.. and this time even with a timely tar-ball, since I'm not traveling any more.
What do we have in it? More than in -rc2. I'm clearly not the only one
who was in Japan for LinuxCon, or something else just made people wake
up.
There's a lot of small one-liners, but a few bigger chunks too: Radeon
DRI updates, some btrfs updates, and fixing Sparc LEON support (and
supporting PCI). Smaller updates to nilfs2 and ceph, and s390 and arm.
Other than that, it's mostly random driver updates all over.
I'm including the shortlog, which is fairly self-explanatory: small
random changes, nothing hugely interesting. It will matter to people
only if it happens to fix that one small nagging issue you had.
But as usual, I'll ask people to give it a good testing,
Linus
---
Al Viro (4):
more conservative S_NOSEC handling
get_net_ns_by_fd() oopses if proc_ns_fget() returns an error
fix return values of l2tp_dfs_seq_open()
lmLogOpen() broken failure exit
Alan Stern (3):
usb-storage: redo incorrect reads
USB: dummy-hcd needs the has_tt flag
USB: CONFIG_USB_GADGET_DUALSPEED is not user-configurable
Alex Deucher (13):
drm/radeon/kms/blit: workaround some hw issues on evergreen+
drm/radeon/kms: add blit support for cayman (v2)
drm/radeon/kms: add missing case for cayman thermal sensor
drm/radeon/kms: fix thermal sensor reading on juniper
drm/radeon/kms: clean up the radeon kms Kconfig
drm/radeon/kms: add family ids for llano APUs
drm/radeon/kms: fill in asic struct for llano
drm/radeon/kms: add llano pci ids
drm/radeon/kms: add support for Llano Fusion APUs
drm/radeon/kms: remove duplicate reg from r600 safe regs
drm/radeon/kms: viewport height has to be even
drm/radeon/kms/atom: fix PHY init
drm/radeon/kms: disable hdmi audio by default
Alexander Holler (2):
arm: omap3: beagle: Ensure msecure is mux'd to be able to set the RTC
bridge: provide a cow_metrics method for fake_ops
Alexey Khoroshilov (1):
usb-gadget: unlock data->lock mutex on error path in ep_write()
Andre Bartke (1):
staging: iio: error case memory leak fix
Andy Lutomirski (3):
efifb: Enable write-combining
efifb: Fix mismatched request/release_mem_region
efifb: Disallow manual bind and unbind
Anirban Chakraborty (1):
qlcnic: Fix bug in FW queue dump
Antti Palosaari (1):
[media] anysee: return EOPNOTSUPP for unsupported I2C messages
Arjan Mels (1):
staging: usbip: bugfix prevent driver unbind
Arnaldo Carvalho de Melo (3):
perf evlist: Remove dependency on debug routines
perf python: Use exception to propagate errors
perf evlist: Don't die if sample_{id_all|type} is invalid
Arne Jansen (3):
btrfs: scrub: errors in tree enumeration
btrfs: reinitialize scrub workers
btrfs: remove unneeded includes from scrub.c
Axel Lin (1):
ARM: OMAP2: Add missing include of linux/gpio.h
Balaji T K (2):
ARM: OMAP4: MMC: no regulator off during probe for eMMC
ARM: OMAP4: MMC: increase delay for pbias
Ben Skeggs (5):
drm/nvc0: recognise 0xdX chipsets as NV_C0
drm/nouveau: don't create accel engine objects when noaccel=1
drm/nouveau: fix vram page mapping when crossing page table boundaries
drm/nouveau: fix leak of gart mm node
drm/nv40: fall back to paged dma object for the moment
Benjamin Herrenschmidt (1):
powerpc: Force page alignment for initrd reserved memory
Bob Peterson (1):
GFS2: Processes waiting on inode glock that no processes are holding
Bruno Prémont (1):
video: Fix use-after-free by vga16fb on rmmod
Bryan Wu (1):
usb: gadget: include <linux/prefetch.h> to fix compiling error
Chris Mason (1):
Btrfs: make sure to recheck for bitmaps in clusters
Chris Metcalf (1):
slub: always align cpu_slab to honor cmpxchg_double requirement
Chris Wilson (5):
drm/i915: s/addr & ~PAGE_MASK/offset_in_page(addr)/
drm/i915: Replace ironlake_compute_wm0 with g4x_compute_wm0
drm/i915/crt: Explicitly return false if connected to a digital monitor
drm/i915: Remove unused enum "chip_family"
drm/i915: Share the common force-audio property between connectors
Christian Borntraeger (2):
[S390] kvm-s390: Fix host crash on misbehaving guests
[S390] kvm-s390: fix stfle facilities numbers >=64
Colin Cross (4):
ARM: omap2+: mux: Allow board mux settings to be NULL
ARM: omap4: Pass core and wakeup mux tables to omap4_mux_init
GPIO: OMAP: fix setting IRQWAKEN bits for OMAP4
GPIO: OMAP: add locking around calls to _set_gpio_triggering
Connor Hansen (2):
iscsi_ibft: iscsi_ibft_find unused variable i
