy adds processing time, not removes it
        <Following patch from Greg KH <greg@kroah.com> himself !!!>
	o [CRITICA] Use dynamically allocated URBs (instead of statically)

IrDA patches on the way...

[Jean Tourrilhes]

	Hi Marcelo,

	The next batch of IrDA patches for 2.4.X. Most of those
patches were integrated in kernel 2.5.09 (6 months ago) and available
for 2.4.X on my web page since, so consider them "tested". I rediffed
and tested the patches on 2.4.20-final.
	Apart from the small important fixes, the new IrTTP
implementation (safer) and the new Donauboe driver that replaces the
obsolete Toshoboe and support more hardware.

	Have fun...

	Jean

-----------------------------------------------------------

[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir241_flow_sched_lap_lmp-6.diff :
-------------------------------
		<Won't compile without ir241_flow_sched_ttp-6.diff>
	o [FEATURE] Reduce LAP Tx queue to 2 packets (from 10)
		Improve latency, reduce buffer usage
	o [FEATURE] LAP Tx queue not full notification (flow start)
		Poll higher layer to fill synchronously LAP window (7 packets)
	o [FEATURE] LMP LSAP scheduler
		Ensure Tx fairness between LSAPs (sockets, IrCOMM, IrNET...)

ir241_flow_sched_ttp-6.diff :
---------------------------
		<Won't compile without ir241_flow_sched_lap_lmp-6.diff>
	o [CORRECT] Fix race condition when starting todo timer
	o [CORRECT] Fix race condition when stopping higher layer
		Higher layer would think it is stopped and us it is started
	o [CORRECT] Give credit even if packets in Tx queue
		If Tx queue was stopped, could starve peer and deadlock
	o [CORRECT] Protect Rx credit update with spinlock
	o [CORRECT] Calculate properly self->avail_credit
		Didn't take into account queued Rx fragments
		Incremented even if Rx frame not delivered to higher layer
		-> would never stop the peer (i.e. not flow control)
		-> could become infinite
	o [CORRECT] Send credit when higher layer reenable receive
		Peer wouldn't restart Tx to us if flow stopped
	o [FEATURE] Implement LAP queue not full notification
		Lower latency, ...
	o [FEATURE] Reduce Tx queue to 8 packets (from 10)
		But make sure we can always send a full LAP window (7)
	o [FEATURE] Fix and optimise TTP flow control
		Make sure peer can always send a full LAP window (7)
		Minimise explicit credit updates (give_credit)
	o [FEATURE] Remove need for todo timer in Tx/Rx paths
		Less potential races, lower latency, lower context switches
		Could not use tasklet because broken API, better anyway ;-)

ir241_dongle_locking.diff :
-------------------------
	o [CORRECT] Load dongle module with irq disabled in irtty
	<Same fix need to go in irport, but irport doesn't work for me>

ir241_lsap_lap_close-2.diff :
---------------------------
		<apply after ir241_flow_sched_lap_lmp-6.diff to avoid fuzz>
	o [CORRECT] Cancel LSAP watchdog when putting socket back to listen
	o [CORRECT] Try to close LAP when closing LSAP still active
	        <Following patch from Felix Tang>
	o [CORRECT] Header fix for compile on Alpha architecture

ir241_irnet_simult_race-2.diff :
------------------------------
	o [CORRECT] Prevent dealock on simultaneous peer IrNET connections
		Only the primary peer will accept the IrNET connection

ir241_smc_msg.diff :
------------------
	        <Following patch from Jeff Snyder>
	o [CRITICA] Release the proper region and not NULL pointer
	o [FEATURE] Fix messages

ir241_donauboe.diff :
-------------------
	        <Following patch from Martin Lucina & Christian Gennerat>
	o [FEATURE] Rewrite of the toshoboe driver using documentation
	o [FEATURE] Support Donau oboe chipsets.
	o [FEATURE] FIR support
	o [CORRECT] Probe chip before opening
	o [FEATURE] suspend/resume support
	o [FEATURE] Numerous other improvements/cleanups
		<Currently, we keep the old toshoboe driver around>

ir241_checker.diff :
------------------
	<Need to apply after ir241_flow_sched_XXX.diff to avoid "offset">
	o [CORRECT] Fix two bugs found by the Stanford checker in IrDA
	o [CORRECT] Fix two bugs found by the Stanford checker in IrCOMM
	o [CORRECT] Fix one bug found by the Stanford checker in ali driver

IrDA patches on the way...

