Linus Torvalds: Linux 2.6.3-rc3

Kernel.org
Mirrors

Full Changelog

More merges, although most of them are architecture updates.
IA64, ppc32/64, SuperH and ARM.

But also some cpufreq, watchdog and ACPI updates.

Linus

—

Summary of changes from v2.6.3-rc2 to v2.6.3-rc3

Alex Williamson:

• ia64: sba_iommu perf tunning and new functionality

Andrew Morton:

• acpi cpu_has_cpufreq() fix
• acpi numa build fix
• Fix buslogic for older gccs
• selinux: Fix bugs in policy loading code
• sh: Update defconfig
• sh: Wrap fb_read/writeX() to __raw_read/writeX()
• sh: RTC fixes
• sh: preempt safe lazy fpu handling
• sh: preempt fixes
• sh: Fix hp680 board support
• sh: hd64461 updates
• sh: Add H8/300 support to sh-sci
• sh: Misc build fixes
• sh: hitfb updates (and accel)
• sh: pvr2fb updates
• Alpha: fix “extern inline” logic for core IO functions
• swap extent merging fix
• Make serial console work for any port
• Fix fadvise() parameter checking
• Suppress reiserfs page allocation wanring
• ppc32: boot and platform fixes
• ppc32: Fix compilation of IBM Spruce &
  !CONFIG_SERIAL_TEXT_DEBUG
• ppc32: use todc time functions for PPC_PREP
• ppc32: IBM 40x and 4xx fixes
• ppc32: PPC4xx cleanup
• ppc32: PPC44x MMU update/fixes
• ppc32: Update IBM Spruce defconfig

Anton Blanchard:

• Fix ppc64 build problem
• add thread_info to oops output
• various xmon cleanups
• cleanup debugger hooks

Bartlomiej Zolnierkiewicz:

• fix build for CONFIG_BLK_DEV_IDEDMA=n

Benjamin Herrenschmidt:

• Export OF device path for PCI devices
• ppc32: Add CONFIG_PPC_PMAC64 when building for G5
• ppc64: Start of PowerMac G5 merge, add all arch and include
  files
• ppc64: Add the Kconfig & Makefile changes related to the
  PowerMac G5 merge
• ppc64: Add some definitions relative to the G5 CPU and POWERMAC
  platform
• ppc64: Add the head.S changes to boot a PowerMac G5
• ppc64: Add support for PowerMacs in the OF client interface
  code (prom.c)
• ppc32: Separate definitions for known vs unknown PowerMac G5
  models
• ppc64: Update the nvram driver to deal with PowerMac G5
• ppc64: Add the PowerMac PCI support
• ppc64: Add the G5 (IBM 970) CPU to the cputable
• ppc64: Add support for PowerMac G5 interrupts
• ppc64: Add the feature_call function pointer to machdep
• ppc64: Remove duplicate (& incorrect) definition of
  kern_add_valid()
• ppc64: Add a missing isync in __hash_page, alloc hash table on
  PowerMac G5
• ppc64: xmon breakpoints are support on PowerMac G5 too
• ppc64: Add support for z85c30 SCCs for low level console
  (PowerMac G5)
• ppc64: Call the PowerMac G5 init routines
• ppc64: Add CPU NAP mode in idle loop on PowerMac G5
• ppc64: Add pciconfig_iobase syscall for 32 bits apps only
• ppc64: Add SMP support for PowerMac G5
• ppc64: Switch off use of polled mode in i2c driver
• ppc64: fix build of pmac “mac-io” IDE driver on 64 bits
  kernel
• ppc64: fix build of ADB driver
• ppc64: Fix a refounting issue in macio_asic
• ppc64: Fix build of via-pmu driver on 64 bits kernel
• ppc64: Fix break handling in pmac_zilog driver, fixes for 64
  bits kernel
• ppc64: Add missing #include, warned on ppc32 and broke build on
  ppc64
• ppc64: Don’t build offb’s code that relies on the BootX
  bootloader on ppc64
• ppc64: Add defconfigs for pSeries and PowerMac G5
• fix rivafb build on ppc64
• New radeonfb
• Fix typo in ppc32 build
• Fix a link conflict between radeonfb and the radeon DRI
• ppc64: CONFIG_PPCPMAC implies
  CONFIGADB_PMU
• ppc64: export clear_user_page
• Fix incorrect kfree in radeonfb

