Summary of changes from v2.6.10 to v2.6.11 ============================================ [PATCH] r8169: add ethtool_ops.{get_regs_len/get_regs} - ethtool_ops.{get_regs_len/get_regs} for r8169; - fix a dubious check: datasheet v1.21 claims on p.44 that io/memory space is exactly 256 bytes wide; - use SET_ETHTOOL_OPS(). Signed-off-by: Francois Romieu [PATCH] r8169: per device receive buffer size Turn the Rx receive buffer size into a per device variable. Signed-off-by: Francois Romieu [PATCH] r8169: code cleanup Cleanup/code removal: - MAX_ETH_FRAME_SIZE is not used; - removal of assertion for impossible condition (if it happens, it will _not_ take long to notice anyway) - introduce rtl8169_release_board() to factor out some code; - rtl8169_init_board: - some variables are not really needed nor do they help read the code; - more explicit name for label; - tp->{Rx/Tx}DescArray: no need to zeroize coherent DMA mapping. Signed-off-by: Francois Romieu [PATCH] r8169: enable MWI - enable Memory Write and Invalidate (disabled after reset); - fix wrong goto. Signed-off-by: Francois Romieu [PATCH] r8169: bump version number Help reviewers realize that the in-kernel driver has evolved lately. Signed-off-by: Francois Romieu [PATCH] r8169: sync the names of a few bits with the 8139cp driver Sync the names of the descriptor with these which are used in the 8139cp driver. Though not exactly identical the descriptors are forward compatible. Signed-off-by: Francois Romieu [PATCH] r8169: comment a gcc 2.95.x bug gcc 2.95.3 bug has been experienced on gcc 2.95.4. Signed-off-by: Francois Romieu [PATCH] r8169: Tx checksum offload SG and IP checksumming support on output. Signed-off-by: Francois Romieu [PATCH] r8169: advertise DMA to high memory Advertise the ability to DMA to high memory. Signed-off-by: Francois Romieu [PATCH] r8169: Rx checksum support Rx IP checksumming support. Signed-off-by: Francois Romieu [PATCH] r8169: vlan support 802.1Q support. Mostly stolen from the 8139cp.c driver. The relevant registers and descriptors bits are identical for both chipsets. Signed-off-by: Francois Romieu [PATCH] eepro100.c iomap conversion [PATCH] r8169: miscalculation of available Tx descriptors The count of available entries in the Tx descriptors ring is badly wrong. Signed-off-by: Francois Romieu [PATCH] r8169: hint for Tx flow control return 1 in start_xmit() when the required descriptors are not available and wait for more room. Signed-off-by: Francois Romieu [PATCH] r8169: TSO support. TSO support. Suggestion of Jeff Garzik. Signed-off-by: Francois Romieu [PATCH] r8169: Mac identifier extracted from Realtek's driver v2.2 Mac identifier extracted from Realtek's driver v2.2. Signed-off-by: Francois Romieu [netdrvr eepro100] fix pci_iomap() args and info msg that follows input: Joydump depends on gameport Signed-off-by: Vojtech Pavlik Patch-by: Olaf Hering [PATCH] r8169: default on disabling PCIDAC Default to disabling PCI DAC as this option appears unsafe on amd64 (original suggestion by Hans-Frieder Vogt ). The driver will typically report PCI System error when something goes wrong. The relevant interrupt is not masked any more and the driver can thus be disabled. Signed-off-by: Francois Romieu [PATCH] mc_filter on big-endian arch On Sat, 19 Jun 2004 17:37:37 -0400, Jeff Garzik wrote: > you would be kind enough to resend the non-via-rhine patches WRT mc_filter? Sure. Patch is for 2.6 (not rediffed, yell if it doesn't apply anymore). Btw, did you pick up the mc_filter patch for 2.4 via-rhine? This untested patch fixes hardware mc filters for tulip_core, winbond, and atp. Hopefully :-). Please review and test. Signed-off-by: Roger Luethi input: Fix oops in gamecon Signed-off-by: Vojtech Pavlik Patch-by: Peter Nelson input: Some HID devices have problems returning the HID class descriptor. Try a few times before giving up. Signed-off-by: Vojtech Pavlik input: Add support for Kensington ThinkingMouse PS/2 protocol. Signed-off-by: Vojtech Pavlik Patch-by: Peter Maydell input: Add ALPS touchpad driver, driver by Neil Brown, Peter Osterlund and Dmitry Torokhov, some fixes by Vojtech Pavlik. Signed-off-by: Vojtech Pavlik Patch-by: Peter Osterlund input: Add AT-compatible rawmode generation for ARM. Signed-off-by: Vojtech Pavlik Patch-by: Woody Suwalski Input: pull common code from psmouse and atkbd into libps2 module Signed-off-by: Dmitry Torokhov Input: add serio_[un]pin_driver() functions so attribute handlers can safely access driver bound to a serio port. Signed-off-by: Dmitry Torokhov Input: atkbd - export extra, scroll, set, softrepeat and softraw as individual keyboard attributes (sysfs) and allow them to be set/changed independently for each keyboard: echo -n "2" > /sys/bus/serio/devices/serio1/set echo -n "1" > /sys/bus/serio/devices/serio1/softrepeat Signed-off-by: Dmitry Torokhov input: More IOWarrior blacklist entries in hid.c, rearranging the blacklist back to alphabetic order. Signed-off-by: Vojtech Pavlik Input: clean up ALPS DualPoint logic Signed-off-by: Dmitry Torokhov input: Tidy up & fix the hid-input.c driver. Dual-wheel A4 mice don't report the phantom button anymore, D-Pads are mapped to Hat-switches, debug can print HID->Input mappings, more mappings added, devices with reports larger than MaxPacketSize work again. Signed-off-by: Vojtech Pavlik input: Remove calls to pm_access() and pm_dev_idle() from input.c, as they're empty functions anyway. Signed-off-by: Vojtech Pavlik Patch-by: Patrick Mochel input: Add LED definitions for PDAs. Signed-off-by: Vojtech Pavlik Signed-off-by: John Lenz Input: correct the the wrong use of "DB9" to the correct name, "DE9". Also, some comments/debugging output is fixed up. Signed-off-by: Jan-Benedict Glaw Signed-off-by: Andrew Morton Signed-off-by: Vojtech Pavlik Input: Add ACPI-based i8042 keyboard and aux controller enumeration; can be disabled by passing i8042.noacpi as a boot parameter. Original code by Bjorn Helgaas , reworked by Dmitry Torokhov , FixedIO support from Hans-Frieder Vogt Signed-off-by: Bjorn Helgaas Signed-off-by: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Vojtech Pavlik input: More comment fixes in lkkbd.c Signed-off-by: Jan-Benedict Glaw Signed-off-by: Vojtech Pavlik Input: add a new signature for ALPS DualPoint found in Dell Inspiron 8500 Signed-off-by: Dmitry Torokhov Input: psmouse - add set_rate and set_resolution handlers to make adding new protocols easier and remove special knowledge from psmouse-base.c Signed-off-by: Dmitry Torokhov Input: synaptics - not only switch to 4-byte client protocol but also revert to 3-byte mode if client selected lower protocol. Signed-off-by: Dmitry Torokhov Input: psmouse - reset mouse before doing intellimouse/explorer probes in case it got confused by earlier probes; switch to streaming mode before setting scale and resolution, otherwise some KVMs get confused. Patch-by: Marko Macek Signed-off-by: Dmitry Torokhov Input: psmouse - export rate, resolution, resetafter and smartscroll (Logitech only) as individual mouse attributes (sysfs) and allow them to be set/changed independently for each mouse: echo -n "100" > /sys/bus/serio/devices/serio0/rate echo -n "200" > /sys/bus/serio/devices/serio0/resolution Signed-off-by: Dmitry Torokhov Input: psmouse - drop PS2TPP protocol (it is handled exactly like PS2PP) to free spot for THINKPS protocol and keep old protocol numbers for binary compatibility with Synaptics/ALPS touchpad driver for X. Signed-off-by: Dmitry Torokhov Input: psmouse - make logips2pp fully decode its protocol packets and not rely on generic handler to finish job. Signed-off-by: Dmitry Torokhov Input: psmouse - explicitely specify packet size instead of relying on protocol numbering scheme. Make protocol detection routines return -1 on failure to keep them consistent with ther rest of the code. Set mouse parameters right in detection routines instead of doing it in psmouse_extensions. Signed-off-by: Dmitry Torokhov [PATCH] r8169: Tx timeout rework Tx timeout rework: - the ring descriptors of the chipset and the ring index of the driver are synced during a reset of the device; - rtl8169_interrupt: rtl8169_hw_reset() replaces the previous stop code. An implicit reset of the device is added but it makes no noticeable difference with the former behavior (i.e.: stop the chipset). Signed-off-by: Francois Romieu [PATCH] r8169: wrong advertisement of VLAN features Removal of an advertisement for VLAN features which is redundant with rtl8169_init_one(). Signed-off-by: Jon Mason [PATCH] r8169: automatic pci dac step down Automatic adjustement of highmem dma feature. The first interruption encountered on systems where the 8169 does not perform PCI DAC correctly seems to always be a PCI error one. When DAC is enabled, the driver tries to issue a complete down/up sequence as an addition to the usual halt of the device. Signed-off-by: Francois Romieu [PATCH] r8169: rtl8169_close() races - close the race with rtl8169_interrupt() which appears when rtl8169_close() uses synchronize_irq()/free_irq(); - netif_poll_disable() allows rtl8169_close() to wait for any pending rtl8169_poll() to complete so it can safely clear the rings. Signed-off-by: Francois Romieu [PATCH] r8169: cleanup Cleanup - timeout message is redundant with net/sched/sch_generic::dev_watchdog; - anti-bloat in rtl8169_get_rx_csum; - format fix. Signed-off-by: Francois Romieu Signed-off-by: Jon Mason [PATCH] iomem annotations in r8169 Signed-off-by: Al Viro [PATCH] r8169: use module_param Use module_param instead of deprecated MDOULE_PARM Signed-off-by: Stephen Hemminger [PATCH] r8169: use netdev_priv Use netdev_priv in a couple of places in realtek 8169 Signed-off-by: Stephen Hemminger [PATCH] via-rhine: use module_param Signed-off-by: Stephen Hemminger [PATCH] via-rhine: free_ring should be static free_ring is a local function Signed-off-by: Stephen Hemminger input: Fix ssize_t prototype mismatch in psmouse and atkbd. Signed-off-by: Vojtech Pavlik [PATCH] ac3200 iomem annotations and fixes annotated, killed isa_... uses by making ioremap() unconditional, fixed the use of isa_... on already remapped address. Signed-off-by: Al Viro [PATCH] ne3210 iomem annotations Signed-off-by: Al Viro Input: i8042 ACPI enumeration - add PNP IDs found in AMD64 laptops. Signed-off-by: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Vojtech Pavlik [PATCH] r8169: always clean Tx desc rtl8169_unmap_tx_skb() can not assume that a Tx ring descriptor belongs to the host as it can be issued during rtl8169_tx_clear() as a part of a recovery process (during Tx timeout for instance). Simple fix: always clean the relevant descriptor entry. Acked-by: Francois Romieu Signed-off-by: Jon Mason [PATCH] r8169: unneeded synchronize_irq() synchronize_irq() is not needed as it is already issued by free_irq(). Signed-off-by: Francois Romieu [PATCH] r8169: netconsole support netconsole support. Signed-off-by: John W. Linville [PATCH] starfire iomem annotations Signed-off-by: Al Viro [PATCH] killed isa_... in 3c507 switched to ioremap() + normal read../write.. Signed-off-by: Al Viro [PATCH] depca iomem annotations Signed-off-by: Al Viro [PATCH] 3c359 iomem annotations Signed-off-by: Al Viro [PATCH] e2100 iomem annotations and fixes added mission ioremap(); driver was using readw() et.al. on non-remapped addresses. Signed-off-by: Al Viro [PATCH] (25/32) lanstreamer iomem annotations Signed-off-by: Al Viro [PATCH] netdev_priv() in arlan Signed-off-by: Al Viro [PATCH] netdev_priv() in netwave_cs Signed-off-by: Al Viro [PATCH] arlan iomem annotations and cleanups iomem annotations + couple of bad implementations of offsetof() replaced with the real thing. Signed-off-by: Al Viro [PATCH] netwave iomem annotations Signed-off-by: Al Viro [PATCH] net/pcmcia iomem annotations Signed-off-by: Al Viro input: Increase ACK timeouts in libps2 in case the RESET_BAT command is used. This should fix most Synaptics "reset failed" cases. Thanks to Keith Packard for the report. Also add some more keyboard IDs, so that unusual keyboards are accepted by libps2 and atkbd. Signed-off-by: Vojtech Pavlik [PATCH] via-rhine iomem annotations, switch to io{read,write} Signed-off-by: Al Viro [PATCH] sundance iomem annotations, switch to io{read,write} Signed-off-by: Al Viro [PATCH] olympic_open() cleanup and fixes Signed-off-by: Al Viro [PATCH] skfp iomem annotations, switch to io{read,write} Signed-off-by: Al Viro [PATCH] lne390 iomem annotations and fixes annotated, killed isa_... uses by making ioremap() unconditional, fixed the use of isa_... on already remapped address. Signed-off-by: Al Viro [PATCH] wavelan_cs iomem annotations Signed-off-by: Al Viro [PATCH] airo iomem annotations Signed-off-by: Al Viro [PATCH] mace iomem annotations - trivial part Signed-off-by: Al Viro Input: couple of whitespace fixes. Signed-off-by: Dmitry Torokhov Input: evdev, joydev, mousedev, tsdev - remove class device and devfs entry when hardware driver disconnects instead of waiting for the last user to drop off. This way hardware drivers can be unloaded at any time. Signed-off-by: Dmitry Torokhov Input: when creating input devices for hardware attached to a serio port properly set input_device->dev pointer so when corresponding class device is created it will show proper links to parent device and driver in sysfs hierarchy. Signed-off-by: Dmitry Torokhov Input: i8042 - allow turning debugging on and off "on-fly" so people do not have to recompile their kernels to provide debug info. Adds new parameter i8042.debug also accessible through sysfs. Signed-off-by: Dmitry Torokhov Input: i8042 - get rid of old style power management handler since APM calls both pm_send and device_suspend. Signed-off-by: Dmitry Torokhov Input: i8042 - get rid of reboot notifier as suspend method should do the job. Signed-off-by: Dmitry Torokhov Input: get rid of pm_dev in input core as it is deprecated and nothing uses it anyway. Signed-off-by: Dmitry Torokhov Input: gscps2 - remove unused statically allocated gscps2_serio_port variable as the port is allocated dynamically. Signed-off-by: Dmitry Torokhov drm: initial core move infrastructure change Initial infrastructure - move old H files to C files also Kconfig and Makefile changes Signed-off-by: Dave Airlie drm: memory allocation patch This removes some unnecessary macros for allocating DRM memory. It doesn't change any functionality. From: Jon Smirl Signed-off-by: Dave Airlie scsi_debug v 1.75 From: Douglas Gilbert - fix highmem data transfers - fix kunmap_atomic() argument - disable clustering - allow every_nth < 0 for error continuously once count is reached - minor sense buffer handling cleanup Signed-off-by: James Bottomley [PATCH] r8169 module_param build fix Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik [PATCH] ray_cs export cleanup From: Arjan van de Ven The ray_cs driver author seemed to have assumed that he needs to exports functions he registers with the core kernel via function pointers, that of course isn't the case so the cleanup below removes this; these functions aren't used anywhere else nor meant to be (they're even static). Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik Input: parkbd - switch to using module_param. Parameter names are parkbd.port and parkbd.mode Signed-off-by: Dmitry Torokhov drm: device minor fixups and /proc fixups This patch fixes up the DRM to do better minor number accounting and /proc directory creation, the old code was buggy in a number of situations with multiple cards, and rather ugly. It is also a step on the way to the drm_core module. From: Jon Smirl and Dave Airlie Signed-off-by: Dave Airlie [PATCH] via-velocity: get rid of unused global Get rid of unused global variable, name the enum instead. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] IBM EMAC Kconfig changes In trying to build for IBM 440GP Eval with CRC32=n, I noticed two things. First, all of the IBM EMAC Kconfig bits are space, not tab indented, and that IBM EMAC doesn't select CRC32 like all of the other enet drivers that need it do. Fix spacing of IBM EMAC Kconfig options. Signed-off-by: Tom Rini Signed-off-by: Jeff Garzik [PATCH] IBM EMAC Kconfig changes: Add 'select CRC32' On Mon, Oct 25, 2004 at 09:12:03AM -0700, Tom Rini wrote: > In trying to build for IBM 440GP Eval with CRC32=n, I noticed two > things. First, all of the IBM EMAC Kconfig bits are space, not tab > indented, and that IBM EMAC doesn't select CRC32 like all of the other > enet drivers that need it do. Add a 'select CRC32' Signed-off-by: Tom Rini Signed-off-by: Jeff Garzik [ACPI] Provide core hotplug support in ACPI Create acpi_bus_trim(), acpi_bus_remove() and acpi_pci_unbind(), The reverse of of acpi_bus_scan(), acpi_bus_add() and acpi_pci_bind() Signed-off-by: Anil S Keshavamurthy Signed-off-by: Len Brown [ACPI] create ACPI hotplug eject interface The kernel when it receives an hardware sci eject request it simply passes this to user mode agent and the agent in turn will offline all the child devices and then echo's 1 onto the eject file for that acpi device. This patch provides the sysfs "eject" interface for the user mode agent to notify the core acpi so that the core acpi can trim its bus which causes .remove function to be called for all child devices. For example for LSB0 which is an ejectable device, we will see /sys/firmware/acpi/namespace/ACPI/_SB/LSB/eject. Signed-off-by: Anil S Keshavamurthy Signed-off-by: Len Brown [ACPI] IA64-specific support for mapping lsapic to cpu array. analogous i386 and x86_64 code TBD Signed-off-by: Anil S Keshavamurthy Signed-off-by: Len Brown IA64 CPU hotplug topology Extend support for dynamic registration and unregistration of the cpu, by implementing and exporting arch_register_cpu()/arch_unregister_cpu(). Also combine multiple implementation of topology_init() functions to single topology_init() in case of ia64 architecture. Signed-off-by: Anil S Keshavamurthy Signed-off-by: Len Brown [ACPI] Extend processor driver to support ACPI-based Physical CPU hotplug Signed-off-by Anil S Keshavamurthy Signed-off-by: Len Brown [ACPI] Initial container driver to support hotplug notifications on ACPI0004, PNP0A05 and PNP0A06 devices. Signed-off-by: Anil S Keshavamurthy Signed-off-by: Len Brown [ALSA] use blacklist/whitelist for (non-)audio Bt878 cards Documentation,PCI drivers,BT87x driver Some Bt87x cards use PCI function 1 for MPEG data instead of audio data, so we blacklist those in the audio driver. Further add a whitelist for cards where audio is known to work (many other cards do not implement the audio connection). Unknown cards can be enabled manually. Signed-off-by: Clemens Ladisch [ALSA] fix sequencer sleeping in interrupt context ALSA sequencer,ALSA<-OSS sequencer Signed-off-by: Clemens Ladisch [ALSA] rearrange OSS SPARC dependencies Sound Core rearrange the SPARC symbols in the OSS dependencies to prevent alsa-driver's mod-deps from throwing up Signed-off-by: Clemens Ladisch [ALSA] snd-usb-usx2y - crash fix for OHCI USB-HCDs USB USX2Y Version: 0.8.6 Work on this started, when rumors spread that OHCI equipped machines would crash. This was due to me missing two facts: 1) Ohci has a bigger usb frame number wrap around. 2) It only supports URB_ISO_ASAP when submitting iso urbs. These issues are fixed now. Signed-off-by: Karsten Wiese Signed-off-by: Jaroslav Kysela [ALSA] au88x0: add resetup dma au88x0 driver Add adbdma_resetup for refreshing the hw page table on pcm start Signed-off-by: Jeff Muizelaar Signed-off-by: Jaroslav Kysela [ALSA] au88x0: fix is-quad oops au88x0 driver Fixes an oops on module removal caused by dereferencing the codec pointer. This is not the best solution, but it is the easiest and fixes things for now. Signed-off-by: Jeff Muizelaar Signed-off-by: Jaroslav Kysela [ALSA] au88x0: set-levels cleanup au88x0 driver Cleanup vortex_EqHw_SetLevels and add a bit of documentation Signed-off-by: Jeff Muizelaar Signed-off-by: Jaroslav Kysela [ALSA] au88x0: sign_invert cleanup au88x0 driver Remove unecessary ' & 0xffff'ing of the result of sign_invert Signed-off-by: Jeff Muizelaar Signed-off-by: Jaroslav Kysela [ALSA] au88x0: name typo au88x0 driver Fix the spelling of my name Signed-off-by: Jeff Muizelaar Signed-off-by: Jaroslav Kysela [ALSA] au88x0: comment and whitespace cleanup au88x0 driver Remove an obsolete comment and cleanup up some whitespace a bit Signed-off-by: Jeff Muizelaar Signed-off-by: Jaroslav Kysela [ALSA] fix data type mismatch in sign_invert au88x0 driver the last sign_invert cleanup introduced a data type mismatch (an unsigned value can never be negative) Signed-off-by: Clemens Ladisch [ALSA] remove old compatibility code USB USX2Y Signed-off-by: Clemens Ladisch [ALSA] remove dead exports ALSA Core,Instrument layer,AD1848 driver,ES1688 driver Alsa currently has tons of dead exports, often with totally unused functions behind them. This removes some of them. Signed-off-by: Christoph Hellwig Signed-off-by: Clemens Ladisch [ALSA] Limit parity error messages BT87x driver Some systems generate tons of PCI parity errors, so shut up when more than 20 have been detected. Signed-off-by: Clemens Ladisch [ALSA] remove kernel version info from proc file ALSA Core The kernel version information isn't necessary for the driver in the kernel tree, so move it to the alsa-driver package. This removes a dependency to . Signed-off-by: Clemens Ladisch [ALSA] fixed emu10k1_fx8010_code_t structure to be less than 8192 bytes EMU10K1/EMU10K2 driver This patch fixes emu10k1_fx8010_code_t structure using indirect pointers to be less than 8192 bytes to follow the ioctl semantics. Signed-off-by: Jaroslav Kysela [PATCH] net/wan/n2.c: remove an unused function Signed-off-by: Adrian Bunk Signed-off-by: Jeff Garzik [PATCH] net/wan/n2.c: remove an unused function Adrian Bunk writes: > The patch below removes an unused function from drivers/net/wan/n2.c A similar thing, for C101 card. Signed-off-by: Krzysztof Halasa Signed-off-by: Jeff Garzik [PATCH] net/typhoon.c: use previously-unused function A response to Adrian Bunk's "remove unused function" cleanup patch. Signed-off-by: Jeff Garzik [PATCH] bonding: remove an unused function Signed-off-by: Adrian Bunk Signed-off-by: Jeff Garzik [PATCH] net/3c505.c: remove unused functions Signed-off-by: Adrian Bunk Signed-off-by: Jeff Garzik [PATCH] net/skfp/smt.c: remove an unused function The patch below removes an unused function from drivers/net/skfp/smt.c Signed-off-by: Adrian Bunk Signed-off-by: Jeff Garzik [PATCH] Add Altimata PHY to sis900 driver The attached patch fixes a long standing detection problem with the sis900 driver. This PHY chip is used on some Pentium 4 with SiS chipset and on the Acer Aspire 1705SMi (at least) notebook. Signed-Off-By: Daniele Venzano Signed-off-by: Jeff Garzik [PATCH] unexport ei_tx_timeout not used by any module Signed-off-by: Jeff Garzik [ALSA] emu10k1 - fixes against the last emufx changes EMU10K1/EMU10K2 driver The indirect pointers are allocated correctly now for default DSP code. Also, one bug in emu10k1_fx8010_code_t has been fixed as well. Signed-off-by: Jaroslav Kysela [PATCH] r8169: endian-swap return of rtl8169_tx_vlan_tag() Endian-swap return of rtl8169_tx_vlan_tag() in rtl8169_start_xmit() Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik [PATCH] r8169: fix RxVlan bit manipulation Fix manipulation of RxVlan bit in rtl8169_vlan_rx_register(), and remove it from rtl8169_vlan_rx_kill_vid(). Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik [PATCH] r8169: simplify trick if() expression Simplify tricky if() expression in rtl8169_vlan_rx_register(). Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik [PATCH] 3c59x: Add MODULE_VERSION Add MODULE_VERSION to 3c59x driver. Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik [PATCH] 8139too: Add MODULE_VERSION Add MODULE_VERSION to 8139too driver. Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik [PATCH] r8169: Add MODULE_VERSION Add MODULE_VERSION to r8169 driver. Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik [PATCH] e100: Add MODULE_VERSION Add MODULE_VERSION to e100 driver. Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik [PATCH] tulip: Add MODULE_VERSION Add MODULE_VERSION to the tulip-based drivers Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik [PATCH] hp100: use netdev_priv (rev 2) Here is a revised version of the hp100 patch sequence. First one just does netdev_priv. Worked with Jean to get these patches tested. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] hp100: use inline for comple usage of dev->priv Make a separate function for the one more complex usage of netdev_priv. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] tlan: use netdev_priv (rev 2) Use netdev_priv Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] tlan: get rid of unneeded global vars (rev 2) The global variable media_map is never used. And the media table media can be static. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] tlan: make inline's static (rev2) Make inline functions static to avoid polluting global namespace. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] tlan: enable faster hash function Cleanout dead code, and use better hash function. The faster hash function was already there, but not turned on by default. Tested hash function for 10 million random addresses. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] xircom_tulip_cb: convert to using module_param Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] prism54 fix resume processing * We need to enable the device on resume. Signed-off-by: Jeff Garzik [PATCH] prism54 sparse fixes * On top of Linus's sparse changes, here is a * fix that further reduces sparse warnings. We are still left with 2 warnings caused by the member "data.pointer" in struct "iwreq_data" being "__user" (from wireless.h). Signed-off-by: Jeff Garzik [ACPI] fix mis-merge in processor.c drm: core/personality split for 2.6 kernel This changeset gets rid of the DRM() macros and implements a core DRM module linked to a per graphics card personality module.. Remove old 2.4 module parameters and switch to 2.6 module parameters From: Jon Smirl and Dave Airlie Signed-off-by: Dave Airlie drm: drm_memory.c missing from build Add drm_memory.c to build. Signed-off-by: Dave Airlie drm: rename fn_tbl to driver as it is no longer a function table This renames the drm_driver_fn to drm_driver and fn_tbl to driver, this name is makes much more sense now. From: Jon Smirl and Dave Airlie Signed-off-by: Dave Airlie drm: move fops into drivers move the drm file operations into the driver. From: Jon Smirl Approved-by: Dave Airlie drm: rearrange some functions for new split This change moves some functions into different C files to align things a bit more correctly... From: Jon Smirl Approved-by: Dave Airlie drm: make pcigart functions inline with these unstatic uninline the kernel wouldn't build with both configured. Signed-off-by: Dave Airlie [PATCH] net/gt96100eth: replace gt96100_delay() with msleep_interruptible() Uses msleep_interruptible() instead of schedule_timeout() in the gt96100_delay() function. Corrects one comment to correspond to the code. Signed-off-by: Nishanth Aravamudan Signed-off-by: Jeff Garzik [PATCH] S2io: cosmetic changes 1. Indentation, change in comment styles, variable name changes etc. 2. Changed the value written to dtx_control register to force XGXS reset. 3. weight parameter(NAPI) changed to 90 for better performance. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: sw bug fixes 1. In free_rx_buffers clearing out RxDs not owned by Xena. 2. In alarm_intr_handler, when a serr error occurs, schedule a task to reset the card rather than stopping Tx queue. 3. In s2io_close freeing IRQ before calling s2io_reset also added a new call to flush queued tasks. This is not done if the s2io_close itself is called from a queued task like s2io_restart_nic. 4. read_eeprom function has been changed such that data to be returned is sent as an input argument and the return value represents a pass/fail. The previous implementation as Randy had pointed out was error prone as on failure it returned -1 which can be interpreted as all ff's, so any data area which contained ff's in the eeprom was likely to be treated as an error. 5. Added a flag "task_flag" to track if the call to s2io_close is coming from the s2io_restart_nic function or from the ifconfig down called by user. 6. Moved register_netdev call from just after setting entry points to the end of the s2io_init_nic function. 7. In s2io.h field added a new member into the s2io_nic structure called "task_flag". Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: optimizations 1. Definitions of LOW and PANIC levels of the Rx buffers have changed. 2. In wait_for_cmd_complete there is no longer a writeq but just a read and wait for strobe bit to reset. 3. In s2io_isr, the isr_lock has been done away with also the NICs interrupt are no longer disabled explicitly on entering the interrupt handler and re-enabled again before leaving it. 4. Also clearing the semaphore "tasklet_status" when exiting erroneously from s2io_isr after failing fill_rx_buffer call. 5. The set/reset Tx Csum function through ethtool was added to the ethtool_ops structure. 6. Added a Rx side error code in the rx_osm_handler function. 7. No longer stopping and waking Tx queue when link state changes in s2io_link function. 8. removed the isr_lock spinlock from the s2io_nic structure. 9. changed parameters which determine thresholds(LOW and PANIC) to replenish Rx buffers. This has been found to result in better performance. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: hardware fixes 1. Xena3's with a set of subsystem IDs had Link LED problems, fixed that specifically for them. 2. To write into the Keyed Mac_Cfg register to enable broadcast, writing two 32 bit writes into it along with a write to the key register rather than a single write to key and a 64 bit write to mac_cfg. This is necessary on 32 bit systems where a writeq(64 bit write) is actually two writel (32 bit writes). 3. Writes to some special registers mentioned in UG is being done by a special macro which defines which 32 bits of the 64 bit register is to be written first. Again this applies only on 32 bit systems. 4. Configured pause frame related water marks and a shared_split value which describes the Max TXDMA related split transaction that can be used without giving room for the Rx transactions. 5. The mac_rmac_err_reg R1 register will be cleared in the interrupt handler itself rather than in the scheduled task as was being done previously. 6. Even on PCC_FB_ECC error the card will be reset by disabling adapter enable bit. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: module loadable parameters 1. Max Txds per List. 2. statistics refresh time 3. pause frame control parameters including gap between two successive frames, threshold watermarks 4. RTI and TTI configuration parameters including ranges, packet counts and timeout periods. For further information please read the section 3.5 of XFrame H/W spec. 5. PCI/PCI-X configuration variables latency_timer, MMRBC and OST. 6. OS offload features TSO (If support available) and checksum offload. 7. If NAPI is not in use, a variable indicate_max_pkts can be used to limit number of Rx side packets processed for one call to Rx Intr handler. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: new txd allocation The attached patch contains a modified scheme for allocating Tx descriptor blocks. More description follows. In the old scheme, the entire Tx descriptor space was allocated in one go. This could cause driver load to fail on systems with low(or scattered) memory. The Tx descriptor blocks are now allocated on per-page basis. A new structure (list_info) has been introduced in nic_t structure to keep track of the physical and virtual addresses of every TxD allocated this way. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: NAPI fix 1. When processing Rx packets, making sure that get offset of ring does not cross the put offset. 2. when NAPI is not in use a new spinlock(put_lock) is used to make sure accessing put offset of ring is atomic. 3. Also introduced a new vaiable put_pos in nic_t to keep track of absolute position of the put pointer of Rx ring. 4. When NAPI is used, fill_rx_buffer is not called from the interrupt handler(s2io_isr) . 5. In s2io_poll, decrementing packets processed is done inside the while loop unlike out side it as was being done last time. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: two buffer mode Attached is the patch for implementing 2-buffer mode on Rx path. On certain systems when a DMA has to happen on an un-aligned memory location performance will take a significant hit. It's standard practice to offset the Rx buffer address by 2 (as Mac header is 14 bytes) so the IP header starts from an aligned location. Obviously using a single Rx buffer both cannot be achieved. Thus XFrame supports something called 2 buffer Rx mode, where in the Rx'ed frame is split into 2 parts, one is the Ethernet header and the other is the Ethernet payload. So now we can allocate proper aligned memory for both buffers, hence the DMA is not slowed down. Also, the Ethernet payload(starting from L3 header) is on an aligned location so OS need not have to do un-aligned accesses to process IP header. To achieve this, the kernel function eth_type_trans functionality has to be partially implemented in the driver itself. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: new functions for card restart The attached patch incorporates Jeff's comments related to creating separate functions for restarting the NIC(without using close and open entry points) and few other comments. Complete list of changes are as follows: 1. Two new functions s2io_card_down() and s2io_card_up() are defined and are called during reset procedure instead of close and open routines. 2. tasklet_status field is now made as unsigned long. 3. On getting serious error, queue is stopped before resetting the card. 4. Removed the check for "queue stopped" in xmit routine. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: 2 buffer mode with copy This patch addresses the comments by Chris Leech about skb->mac.ethernet resulting in NULL dereference with the old method of implementing 2 buffer mode. The new method performs a copy of the MAC header to the head of the payload. This is a stop-gap measure till the fragmented skb receive feature in the kernel is made functional. Also, using GFP_KERNEL flag for buffer0, buffer1 memory allocation instead of GFP_ATOMIC. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: modified loadable parameters Attached is the patch to implement module loadable parameters as per new API. Following is the list of changes. 1. Used new module_param() API. 2. List of variables for tx_fifo_len and rx_ring_sz replaced with array. 3. Some of the module parameters which can be set thru setpci command have been removed, such as latency_timer, max_read_byte_cnt, max_split_transactions. 4. Other parameters which were felt to be not required, such as rx_prio, tx_prio have been removed. 5. Interrupt moderation parameters(such as tx_urange_*) are no longer module loadable parameters since they can be configured thru' a separate patch available to customers. 6. Changed default max_read_byte_count to 1024. 7. If scatter-gather is enabled, checksum is enabled too. 8. Not verifying if module loadable parameters are within valid range (verify_load_param() removed). Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik [PATCH] S2io: styling Attached is the patch to implement comments about styling and few other changes. Following is list of changes. 1. Incorporated Randy's comment about C99 format for s2io_driver structure initialization. 2. Driver version displayed at load time. 3. If initialization failed in s2io_init_nic(), appropriate error codes are returned. 4. #ifdef SET_ETHTOOL_OPS removed in couple of places. Signed-off-by: Raghavendra Koushik Signed-off-by: Ravinandan Arakali Signed-off-by: Jeff Garzik drm: move ati_pcigart into drm core This moves the ati_pcigart code into the drm core. From: Jon Smirl Signed-off-by: Dave Airlie [PATCH] fix platform_rename_gsi related ia32 build breakage "platform_rename_gsi" does not exist unless CONFIG_ACPI_BOOT is defined. Fix the fix. Silly #endif placement problem. [ACPI] CPU hotplug, use kobject_hotplug(), kobject_register() Signed-off-by: Anil S. Keshavamurthy Signed-off-by: Len Brown drm: fix warning for missing vunmap In file included from drivers/char/drm/drmP.h:795, from drivers/char/drm/drm_dma.c:36: drivers/char/drm/drm_memory.h: In function `drm_ioremapfree': drivers/char/drm/drm_memory.h:191: warning: implicit declaration of function `vunmap' Signed-off-by: Andrew Morton Signed-off-by: Dave Airlie drm: fix Kconfig dependency fixup DRM/AGP Kconfig inter-dependency... From: Roman Zippel Signed-off-by: Dave Airlie [ALSA] replace schedule_timeout() with msleep() CS4231 driver Uses msleep() instead of schedule_timeout() to guarantee the task delays as expected. This lead to several related changes, as the current code assumes the value of HZ is 100. Use timeout as an iteration variable to count out how many 10ms delays should be used. Signed-off-by: Nishanth Aravamudan Signed-off-by: Clemens Ladisch [ALSA] replace schedule_timeout() with msleep() SPARC cs4231 driver Uses msleep() instead of schedule_timeout() to guarantee the task delays as expected. Signed-off-by: Nishanth Aravamudan Signed-off-by: Clemens Ladisch [ALSA] emu10k1 - another attempt to correct the new emufx DSP code EMU10K1/EMU10K2 driver Signed-off-by: Jaroslav Kysela [ALSA] Added SNDRV_HWDEP_IFACE_BLUETOOTH ALSA Core Signed-off-by: Marcel Holtmann Signed-off-by: Jaroslav Kysela [ALSA] remove snd_seq_simple_id Instrument layer,GUS Library,Trident driver remove uses of the snd_seq_simple_id symbol because it is no longer exported Signed-off-by: Clemens Ladisch [ALSA] Fixed the description of module_parm_array() Documentation Fixed the description about module_param_array() for the latest change. Signed-off-by: Takashi Iwai [ALSA] removes unneeded spin_lock_irqsave()s from snd-es1968 ES1968 driver spin_lock_irqsave(&chip->reg_lock) was called a second time in sequence from snd_es1968_bob_start() called from es1968_measure_clock(). While this didn't cause harm on my UP laptop with mainline kernels, it made 'insmod snd-es1968' hang on kernel 2.6.9-mm1-RT-V0.6.9. The patch assumes that 2 callpaths don't need explicit spinlock protection: 1: The trigger callback, because it is called with IRQs disabled. 