[Jean Tourrilhes]

	Hi Marcelo,

	First, an apology. The changes were not as bad as I though and
I only had to fix a few patches (but I could not verify that
yesterday).
	This is my IrDA update for 2.4.20. Those IrDA patches are
mostly the same collection that went in 2.5.8-pre2. It fixes some
important bug, or are trivally obvious fixes. I also added some new
drivers that were added in 2.5.22. All stuff really safe.
	My remaining patch queue is still pretty full (including
Stanford checker bugs - see my web page), but that will wait for
later.
	Patches tested on 2.4.19 and recompiled on 2.4.20-pre1.

	Have fun...

	Jean

-----------------------------------------------------------

[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir240_trivial_fixes-3.diff :
--------------------------
	o [CORRECT] Handle signals while IrSock is blocked on Tx
	o [CORRECT] Fix race condition in LAP when receiving with pf bit
	o [CRITICA] Prevent queuing Tx data before IrComm is ready
	o [FEATURE] Warn user of common misuse of IrLPT

ir240_sys_max_tx-2.diff :
-----------------------
	o [FEATURE] Allow tuning of Max Tx MTU to workaround spec contradiction

ir240_irnet_disc_ind_again.diff :
-------------------------------
	o [CORRECT] Correct fix for IrNET disconnect indication :
	  if socket is not connected, don't hangup, to allow passive operation

ir240_discovery_fixes.diff :
--------------------------
	<Need to apply after ir240_trivial_fixes-3.diff to avoid "offset">
	o [FEATURE] Propagate mode of discovery to higher protocols
	o [CORRECT] Disable passive discovery in ircomm and irlan
	  Prevent client and server to simultaneously connect to each other
	o [CORRECT] Force expiry of discovery log on LAP disconnect

ir240_usb_disconnect-2.diff :
----------------------------------
	o [CRITICA] Fix race condition between disconnect and the rest
	o [CRITICA] Force synchronous unlink of URBs in disconnect
	o [CRITICA] Cleanup instance if disconnect before close

ir240_nsc_ob6100.diff :
---------------------
        <Following patch from Kevin Thayer>
	o [FEATURE] Handle what is probably a new variant of NSC chip

ir240_irtty_stats.diff :
----------------------
        <Following patch from Frank Becker>
	o [FEATURE] Update dev tx stats at the right time

ir240_expiry_fix.diff :
---------------------
	o [CORRECT] Make discovery expiry work properly for non default
		discovery period/timeout

ir240_mcp2120_act200l_ma600_drivers-2.diff :
------------------------------------------
	        <Following patch from Felix Tang>
	o [FEATURE] Dongle driver for mcp2120/crystal hardware
	        <Following patch from Shimizu Takuja/Gerhard Bertelsmann>
	o [FEATURE] Dongle driver for ActiSys 200L hardware
	        <Following patch from Leung/me>
	o [FEATURE] Dongle driver for Mobile Action MA600 hardware

IrDA patches on the way...

[Jean Tourrilhes]

	Hi Jeff,

	I was fortunate enough to receive some very important driver
updates that I had been waiting for a while, so I'm sending you the
current content of my patch queue.
	Also included is patch for bugs found by the Stanford checker
and an update of MAINTAINERS to point to to proper mailing list. Also,
now the non-modular init is 100% proper.
	Patches tested on 2.5.25.

	Have fun...

	Jean

-----------------------------------------------------------

[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir255_hashbin_fixes-2.diff :
--------------------------
	o [CRITICA] Remove correct IAS Attribute/Object even if name is dup'ed
	o [CORRECT] Make irqueue 64 bit compliant (__u32 -> long)
	o [FEATURE] Don't use random handle for IrLMP handle, use self
		Remove dependancy on random generator in stack init

