Linux 2.6.10-rc3

Linux 2.6.10-rc3

[Linus Torvalds]

Ok, it's out there in all the normal places, and here's the shortlog for
the thing.

Mostly a lot of small fixes, although the MIPS update is pretty sizeable 
simply because it's been a while. 

ACPI updates and a new i2c driver, mtd, arm, uml updates.. fbdev and 
sparse fixes. And a lot of other small things better just described by the 
changelogs.

Please do test this - and don't send me anything but bug-fixes. Let's aim
for a real 2.6.10 before xmas (or hanukkah, or whatever your favourite
holiday happens to be).

		Linus

-----

Summary of changes from v2.6.10-rc2 to v2.6.10-rc3
============================================

Adam J. Richter:
  o sysfs: fix sysfs_dir_close memory leak

Adrian Bunk:
  o kill lockd_syms.c
  o SCSI_QLOGIC_1280_1040 dependencies
  o SCSI: aha1542.c: make some code static
  o SCSI atp870u.c: make a needlessly global function
  o SCSI dc395x.c: make a function static
  o SCSI: fdomain.c: make a struct static
  o SCSI qla1280: some firmware files cleanups
  o PCI Hotplug: remove unused drivers/pci/hotplug/pciehp_sysfs.c
  o ISDN divert_init.c build fix
  o ST_partstat multiple definition
  o remove outdated OSS Changelogs
  o [PCMCIA] make cardbus_type static

Al Borchers:
  o USB Gadget: gadget serial documentation

Alan Stern:
  o sd.c: adjust READ_CAPACITY for broken devices

Alasdair G. Kergon:
  o device-mapper: dm-crypt fix for zero-length key
  o device-mapper: Fix some DMERR macro usage
  o device-mapper: Add DM_TARGET_MSG
  o device-mapper: Allow referencing by device number

Alexander Kern:
  o USB quirk section fix

Alexander Viro:
  o alpha sysrq compile fix
  o sparse: add -m64 to CHECKFLAGS on alpha and sparc64
  o sparc io.h annotations and fixes
  o pcilynx iomem annotations
  o tgafb iomem annotations
  o iphase iomem annotations
  o 64bit portability fixes
  o ad1889 annotations and fixes
  o drivers/media annotations
  o (1/4) eicon iomem annotations and fixes
  o (2/4) eicon iomem annotations and fixes
  o (3/4) eicon iomem annotations and fixes
  o (4/4) eicon iomem annotations and fixes
  o pc300 portability fixes
  o i2o iomem annotations
  o cpqphp_nvram iomem annotations
  o misc __user annotations
  o more C99 initializers
  o drivers/w1 iomem annotations
  o sound/oss iomem annotations
  o zoran fix
  o imsttfb iomem annotations
  o pcbit fix
  o dpt_i2o partial iomem annotations
  o 3w-9xxx iomem annotations
  o ncr iomem annotations
  o isdn_divert annotations
  o fix for breakage introduced in hgafb.c
  o more sparc64 io.h annotations
  o fore200e iomem annotations
  o sunzilog iomem annotations
  o mesh iomem annotations
  o tpam annotations and cleanups
  o (1/12) bw2 iomem annotations
  o (2/12) cg14 iomem annotations
  o (3/12) cg3 iomem annotations
  o (4/12) cg6 iomem annotations
  o (5/12) controlfb iomem annotations
  o (6/12) hgafb iomem annotations
  o (7/12) leo iomem annotations
  o (8/12) offb iomem annotations
  o (9/12) p9100 iomem annotations
  o (10/12) platinumfb iomem annotations
  o (11/12) tcx iomem annotations
  o (12/12) valkyriefb iomem annotations
  o i2c iomem annotations
  o pmac iomem annotations
  o pmac sound iomem annotations
  o pmac_zilog iomem annotations
  o sunbpp iomem annotations
  o misc sparc iomem annotations
  o partial fc4 iomem annotations
  o iomem annotations and isa_-ectomy: media/video/pms.c
  o mtd/maps trivial annotations
  o misc iomem annotations
  o isa_check_signature() finally gone
  o t128 iomem annotations and isa_-ectomy
  o dtc iomem annotations and isa_-ectomy
  o seagate iomem annotations, cleanup and isa_-ectomy
  o wd7000 iomem annotations and fixes
  o swim3 __user annotation
  o misc drivers/atm iomem annotations and NULL noise removal
  o iomem annotations and fixes + isa_-ectomy in msnd
  o Work around devices with bogus media change indication on the first
    open

Andi Kleen:
  o x86-64: Fix get_user_pages access to vsyscall page
  o x86-64: fix boot crash on VIA systems
  o x86_64: fix async IPIs
  o x86_64: fix vsyscalls
  o x86_64: fix interrupt routing with nosmp
  o x86_64: fix early oops printing
  o x86_64: increase timer fallback threshold
  o PCI: Add sysfs file to map PCI busses to cpus
  o PCI: Disable mmconfig on AMD CPUs
  o x86_64: Fix lost edge triggered irqs on UP kernel

Andrea Arcangeli:
  o fix for mpol mm corruption on tmpfs
  o mempolicy can select the wrong policy

Andreas Gruenbacher:
  o compat_sys_fcntl[64] contain superfluous, buggy test

Andreas Herrmann:
  o s390: zfcp host adapter
  o s390: zfcp act enhancements

Andrew Morton:
  o sound_alloc_dmap memory allocation warning suppression
  o vmscan: ignore swap token when in trouble
  o x86 current_stack_pointer warning fix
  o documentation: nmi_watchdog.txt update
  o x86_64: duplicated patch
  o e100: early reset fix
  o dont deprecate MODULE_PARM
  o e100 deadlock fix
  o generic_make_request stack savings
  o revert the "dio handle eof" fix
  o blkdev_get_blocks(): handle eof

Andrew Patterson:
  o cciss: Off-by-one error causing oops in CCISS_GETLUNIFO ioctl

Andrew Walrond:
  o fbdev: Fix rivafb breakage (typo introduced by NV IO access
    cleanups)

Andries E. Brouwer:
  o fix appletalk locking
  o dm_init unresolved reference to _exits
  o net: scheduling policing fix

Anton Blanchard:
  o ppc64: ratelimit some rtas errors
  o ppc64: Use pci_device_to_OF_node
  o ppc64: avoid 32bit only syscalls in unistd.h
  o ppc64: pci cleanup
  o ppc64: remove phb_set_model
  o ppc64: make fixup_winbond_82c105 pseries specific
  o ppc64: remove duplication in pci_alloc_*
  o ppc64: OF overrides for pci_probe_only, pci_assign_all_buses
  o ppc64: remove BUG()s in pcibios_fixup_bus
  o ppc64: get_phb_reg_prop only required on python PCI machines
  o ppc64: alloc_bootmem returns void *
  o ppc64: linux,rtas* fixes
  o ppc64: Reserve kernel memory in kernel instead of wrapper
  o ppc64: linux,tce* changes
  o Allow multiple cpus in irq affinity call