2. PM's suspend/resume callbacks, because those are called while ortdinary user processes are frozen. Thus the spin_lock_irqsave(&chip->reg_lock) calls in snd_es1968_bob_start() / snd_es1968_bob_stop() are not needed. Signed-off-by: Karsten Wiese Signed-off-by: Takashi Iwai [ALSA] Add subvendor ID to the pci id table of vx222 driver Digigram VX222 driver The subsystem ID is added to the pci id table of vx222 driver to make the matching more strict since it (PLX) conflicts with other devices. Signed-off-by: Takashi Iwai [ALSA] Fixed issues with Abit AV8 VIA82xx driver Added Abit AV8 sound card to the white list to use VIA_DXS_NO_VRA by default. This resolves issues with programs wanting to use 41k streams. It also fixes gstreamer issue with alsasink module interaction. Signed-off-by: Jerone Young Signed-off-by: Takashi Iwai [ALSA] [emu10k1] add interval timer support EMU10K1/EMU10K2 driver Signed-off-by: Lee Revell Signed-off-by: Takashi Iwai [ALSA] emu10k1 - fixed remaining problems with new DSP code loading EMU10K1/EMU10K2 driver Signed-off-by: Jaroslav Kysela [ALSA] handle missing control bitmap when parsing MUDs USB generic driver The AudioTrak Maya44 USB has a mixer unit descriptor without a bmControl field; handle this as if all bits are zero. Signed-off-by: Clemens Ladisch [ALSA] read bmControls array in correct order USB generic driver The driver used some code from audio.c that reads the bmControl array backwards; this would not work here as we get a pointer to the beginning of the array. Signed-off-by: Clemens Ladisch [ALSA] fix parsing of mixer unit descriptors USB generic driver MU descriptor parsing code completely rewritten; the old code confused the number of input audio channel clusters and the number of input channels. Furthermore, check all bmControls bits so that mixer controls are created even if the first output channel doesn't have a control. Signed-off-by: Clemens Ladisch drm: remove use of drm_agp use agp backend directly. This removes the inter module stuff between the DRM and AGP. Signed-off-by: Dave Airlie [ALSA] nonblock_open=1 by default for OSS PCM API emulation Documentation,ALSA<-OSS emulation Signed-off-by: Jaroslav Kysela drm: in-correct locking in intel drms The locking in the intel drms is incorrect it doesn't check the current context owns the lock, just that someone does. This could allow strange things to happen with multiple clients. From: Stefan Dirsch Approved-by: Dave Airlie ALSA 1.0.7 [PATCH] sk98: no explicit module ref counting needed Network driver's shouldn't be doing module reference counting on 2.6. This is a bad replication of 2.4 behaviour Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] sk98: local variable can be constant The table OnesHash is only used in this file, and should be made a local constant. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] sk98: /proc interface related changes The /proc intereface can be cleaned up a lot: + Pass the device to through the proc data/seq_private hook. + No longer needs to have a list of network devices + use seq_printf directly rather than indirection into a buffer + failure to create proc directory or interface should not be fatal. + if PROC_FS not configured, let the stubs cause the necessary code elimination rather than using #ifdef Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] sk98: use netdev_priv Use netdev_priv when dereferenceing net_device private Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] sk98: use module_param Use new format module_param rather than now deprecated MODULE_PARM Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] sk98: add netpoll console support Add netpoll console support From: Michal Schmidt Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] sk98: basic ethtool support The basic stuff comes from the newer code from SysKonnect, but I redid it using ethtool_ops and a cleaner way of doing the stats (from e100) Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] sk98: ethtool phy support The basic stuff comes from the newer code from SysKonnect, but I redid it using a per board timer rather than a global timer. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] sk98: ethtool pause param support The basic stuff comes from the newer code from SysKonnect, but there code had ugly way to find the PNMI instance. This driver seems full of places where they do unnecessary board/port bookeeping (*sigh*) Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik [PATCH] Another big 6pack patch Below another 6pack patch which I've got sitting in my repository for a while already. The patch does: o Update my callsign o Cleanup the internal for synchronization with a 6pack TNC a little bit so it won't result in heaps and piles of messages in syslogs as it used to do. o Does paranoia checks for the maximum allowable packet size on transmit. There was a slight chance this might have resulted in a crash in just the right configuration. o Verifies the address family for the SIOCSIFHWADDR network interface ioctl. o Tries to do proper locking so nothing will access dev->dev_addr in midflight while changing the layer 2 address. o Uses an intermediate buffer for the SIOCSIFHWADDR ioctl so an error in copy_from_user can't result in a half-modified MAC address. o Swaps the arguments of decode_prio_command, decode_std_command and decode_data such that for consistency the struct sixpack pointer will be the first argument like everywhere else. Signed-off-by: Jeff Garzik [PATCH] convert sis900 to new style parameters Signed-off-by: Jeff Garzik [sound/oss i810_audio] use module_param() Also, set MODULE_AUTHOR and correct module name in a macro. [sound/oss] use module_param() in soundcard.c and uart401.c [libata sata_via] minor cleanups Preparation for addition of VT6421 support. Mostly moving bits of code into discrete functions. [libata sata_via] add support for VT6421 SATA [libata sata_uli] add 5281 support, fix SATA phy setup for others Contributed by Peer Chen @ ULi and tested by a user. JFFS2: Add support for bizarre NOR flash with ECC. Signed-off-by: Josh Boyer Signed-off-by: David Woodhouse MTD: Provide XIP support for Intel flash chips. This allows for MTD support to be used on flash memory which is also used for XIP purposes, either XIP kernel or XIP userspace. The whole idea is to relocate functions actually modifying the flash state away from array mode to ram and run them with interrupt disabled. When the flash needs some time to complete a certain operation, we poll the processor for pending (but still masked) interrupts, and when they occur we suspend the flash operation in order to unmask interrupts and let the system run again. Signed-off-by: Nicolas Pitre Signed-off-by: David Woodhouse MTD: NAND driver updates - Support 2048-byte HW ECC (from Juha Yrjölä ) - Allow board drivers to provide pattern for bad block scanning Signed-off-by: Thomas Gleixner Signed-off-by: David Woodhouse JFFS2: Move very noisy debugging messages from level 1 to level 2. Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse update the fc_transport_class to use a workqueue instead of a timeout The amount of work that has to be done in the timeout routines is really a bit much if there's a large number of LUNS. Plus the device_for_each_child needs user context to get the bus semaphore, so the solution is to migrate them from a timer to delayed work. There's still a race here in that the timer may still be ticking when the device is destroyed ... To fix this, I think we may need to introduce a destroy callback to the transport class. Signed-off-by: James Bottomley SCSI: updates to constants.c From: Douglas Gilbert - bring opcode names, asc/ascq strings and sense format into line with SPC-3 rev 21 (22 September 2004) - drop SCSI-1 sense buffer decoding [still output it in hex] - opcode names include those that depend on service actions including variable length commands ** - decodes both fixed and descriptor sense data formats - use KERN_INFO on printk()s that start on new lines - flag vendor specific asc and acsq codes - print all bytes of a cdb after the name (previously skipped the first byte) - cleanup file, tab to 8 spaces Signed-off-by: James Bottomley SCSI: Add transport destructors From: James.Smart@Emulex.Com This patch adds host/target/sdev destructor functions to the transport template. The FC transport is updated to utilize them to cancel any outstanding timer. It slightly rearranges the slave_xxx functions so the transport is always involved prior to the LLDD. Minor rejection fixes and Signed-off-by: James Bottomley JFFS2: Add notes on inocache_lock spinlock to README.Locking Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse Email address update. The work address is increasingly unreliable and incompetently run. Time to remove all visible instances of it and rely only on one which isn't run by crack-monkeys. Signed-off-by: David Woodhouse [PATCH] iSCSI transport class The attached patch adds an iSCSI transport class. It was built against scsi-misc-2.6. It allows the software/virtual iSCSI driver to remove at least one of its ioctl commands. The patch also assumes the no multipath/failover rule in llds applies to iSCSI drivers, so when we remove our portal/portal_group failover support it will allow us to kill all our ioctl commands execpt the session creation one (that will take more thought). We would like to build some functions like the unblock/block capabilites into the iscsi transport class like James Smart has done with FC class. but initially we would like to begin with this smaller patch that only provides an iscsi sysfs interface. Signed-off-by: James Bottomley SCSI: descriptor sense format, mid-level - generalize sense data logic to cope with both fixed and descriptor format - use KERN_INFO on most printk()s to limit console noise - retire mid-level usage of sense_class(), sense_error() and sense_valid() macros which are SCSI-1 remnants. Now only cpqfcTSinit.c seems to use them Signed-off-by: James Bottomley SCSI: Add missing state transition BLOCK->OFFLINE From: James.Smart@Emulex.Com Signed-off-by: James Bottomley JFFS2: Remove definition of obsolete struct jffs2_scan_info We no longer use this, since we rewrote the mount code to behave entirely differently and not build up all the trees on mount. Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse MTD: Fix chip ident definition for AMD 29F002T devices. They were missing unlock address and command set fields. From: David Vrabel Signed-off-by: David Woodhouse [PATCH] iscsi_transport build fix drivers/scsi/scsi_transport_iscsi.c:176: structure has no member named `sin6_addr' Older gcc's dont support anonymous unions. Signed-off-by: Andrew Morton Signed-off-by: James Bottomley JFS: fix race in jfs_commit_inode There was a race that resulted in old, deleted inodes being written to disk after the inode number had been reused. jfs_commit_inode needs to verify that the inode is still linked and dirty before committing it. Signed-off-by: Dave Kleikamp [PKT_SCHED]: Allow using nfmark as key in U32 classifier. Signed-off-by: Catalin(ux aka Dino) BOIE Signed-off-by: David S. Miller [NET]: Add alloc_skb_from_cache. This serves two purposes: firstly, we like to allocate page-sized skbs as this means we zero-copy transfer of network buffers between guest operating systems. Secondly, it enables us to have a cache of pages that have been used for network buffers that we can be more lax about scrubbing when they change VM ownership (since they could be sniffed on the wire). Signed-off-by: David S. Miller [NET]: Kill drivers/net/net_init.c After the last patch only three routines are left in this file, but all of the fir into net/core/dev.c much better: - {un,}register_netdev are just wrappers around {un,}register_netdevice from dev.c - alloc_netdev's counterpart, free_netdev is in dev.c aswell. So move over the remaining contents and add some kerneldoc comments describing the functions. Signed-off-by: David S. Miller [NET]: Use local_softirq_pending instead of softirq_pending in netif_rx_ni Some architectures can optimize local_softirq_pending much better than softirq_pending(smp_processor_id()), and for all others the former is just a macro expanding to the later. Also this is the last use of softirq_pending() in common code, once this is in we can soon kill the notation of beeing able to query other cpus softirq pending count. Signed-off-by: David S. Miller MTD: Fix JEDEC probe of chips which don't require unlock sequence Signed-off-by: Thayne Harbaugh Signed-off-by: David Woodhouse JFFS2: Remove obsolete structure definitions and update comments. New eyes are looking over ancient code and comments which were put in as the design was first coming together... and which don't make even the slightest bit of sense any more. Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse JFFS2: Discard dirents which point to non-existent inodes. If a directory entry refers to an inode which doesn't actually exist, we weren't marking it obsolete, so it was still visible in the file system, and would give EIO if you ever tried to stat it. Once upon a time, perl -e 'unlink' and rm -f would manage to unlink such things but nowadays they both try to stat it first and stupidly refuse to even attempt the unlink if the stat fails, and this is more of a problem. So we throw it away ourselves. Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse JFFS2: Fix memory leak if jffs2_scan_medium() fails. We weren't releasing all the temporary dirent structures we may have built up during the first part of the scan. Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse JFFS2: Fix race on read access to NAND write-buffer. With SMP or preempt, we could attempt to read data from the wbuf while it was being updated. Introduce a new rwsem to prevent this, and update the documentation accordingly Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse MTD NOR chip drivers: use msleep() Description: Use msleep() instead of schedule_timeout() to guarantee the task delays as expected. Signed-off-by: Nishanth Aravamudan Signed-off-by: Maximilian Attems Signed-off-by: David Woodhouse MTD: Use msleep() in cfi_udelay() helper function. Signed-off-by: David Woodhouse drm: move the enable device before filling in device info This moves the pci_enable_device to before the device info is filled out as without routeirq this goes wrong.. Thanks to Ralf Gerbig for testing this. Signed-off-by: Dave Airlie [IA64] convert to use CONFIG_GENERIC_HARDIRQS Convert ia64 to use generic irq handling code. sn2 fixes and testing by Jesse Barnes Signed-off-by: Tony Luck [EBTABLES]: Add userspace logging via netlink socket. The patch below adds the ebtables ulog watcher, that sends packets to userspace. It is based on ipt_ULOG.c. The complete packet, including Ethernet header, is sent to userspace. The in and out bridge ports are also sent to userspace. Logging is of course not restricted to IP packets. An example of a userspace program that receives and parses packets sent by the ulog watcher is in the ebtables CVS tree under examples/ulog/ Signed-off-by: Bart De Schuymer Signed-off-by: David S. Miller [PKT_SCHED]: Clean up tcf_action_init memory handling Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller [XFRM]: Export xfrm_policy_delete() Will be needed for DCCP. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller [ARCNET]: Fixes. As previously reported the ArcNet driver didn't work with Preempt and SMB on. They do now. I have changed the locking system from being a global arcnet lock to being a lock per device. I used the lock in dev->hard_start_xmit = arcnet_send_packet. Furthermore I added the 'CAP mode' encapsulation. As far as I see it it is the only encapsulation which actually makes ArcNet usefull over ethernet. Previously, the driver just ignored the hardware transmit status, now you can get hardware acknowledge and excessive nacks back to userspace via a raw socket. The capmode.c is nearly just a copy of arc-rawmode.c. The difference is that it inserts a ack_tx() handle into the general driver framework. Signed-off-by: Andrew Morton Signed-off-by: David S. Miller [X25]: When receiving a call, check listening sockets for matching call user data. If a listening socket sets call user data, ensure it only receives calls with matching call user data. Also ensure incoming calls with matching call user data dont go to another listening socket. Signed-off-by: Andrew Hendry Signed-off-by: Andrew Morton Signed-off-by: David S. Miller [X25]: Remove unused header files. Signed-off-by: Andrew Hendry Signed-off-by: Andrew Morton Signed-off-by: David S. Miller [NET]: __sock_create() cleanup The 'i' variable in net/socket.c::__sock_create() is not necessary. It's have been used to store error codes but there is an 'err' variable already. Signed-off-by: Luiz Capitulino Signed-off-by: Andrew Morton Signed-off-by: David S. Miller [PATCH] feed eata.c through Lindent the driver oopses on load for me currently, but to debug it I need to actually be able to read the source.. Signed-off-by: James Bottomley convert eata to pass pointers around From: Christoph Hellwig pass pointers to Scsi_Host and the private data around instead of using indices into global arrays all over the place. Signed-off-by: James Bottomley [PATCH] kill scsi_syms.c The patch below removes scsi_syms.c and moves the EXPORT_SYMBOL's to the files where the actual functions are. Signed-off-by: Adrian Bunk Signed-off-by: James Bottomley [PATCH] kill locking around scsi_done() Lets show the mid layer as a good example and remove the host lock around the scsi_done() call. Signed-off-by: Jens Axboe Signed-off-by: James Bottomley ===== drivers/scsi/scsi.c 1.148 vs edited ===== scsi: LLDD dynamic scan aids From: James.Smart@Emulex.Com With rejection fixes around the removal of scsi_syms.c Signed-off-by: James Bottomley fixup dynamic scan aids EXPORT_SYMBOL mismerge OK, so I exported the wrong symbols ... Signed-off-by: James Bottomley LSI Logic - SAS and FC PCI ID's From: Moore, Eric Dean Here are new PCI ID's for SAS and Fibre Channel LSI Logic controllers. Signed-off-by: Eric Moore Signed-off-by: James Bottomley [PATCH] SCSI aic7xxx_old.c: make a function static The patch below makes the needlessly global function aic7xxx_info static. Signed-off-by: Adrian Bunk Signed-off-by: James Bottomley [PATCH] remove bouncing email address of Deanna Bonds The patch below (applies against both 2.4 and 2.6) removes the bouncing email address of Deanna Bonds (I didn't find a nmore recent address). Signed-off-by: Adrian Bunk Signed-off-by: James Bottomley [NETFILTER]: Release dst_entry in PRE_ROUTING after NAT Fixes NAT on loopback. Signed-off-by: Patrick McHardy [NETFILTER]: Remove CONFIG_IP_NF_NAT_LOCAL config option Signed-off-by: Patrick McHardy [NETFILTER]: Save a level of indentation in icmp_reply_translation Signed-off-by: Patrick McHardy [NETFILTER]: Apply PRE_ROUTING manips in LOCAL_OUT for locally generated icmp errors Locally generated ICMP errors never hit PRE_ROUTING. Fixes invalid addressed ICMP errors for SNATed packets. Signed-off-by: Patrick McHardy [NETFILTER]: Verify NAT manips have been applied before reversing them in icmp_reply_translation ICMP errors may be generated for packets that don't have all NAT manips applied yet. Verify manips have been applied before reversing them. Signed-off-by: Patrick McHardy MTD: Fix RedBoot FIS table detection. The redboot partitioning code currently searches for a FIS table entry named "RedBoot" (including the trailing \0) in one of the first three entries to try and verify that the sector it is looking at really is a FIS partition table. Firstly it fails when RedBoot is stored in some other Flash chip such as a boot PROM, in this case there is no "RedBoot" entry in the partition table. However there will always be a "FIS directory" entry in any valid FIS directory. Secondly it can fail since the RedBoot entry is not always in the first 3 slots -- this can happen for example on an x86 based platform where the processor expects the boot device to be at the top of memory rather than the bottom and so RedBoot is at the end of the flash device. Equally when using "FIS directory" as the entry to search for it very likely is not in the first three since the directory is typically at the end of the flash. Signed-off-by: Ian Campbell Signed-off-by: David Woodhouse MTD: mtdpart_setup() is used from platform code. Remove 'static'. Signed-off-by: Jarkko Lavinen Signed-off-by: David Woodhouse MTD: Make phram work again o Add simple usage example. o Fix up unit handling (k, M. G). o Use correct size limit for setup string. o Remove pointless printk message. o Fix mtdblock to not allocate a read-modify-write buffer for CAP_RAM devices (and fail to do so, returning an error). Signed-off-by: Jörn Engel Signed-off-by: David Woodhouse MTD: Fix oops on erase in NFTL/INFTL (again). Only this time, set the field we were dereferencing _after_ we zero it not before. Signed-off-by: Kalev Lember Signed-off-by: David Woodhouse MTD: Fix timing setup for NAND flash on Samsung S3C2410. Spotted by Shannon Holland. Signed-off-by: Ben Dooks Signed-off-by: David Woodhouse MTD: Support NOR and NAND flash on Sharp SL Series PDAs. Signed-off-by: Richard Purdie Signed-off-by: David Woodhouse MTD: NAND flash simulator We want to contribute the NAND flash simulator which we successfully use when developing JFFS2. It is very useful tool which helps to develop without real flash. It supports several options which allow to select which flash type to emulate (just flash IDs which are in nand_ids.c file). See modinfo for other options. Signed-off-by: Artem B. Bityuckiy Signed-off-by: David Woodhouse JFFS2: Fix oops in read_inode. If the node with lowest version in a file was in the range 0xfffff000 onwards, we oopsed because our rounding was bogus. Signed-off-by: David Woodhouse JFFS2: Various fixes for recent RAM use reduction. We recently started trying to merge the jffs2_raw_node_ref structures for adjacent obsolete nodes. Fix a bunch of problems in that: - Free inode cache structure for a dead inode which is now completely gone. - Fix races vs. erase code which also walks the same lists. - Fix BUG() when trying to merge with a node which still belongs to an inode. Signed-off-by: David Woodhouse JFFS2: Initialise bad_count for each eraseblock correctly. Patch from Estelle Hammache Signed-off-by: David Woodhouse JFFS2: Split eraseblock refiling into separate function ...in preparation for further cleanups from Estelle Hammache which will need to call it from elsewhere. Signed-off-by: David Woodhouse JFFS2: Allow NAND driver to disable virtual eraseblocks. In order to keep the RAM usage down with large devices and smaller erase block sizes, we were using blocks in JFFS2 larger than the physical erase size. This means that bad blocks lose a lot more space though; allow it to be disabled. Signed-off-by: Thomas Gleixner Signed-off-by: David Woodhouse rslib: Spelling fixes. Some of the fixes from Joe Perches got missed out of the last update. Signed-off-by: Thomas Gleixner Signed-off-by: David Woodhouse NULL noise removal, missing __iomem in a couple of declarations, removal of bogus cast to void * in iounmap() calls. Signed-off-by: Al Viro Signed-off-by: David Woodhouse [ALSA] Fix the missing line in the patch for hdsp accurate_ptr RME HDSP driver Fixed missing line in the patch for hdsp accurate_ptr. Signed-off-by: Takashi Iwai [ALSA] Fix WM8770 Init ICE1712 driver Corrects WM8770 ADC mux initialization Signed-off-by: Peter Christensen Signed-off-by: Takashi Iwai [ALSA] AC97 quirks for Dell Intel8x0 driver Added ac97 quirks for some dell machines. Signed-off-by: Takashi Iwai [ALSA] fix sleep in atomic during prepare callback Intel8x0 driver Fixed the sleep in spinlock during prepare callback. This happened only on Nforce chips. Signed-off-by: Takashi Iwai [ALSA] Add pci_save_state() in suspend ALSA Core Added pci_save_state() in the ALSA suspend core callback since its call was removed from the PCI core driver in the recent version. Signed-off-by: Takashi Iwai [ALSA] add Line/Headphone jack detection for AD1981A/B AC97 Codec Core Line/Headphone Jack Sense mixer switches are added to AD1981A/B. HP sense is activated as default. Signed-off-by: Takashi Iwai [ALSA] fix iomem mmap PCM Midlevel The patch adds the definition vm_private_data again to snd_pcm_lib_mmap_iomem(). It got lost during the rewrite of the mmap stuff. Signed-off-by: Martin Langer Signed-off-by: Takashi Iwai [ALSA] fix