vlsi_ir-2.5.24-v0.4-patch :
-------------------------
	        <Following patch from Martin Diehl>
        * merge+sync with changes from recent kernels: pci_[sg]et_drvdata,
          __devexit_p, netdev->last_rx, irda header cleanup
        * add netdev tx_timeout which re-initializes the whole thing
        * add power management support consistent with pci driver api
        * major rework of the ring descriptor operations
        * make correct usage of consistent and streaming pci dma api
        * nuke last virt_to_bus() and friends
        * support MIR/FIR highspeed interaction pulse (SIP)
        * review all paths for packet-size issues (rx and tx)
        * fix an old issue requiring hw powercycle caused by a race
          between IrLAP and hardware when switching _back_ to default
          speed at LAP disconnect. This was opened by the complete async
          behaviour of netdev->xmit but didn't happen before your latency
          improvements went into the stack.
        * add driver status readout under /proc/driver/vlsi_ir/irda%
          For 2.5, this will probably go into driverfs once things have
          stabilized.
        * fix potential deadlock in speed changing code
        * make identical driver working for both 2.4 and 2.5
        * add __attribute__((packed)) to hardware-exposed struct
        * add suggested pci_dma_prep_single() to flush cpu cache before
          streaming dma buffer gets reused for busmastering

ir255_donauboe.diff :
-------------------
	        <Following patch from Martin Lucina & Christian Gennerat>
	o [FEATURE] Rewrite of the toshoboe driver using documentation
	o [FEATURE] Support Donau oboe chipsets.
	o [FEATURE] FIR support
	o [CORRECT] Probe chip before opening
	o [FEATURE] suspend/resume support
	o [FEATURE] Numerous other improvements/cleanups
		<Currently, we keep the old toshoboe driver around>
	o [FEATURE] Config.help for ma600 driver (unrelated ;-)

ir255_checker.diff-2 :
--------------------
	o [CORRECT] Fix two bugs found by the Stanford checker

ir255_nsc_speed-4.diff :
----------------------
	o [FEATURE] Cleanly change speed back to 9600bps
	o [CORRECT] Change speed under spinlock/irq disabled

ir255_comments.diff :
-------------------
	o [FEATURE] Update MAINTAINERS file
	o [FEATURE] Update OHCI comment in irda-usb

IrDA patches on the way...

[Jean Tourrilhes]

        Hi Jeff,

	More fixes. Thanks to Stanford. Also, fix previous fixes.
	Patch done on 2.5.20 and verified/tested on 2.5.21.

	Regards,

	Jean

-----------------------------------------------------------

[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir250_headers_init-2.diff :
-------------------------
	o [FEATURE] Use new kernel init/exit style, should fix static builds
	o [FEATURE] Reduce header dependancies
						Before	After
		net/irda/.depend		14917	13617 B
		drivers/net/irda/.depend	16134	14293 B
		irda full recompile		3'13	3'10

ir250_act200l_ma600_drivers.diff :
--------------------------------
	        <Following patch from Shimizu Takuja/Gerhard Bertelsmann>
	o [FEATURE] Dongle driver for ActiSys 200L hardware
	        <Following patch from Leung/me>
	o [FEATURE] Dongle driver for Mobile Action MA600 hardware

ir250_cache_wait_data-2.diff :
----------------------------
	o [CRITICA] Fix one instance were we forgot to clear LSAP cache
	o [CORRECT] Fix a bogus conversion to wait_event()
		The socket closure would never propagate to the app

ir250_checker.diff :
------------------
	o [CORRECT] Fix two bugs found by the Stanford checker in IrCOMM

ir250_usb_cleanup-2.diff :
------------------------
	o [FEATURE] Update various comments to current state
	o [CORRECT] Handle properly failure of URB with new speed
	o [CORRECT] Don't test for (self != NULL) after using it (doh !)
	o [FEATURE] Other minor cleanups

IrDA patches on the way...

[Jean Tourrilhes]

	Hi Jeff,

	Time to dump my current batch of IrDA patches in the
kernel. That will help the current "warning message" squashing
party. Tested on 2.5.15.
	Regards,

	Jean

-----------------------------------------------------------

[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir253_smc_msg.diff :
------------------
	        <Following patch from Jeff Snyder>
	o [CRITICA] Release the proper region and not NULL pointer
	o [FEATURE] Fix messages

ir253_long_set_bit.diff :
-----------------------
	        <Following patch from Paul Mackerras>
	o [CORRECT] Argument of set_bit and friends should be unsigned long
		Should fix all compile warnings ;-)

ir253_lsap_cache_fix.diff :
-------------------------
	        <Following patch from Christoph Bartelmus, mangled by me>
	o [CORRECT] replace the global LSAP cache with LSAP caches private
		to each LAP.
		Fix a bug where two simultaneous connections from two devices
		using the same LSAPs would get mixed up.
		Should also improve performance in similar cases. 