Antonino Daplas:
  o fbdev: Fix for using >16 pixel wide font in fb console
  o fbdev: Support for bigger than 16x32 fonts in softcursor
  o fbdev: Support for bigger than 16x32 fonts in rivafb cursor
  o fbcon: Disable fbcon cursor if vt softcursor is enabled
  o fbdev: Allow mode change even if EDID block is not found
  o fbdev: Fix cursor in doublescan mode in atyfb
  o fbdev: Fix typo in atyfb
  o fbdev: Change the find_mode behavior
  o rivafb: fix broken burst length calculation
  o fbdev: Fix screen corruption in neofb
  o fbdev: Fix lockup when switching to/from X/console
  o fbdev: Fix module_param in rivafb
  o fbdev: Fix crash if fb_set_var() called before
    register_framebuffer()
  o fbdev: fix wrong colors at 16 bpp in tridentfb

Aristeu Sergio Rozanski Filho:
  o i2c-elektor: get rid of cli/sti
  o [2/2] i2c-elektor: adding missing casts
  o i2c-ite: get rid of cli()/sti()

Armijn Hemel:
  o USB: add ati_remote.c device id

Arnaldo Carvalho de Melo:
  o [NET] Assign inet transport sockets to the right module

Bartlomiej Zolnierkiewicz:
  o [ide] update CRISv10 IDE driver
  o [ide] no need to alloc sg_table in CRISv10 IDE driver
  o [ide] small IDE cleanups
  o [ide] fix /proc/ide/hd?/settings to not spam logs

Ben Dooks:
  o [ARM PATCH] 2234/3: S3C2410 - new serial driver (1/4)
  o [ARM PATCH] 2241/1: S3C2410 - default configuration update
  o [ARM PATCH] 2242/1: BAST - default configuration update
  o [ARM PATCH] 2243/1: BAST - move pm init to init_machine
  o [ARM PATCH] 2246/1: S3C2410 - rename i2c depending on 2410/2440
  o [ARM PATCH] 2247/1: S3C2410 - serial low-level updates
  o [ARM PATCH] 2248/1: S3C2410 - missing serial config in
    arch/arm/mach-s3c2410/Kconfig
  o [ARM PATCH] 2251/1: S3C2410 - system timer rename [cosmetic]
  o [ARM PATCH] 2249/1: S3C2410 - update help for
    arch/arm/Kconfig.debug
  o [ARM PATCH] 2256/1: S3C2410 - dma load fixes
  o [ARM PATCH] 2263/1: S3C2410 - gpio pin config fixes
  o [ARM PATCH] 2272/1: S3C2410 - rtc should check for <0 on alarm set
  o [ARM PATCH] 2273/1: S3C2410 - timex.h CLOCK_TICK_RATE fix
  o [ARM PATCH] 2275/1: S3C2410 - serial rx fifo full check
  o [ARM PATCH] 2284/1: S3C2410 - core device registration update
  o [ARM PATCH] 2285/1: S3C2410 - regs-sdi.h fixes

Benjamin Herrenschmidt:
  o [SUNZILOG]: Update timeout when setting termios
  o ppc64: Fix default command line
  o ppc64: Fix typo when parsing isa "reg" properties
  o ppc64: pci_bus_to_host() simplification
  o ppc64: Fix early serial setup baud rate
  o del_timer() vs. mod_timer() SMP race
  o ppc32: Fix an IRQ issue with cpufreq

Bjorn Helgaas:
  o Fix ia64 flush_tlb_page build error
  o early uart console support
  o move HCDP/PCDP to early uart console
  o [IA64] iosapic.c: don't direct interrupts to offline cpus
  o [IA64] Bigsur config: Add CONFIG_SERIAL_8250_ACPI so we can find
    serial devices
  o DAC960: Don't look at PCI_Device->irq before calling
    pci_enable_device()

Bob Breuer:
  o [CG14]: Fix NULL sbus_dev handling and colormap setup

Bob Tracy:
  o sym53c500_cs driver update

Bodo Stroesser:
  o uml: don't rule out syscall_nr == 0
  o uml: redundant code removal from signal delivery
  o uml: redundant argument removal from handle_signal
  o uml: handle_signal simplification
  o uml: fix setting of interrupted syscall return value
  o uml: make signal frame construction more resemble x86
  o uml: fix signal mask on delivery error
  o uml: Don't delay segfaults

Brian Gerst:
  o Regparm for x86 machine check handlers
  o fastcall fixes for x86 smp interrupts

Brian Haley:
  o [IPV6] improve ipv6_ifa_notify() readability

Brian King:
  o sg: Fix oops of sg_cmd_done and sg_release race

Cal Peake:
  o Documentation/kernel-parameters.txt: scsi param updates
  o fix typo in init/Kconfig
  o fix typo in init/Kconfig

Carsten Otte:
  o s390: dcss segments

Chris Wright:
  o setup_arg_pages can insert overlapping vma
  o a.out: error check on set_brk

Christoph Hellwig:
  o ppc64: reduce ifdef clutter in arch/ppc64/kernel/sysfs.c
  o ppc64: cleanups hpte_init_native, kill warning for !PSERIES builds
  o [IA64] remove dead wood from asm-ia64/hardirq.h
  o allow NFS exports of EFS filesystems
  o [ARM] kill unused call_irq()

Christoph Lameter:
  o mmtimer driver update

Colin Leroy:
  o Switch therm_adt746x to new module_param

Cornelia Huck:
  o s390: common i/o layer
  o s390: common i/o layer

Daniel Drake:
  o Permit LOG_SENSE and LOG_SELECT in SG_IO command table

Daniel Ritz:
  o USB: Add some help text for touchkitusb
  o USB touchkitusb: module_param to swap axes
  o yenta: don't enable read prefetch on older o2 bridges

Dave Kleikamp:
  o radix_tree_delete() fix

David Brownell:
  o USB: fix Genesys GL880S EHCI
  o USB: usb_sg_*() unlink deadlock fix
  o USB: "sparse -Wcontext" and USB HCDs
  o USB: ax8817x/usbnet, no GFP_KERNEL blocking in_irq

David Eriksson:
  o Re: The "ipaq" module: Updated list of vendor/product IDs

David Fries:
  o USB: fix for HID field index

David Gibson:
  o ppc64: Kill unused KRANGE_{START,END} macros

David Howells:
  o Fork fix fix

David Mosberger:
  o [IA64] speedup ptrace by avoiding kernel-stack walk

David S. Miller:
  o [SPARC64]: Update defconfig
  o [AF_UNIX]: Serialize dgram read using semaphore just like stream
  o [IPV6]: Temp fix for ipv6 link-local address problem
  o [TG3]: Update driver version and reldate
  o [SPARC64]: Couple of do_sparc64_fault fixes
  o [SPARC64]: Two io_remap_page_range() fixes
  o [SPARC64]: Do not set VM_LOCKED on I/O mapped areas
  o [SPARC64]: Update defconfig
  o [IPV6]: Set sk_prot early enough in inet6_create()
  o [SPARC]: Fix serial console handling