MIDI GS chorus/reverb mode ALSA sequencer Fixed the parsing of MIDI GS chorus/reverb mode SYSEX messages. They were swapped. Signed-off-by: Takashi Iwai [ALSA] fix chorus/reverb FX loader EMU8000 driver Fixed the chorus/reverb FX loader callback. The header bytes must be eliminated. Signed-off-by: Takashi Iwai [ALSA] minor send routing cleanup EMU10K1/EMU10K2 driver Use snd_emu10k1_compose_audigy_fxrt* macro where appropriate. Also add a few comments to explain send routing structure. Signed-off-by: Lee Revell Signed-off-by: Takashi Iwai [ALSA] whitespace cleanup EMU10K1/EMU10K2 driver Signed-off-by: Lee Revell Signed-off-by: Takashi Iwai [ALSA] Add pci_disable_device() to removal and error paths ALS4000 driver,ATIIXP driver,ATIIXP-modem driver,AZT3328 driver BT87x driver,CMIPCI driver,CS4281 driver,ENS1370/1+ driver ES1938 driver,ES1968 driver,FM801 driver,Intel8x0 driver Intel8x0-modem driver,Maestro3 driver,RME32 driver,RME96 driver SonicVibes driver,VIA82xx driver,ALI5451 driver,CS46xx driver EMU10K1/EMU10K2 driver,ICE1712 driver,ICE1724 driver,KORG1212 driver MIXART driver,NM256 driver,RME HDSP driver,RME9652 driver Trident driver,Digigram VX222 driver,YMFPCI driver pci_disable_device() is called properly in the removal and error paths. Also, the pci_set_master() is added to the resume callbacks if missing (just to be sure). Signed-off-by: Takashi Iwai [ALSA] Addition of pci_disable_device() and cleanup Documentation Added pci_disable_device() in the removal and error paths. Replaced with C-style comments in many places. Signed-off-by: Takashi Iwai [ALSA] via82xx: Enable DXS on ABIT KV8 Pro VIA82xx driver From Marko Kreen : Enable DXS for via audio chip on ABIT KV8 Pro. Works for me. Signed-off-by: Takashi Iwai [ALSA] sort DXS whitelist VIA82xx driver DXS whitelist entries are sorted. Signed-off-by: Takashi Iwai [ALSA] AD18xx/19xx resume fix AC97 Codec Core Added resume callback so that the codec-specific resume code can be called properly. Moved AD-specific initialization code into it. Signed-off-by: Takashi Iwai [ALSA] fix display of send routing in /proc EMU10K1/EMU10K2 driver The emu10k1's send routing is a per channel setting. The emu10k1 has 4 sends per channel, the Audigy 8. Currently we only display the first 4 sends for the first channel. This patch updates the /proc file to display the send routing for all 64 channels, and to display the last 4 sends if present. Signed-off-by: Lee Revell Signed-off-by: Takashi Iwai [ALSA] ALSA PCI drivers: misc cleanups EMU10K1/EMU10K2 driver,Trident driver,YMFPCI driver,AZT3328 driver SonicVibes driver,AC97 Codec Core,CS46xx driver,KORG1212 driver RME HDSP driver,RME9652 driver The patch below does the following cleanups under sound/pci/ : - make some needlessly global code static - remove the following unused EXPORT_SYMBOL's: - trident/trident_main.c: snd_trident_clear_voices - trident/trident_main.c: snd_trident_synth_bzero - remove the following unused global functions: - azt3328.c: snd_azf3328_mixer_read - emu10k1/io.c: snd_emu10k1_sum_vol_attn - trident/trident_main.c: snd_trident_detach_synthesizer - trident/trident_memory.c: snd_trident_synth_bzero Signed-off-by: Adrian Bunk Signed-off-by: Takashi Iwai [ALSA] Fixes the 'It disables the right channel' bug EMU10K1/EMU10K2 driver Signed-off-by: James Courtier-Dutton Signed-off-by: Takashi Iwai [ALSA] Support for Audigy2 Value SB0400 EMU10K1/EMU10K2 driver This adds support for the SB0400 version of the Audigy2 Value, that uses the new CA0108 chip. Signed-off-by: James Courtier-Dutton Signed-off-by: Takashi Iwai [ALSA] make some code static Sound Core PDAudioCF driver The patch below makes some needlessly global code static. Signed-off-by: Adrian Bunk Signed-off-by: Takashi Iwai [ALSA] check CONFIG_COMPAT for snd-ioctl32 ALSA Core check CONFIG_COMPAT for selection of snd-ioctl32 module. Signed-off-by: Takashi Iwai [ALSA] add register dump to proc EMU10K1/EMU10K2 driver The register dump proc files are added. Enabled only when CONFIG_SND_DEBUG=y. Signed-off-by: James Courtier-Dutton Signed-off-by: Takashi Iwai [ALSA] Fixed problem with changing size of etram EMU10K1/EMU10K2 driver It doesn't work to double the etram size, because of a bad comparison statement. Signed-off-by: Mikael Magnusson Signed-off-by: Takashi Iwai [ALSA] [trivial] Fix compile warnings ALSA Core,CS46xx driver,MIXART driver Fix trivial compile warnings. Signed-off-by: Takashi Iwai [ALSA] Fix the detection of Audigy2 ZS EMU10K1/EMU10K2 driver Fix the detection of the older model of Audigy2 ZS. Signed-off-by: Takashi Iwai [ALSA] fix weird placement of static keyword in sound/core/pcm_memory.c PCM Midlevel This patch moves the 'static' keyword to the beginning of the declaration to eliminate the following warning when building with gcc -W sound/core/pcm_memory.c:40: warning: 'static' is not at beginning of declaration This has no actal imact on the code, but it's one less warning to sift through when looking for potential trouble-code with -W I have a hard time thinking of a reason to not apply this trivial patch :) Signed-off-by: Jesper Juhl Signed-off-by: Takashi Iwai [ALSA] misc clean up Intel8x0 driver Clean up the module init code after removal of midi/joystick support. Signed-off-by: Takashi Iwai [ALSA] ALSA ISA drivers: misc cleanups AD1848 driver,CS4231 driver,ES1688 driver,GUS Library,SB drivers Wavefront drivers,ES18xx driver,OPL3SA2 driver,AD1816A driver Opti9xx drivers,SB16/AWE driver,SB8 driver The patch below makes cleanups under sound/isa/ including: - make needlessly global code static - ad1816a/ad1816a_lib.c: much code was unused starting with the global function snd_ad1816a_timer - removed EXPORT_SYMBOL's: - cs423x/cs4231_lib.c: snd_cs4231_outm - es1688/es1688_lib.c: snd_es1688_mixer_read Signed-off-by: Adrian Bunk Signed-off-by: Takashi Iwai [ALSA] Disable 'IEC958 Input Monitor' switch for ALC codecs AC97 Codec Core 'IEC958 Input Monitor' switch for ALC codecs produces only the cracky noises, and KDE seems to turn this on as default in the initialization. So, better to remove this switch. Signed-off-by: Takashi Iwai [ALSA] alternate ALS0200 ident string ALS100 driver The patch adds an alternate ALS0200 ident string Signed-off-by: Pang Lih Wuei Signed-off-by: Takashi Iwai [ALSA] alternate CS4235 ident string CS4236+ driver The patch adds an alternate CS4235 ident string Signed-off-by: Pang Lih Wuei Signed-off-by: Takashi Iwai MTD: Use select in Kconfig where appropriate. Signed-off-by: Adrian Bunk Signed-off-by: David Woodhouse JFFS2: Include vmalloc.h to fix compile warning. The previous fixes to use vmalloc for the eraseblock array missed it. Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse JFFS2: jffs2_get_inode_nodes(): Remove gratuitous memset on new nodes. We're about to set every field in the node structure. No need to spend time filling it with zeroes. This is a fairly hot path. Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse [IPV4]: FIB cleanup, rtmsg_fib() Based largely upon a patch by Robert Olsson. Abstract out rtmsg_fib() so that it does not depend upon fib_hash internal datastructures, move it to fib_semantics.c Signed-off-by: David S. Miller [IPV4]: FIB cleanup, fib_find_alias() Abstract out fib_node arg usage in fib_find_alias(), move to fib_semantics.c Signed-off-by: David S. Miller [IPV4]: FIB cleanup, fib_detect_death() Remove dependancy upon fib_hash specific fn_hash_select_dflt and move it over to fib_semantics.c Signed-off-by: David S. Miller [CRYPTO]: Make some code static This patch below makes some needlessly global code static. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller [CRYPTO]: Standalone VIA PadLock driver. Signed-off-by: David S. Miller [RTNETLINK]: Link attribute modification by interface name. This patch allows modification of link attributes by the interface name, avoids the requirement of translating a name to an ifindex first, and provides better atomicity to userspace. It works by setting the ifindex to a negative number and provide the interface name via the IFLA_IFNAME TLV and let the kernel lookup the device by name instead of ifindex. Changing the interface name will not work using this method because IFLA_IFNAME also transports the new interface name. The patch also fixes a possible source for bugs if IFNAMSIZ is ever changed to a number not aligned to RTA_ALIGNTO. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller JFS: add security and trusted xattrs Signed-off-by: Dave Kleikamp [PATCH] via-rhine: WOL band-aid After I disabled legacy WOL (i.e. controlled by EEPROM rather than driver) in 2.6.9, several people reported regressions. Legacy WOL had worked for them, but now it didn't anymore. The Right Way (TM) to fix this will get the driver to set up working WOL for all hardware, but a simpler solution will have to do for the time being: If a user requests magic packet WOL, the driver re-enables legacy WOL. Yeah, I know it's cheating. This version applies against -mm. I suggest to put it there for testing and into 2.6.11 if feedback is good. Thanks to Pavel Ruzicka for testing. Roger Signed-off-by: Roger Luethi Signed-off-by: Jeff Garzik JFS: speed up nointegrity mount No need to flush metadata to advance the journal's sync-point if we're not really writing to the journal Signed-off-by: Dave Kleikamp [ACPI] fix VIA IRQ issue by enabling VIA quirk http://bugzilla.kernel.org/show_bug.cgi?id=3319 Signed-off-by: David Shaohua Li Signed-off-by: Len Brown [ACPI] S3 resume using RTC http://bugzilla.kernel.org/show_bug.cgi?id=1320 By: Patrick Mochel, Karol Kozimor, Shaohua Li Signed-off-by: Len Brown [ACPI] fix "Error getting context for object" warning http://bugzilla.kernel.org/show_bug.cgi?id=3805 Signed-off-by: David Shaohua Li Signed-off-by: Len Brown [ACPI] add "acpi_fake_ecdt" workaround for Gateway: ex_access_region Region EmbeddedControl(3) has no handler http://bugzilla.kernel.org/show_bug.cgi?id=1690 Signed-off-by: David Shaohua Li Signed-off-by: Len Brown [ACPI] handle GPE sharing between button and lid http://bugzilla.kernel.org/show_bug.cgi?id=3518 Signed-off-by: David Shaohua Li Signed-off-by: Len Brown [ACPI] fixes from stack consumption audit http://bugzilla.kernel.org/show_bug.cgi?id=2901 Signed-off-by: Luming Yu Signed-off-by: Len Brown build fix [ACPI] 32-bit EC access http://bugzilla.kernel.org/show_bug.cgi?id=1744 Signed-off-by: Luming Yu Signed-off-by: Len Brown Rename SCSI ChangeLog to reflect its venarability The last time it was modified was 1997, so add the date range of applicability to its name. Signed-off-by: James Bottomley build fix [PATCH] Move MCA_bus to linux/mca.h - Move MCA_bus declaration from to - Define it in mca.c, not setup.c - EXPORT_SYMBOL it at the site of its definition. - Fix up random files to include for the use of the MCA_bus symbol - Delete some unnecessary ifdefs. - Delete some unneeded comments. Signed-off-by: James Bottomley [PATCH] i386 mca.c: small cleanups The patch below contains the following cleanups: - make spinlock mca_lock static - make struct mca_standard_resources static Signed-off-by: Adrian Bunk Signed-off-by: James Bottomley [ACPI] ACPICA 20041119 from Bob Moore Fixed a problem in acpi_ex_convert_to_integer where new integers were not truncated to 32 bits for 32-bit ACPI tables. This routine converts buffers and strings to integers. Implemented support to store a value to an Index() on a String object. This is an ACPI 2.0 feature that had not yet been implemented. Implemented new behavior for storing objects to individual package elements (via the Index() operator). The previous behavior was to invoke the implicit conversion rules if an object was already present at the index. The new behavior is to simply delete any existing object and directly store the new object. Although the ACPI specification seems unclear on this subject, other ACPI implementations behave in this manner. (This is the root of the AE_BAD_HEX_CONSTANT issue.) Modified the RSDP memory scan mechanism to support the extended checksum for ACPI 2.0 (and above) RSDPs. Note that the search continues until a valid RSDP signature is found with a valid checksum. Signed-off-by: Len Brown [ACPI] ACPICA 20041203 from Bob Moore and Alexey Starikovskiy The low-level field insertion/extraction code (exfldio) has been completely rewritten to eliminate unnecessary complexity, bugs, and boundary conditions. Fixed a problem in the ToInteger, ToBuffer, ToHexString, and ToDecimalString operators where the input operand could be inadvertently deleted if no conversion was necessary (e.g., if the input to ToInteger was an Integer object.) Fixed a problem with the ToDecimalString