IrDA patches on the way...

[Jean Tourrilhes]

	Hi Jeff,

	Would you mind to push the following batch of patches to Linus ?
	The first two patches feature a partial rewrite of the TTP
layer that fixes lots of things. Has been on my web page for over a
month and tested intensively.
	The remaining patches are less interesting. You may have
followed discussions about the wait_event stuff ;-)
	Regards,

	Jean

--------------------------------------------------------

[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir258_flow_sched_lap_lmp-6.diff :
-------------------------------
		<Won't compile without ir258_flow_sched_ttp-6.diff>
	o [FEATURE] Reduce LAP Tx queue to 2 packets (from 10)
		Improve latency, reduce buffer usage
	o [FEATURE] LAP Tx queue not full notification (flow start)
		Poll higher layer to fill synchronously LAP window (7 packets)
	o [FEATURE] LMP LSAP scheduler
		Ensure Tx fairness between LSAPs (sockets, IrCOMM, IrNET...)

ir258_flow_sched_ttp-6.diff :
---------------------------
		<Won't compile without ir258_flow_sched_lap_lmp-6.diff>
	o [CORRECT] Fix race condition when starting todo timer
	o [CORRECT] Fix race condition when stopping higher layer
		Higher layer would think it is stopped and us it is started
	o [CORRECT] Give credit even if packets in Tx queue
		If Tx queue was stopped, could starve peer and deadlock
	o [CORRECT] Protect Rx credit update with spinlock
	o [CORRECT] Calculate properly self->avail_credit
		Didn't take into account queued Rx fragments
		Incremented even if Rx frame not delivered to higher layer
		-> would never stop the peer (i.e. not flow control)
		-> could become infinite
	o [CORRECT] Send credit when higher layer reenable receive
		Peer wouldn't restart Tx to us if flow stopped
	o [FEATURE] Implement LAP queue not full notification
		Lower latency, ...
	o [FEATURE] Reduce Tx queue to 8 packets (from 10)
		But make sure we can always send a full LAP window (7)
	o [FEATURE] Fix and optimise TTP flow control
		Make sure peer can always send a full LAP window (7)
		Minimise explicit credit updates (give_credit)
	o [FEATURE] Remove need for todo timer in Tx/Rx paths
		Less potential races, lower latency, lower context switches
		Could not use tasklet because broken API, better anyway ;-)

ir258_wait_event_fixes-2.diff :
-----------------------------
	        <Following patch from Martin Diehl, mangled by me>
	o [FEATURE] Replace interruptible_sleep_on() with wait_event().
		Most races were taken care off, but cleaner anyway

ir258_mcp2120_driver.diff :
-------------------------
	        <Following patch from Felix Tang>
	o [FEATURE] Dongle driver for mcp2120/crystal hardware

ir258_dongle_locking.diff :
-------------------------
	o [CORRECT] Load dongle module with irq disabled in irtty
	<Same fix need to go in irport, but irport doesn't work for me>

ir258_lsap_lap_close-2.diff :
---------------------------
		<apply after ir258_flow_sched_lap_lmp-6.diff to avoid fuzz>
	o [CORRECT] Cancel LSAP watchdog when putting socket back to listen
	o [CORRECT] Try to close LAP when closing LSAP still active
	        <Following patch from Felix Tang>
	o [CORRECT] Header fix for compile on Alpha architecture

ir258_irnet_simult_race-2.diff :
------------------------------
	o [CORRECT] Prevent dealock on simultaneous peer IrNET connections
		Only the primary peer will accept the IrNET connection

IrDA patches on the way...

[Jean Tourrilhes]

        Hi Linus,

	Do I need to say more ? A new batch of IrDA patches that have
been waiting on my web page and tested intensively with 2.5.7-SMP
(Marcelo will have to wait for his turn).
	This mostly include some race conditions in LAP and discovery,
correcting my previous IrNET fix and final fixing of the USB driver.
	Not included in this batch is the LSAP scheduler and my
partial rewrite of IrTTP, that will have to wait. Martin also promised
me a new version of the vlsi driver.
        Have fun...