David Woodhouse:
  o RS library spelling fixes
  o MTD: Intel flash chip driver locking fixes
  o MTD: DiskOnChip drivers should no longer require old docecc code
  o MTD map/device driver cleanups -- remove bogus __iomem casts
  o MTD: Fix suspend/resume on Intel flash chip driver
  o MTD NAND drivers: cleanup MODULE_PARAM and bogus __iomem casts
  o MTD: DiskOnChip driver fixes: MODULE_PARAM and __iomem, and fix RS
    init
  o MTD: Fix oops after erase in NFTL/INFTL (DiskOnChip translation
    layers)
  o MTD: Fix Pb1550 board NAND driver to not read write-only registers
  o MTD: Compile fix for the typo fix in ixp2000 map driver
  o MTD: Avoid false positives in CFI probe due to floating data bus
  o MTD: Fix detection of hardware partitions in Intel flash chips
  o JFFS2: locking fixes
  o JFFS2: Remove redundant 'ino' arg from jffs2_get_inode_nodes()
  o JFFS2: make sync() actually work by providing ->sync_fs method
  o JFFS2: jffs2_fs_i.h needs <asm/semaphore.h>
  o JFFS2: fix printk argument type warning
  o MTD: some cleanups
  o MTD: Fix memory leak in FTL translation layer

Deepak Saxena:
  o [ARM PATCH] 2252/1: Fix IXP4XX timer interrupt implementation
  o [ARM PATCH] 2253/1: Fix IXP4xx PCI config cycle routines
  o [ARM PATCH] 2254/1: Fix ixp4xx-regs.h PCI config address typo
  o [ARM PATCH] 2258/1: Add missing IXP2000 Makefile.boot file
  o [ARM PATCH] 2262/1: Various IXP2000 typo fixes and comment cleanups
  o [ARM PATCH] 2268/1: Update Documentation/arm/Booting
  o [ARM PATCH] 2270/1: [Trivial] Remove reference to head-armv.S
  o [ARM PATCH] 2255/1: Add IXDPG425 platform support
  o [ARM PATCH] 2257/1: Add I2C device to IXDP2x01 platforms
  o [ARM PATCH] 2259/1: Rip out ixp2000 IRQ_ERR_STATUS demultiplexing
  o [ARM PATCH] 2260/1: Rename IXP2000_IRQ_SWI to reduce user confusion
  o [ARM PATCH] 2261/1: Cleanup use of ixp_reg_write in
    arch/arm/mach-ixp2000

Dely Sy:
  o PCI Hotplug: Add pci_enable_device() in hot-plug drivers

Dmitry Krivoschokov:
  o USB Gadget: add and use gadget_is_pxa27x()

Dmitry Torokhov:
  o i8k: fix 'power_status' sysfs permissions
  o v4l: fix permissions on module parameters exported via sysfs

Domen Puncer:
  o video: semicolon bug in atyfb_base.c

Don Fry:
  o pcnet32: added pci_disable_device

Edward Falk:
  o Documentation for IDE and CDROM ioctls
  o fix typo in cdrom.c

Eric Brower:
  o [SPARC]: Remove unnecessary pm_idle comment

Eric Rossman:
  o s390: crypto driver

Evgeniy Polyakov:
  o w1: do not stop and oops if netlink socket was not allocated
  o w1: make W1_DS9490_BRIDGE available
  o drivers/w1/dscore: fix the inline mess
  o W1: check nls in return path

Fenghua Yu:
  o [IA64] add cpu_relax() in the body of spin loops
  o add cpu_relax() in spin loops & clean up barrier()

Gabriel Paubert:
  o I2C: minor comment fix

Geert Uytterhoeven:
  o M68k: Update defconfigs for 2.6.10-rc1
  o M68k: Add 3 missing syscalls
  o 68851 MMU: Fix harmless typo in the MMU configuration code
  o Sun-3: Fix link error
  o fm2fb: Update Steffen A. Mork's email address
  o M68k I/O: Move HP300 I/O macros close to other I/O macros again
  o M68k: Update defconfigs for 2.6.10-rc2
  o [DIO]: Fix typo in dio_resource_len()
  o M68k: Update Atari defconfig (enable Ethernet and MII)
  o M68k HP Lance Ethernet: Fix leaks on probe/removal
  o M68k: Update HP300 defconfig (enable DIO and HP Lance Ethernet)
  o M68k Ethernet drivers depend on NET_ETHERNET
  o M68k HP Lance Ethernet depends on DIO bus support

Gerald Schaefer:
  o s390: monreader docu
  o s390: z/VM monitor stream

Gerd Knorr:
  o fix kobject varargs bug
  o video-buf oops/crash fixes
  o v4l: disable unused function
  o v4l: more modparam
  o tuner update

Giuseppe Sacco:
  o gbefb.c build fix

Grant Grundler:
  o [IA64] perfmon: fix double end-of-comment in previous checkin

Greg Kroah-Hartman:
  o I2C: make fixup_fan_min static in adm1026 driver
  o PCI Hotplug: fix warning compile issue in cpqphp driver
  o USB: fix dev_dbg() call in visor.c
  o USB: fix oops in io_edgeport.c driver
  o USB: minor Makefile fix
  o PCI: fix build warning in pci-sysfs.c
  o USB: move a internal usbfs only structure out of a public header
    file
  o Driver Core: restore comment in kobject_uevent.c
  o Add documentation about why the in-kernel api is the way it is

Guennadi Liakhovetski:
  o tmascsim: (resend updated) track_queue_full
  o refactor tmscsim inititalization code

Guido Guenther:
  o fbdev: Add NV30 pci_id and cleanup of probe error returns

Haroldo Gamal:
  o smbfs: Bug #3758 - Broken symlinks on smbfs

Heiko Carstens:
  o s390: remove zfcp hba api callbacks

Herbert Xu:
  o [NETLINK]: Fix mc_list operations
  o [IPV6]: Fix xfrm6_tunnel_check_size mtu calc
  o [IPV4/IPV6]: Remove frag_list check from output path
  o [NETFILTER]: Apply ipsec to ipt_REJECT packets

Hideaki Yoshifuji:
  o [IPV6] Fix possible dead-lock in ipv6_create_tempaddr()
  o [IPV6] Fix a race when dad completed during shutting down its owner
    interface
  o [IPV6] Stop DAD during shutting down the interface
  o [IPV6] Clean-up locking in ipv6_add_addr()
  o [IPV6]: Fix races in ip6_route_{input,output}()

Hirofumi Ogawa:
  o cont_prepare_write() fix

Hirokazu Takata:
  o m32r: update for m32r-g00ff
  o m32r: CF boot support for Mappi2
  o m32r: update defconfig files
  o m32r: update dot.gdbinit files
  o m32r: Fix build error of  arch/m32r/mm/fault.c
  o m32r: Kconfig.debug support
  o m32r: Fix a boot hang of UP kernel
  o m32r: make zImage a default build target
  o m32r: io_xxxxx.c cleanups
  o media: Update drivers/media/video/arv.c