and ToHexString where an incorrect exception code was returned if the resulting string would be > 200 chars. AE_STRING_LIMIT is now returned. Fixed a problem with the Concatenate operator where AE_OK was always returned, even if the operation failed. Fixed a problem in oswinxf (used by AcpiExec and iASL) to allow > 128 semaphores to be allocated. Signed-off-by: Len Brown JFS: flush new iag from bd_inode's mapping This is a fix to help jfs work with grub. A new IAG is created in the bd_inode's mapping, but subsequently modified in a different mapping. We should invalidate the former page to keep grub from using that cached page. It isn't useful to have it cached anyway, since jfs will never access it again through that mapping. Signed-off-by: Dave Kleikamp drivers/block/floppy: kill #include linux/version.h Appears to be unnecessary. [ACPI] remove duplicate _PDC #defines resulting from mis-merge Signed-off-by: Zhenyu Z. Wang Signed-off-by: Len Brown [ACPI] fix polarity of CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI message Signed-off-by: Len Brown [IA64-SGI] Add support for a future SGI chipset (shub2) 1of4 This patch changes the SN macros for calulating the addresses of shub MMRs. Functionally, shub1 (current chipset) and shub2 are very similar. The primary differences are in the addresses of MMRs and in the location of the NASID (node number) in a physical address. This patch adds the basic infrastructure for running a single binary kernel image on either shub1 or shub2. Signed-off-by: Jack Steiner [IA64-SGI] Add support for a future SGI chipset (shub2) 2of4 This patch adds the addresses of shub2 MMRS to the shub_mmr header file. During boot, a SAL call is made to determine the type of the shub. Platform initialization sets the appropriate MMR addresses for the platform. A new macro (is_shub1() & is_shub2()) can be used at runtime to determine the type of the shub. Signed-off-by: Jack Steiner [IA64-SGI] Add support for a future SGI chipset (shub2) 3of4 Change the IPI & TLB flushing code so that it works on both shub1 & shub2. Signed-off-by: Jack Steiner [IA64-SGI] Add support for a future SGI chipset (shub2) 4of4 Change the code that manages the LEDs so that it works on both shub1 & shub2. Signed-off-by: Jack Steiner [IA64] Cachealign jiffies_64 to prevent unexpected aliasing in the caches. On large systems, system overhead on cpu 0 is higher than on other cpus. On a completely idle 512p system, the average amount of system time on cpu 0 is 2.4% and .15% on cpu 1-511. A second interesting data point is that if I run a busy-loop program on cpus 1-511, the system overhead on cpu 0 drops significantly. I moved the timekeeper to cpu 1. The excessive system time moved to cpu 1 and the system time on cpu 0 dropped to .2%. Further investigation showed that the problem was caused by false sharing of the cacheline containing jiffies_64. On the kernel that I was running, both jiffies_64 & pal_halt share the same cacheline. Idle cpus are frequently accessing pal_halt. Minor kernel changes (including some of the debugging code that I used to find the problem :-( ) can cause variables to move & change the false sharing - the symptoms of the problem can change or disappear. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck [IA64] per cpu MCA/INIT save areas Linux currently has one MCA & INIT save area for saving stack and other data. This patch creates per cpu MCA save areas, so that each cpu can save its own MCA stack data. CPU register ar.k3 is used to hold a physical address pointer to the cpuinfo structure. The cpuinfo structure has a physical address pointer to the MCA save area. The MCA handler runs in physical mode and the physical address pointer avoids the problems associated with doing the virtual to physical translation. The per MCA save areas replace the global areas defined in arch/ia64/kernel/mca.c for MCA processor state dump, MCA stack, MCA stack frame, and MCA bspstore. The code to access those save areas is updated to use the per cpu save areas. No changes are made to the MCA flow, ie all the old locks are still in place. The point of this patch is to establish the per cpu save areas. Additional usage of the save areas, such as enabling concurrent INIT or MCA handling, will be the subject of other patches. Signed-off-by: Russ Anderson Signed-off-by: Tony Luck [IA64] hardirq.h: Add declaration for ack_bad_irq(). Cleanup a warning from my irq merge. Signed-off-by: Tony Luck SCSI: Quieten the incorrect state change message Make it a scsi logging message instead (under ERROR_RECOVERY) Signed-off-by: James Bottomley MTD: Increase nand_write_page() ECC buffer size to cope with 12-byte ECC. Signed-off-by: Jarkko Lavinen Signed-off-by: David Woodhouse JFFS2 locking fix: Don't hold references to obsolete nodes without lock. Signed-off-by: David Woodhouse MTD NAND flash simulator update. - Default to 8MiB instead of 64MiB - Use generic LP NAND command definition - Other minor cleanup Signed-off-by: Artem Bityuckiy Signed-off-by: David Woodhouse MTD: phram device cleanup. - Remove old slram-like interface - Fewer memory allocations - Saner usage example Signed-off-by: Jörn Engel Signed-off-by: David Woodhouse MTD: AMD/Fujitsu flash driver cleanup Use get_chip() helper function instead of open-coding it. Signed-off-by: David Vrabel Signed-off-by: David Woodhouse MTD: Use JEDEC probe for flash chips on Ebony board. Signed-off-by: Heikki O Lindholm Signed-off-by: David Woodhouse MTD: New mapping driver for IBM 405GP 'Walnut' board. Signed-off-by: Heikki O Lindholm Signed-off-by: David Woodhouse MTD XIP support: allyesconfig compile fix. People insist on turning stuff on which doesn't make sense. Make it compile in the case where the platform doesn't provide the necessary XIP primitives. Signed-off-by: Nicolas Pitre Signed-off-by: David Woodhouse [IA64] Add TIF_SIGDELAYED, delay a signal until it is safe Some of the work on recoverable MCA events has a requirement to send a signal to a user process. But it is not safe to send signals from MCA/INIT/NMI/PMI, because the rest of the kernel is an unknown state. This patch adds set_sigdelayed() which is called from the problem contexts to set the delayed signal. The delayed signal will be delivered from the right context on the next transition from kernel to user space. If TIF_SIGDELAYED is set when we run ia64_leave_kernel or ia64_leave_syscall then the delayed signal is delivered and cleared. All code for sigdelayed processing is on the slow paths. A recoverable MCA handler that wants to kill a user task just does set_sigdelayed(pid, signo, code, addr); Signed-off-by: Keith Owens Signed-off-by: Tony Luck [IA64] Clear all corrected records as they occur Because MCA events are not irq safe, they cannot be logged via salinfo_decode at the time that they occur. Instead kernel salinfo.c runs a timer every few minutes to check for and to clear corrected MCA records. If a second recoverable MCA occurs on the same cpu before salinfo_decode has cleared the first record then OS_MCA reads the record for the first MCA from SAL, which passes invalid data to the MCA recovery routines. This patch treats all corrected records the same way, by clearing the records from SAL as soon as they occur. CMC and CPE records are cleared as they are read. Recoverable MCA records are cleared at the time that we decide they can be corrected. If salinfo_decode is not running or is backlogged then we lose some logging, but that has always been the case for corrected errors. Signed-off-by: Keith Owens Signed-off-by: Tony Luck [IA64] Drop SALINFO_TIMER_DELAY from 5 minutes to 1 minute Experience with recoverable MCA events shows that a poll interval of 5 minutes for new MCA/INIT records is a bit too long. Drop the poll interval to one minute. Signed-off-by: Keith Owens Signed-off-by: Tony Luck [AGPGART] amd64-agp.c replace pci_find_device with pci_get_device As pci_find_device is going away soon I have converted this file to use pci_get_device instead. I have compile tested it. If anyone has this hardware and could test it that would be great. Signed-off-by: Hanna Linder Signed-off-by: Dave Jones [AGPGART] generic.c: replace pci_find_device with pci_get_device As pci_find_device is going away soon I have converted this file to use pci_get_device instead. I have compile tested it. If anyone has this hardware and could test it that would be great. Signed-off-by: Hanna Linder Signed-off-by: Dave Jones [AGPGART] intel-agp.c: replace pci_find_device with pci_get_device As pci_find_device is going away soon I have converted this file to use pci_get_device instead. I have compile tested it. If anyone has this hardware and could test it that would be great. Signed-off-by: Hanna Linder Signed-off-by: Dave Jones [AGPGART] intel-mch-agp.c: replace pci_find_device with pci_get_device As pci_find_device is going away soon I have converted this file to use pci_get_device instead. I have compile tested it. If anyone has this hardware and could test it that would be great. Signed-off-by: Hanna Linder Signed-off-by: Dave Jones [AGPGART] isoch.c: replace pci_find_device with pci_get_device As pci_find_device is going away soon I have converted this file to use pci_get_device instead. for_each_pci_dev is just a macro wrapper around pci_get_device. I have compile tested it. If anyone has this hardware and could test it that would be great. Signed-off-by: Hanna Linder Signed-off-by: Dave Jones [AGPGART] Simplify global_cache_flush on_each_cpu does the right thing in the UP case, so we can kill those ugly ifdefs. Signed-off-by: Dave Jones [AGPGART] Add support for ALI M1681/M1683 Signed-off-by: Dave Jones [AGPGART] Fix TLB flushing issues with change_page_attr() Calls to change_page_attr() need an explicit call to global_flush_tlb() afterwards. The AGP code didn't do this in a number of cases. This patch makes map_page_into_agp/unmap_page_from_agp do the calls themselves, which takes care of most of the problem. The Intel AGP driver also has some slightly different calls to what map_page_into_agp() does, as it changes 4 contiguous pages. Introduce explicit flushes afterwards there too. Thanks to Alan Cox for pointing this out. Signed-off-by: Dave Jones [PATCH] add class_device to miscdevice Currently misc_register() throws away the return from class_simple_device_add(). This makes it impossible to get to the class_device of the directories in /sys/class/misc and, for example, thus impossible to add attributes to those directories. Attached patch adds a class_device structure to the miscdevice structure and assigns to it the value returned from class_simple_device_add() in misc_register(), thus caching the value and allowing us to f.e. later call class_device_create_file(). We need this for inotify, but I can see plenty of other misc. devices wanting this and consider it missing but required functionality. Add the class_device structure to miscdevice so that we can add sysfs attributes to /sys/class/misc/foo Signed-Off-By: Robert Love Signed-off-by: Greg Kroah-Hartman [PATCH] misc: remove device.h #include from miscdevice.h Signed-off-by: Greg Kroah-Hartman Documentation: fix some grammer in the stable_api_nonsense.txt file Thanks to Andries Brouwer for pointing this out. [PATCH] select HOTPLUG The patch below changes all dependencies on HOTPLUG to selects. The help text of HOTPLUG is adjusted in a way, that manually selecting it is only required for external modules. If an option already depends on PCMCIA or selects FW_LOADER an explicit select of HOTPLUG is not required. Signed-off-by: Adrian Bunk Signed-off-by: Greg Kroah-Hartman [PATCH] misc: remove miscdevice.h from pci hotplug drivers as they do not need it. Signed-off-by: Greg Kroah-Hartman [PATCH] w1: Documentation bits for generic w1 behaviour. Documentation bits for generic w1 behaviour. Signed-off-by: Evgeniy Polyakov Signed-off-by: Greg Kroah-Hartman [PATCH] I2C: use chip driver name to request regions > Trivial patch against 2.6.10-rc3-bk2. > When request_region is called name is set to "", use module name. Correct. I even think we should do the same for other chip drivers. For now they all use an arbitrary string. Using the driver's name would be both more consistent and more efficient (spares some bytes of memory). Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman [PATCH] I2C: i2c-algo-bit should support I2C_FUNC_I2C Signed-off-by: Ian Campbell Signed-off-by: Greg Kroah-Hartman [PATCH] I2C: i2c-algo-bit should support I2C_FUNC_I2C > Very few drivers seem to support the I2C_FUNC_I2C functionality, is > there a reason for that? Yes, most bus drivers are for SMBus, not I2C, masters. SMBus is a subset of I2C. These SMBus master are fully I2C-capable, although in most cases it doesn't matter. Most chip drivers are for SMBus clients as well. Almost all hardware monitoring chips are SMBus devices. So it's not surprising not to see I2C_FUNC_I2C widely used. > I have an I2C bus on my platform constructed from a couple of GPIO lines > using the i2c-algo-bit driver. The device on the bus is a DS1307 I2C RTC > and the driver for that currently checks for > I2C_FUNC_SMBUS_WORD_DATA | I2C_FUNC_SMBUS_WRITE_BYTE > however the datasheet suggests it is a simple i2c device with none of > this smbus stuff, Russell King queried this here > http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=2021/1 First, note that all SMBus commands are valid I2C transfers. In fact, the SMBus specification doesn't do much except put names on specific I2C transfers, and sometimes give meanings to the data being transfered. As a result, "I2C" clients, although not specifically SMBus-compatible, may enjoy SMBus commands such as SMBUS_READ_BYTE, SMBUS_READ_BYTE_DATA and SMBUS_WRITE_BYTE_DATA. One typical example of that are EEPROMs and the eeprom driver. It relies on i2c_smbus_write_byte and i2c_smbus_read_byte (BTW I just noticed that it doesn't properly check for these functionalities... have to fix that) because it's so much easier to call these standard functions than rewrite I2C code manually. Also, this let us access the EEPROMs on SMBus (non-I2C) busses. Now, note that Russell is not quite correct in is assertion: "Do we really require SMBUS functionality, or is i2c functionality sufficient?" It's actually the other way around. SMBus puts restrictions on what a valid I2C transfer is. "Do we need the full I2C functionality or is the SMBus subset sufficient?" would make more sense. As for the exact functionality you need to check, let's just see how you access the bus. As far as I can see (providing that the code below Russell's comments is still valid), you rely on: * i2c_smbus_write_byte_data * i2c_smbus_read_byte_data * i2c_transfer So yo *need* to check for the availability of I2C_FUNC_SMBUS_BYTE_DATA on the adapter (which is part of I2C_FUNC_SMBUS_EMUL so any bus driver using that, including any relying on i2c-algo-bit, will work with your client driver) for the first two. And you also need to check for I2C_FUNC_I2C for the third one. Of course, any adapter with I2C_FUNC_I2C will be able to do SMBus byte data transfers, but since you do not use i2c_transfer to do them, you need to check the functionality separately (I think). Also, I think that what you do with i2c_transfer is similar to I2C_FUNC_SMBUS_READ_I2C_BLOCK, which is supported by some SMBus (non-I2C) masters. If you convert your code to use i2c_smbus_read_i2c_block_data, then you don't rely on the full I2C capatbilities of the bus, which means that your chip driver will be useable on more plateforms. That said, note that this feature is unimplemented on most SMBus master drivers as of now, and broken on a number of others (but I guess we would start paying more attention to them if there were more users for this function). > If I change it to a check for I2C_FUNC_I2C and change the algo-bit > driver to declare I2C_FUNC_I2C then the driver continues to work fine. You are right that i2c-algo-bit should declare itself I2C_FUNC_I2C capable. I even think that every bus being I2C_FUNC_SMBUS_EMUL capable is very likely to be I2C_FUNC_I2C capable. This means that other algorithms (ite, pcf, maybe pca) could most probably be declared I2C_FUNC_I2C capable as well. Can anyone confirm? > Given the above, is the following patch appropriate, or is there > something about the relationship between i2c and smbus that I don't > understand. I admit that the relationship between I2C and SMBus is somewhat tricky, it took me some time to get it and even then I am sometimes not sure to understand exactly what implies what ;) So we cannot blame you for not getting it at first. I hope I helped make things a little clearer. If not I welcome questions. Whether or not you change your code to use SMBus only in your driver to make it more widely useable, your patch to i2c-algo-bit is valid, I am signing it too and will apply it to the 2.4 version of the driver as well. Signed-off-by: Ian Campbell Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman [PATCH] it87: /proc/ioports fix When request_region is called name is set to "", use module name. Signed-off-by: Domen Puncer Signed-off-by: Greg Kroah-Hartman [PATCH] USB: serial driver for TI USB 3410/5052 chips (3/3) * New USB serial driver for devices based on the TI USB 3410 and 5052 chips. (3/3) Signed-off-by: Al Borchers Signed-off-by: Greg Kroah-Hartman [PATCH] USB: serial driver for TI USB 3410/5052 chips (1/3) * New USB serial driver for devices based on the TI USB 3410 and 5052 chips. (1/3) * Changed to use circ_buf.h. Signed-off-by: Al Borchers Signed-off-by: Greg Kroah-Hartman [PATCH] USB: serial driver for TI USB 3410/5052 chips (2/3) * New USB serial driver for devices based on the TI USB 3410 and 5052 chips. (2/3) Signed-off-by: Al Borchers Signed-off-by: Greg Kroah-Hartman [PATCH] USB: fix sparse and compiler warnings in ti_usb_3410_5052.c Signed-off-by: Greg Kroah-Hartman [PATCH] USB: SN9C10x driver updates SN9C10x driver updates. Changes: + new, - removed, * cleanup, @ bugfix, = sync with kernels * SN9C10x system clock fine-tuning when switching from native to compressed format and viceversa for each image sensor + Add sn9c102_i2c_try_raw_read() + Frame header available from sysfs interface + Documentation updates: new "Video frame formats" paragraph, new entries in "Credits" and frame header description + Plugin's for HV7131D and MI-0343 image sensors Signed-off-by: Luca Risolia Signed-off-by: Greg Kroah-Hartman [PATCH] USB Storage: support 'bulk32' devices This patch implements support for what we call "bulk32" devices. These are devices that use the BBB transport mechanism with the slight modification that the CBW is padded to 32 bytes (instead of the standard 31 bytes). Signed-off-by: Theodore Kilgore Signed-off-by: Phil Dibowitz Signed-off-by: Matthew Dharm Signed-off-by: Greg Kroah-Hartman USB: delete the tiglusb driver as it's not needed You can use a userspace program and libusb instead. The author asked for this driver to be removed because of this. Signed-off-by: Greg Kroah-Hartman [PATCH] USB Storage: Remove MODE_XLATE flag from unusual_devs.h This patch removes all MODE_XLATE flags from unusual_devs.h. Since the file is no longer close to in sync with the 2.4 version, Alan and I agreed it's reasonable to remove all of these now. Signed-off-by: Phil Dibowitz Signed-off-by: Greg Kroah-Hartman [PATCH] USB: update drivers/usb/README This just adds a bit more info to drivers/usb/README, mostly just pointing to where documentation is to be found (including current kerneldoc). Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB Storage: Remove old XLATE-only entries from unusual_devs.h This patch removes all entries from unusual_devs.h that appear to have only been there for the MODE_XLATE flag which was removed in my previous patch. Signed-off-by: Phil Dibowitz Signed-off-by: Greg Kroah-Hartman [PATCH] usb-serial: add tty_hangup on disconnect When a USB serial device is disconnected, user applications performing a read() now receive an error code, rather than waiting indefinitely. The included patch is originally from Al Borchers, massaged to apply to 2.6.9 and 2.6.10-rc2. I've tested it on 2.6.9, but not on 2.6.10-rc2. Al Borcher's original post: http://www.mail-archive.com/linux-usb-devel@lists.sourceforge.net/msg27475.html Signed-off-by: Edwin Olson (eolson@mit.edu) Signed-off-by: Greg Kroah-Hartman [PATCH] ftdi_sio: Add sysfs attributes for event character and latency Setting the event character and latency timer can greatly improve performance for some applications which use the ftdi_sio module (a serial->USB converter). The following patch adds sysfs attributes, exposing these configuration registers. Signed-off-by: Edwin Olson (eolson@mit.edu) Signed-off-by: Greg Kroah-Hartman [PATCH] USB: usb_dev->ep[] not usb_dev->epmaxpacket (1/15) This starts updating the usbcore interface to use endpoints in places it previously used pipes or other representations of the endpoint. - add new arrays of "struct usb_host_endpoint" pointers, matching current config and altsetting - get rid of the two epmaxpacket[] arrays; they duplicate information that's now readily accessible from the usb_host_endpoint. - resolve a FIXME by removing a function that only existed because the usb_host_endpoint wasn't readily accessible. It also removes most of an old rant about pipes, trimming it down so only the important bits remain. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: usbcore changes for usb_dev->ep[](2/15) This patch updates usbcore to match interface changes in the previous patch. - udev->ep[] arrays are updated during config change events and as needed during enumeration; - usb_epnum_to_ep_desc() vanishes; - so do the udev->epmaxpacket[] arrays. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: usbfs changes for usb_dev->ep[] (3/15) Updates usbfs to stop using usb_epnum_to_ep_desc(). In the process, it acquires better error detection on its urb submit path. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: ALSA and usb_dev->ep[] (4/15) Makes an ALSA audio driver stop referencing the udev->epmaxpacket[] arrays. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: auerswald and usb_dev->ep[] (5/15) Update the auerswald driver to stop using the now-vanished usb_epnum_to_ep_desc() function. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: CRIS HCD and usb_dev->epmaxpacket (6/15) Makes the CRIS HCD stop referencing the udev->epmaxpacket[] arrays. Also makes it stop providing device allocate/deallocate routines; this HCD doesn't need them, and a later patch will remove that API. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: EHCI HCD and usb_dev->epmaxpacket (7/15) Makes EHCI stop referencing the udev->epmaxpacket[] arrays. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: usbtest and usb_dev->epmaxpacket (8/15) Makes usbtest stop referencing the udev->epmaxpacket[] arrays. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: HCD/usb_bus interface cleanup (9/15) This changes the usbcore interfaces provided to HCDs: - Remove usb_device->hcpriv and it allocation/deallocation hooks - Replace struct hcd_dev with more appropriate per-endpoint state - Update HCD apis to use usb_host_endpoint in key places Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: remove some now-unused HCD infrastructure (10/15) This removes the code supporting usb_device->hcpriv allocation/deallocation, and hcd_dev-dev_list, from usbcore and the HCDs. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: maintain usb_host_endpoint.urb_list (11/15) This patch changes the HCD glue code to use the URB queue now kept in usb_host_endpoint, and matching HCD API changes. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: EHCI and HCD API updates (12/15) Updates the EHCI HCD to match API updates. This includes both new changes (struct hcd_dev gone) and older ones (endpoints now properly enabled/disabled, so "re"init paths aren't needed). Hmm, the ISO stuff could avoid that lookup now too. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: OHCI and HCD API changes (13/15) Updates the OHCI HCD to match API updates. This includes both new changes (struct hcd_dev gone) and older ones (endpoints now properly enabled/disabled, so "re"init paths aren't needed). Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: UHCI and HCD API change (14/15) UHCI changes to match the updated HCD glue calls. Since it's handed the relevant endpoint queue on a silver platter, the driver no longer needs to search anything to find the queue entries. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [PATCH] USB: better messages for "no-IRQ" cases (15/15) This changes the usbcore message about HCD IRQ problems so it makes sense on systems without ACPI or an APIC. It also updates the comments; the issue doesn't appiear only with PCI, and with the recent enumeration changes it doesn't happen just with set_address either. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman [AGPGART] Fix up PCI posting bugs. Also remove a pointless wrapper. Signed-off-by: Dave Jones [AGPGART] More PCI Posting bugs. Signed-off-by: Dave Jones [AGPGART] Another missing PCI Posting bugfix. Signed-off-by: Dave Jones [AGPGART] Remove unnecessary parenthesis on return statements. Si