        Jean

-------------------------------------------------------

[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir256_w83977af_bus_to_virt.diff :
-------------------------------
	o [CRITICA] Fix w83977af_ir FIR drivers for new DMA API
	<PCI FIR drivers are still broken and need fixing>

ir257_trivial_fixes-3.diff :
--------------------------
	o [CORRECT] Handle signals while IrSock is blocked on Tx
	o [CORRECT] Fix race condition in LAP when receiving with pf bit
	o [CRITICA] Prevent queuing Tx data before IrComm is ready
	o [FEATURE] Warn user of common misuse of IrLPT

ir257_sys_max_tx-2.diff :
-----------------------
	o [FEATURE] Allow tuning of Max Tx MTU to workaround spec contradiction

ir256_irnet_disc_ind_again.diff :
-------------------------------
	o [CORRECT] Correct fix for IrNET disconnect indication :
	  if socket is not connected, don't hangup, to allow passive operation

ir257_discovery_fixes.diff :
--------------------------
	<Need to apply after ir257_trivial_fixes-3.diff to avoid "offset">
	o [FEATURE] Propagate mode of discovery to higher protocols
	o [CORRECT] Disable passive discovery in ircomm and irlan
	  Prevent client and server to simultaneously connect to each other
	o [CORRECT] Force expiry of discovery log on LAP disconnect

ir257_usb_disconnect_atomic-2.diff :
----------------------------------
	o [CRITICA] Fix race condition between disconnect and the rest
	o [CRITICA] Force synchronous unlink of URBs in disconnect
	o [CRITICA] Cleanup instance if disconnect before close
        <Following patch from Martin Diehl>
	o [CRITICA] Call usb_submit_urb() with GPF_ATOMIC

ir257_nsc_ob6100.diff :
---------------------
        <Following patch from Kevin Thayer>
	o [FEATURE] Handle what is probably a new variant of NSC chip

ir257_irtty_stats.diff :
----------------------
        <Following patch from Frank Becker>
	o [FEATURE] Update dev tx stats at the right time

IrDA patches on the way...

[Jean Tourrilhes]

	Hi,

	You know the story : another set of IrDA patches are ready to
be incorporated in the Linux kernel. One one hand, minor cleanups of
the stack, one the other hand nasty bugs in IrLPT and IrLAN (that
could crash your kernel).
	Patches are mostly obvious and have been tested on 2.4.18-pre2
and 2.5.2-pre10 (because I still want to keep 2.4.X and 2.5.X in
sync). Some patches will apply to 2.5.X with harmless line
differences. The last patch, ir248_config-3.diff is only to sync 2.4.X
with 2.5.X (Marcelo missed one bit).
        Have fun...

        Jean


[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir247_irlan_dynalloc-2.diff :
---------------------------
	o [CRITICA] Don't pass the same skb to two different state machines
        <Following patch from Martin Diehl, modified by me>
	o [CRITICA] Don't flag netdev as NETIF_F_DYNALLOC, because not kmalloc

ir247_lpt_fix.diff :
------------------
	o [CRITICA] Provide a valid skb when calling irlmp_connect_response()
	o [FEATURE] Display something meaningfull in /proc/net/irda/ircomm

ir247_debug_warn.diff :
---------------------
	<Patch from Kai Germaschewski>
	o [FEATURE] Remove compiler warning when not using DEBUG

ir247_drv_region-2.diff :
-----------------------
	<Patch from Steven>
	o [CORRECT] Cleanup check_region()/request_region() in various drivers

ir248_config-3.diff :
-------------------
	<Already in 2.5.1, for 2.4.18 only>
	o [FEATURE] Remove useless and confusing config option

IrDA patches on the way...

[Jean Tourrilhes]

	Hi,

	Classical ritual : I'm going to send a bunch or IrDA patches
for inclusion in 2.4.17 and 2.5.1 kernel. Patch have been around on
IrDA mailing list and torture tested on 2.4.17-pre1, so are ready to
go in.
	Nothing really major this time. 2 fixes for cellular phones
(MIR and min_tx_turn_time), one crasher (self->tsap->lsap->lap) and a
bunch of cleanups.
	No major work on the IrDA stack is planned as yet for 2.5.X,
and most people seem happy with the stack as it is. I'll keep the
stack in <bugfix/minor cleanup/new driver> mode and try to keep 2.4.X
and 2.5.X identical to simplify maintainance until something big
happens (I've got some catch up to do on 802.11).
	Have fun...