Holger Freyther:
  o [ARM PATCH] 2276/1: [PATCH] SIMpad: make simpad.c compile
  o [ARM PATCH] 2277/1: [PATCH] SIMpad: fix warnings emitted by the
    compiler
  o [ARM PATCH] 2278/1: [PATCH] SIMpad: add a default config
  o [ARM PATCH] 2279/1: [PATCH] SIMpad: Add a mq200 device to the
    platform bus
  o [ARM PATCH] 2280/1: [PATCH] SIMpad: Change maintainer to me
  o [ARM PATCH] 2283/1: SA1100 USB Config options

Horst Hummel:
  o s390: dasd driver

Hugh Dickins:
  o low discontig highmem_start_page
  o tmpfs free_inodes leak
  o mlock-vs-VM_IO hang fix

Iacopo Spalletti:
  o Add PCI-quirks for ASUS M6Ne notebook

Ian Campbell:
  o Avoid deadlock in smc91x driver

Ian Pratt:
  o [IPV4]: Missing pskb_may_pull in icmp_filter

Ingo Molnar:
  o sched: fix ->nr_uninterruptible handling bugs
  o floppy boot-time detection fix
  o acpi_processor_idle() latency fix

Jakub Jelínek:
  o binfmt_elf: handle p_filesz == 0 on PT_INTERP section

James Bottomley:
  o SCSI: Fix Bug 3753 (multiple definition of ST_partstat)
  o SCSI: fix USB forced remove oops
  o Change MCA maintainer

James Morris:
  o [AF_UNIX]: Fix SELinux crashes with SOCK_SEQPACKET
  o [AF_UNIX]: Don't lose ECONNRESET in unix_seqpacket_sendmsg()

Jamie Lenehan:
  o SCSI dc395x.c: store pci device pointer
  o SCSI dc395x.c: Fix type for irq and io ports
  o SCSI dc395x.c: Call pci_disable during cleanup

Jamie Lokier:
  o revert recent futex_wait fix

Jan Kara:
  o Minor fix of inequalities in the quota code
  o Add missing DQUOT_OFF

Jean Delvare:
  o I2C: Do not register useless smsc47m1
  o I2C: Fixes to the i2c-amd756-s4882 driver
  o I2C: Cleanups to the recent smbus functions removal
  o I2C: More verbose w83l785ts driver
  o I2C: Add support for the nForce2 Ultra 400 to i2c-nforce2
  o I2C: macintoch/therm_* drivers cleanups

Jeff Dike:
  o uml: signal bug fix
  o uml: 64-bit cleanups in the system calls
  o uml: 64-bit type cleanups
  o uml: fix definitions of pte_unmap_*
  o uml: LFS 64-bit cleanups
  o uml: Remove unused declaration
  o uml: remove some dead code
  o uml: defconfig update

Jeff Garzik:
  o [libata] fix DocBook bugs
  o [libata ahci] minor fixes
  o [libata docs] add chapter on libata driver API

Jeff Mahoney:
  o selinux: cache not freed if load_policy fails; reload BUG's

Jeff Scheel:
  o ppc64: iSeries legacy model emulation of PURR

Jens Axboe:
  o io context leak on queue drain
  o 3ware bad queuecommand returns
  o ide-scsi bad queuecommand return
  o megaraid bad queuecommand return
  o ncr53c8xx bad queuecommand return
  o nsp32 bad queuecommand return
  o aacraid bad queuecommand return
  o megaraid2 bad queuecommand return
  o nsp_cs bad queuecommand return
  o cfq-iosched: fix allocation increment race #3
  o bio: fix leak in failure case in bio_copy_user()
  o cfq-iosched: kill show_status sysfs entry

Jeremy Fitzhardinge:
  o Buffer overrun in arch/x86_64/sys_ia32.c:sys32_ni_syscall()

Jesper Juhl:
  o [SCTP]: Fix static inline declarations
  o remove errornous semicolon in
    arch/i386/kernel/traps.c::do_general_protection
  o [NET]: Fix inline keyword usage in skbuff.c

Jesse Barnes:
  o [IA64] fix phys. address conversion in ia64_pal_tr_read
  o correct copyright in arch/ia64/kernel/domain.c

Joe Korty:
  o fix uninitialized variable in waitid(2)

John W. Linville:
  o [VLAN]: change_mtu should return 0 on success
  o tulip: make tulip_stop_rxtx() wait for DMA to fully stop

Juerg Billeter:
  o Don't remove /sys in initramfs

Justin Thiessen:
  o I2C: add adm1026 chip driver

Kai Mäkisara:
  o "mt-st tell" fails in 2.6.10-rc1

Karsten Keil:
  o i4l: fix deadlock in CAPI code, reenable SMP

Len Brown:
  o [ACPI] acpi_pci_irq_enable() now returns 0 on success
  o fix non-ACPI IOAPIC build
  o Merge
  o [ACPI] handle out of spec EC bit width
    http://bugzilla.kernel.org/show_bug.cgi?id=1744
  o [ACPI] migrate to seq_file() interface
    http://bugzilla.kernel.org/show_bug.cgi?id=3333
  o [ACPI] automatic workaround for NFORCE timer-override BIOS bug
    http://bugzilla.kernel.org/show_bug.cgi?id=3551
  o [ACPI] clean up the NFORCE BIOS bug workaround delete now obsolete
    dmi_scan entries fix build for ACPI & !IOAPIC
  o [ACPI] fix warnings for 64-bit video build (Luming Yu)
  o Cset exclude:
    torvalds@ppc970.osdl.org|ChangeSet|20041111002817|28673 Cset
    exclude:
    acme@conectiva.com.br[torvalds]|ChangeSet|20041111002501|29509
  o [ACPI] remove acpi_ksyms.c
  o [ACPI] remove acpi_ksyms.c
  o [ACPI] add #ifdef ACPI_FUTURE_USAGE
  o [ACPI] survive a BIOS that erroneously supplies multiple MADTs
    http://bugzilla.redhat.com/beta2/show_bug.cgi?id=135449
  o [ACPI] fix build errors resulting from auto-merge
  o [ACPI] platform_rename_gsi() is no longer limited to ACPI specific
    code, so call it ioapic_renumber_irq().
  o [ACPI] fix IRQ assignment regression with CONFIG_PNPACPI=y
    http://bugzilla.kernel.org/show_bug.cgi?id=3762
  o [ACPI] fix reboot on poweroff regression due to enabled wakeup GPEs
    http://bugzilla.kernel.org/show_bug.cgi?id=3669
  o [ACPI] IPMI must supply the address of its GPE handler to install
    or remove it
  o Cset exclude:
    len.brown@intel.com[lenb]|ChangeSet|20041109085620|42985
  o [ACPI] disable LAPIC at reboot and poweroff if Linux forced it on
    http://bugzilla.kernel.org/show_bug.cgi?id=3643
  o [ACPI] update C-state limiting patch