Chas Williams:

• [ATM]: prevent userspace compilation errors with
  glibc-kernheaders
• [ATM]: [he] unconditionalize extra pci reads to flush posted
  writes
• [ATM]: [clip] delay /proc/net/atm/arp creation

Christoph Hellwig:

• [XFS] make sure i_size_write is called under i_sem
• back out fbdev sysfs support

Dave Jones:

• [CPUFREQ] Silence powernow-k7 when built as a module
• [CPUFREQ] fix up CPU detection in p4-clockmod <Dominik
  Brodowski>
• [CPUFREQ] Geode register fixes
• [CPUFREQ] Fix an oops unloading p4-clockmod
• [CPUFREQ] convert powernow-k8 to use frequency tables [1/5] Add
  a struct cpufreq_frequency_table, fill it with content, and use it
  for
• [CPUFREQ] convert powernow-k8 to use frequency tables
  [2/5]
• [CPUFREQ] convert powernow-k8 to use frequency tables [3/5]
  Keep *ppst local to the only function which needs it any
  longer.
• [CPUFREQ] convert powernow-k8 to use frequency tables [4/5]
  Remove the *ppst table, and remove an unneccessary
  forward-declaration
• [CPUFREQ] convert powernow-k8 to use frequency tables [5/5]
  Move the table verification to an extra function.
• [CPUFREQ] powernow-k8 printk cleanups from Pavel
• [CPUFREQ] Pentium-4-M detection fix for speedstep-lib From
  Dominik..
• [CPUFREQ] Fix deadlock in userspace governor
• [CPUFREQ] Fix off-by-1000 error in longhaul
• [CPUFREQ] Remove bogus scaling from longhaul driver freqency
  tables

Dave Kleikamp:

• JFS: rename should update mtime on source and target
  directories
• JFS: Threads should exit with complete_and_exit

David Mosberger:

• ia64: Based on patch by Stephane Eranian: Make fpswa version
  info available via /proc/efi/fpswa, rather than printing it at boot
  time.
• ia64: Fix some more warnings caused by casts used as
  l-values
• ia64: Update defconfig
• ia64: Correct init_task.rbs_bot value (not that it
  matters)
• ia64: Drop some unneeded __KERNEL_SYSCALL__ defines (found by
  Dave Jones) and an unnecessary include of
  <linux/config.h>.
• ia64: Update toolchain-flags with a check for working .align
  inside a now that there is a fixed GAS.

David S. Miller:

• [SPARC64]: Fix exception remaining length calcs in VIS copy
  routines
• Cset exclude: davem@nuts.davemloft.net|ChangeSet|20040212080313|45938

Ingo Molnar:

• open writecount scalability cleanup

Jack Steiner:

• ia64: Enable cpu_vm_mask maintenance and improve SN2 TLB
  flushing

James Simmons:

• framebuffer GPM corruption fix

Jeff Garzik:

• Bump libata, ata_piix to version 1.0
• Cset exclude: jgarzik@redhat.com|ChangeSet|20040213172720|60184

Jens Axboe:

• DVD-R capability flag set incorrectly, /proc formatting
  fix

Jeroen Vreeken:

• [AX25]: Fix locking in ax25_rt_free()

Jesse Barnes:

• ia64: kill misc. warnings

Joe Thornber:

• dm: block size bug with 64 bit devs

Jun Nakajima:

• Remove the assumption that the number of the sibling is 2

Keith M. Wesolowski:

• [SPARC32]: Take parisc atomic_t implementation so they are full
  32-bits
• [SPARC32]: Stub DMA routines to fix the build
• [SPARC32]: Mask PIL in local_irq operations

Keith Owens:

• ia64: Avoid deadlock when using printk() for MCA and INIT
  records
• ia64: mca.c cleanup – Delete all record printing code, moved to
  salinfo_decode in user space
• ia64: mca.c cleanup – Mark variables and functions static where
  possible
• ia64: mca.c cleanup – Delete dead variables and functions
• ia64: mca.c cleanup – Reorder to remove the need for forward
  declarations and to consolidate related code
• ia64: mca.c cleanup – Bjorn’s printk cleanup
• ia64: mca.c – pass irq_safe around
• ia64: mca.c – Fix the “did we recover from MCA test” and move
  it up
• ia64: Periodically forward MCA or INIT records to
  user-level
• ia64: Delete redundant ia64_mca_check_errors()
• ia64: Correct ifdef for srat_num_cpus

Kenneth W. Chen:

• ia64: fix ld.a emulation
• ia64: remove unused cpucount variable

Len Brown:

• [ACPI] nforce2 timer lockup from Maciej W. Rozycki
• [ACPI] NUMA build fix — NR_MEMBLKS is now NR_NODE_MEMBLKS
• [ACPI] delete mention of stale config option ACPI_HT_ONLY
• [ACPI] don’t register disabled processors — it just confuses
  people
• [ACPI] clarify error message in processor.c
• [ACPI] interrupt over-ride fix from i386 (Maciej W.
  Rozycki)

Linus Torvalds:

• Make <linux/compiler.h> a bit more palatable to user
  program inclusion.
• Fix bogus mode bit testing by smbfs
• This reverts the mmap address hint usage for now
• ppc64: remove autogenerated file, and incorrect header
  inclusion
• Fix “bus_for_each_dev()” and “bus_for_each_drv()”, which did
  not correctly handle the “restart from this device/driver” case,
  and caused oopses with ieee1394.
• Fix broken ppc64 kernel debugger call
• Make G5 defconfig a bit saner. In particular, we want firewire
  and we do _not_ want the broken MACZILOG serial driver.
• Remove stale “xmon.h” include
• Linux 2.6.3-rc3

Marcel Holtmann:

• [Bluetooth] Support for tracking the voice setting
• [Bluetooth] Fix race for incoming connections
• [Bluetooth] Fix error handling for not connected socket
• [Bluetooth] Fix several copy_to_user() and reference counting
  glitches
• [Bluetooth] Fix non-blocking socket race conditions

Martin Hicks:

• ia64: Add EXPORT_SYMBOL for SN2 physical_node_map
• ia64: don’t call note_interrupt() for per-CPU irqs

Martin J. Bligh:

• NUMA build fix drivers/acpi/numa.c is IA64 only for now —
  enforce it.

Matthew Wilcox:

• [WATCHDOG] v2.6.2 watchdog-architecture-cleanup

Roman Zippel:

• fix FB_RADEON_I2C dependency

Russell King:

• [ARM] Use __attributeused__ rather than
  _attribute__((used))
• [ARM] Allow sub-architectures to provide their own
  sched_clock()
• [ARM] Fix couple of compiler warnings
• [ARM] Add DMA mask for SA11x0 MCP device
• [ARM] Improve help for CONFIG_ARM_THUMB

Rusty Russell:

• Shut up about the damn modules already

Sridhar Samudrala:

• [SCTP] Use __get_free_pages() to allocate ssnmap
• [SCTP] Fix SCTP_INITMSG set socket option so that a parameter
  with 0 value will not change its current value.
• [SCTP] Fix sctp_getladdrs()/sctp_getpaddrs() API so that the
  port value in the returned addresses is in network byte order.

Wim Van Sebroeck:

• [WATCHDOG] v2.6.2 shwdt-cleanup
• [WATCHDOG] v2.6.2 watchdog-module_*-update
• [WATCHDOG] v2.6.2 acquirewdt-cleanup
• [WATCHDOG] v2.6.2 indydog-v0.3_update
• [WATCHDOG] v2.6.2 i8xx_tco-v0.06_update
• [WATCHDOG] v2.6.2 watchdog-Kconfig-patch
• [WATCHDOG] v2.6.2 indydog-Kconfig+Makefile-patch
• [WATCHDOG] v2.6.2 pcwd_pci-watchdog
• [WATCHDOG] v2.6.2 arch-[m68k/sparc/sparc64]-Kconfig-patch