	Jean

[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir245_vlsi_mir.diff :
-------------------
        <Patch from Martin Diehl>
	o [CORRECT] When MIR is requested, enable MIR on the chip, not FIR

ir245_sysctl-2.diff :
-------------------
	o [CORRECT] Check min/max boundary on various sysctl
	o [FEATURE] Add min_tx_turn_time sysctl and make it default to 10
	  This fixes problem with some broken cellular phones
	o [FEATURE] Simplify retry_count tests
	o [FEATURE] Allow to set the "no activity" event timeout via sysctl
	o [FEATURE] "no activity" event generated each multiple of timeout
	  and not only the first time. This allow apps to know is connectivity
	  is restored (they don't receive anymore the event).

ir245_af_ias.diff :
-----------------
	o [CORRECT] Restrict write access to IAS database to ROOT. Users can
	  only write IAS object attached to their own socket. This avoid apps
	  polluting each other.
	o [FEATURE] Empty IAS classname will refer to object attached to
	  present socket.

ir245_clean_listen.diff :
-----------------------
	o [FEATURE] Put the code to set the socket back to listen mode in
	  common functions instead of cut'n'pasted all over the place.
	o [CRITICA] When setting a socket back to listen mode, set
	  self->tsap->lsap->lap = NULL; to avoid crashing is LAP is
	  removed before the socket.
	o [CORRECT] Fix disconnect event generation. Doh !

ir247_config-2.diff :
-------------------
	o [FEATURE] Remove CONFIG_IRDA_OPTIONS. A menu for only 3 items !
	o [FEATURE] Set various CONFIG_IRDA options default to YES
	o [FEATURE] Advertise more aggressively usage of those options
	  in Configure.help. Too many bug reports of people not using those.
	o [FEATURE] Remove obsolete Configure.help entries

IrDA patches on the way...

[Jean Tourrilhes]

	Hi,

	After an important user reported a crash of the IrDA stack, I
did an audit of the disconnect/close code for race conditions. I also
kept in mine to looks for potential skb leaks. Well, that was some
work, but my quest was quite successfull ;-)
	I also fixed a connection setup deadlock that was an old
friend eluding me. And I've been sent an endianess patch that is
necessary for operation on PPC.
	Patches have been on the IrDA mailing list for one week and
have suffered my torture test since, so they are ready to go ;-)

	Regards,

	Jean


[FEATURE] : Add a new feature to the IrDA stack
[CORRECT] : Fix to have the correct/expected behaviour
[CRITICA] : Fix potential kernel crash

ir243_endian_fix.diff :
---------------------
	<Patch from Michel Denzer and Benjamin Herrenschmidt>
	o [CORRECT] Fix endianess in IrDA parameters management
	  Without this, the IrDA stack doesn't work on PPC :-(

ir243_usb_descr.diff :
--------------------
	<First part is from Martin Diehl>
	o [CORRECT] Use specific get class descriptor and not the standard one
	o [FEATURE] Reduce verbosity of the driver

ir243_lap_lmp_races-2.diff :
--------------------------
	o [CRITICA] Avoid race condition in LSAP closure
	o [CORRECT] Remove potential skb leaks
	o [CORRECT] Add MEDIA_BUSY_TIMER_EXPIRED event to avoid stack deadlock
	o [CORRECT] Do a proper disconnect on RECV_RD_RSP
	o [FEATURE] Cleanup handling of pending LAP connect & Ultra Tx
	o [FEATURE] Add small busy delay after discovery indication
	o [FEATURE] Remove discovery event postponing kludge
	o [FEATURE] cleanups, comments
	o [FEATURE] Reduce verbosity of the stack

ir243_ttp_sock_races-2.diff :
---------------------------
	o [CRITICA] Avoid race condition in TSAP closure
	o [CRITICA] Avoid race condition in socket closure
	o [CRITICA] Avoid various race conditions in IrNET
	o [CORRECT] Remove potential skb leaks
	o [CORRECT] Remove IrNET discovery log leak
	o [CORRECT] Don't export inline function (irda_lock)
	o [FEATURE] Use skb_orphan() instead of skb_clone() + kfree_skb()
	o [FEATURE] cleanups, comments
	o [FEATURE] Reduce verbosity of the stack

ir245_help_obex.diff :
--------------------
	o [FEATURE] Looks like somebody got confused about what IrOBEX is ;-)