Linus Torvalds:
  o Fix floppy driver lock-up when you have an irq storm
  o Fix reading /proc/<pid>/mem when parent dies
  o x86: clean up ptrace single-stepping, make PT_DTRACE exact
  o x86: make TF handling at signals consistent
  o acpi: don't disable PCI irq links that were active at boot
  o acpi: disable PCI links at boot again, fix ELCR
  o Linux 2.6.10-rc3

Maciej W. Rozycki:
  o i386: apic_printk() used before initialized

Magnus Damm:
  o [ide] update documentation for ide params
  o [ide] "ide=nodma" printout fix
  o documentation: nohighio
  o [IPV4]: Use schedule_timeout() instead of jiffies polling in
    ipconfig

Maneesh Soni:
  o fix oops in sysfs_remove_dir()

Manfred Schwarb:
  o [ATM]: Force -n option in gzip invocation
  o [DECNET]: dn_neigh.c needs linux/module.h

Manfred Spraul:
  o proc_pid_status() oops fix

Marc Leeman:
  o make number of ramdisks Kconfigurable

Marcel Holtmann:
  o [Bluetooth] Update copyright information
  o [Bluetooth] Correct locking for zero SCID responses
  o [Bluetooth] Add support for L2CAP secure mode
  o [Bluetooth] Check for L2CAP reliability
  o fix unnecessary increment in firmware_class_hotplug() and USB core

Mark Fortescue:
  o [SPARC]: In cg3 driver, access control reg using byte not long IOs

Mark Haverkamp:
  o 2.6 aacraid: Interrupt function cleanup
  o 2.6 aacraid: rx check health function update

Markus Lidel:
  o i2o: changed code with BUG() to BUG_ON()
  o i2o: remove unused code and make needlessly global code static
  o i2o: changed old queueing code with wait_event API
  o i2o: converted SPIN_LOCK_UNLOCKED into spin_lock_init()

Martin Schwidefsky:
  o s390: 3270 console
  o s390: core changes

Masaki Chikama:
  o USB: new defice for usb serial pl2303

Matt Porter:
  o ppc32: Fix uninitialized PPC40x vars
  o Fix warnings in ibm_emac driver

Matthew Dharm:
  o USB Storage: fixes to usb-storage scanning thread
  o USB Storage: Force INQUIRY length to be 36
  o USB Storage: Remove unnecessary state testing

Matthieu Castet:
  o [WATCHDOG] i8xx_tco.c-request_region-patch

Maximilian Attems:
  o [PCMCIA] replace schedule_timeout() with msleep()
  o [VLAN]: Handler register_netdevice_notifier() errors
  o [ATM]: Handle register_netdevice_notifier() errors in mpc.c

Michael Chan:
  o [TG3]: 5753 support and a bug fix

Michael Kerrisk:
  o RLIMIT_MEMLOCK accounting of shmctl() SHM_LOCK is broken

Michael Obster:
  o mc146818rtc.h include fix

Michal Rokos:
  o [PCMCIA] Exclude uneeded code when ! CONFIG_PROC_FS

Michal Schmidt:
  o md: fix jiffies handling in md.c

Mike Christie:
  o Fix badness in scsi_lib.c

Miles Bader:
  o Remove duplicate safe_for_read(READ_BUFFER) entry in scsi_ioctl.c
  o Remove duplicate safe_for_read(READ_BUFFER)

Mitchell Blank Jr.:
  o [NET]: Missing security_*() check in net/compat.c

Nathan Scott:
  o Fix an XFS direct I/O deadlock

Neil Brown:
  o kNFSd: fix d_find_alias brokenness
  o md: Fix problem with unsigned variable going "negative" in linear.c
  o knfsd: svcrpc: fqdn length fix
  o md: fix careless bug in raid10

Nick Piggin:
  o mm: tune the page allocator thresholds

Nicolas Pitre:
  o [ARM PATCH] 2267/1: don't fiddle with GPDR/GAFR directly
  o [ARM PATCH] 2271/3: MMC for Mainstone/PXA27x
  o [ARM PATCH] 2287/1: remove bogus EXPORT_SYMBOL(*)
  o [ARM PATCH] 2288/1: unlink MMC DMA on driver exit

Oleg Nesterov:
  o uninline do_trap(), remove get_cr2()

Olof Johansson:
  o ppc64: Make pci_alloc_consistent() conform to API docs
  o ppc64: Make early processor spinup based on physical ids

Paolo 'Blaisorblade' Giarrusso:
  o uml: update some copyrights
  o uml: partial KBUILD_OUTPUT fix
  o akpm has moved

Patrick Caulfield:
  o [DECNET]: Remove DECNET_SIOCGIFCONF
  o [DECNET]: Typo in accept causes OOPS

Patrick McHardy:
  o [NETFILTER]: Handle nonlinear skbs in ip_queue/ip6_queue
  o [NET]: Move rx timestamp functions to net/core/dev.c
  o [NETFILTER]: Enable rx timestamps in ip_queue/ip6_queue
  o [NETFILTER]: associate locally generated icmp errors with conntrack
    of original packet
  o [NETFILTER]: Fix invalid tcp/udp checksums within NATed icmp errors
  o [SCTP]: Fix inetaddr notifier chain corruption
  o [XFRM]: Fix endless loop in xfrm_policy_insert

Paul Mackerras:
  o PPC64 call ibm,os-term only if its available
  o PPC64 rtasd: window when error_log_cnt could get zeroed
  o Fix pmac_zilog.c so it compiles again
  o Do power_state conversion for mesh.c
  o Add __iomem annotations to drivers/scsi/mac53c94.c
  o power_state and __iomem for mediabay.c
  o __iomem annotations for swim3.c
  o Multilink fix for ppp_generic.c
  o ppc64: move emulate_step to arch/ppc64/lib
  o ppc64: fix compilation with recent toolchains
  o ppc64: remove the volatile from cpus_in_xmon
  o ppc64: fix hang on legacy iSeries

Paul Ortyl:
  o USB Storage: Unusual_dev entry for tekom/yakumo

Paulo Marques:
  o USB: add PID to ftdi_sio.c

Pete Zaitcev:
  o ub: flag day - major 180
  o ub: oops with preempt ("Sahara Workshop")

Peter Chubb:
  o [ARM PATCH] 2269/2: Updated Pleb-1 support patch for Linux 2.6

Petko Manolov:
  o USB: pegasus endian fixes

Phil Dibowitz:
  o USB Storage: Add unusual_devs entry for another yakumo camera

Phil Oester:
  o [NETFILTER]: revert MASQUERADE optimization for mostly static IPs

Prasanna Meda:
  o unlocked access to task->comm
  o /proc/cmdline missing mmput
  o sys_set/getpriority PRIO_USER semantics fix and optimisation

Prasanna S. Panchamukhi:
  o kprobes: dont steal interrupts from vm86

Ralf Bächle:
  o MIPS updates

Randy Dunlap:
  o x86_64 hpet: fix function warning
  o [SCTP]: Fix printk arg type
  o PCI Hotplug: cpcihp_generic: fix module_param data type
  o cdrom: handle SYSCTL without PROC_FS
  o usb-storage should enable scsi disk in Kconfig
  o eth1394: use SET_NETDEV_DEV() for udev
  o VISWS: prevent APM

Roger Luethi:
  o USB: visor: Always do generic_startup
  o USB visor: Don't count outstanding URBs twice

Roland Dreier:
  o cdev_init: zero out cdev before kobject_init()
  o linux/mount.h: add atomic.h and spinlock.h #includes

Rolf Eike Beer:
  o PCI Hotplug: clean up rpaphp_pci.c::rpaphp_find_pci_dev
  o PCI: fix Documentation/pci.txt inconsistency

Russell King:
  o [SERIAL] s3c2410: remove duplicate include
  o [PCMCIA] Don't place Cardbus bridges into D3 state on suspend
  o parport_pc CONFIG_PCI=n build fix
  o [ARM] Sparse fixes
  o [ARM] Add missing ecard_resource_flags() macro
  o [ARM] Fix VFP NaN flag handling
  o smc91c92_cs outw() fix
  o [SERIAL] imx: remove two unnecessary includes
  o [ARM] sa1111: don't reference dev->power.saved_state if CONFIG_PM
    is unset
  o [ARM] icside: ensure interfaces are probed and correctly setup
  o [ARM] ARMv6 always selects correct user operations at runtime
  o [ARM] omap: remove unnecessary linux/device.h include

Rusty Russell:
  o [NETFILTER]: Cleanup find_appropriate_src() Fix
  o [NETFILTER]: Fix stack leakage in iptables
  o Remove Futex Warning
  o Fix parameter handling in ibm_acpi.c
  o Fix occasional stop_machine() lockup with > 2 CPUs

Sam Ravnborg:
  o kbuild: Major update of Documentation/kbuild/modules.txt

Santiago Leon:
  o make ibmveth link always up

Scott Feldman:
  o [PKTGEN]: Clean error count before each run

Sreenivas Bagalkote:
  o megaraid 2.20.4.1 Driver

Srihari Vijayaraghavan:
  o [ide] remove RICOH CD-R/RW MP7083A from DMA blacklist

Stefan Bader:
  o s390: core changes

Stefan Weinhuber:
  o s390: dasd driver

Steffen A. Mork:
  o Make dss1_divert ISDN module work on SMP again
  o fix dss1_divert fixes

Stelian Pop:
  o sonypi: return an error from sonypi_camera_command() if the camera
    isn't enabled

Stephen D. Smalley:
  o SELinux: map Unix seqpacket sockets to appropriate security class

Stephen Hemminger:
  o [RANDOM]: Remove TCP MD4 code if not CONFIG_INET
  o usb_unlink_urb: ratelimit warning
  o [UDP]: Select handling of bad checksums

Stephen Rothwell:
  o ppc64 iSeries: don't share request queues in viocd
  o ppc64: add missing braces to rtc driver
  o ppc64 iSeries: PURR emulation fix
  o ppc64 iSeries: fix viodasd remove

Stéphane Eranian:
  o [IA64] perfmon: enable interrupts around semaphore call

Sylvain Meyer:
  o fbdev: Add vram option to intelfb

Thomas Gleixner:
  o Lock initializer unifying Batch 2 (SCSI)
  o yenta_socket.c: Fix missing pci_disable_dev

Thomas Graf:
  o [GNET_STATS]: kernel-api doc for gnet stats/estimator

Thomas Leibold:
  o I2C: i2c-nforce2.c add support for nForce3 Pro 150 MCP

Thomas Spatzier:
  o s390: network driver
  o s390: qeth network driver

Tim T. Murphy:
  o serial: add support for Dell Remote Access Card 4

Tom Rini:
  o ppc32: Fix Motorola Sandpoint builds
  o ppc32: Fix CONFIG_8260 and CONFIG_BLK_DEV_INITRD
  o x86_64: only single-step into signal handlers if the tracer asked
    for it
  o ppc32: Have the 8260 board-hook happen a bit later
  o ppc32: Fix __iomem warnings in TODC code

Tony Lindgren:
  o [ARM PATCH] 2201/1: OMAP timer 1/2: Clean-up MPU timer

Tony Luck:
  o [IA64] Allocate syscall #1270 for waitid syscall
  o [IA64] Add add_key, request_key, keyctl syscalls
  o [IA64] if idle doesn't halt, it should at least relax

Vadim Lobanov:
  o swsusp kconfig: Change in wording

Volker Sameske:
  o s390: zfcp read-only lun sharing

William Lee Irwin III:
  o parport_pc warning fixes

Yasuyuki Kozakai:
  o [NETFILTER]: Fix multiple bugs in ipv6header match
  o [NETFILTER]: introduce skb_header_pointer() to ipv6header match
  o [NETFILTER]: Make eui64 match usuable in FORWARD chain

Yoshinori Sato:
  o H8/300: /proc/cpuinfo typo fix
  o H8/300: signal handling update
  o H8/300: read{b,w,l} / write{b,w,l} error fix
  o H8/300: vmlinux.lds.S update
  o CONFIG_UNIX98_PTY=n warning fix

Zou Nanhai:
  o ia64/x86_64/s390 overlapping vma fix

Re: Linux 2.6.10-rc3

[Soeren Sonnenburg]

On Sat, 04 Dec 2004 00:19:48 +0000, Linus Torvalds wrote:

> Ok, it's out there in all the normal places, and here's the shortlog for
> the thing.

as -rc2 this one is still oopsing on resume from sleep on powerpc ...
kernel versions 2.6.X  (X<=9) used to work fine...

Soeren

Re: Linux 2.6.10-rc3

[Alex Romosan]

Soeren Sonnenburg <kernel@nn7.de> writes:

> as -rc2 this one is still oopsing on resume from sleep on powerpc
> ...

well, it's still more than my thinkpad which doesn't want to wake up
from sleep anymore.

> kernel versions 2.6.X  (X<=9) used to work fine...

same here. it started with -rc1.

--alex--

-- 
| I believe the moment is at hand when, by a paranoiac and active |
|  advance of the mind, it will be possible (simultaneously with  |
|  automatism and other passive states) to systematize confusion  |
|  and thus to help to discredit completely the world of reality. |

Re: Linux 2.6.10-rc3

[Ari Pollak]

Alex Romosan wrote:
> well, it's still more than my thinkpad which doesn't want to wake up
> from sleep anymore.

My thinkpad will resume fine if I remove the intel8x0 and intel8x0m ALSA 
modules before going into suspend - works with both APM and ACPI, though 
I don't really use ACPI suspend because the battery drains like crazy.

Re: Linux 2.6.10-rc3

[Alex Romosan]

Ari Pollak <aripollak@gmail.com> writes:

> Alex Romosan wrote:
>> well, it's still more than my thinkpad which doesn't want to wake up
>> from sleep anymore.
>
> My thinkpad will resume fine if I remove the intel8x0 and intel8x0m
> ALSA modules before going into suspend - works with both APM and ACPI,
> though I don't really use ACPI suspend because the battery drains like
> crazy.

i saw there were some changes to alsa cvs having to do with the new
pci device handling. i'll reconfigure the kernel with alsa as modules
and try alsa cvs to see if that makes any difference. thanks.

--alex--

-- 
| I believe the moment is at hand when, by a paranoiac and active |
|  advance of the mind, it will be possible (simultaneously with  |
|  automatism and other passive states) to systematize confusion  |
|  and thus to help to discredit completely the world of reality. |

Re: Linux 2.6.10-rc3

[Ari Pollak]

I looked at ALSA driver CVS a couple of days ago to see if anything has 
changed since the last release (included in 2.6.10-rc1+). The only thing 
significant that I found in the CVS intel8x0 driver, which added 
pci_disable_device() calls, didn't seem to help the resume problem when 
I applied this change to my kernel sources.

Alex Romosan wrote:
> i saw there were some changes to alsa cvs having to do with the new
> pci device handling. i'll reconfigure the kernel with alsa as modules
> and try alsa cvs to see if that makes any difference. thanks.
> 
> --alex--
> 

Re: Linux 2.6.10-rc3

[Martin Josefsson]

[-- Attachment #1: Type: text/plain, Size: 1494 bytes --]

On Sat, 2004-12-04 at 18:40, Alex Romosan wrote:
> Ari Pollak <aripollak@gmail.com> writes:
> 
> > Alex Romosan wrote:
> >> well, it's still more than my thinkpad which doesn't want to wake up
> >> from sleep anymore.
> >
> > My thinkpad will resume fine if I remove the intel8x0 and intel8x0m
> > ALSA modules before going into suspend - works with both APM and ACPI,
> > though I don't really use ACPI suspend because the battery drains like
> > crazy.
> 
> i saw there were some changes to alsa cvs having to do with the new
> pci device handling. i'll reconfigure the kernel with alsa as modules
> and try alsa cvs to see if that makes any difference. thanks.

This has been discussed in an earlier thread, I was hoping the ALSA
people would submit the patch but that hasn't happened.
I'll submit it in a separate mail.

Try the patch below, it should fix the problem.

--- linux/sound/core/init.c	8 Nov 2004 11:37:08 -0000	1.48
+++ linux/sound/core/init.c	12 Nov 2004 13:56:32 -0000
@@ -782,12 +782,15 @@<br>
 int snd_card_pci_suspend(struct pci_dev *dev, u32 state)
 {
 	snd_card_t *card = pci_get_drvdata(dev);
+	int err;
 	if (! card || ! card->pm_suspend)
 		return 0;
 	if (card->power_state == SNDRV_CTL_POWER_D3hot)
 		return 0;
 	/* FIXME: correct state value? */
-	return card->pm_suspend(card, 0);
+	err = card->pm_suspend(card, 0);
+	pci_save_state(dev);
+	return err;
 }

 int snd_card_pci_resume(struct pci_dev *dev)


-- 
/Martin

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Linux 2.6.10-rc3

[Alex Romosan]

Martin Josefsson <gandalf@wlug.westbo.se> writes:

> This has been discussed in an earlier thread, I was hoping the ALSA
> people would submit the patch but that hasn't happened.
> I'll submit it in a separate mail.
>
> Try the patch below, it should fix the problem.
>
> --- linux/sound/core/init.c	8 Nov 2004 11:37:08 -0000	1.48
> +++ linux/sound/core/init.c	12 Nov 2004 13:56:32 -0000
> @@ -782,12 +782,15 @@<br>
>  int snd_card_pci_suspend(struct pci_dev *dev, u32 state)
>  {
>  	snd_card_t *card = pci_get_drvdata(dev);
> +	int err;
>  	if (! card || ! card->pm_suspend)
>  		return 0;
>  	if (card->power_state == SNDRV_CTL_POWER_D3hot)
>  		return 0;
>  	/* FIXME: correct state value? */
> -	return card->pm_suspend(card, 0);
> +	err = card->pm_suspend(card, 0);
> +	pci_save_state(dev);
> +	return err;
>  }
>
>  int snd_card_pci_resume(struct pci_dev *dev)

thank you. the laptop wakes up now but i get the following when it
resumes (this is the output from dmesg):

Back to C!
Warning: CPU frequency is 1600000, cpufreq assumed 600000 kHz.
MCE: The hardware reports a non fatal, correctable incident occurred on CPU 0.
Bank 1: f200000000000115
PCI: Setting latency timer of device 0000:00:1d.0 to 64
PCI: Setting latency timer of device 0000:00:1d.1 to 64
PCI: Setting latency timer of device 0000:00:1d.2 to 64
PCI: cache line size of 32 is not supported by device 0000:00:1d.7
ehci_hcd 0000:00:1d.7: USB 2.0 restarted, EHCI 1.00, driver 26 Oct 2004
ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 6 (level, low) -> IRQ 6
ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 5 (level, low) -> IRQ 5
PCI: Setting latency timer of device 0000:00:1f.5 to 64
ACPI: PCI interrupt 0000:02:01.0[A] -> GSI 4 (level, low) -> IRQ 4
psmouse.c: TouchPad at isa0060/serio1/input0 lost sync at byte 1
scheduling while atomic: sleepbtn.sh/0x00000001/3201
 [<c0103557>] dump_stack+0x17/0x20
 [<c030e4e2>] schedule+0x522/0x530
 [<c030e97b>] schedule_timeout+0x5b/0xb0
 [<c011fbaf>] msleep+0x2f/0x40
 [<e1079387>] ehci_hub_resume+0xd7/0x1c0 [ehci_hcd]
 [<e10cce6a>] hcd_hub_resume+0x1a/0x20 [usbcore]
 [<e10c9ff2>] usb_resume_device+0xa2/0xc0 [usbcore]
 [<c022c50a>] resume_device+0x1a/0x20
 [<c022c5b7>] dpm_resume+0xa7/0xb0
 [<c022c5d4>] device_resume+0x14/0x30
 [<c012f27b>] suspend_finish+0xb/0x30
 [<c012f2ed>] enter_state+0x4d/0x70
 [<c01f6bd6>] acpi_suspend+0x3d/0x4b
 [<c01f6cbb>] acpi_system_write_sleep+0x64/0x76
 [<c014e042>] vfs_write+0xa2/0x100
 [<c014e151>] sys_write+0x41/0x70
 [<c0102ffb>] syscall_call+0x7/0xb
Restarting tasks... done
agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
agpgart: Putting AGP V2 device at 0000:00:00.0 into 4x mode
agpgart: Putting AGP V2 device at 0000:01:00.0 into 4x mode
[drm] Loading R200 Microcode

not sure if this means anything bad or not... i use acpi.

--alex--

-- 
| I believe the moment is at hand when, by a paranoiac and active |
|  advance of the mind, it will be possible (simultaneously with  |
|  automatism and other passive states) to systematize confusion  |
|  and thus to help to discredit completely the world of reality. |

Re: Linux 2.6.10-rc3

[Martin Josefsson]

On Sat, 4 Dec 2004, Alex Romosan wrote:

> thank you. the laptop wakes up now but i get the following when it
> resumes (this is the output from dmesg):
>
> scheduling while atomic: sleepbtn.sh/0x00000001/3201
>  [<c0103557>] dump_stack+0x17/0x20
>  [<c030e4e2>] schedule+0x522/0x530
>  [<c030e97b>] schedule_timeout+0x5b/0xb0
>  [<c011fbaf>] msleep+0x2f/0x40
>  [<e1079387>] ehci_hub_resume+0xd7/0x1c0 [ehci_hcd]
>  [<e10cce6a>] hcd_hub_resume+0x1a/0x20 [usbcore]
>  [<e10c9ff2>] usb_resume_device+0xa2/0xc0 [usbcore]
>  [<c022c50a>] resume_device+0x1a/0x20
>  [<c022c5b7>] dpm_resume+0xa7/0xb0
>  [<c022c5d4>] device_resume+0x14/0x30
>  [<c012f27b>] suspend_finish+0xb/0x30
>  [<c012f2ed>] enter_state+0x4d/0x70
>  [<c01f6bd6>] acpi_suspend+0x3d/0x4b
>  [<c01f6cbb>] acpi_system_write_sleep+0x64/0x76
>  [<c014e042>] vfs_write+0xa2/0x100
>  [<c014e151>] sys_write+0x41/0x70
>  [<c0102ffb>] syscall_call+0x7/0xb

That's an usb2.0 bug, the ehci driver sleeps when it can't sleep.

/Martin

Re: Linux 2.6.10-rc3

[Giridhar Pemmasani]

Alex Romosan wrote:

> Ari Pollak <aripollak@gmail.com> writes:
> 
> i saw there were some changes to alsa cvs having to do with the new
> pci device handling. i'll reconfigure the kernel with alsa as modules
> and try alsa cvs to see if that makes any difference. thanks.
> 

I have been using the following patch for a while with no problems:


--- linux/sound/pci/intel8x0.c 2004-12-04 16:37:11.000000000 -0500
+++ linux/sound/pci/intel8x0.c 2004-12-04 04:47:14.000000000 -0500
@@ -2279,6 +2279,8 @@
  for (i = 0; i < 3; i++)
   if (chip->ac97[i])
    snd_ac97_suspend(chip->ac97[i]);
+ pci_save_state(chip->pci);
+ pci_disable_device(chip->pci);
  snd_power_change_state(card, SNDRV_CTL_POWER_D3hot);
  return 0;
 }
@@ -2289,6 +2291,7 @@
  int i;
 
  pci_enable_device(chip->pci);
+ pci_restore_state(chip->pci);
  pci_set_master(chip->pci);
  snd_intel8x0_chip_init(chip, 0);
 

Re: Linux 2.6.10-rc3

[Paul Ionescu]

On Sat, 04 Dec 2004 22:42:11 +0100, Martin Josefsson wrote:
> That's an usb2.0 bug, the ehci driver sleeps when it can't sleep.

I have the message bellow without ehci-hcd module loaded, so I took out
USB2 problem, but this one I don't know what it means:


Warning: CPU frequency is 1700000, cpufreq assumed 600000 kHz.
Debug: sleeping function called from invalid context at mm/slab.c:2063
in_atomic():0[expected: 0], irqs_disabled():1
 [<0211cbcb>] __might_sleep+0x7d/0x8a
 [<0214bf9f>] __kmalloc+0x42/0x7d
 [<021f48e9>] acpi_os_allocate+0xa/0xb
 [<0220878a>] acpi_ut_allocate+0x2e/0x52
 [<02208721>] acpi_ut_initialize_buffer+0x41/0x7c
 [<02205474>] acpi_rs_create_byte_stream+0x23/0x3b
 [<02206976>] acpi_rs_set_srs_method_data+0x1b/0x9d
 [<0211b101>] recalc_task_prio+0x128/0x133
 [<0220e15c>] acpi_pci_link_set+0xfe/0x176
 [<0220e4e0>] irqrouter_resume+0x1c/0x24
 [<0224366a>] sysdev_resume+0x3e/0xa5
 [<02246564>] device_power_up+0x5/0xa
 [<0213db9a>] suspend_enter+0x25/0x2d
 [<0213dc08>] enter_state+0x3f/0x5e
 [<0220ad54>] acpi_suspend+0x28/0x34
 [<0220b7c4>] acpi_system_write_sleep+0x5c/0x6d
 [<02179769>] locate_fd+0x5c/0x78
 [<02165c82>] vfs_write+0xb6/0xe2
 [<02165d4c>] sys_write+0x3c/0x62
PCI: Setting latency timer of device 0000:00:1d.0 to 64

Re: Linux 2.6.10-rc3

[David Brownell]

> From:       Martin Josefsson <gandalf () wlug ! westbo ! se>
> Date:       2004-12-04 21:42:11
> 
> On Sat, 4 Dec 2004, Alex Romosan wrote:
> 
> > thank you. the laptop wakes up now but i get the following when it
> > resumes (this is the output from dmesg):
> >
> > scheduling while atomic: sleepbtn.sh/0x00000001/3201
> ...
> That's an usb2.0 bug, the ehci driver sleeps when it can't sleep.

Who changed it so that context was no longer allowed to sleep???

That's a very recent change ... I've done a fair amount of testing
in previous kernels and _never_ got that message on that path.

Why was that changed?  Are you sure it's not just a bug higher up
in the call stack?  Classically(*), both suspend() and resume()
methods are called in contexts that can sleep, so that's a big
change I'd expect to impact other drivers too.  In fact that'd
explain a lot of other messages I saw reported on the list...

- Dave

(*) Since APM days if not before.

Re: Linux 2.6.10-rc3

[Benjamin Herrenschmidt]

> Why was that changed?  Are you sure it's not just a bug higher up
> in the call stack?  Classically(*), both suspend() and resume()
> methods are called in contexts that can sleep, so that's a big
> change I'd expect to impact other drivers too.  In fact that'd
> explain a lot of other messages I saw reported on the list...

Yes, lots of drivers are expected to be able to schedule in suspend and
resume requests. If that was broken upstream, then this is a big BUG.

> - Dave
> 
> (*) Since APM days if not before.
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
-- 
Benjamin Herrenschmidt <benh@kernel.crashing.org>

Re: Linux 2.6.10-rc3

[Russell King]

On Sat, Dec 04, 2004 at 07:03:55PM -0800, David Brownell wrote:
> > From:       Martin Josefsson <gandalf () wlug ! westbo ! se>
> > Date:       2004-12-04 21:42:11
> > 
> > That's an usb2.0 bug, the ehci driver sleeps when it can't sleep.
> 
> Who changed it so that context was no longer allowed to sleep???

suspend and resume methods must be able to sleep because you may
need to talk to external hardware, wait for queues to drain, etc
which may in turn require kernel threads to run.

We must be able to sleep in suspend/resume methods.  PCMCIA requires
it, as do other subsystems.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 PCMCIA      - http://pcmcia.arm.linux.org.uk/
                 2.6 Serial core