ide-cd: signedness warning fix again
Dan Carpenter (6):
ptp: Return -EFAULT on copy_to_user() errors
ptp: Fix some locking bugs in ptp_read()
xen: off by one errors in multicalls.c
drm/i915: fix if statement in ivybridge irq handler
Staging: mei: fix debug code
netpoll: call dev_put() on error in netpoll_setup()
Dan Williams (3):
option: add Zoom 4597 modem USB IDs
option: add Alcatel X200 to sendsetup blacklist
option: add Prolink PH300 modem IDs
Daniel Drake (1):
libertas_sdio: handle spurious interrupts
Daniel Haid (1):
drm/radeon/kms: fix for radeon on systems >4GB without hardware iommu
Daniel Hellstrom (6):
sparc32: added CONFIG_PCIC_PCI Kconfig setting
sparc32,leon: added LEON-common low-level PCI routines
sparc32,leon: add GRPCI2 PCI Host driver
sparc32, sun4m: bugfix in SMP IPI traphandler
sparc32, leon: bugfix in LEON SMP interrupt init
dl2k: EEPROM CRC calculation wrong endianess on bigendian machine
Daniel T Chen (2):
ALSA: hda: Fix quirk for Dell Inspiron 910
ALSA: hda: Fix inaudible internal speakers on CyberpowerPC Gamer
Xplorer N57001 laptop
Daniel Vetter (2):
drm/i915: Only print out the actual number of fences for i915_error_state
drm/915: fix relaxed tiling on gen2: tile height
Darren Salt (1):
CIFS ACL support needs CONFIG_KEYS, so depend on it
Dave Airlie (2):
drm/radeon: fix oops in ttm reserve when pageflipping (v2)
drm/radeon/kms: set family for use in parser.
Dave Jones (1):
netfilter: use unsigned variables for packet lengths in ip[6]_queue.
David Howells (4):
MN10300: Fix one of the kernel debugger cacheflush variants
MN10300: die_if_no_fixup() shouldn't use get_user() as it
doesn't call set_fs()
MN10300: Add missing _sdata declaration
linux/seqlock.h should #include asm/processor.h for cpu_relax()
David S. Miller (3):
sparc: PCIC_PCI needs SPARC32 dependency
net: Rework netdev_drivername() to avoid warning.
irda: iriap: Use seperate lockdep class for irias_objects->hb_spinlock
David Sterba (1):
btrfs: fix unlocked access of delalloc_inodes
Dmitry Torokhov (1):
USB: xhci - fix interval calculation for FS isoc endpoints
Eric Dumazet (3):
netfilter: add more values to enum ip_conntrack_info
af_packet: prevent information leak
net: pmtu_expires fixes
FUJITA Tomonori (1):
swiotlb: Export swioltb_nr_tbl and utilize it as appropiate.
Felipe Balbi (1):
MAINTAINERS: add a maintainer to Gadget Framework
Florian Fainelli (1):
x86: devicetree: Add missing early_init_dt_setup_initrd_arch stub
Francisco Jerez (1):
drm/nv17-nv40: Fix modesetting failure when pitch == 4096px (fdo
bug 35901).
Frederic Weisbecker (1):
perf python: Fix argument name list of read_on_cpu()
Geert Uytterhoeven (2):
sparc: Do not select GENERIC_HARDIRQS_NO_DEPRECATED
spi/rtc-m41t93: Use spi_get_drvdata() for SPI devices
Govindraj.R (2):
OMAP2+: mux: fix compilation warnings
spi/omap2: fix uninitialized variable
Grant Likely (2):
net: fix smc91x.c device tree support
MAINTAINERS: Saying goodbye to David Brownell
Greg Kroah-Hartman (4):
Staging: altera: move .h file to proper place
Staging: remove STAGING_EXCLUDE_BUILD option
Revert "USB: option: add ID for ZTE MF 330"
Staging: cx23885: fix include of altera.h
Guan Xuetao (5):
unicore32: move rtc-puv3.c to drivers/rtc directory
unicore32: change zImage physical address, though it's PIC codes
unicore32: add KBUILD_DEFCONFIG with unicore32_defconfig (old
debug_defconfig)
unicore32: change PERCPU to PERCPU_SECTION
unicore32: using generic-y format for one line asm-generic files
GuoWen Li (1):
ftrace: Fix possible undefined return code
H Hartley Sweeten (2):
gpio/samsung: make Kconfig options def_bool
ep93xx_eth: Update MAINTAINERS
Hans Petter Selasky (1):
[media] Make nchg variable signed because the code compares this
variable against negative values
Hans de Goede (1):
drm/i915: Add a no lvds quirk for the Asus EeeBox PC EB1007
Heiko Carstens (2):
KVM: add missing void __user * cast to access_ok() call
net: cpu offline cause napi stall
Heiko Stuebner (3):
ARM: S3C24xx: Fix missing struct for s3c2410_dma_chan
ARM: S3C24XX: Move s3c24xx_irq_syscore_ops to plat-s3c24xx/irq.c
ARM: S3C2410: remove the now empty mach-s3c2410/irq.c
Ian Armstrong (3):
[media] ivtv: Make two ivtv_msleep_timeout calls uninterruptable
[media] ivtvfb: Add sanity check to ivtvfb_pan_display()
[media] ivtv: Internally separate encoder & decoder standard setting
Igor Grinberg (3):
arm: omap3: cm-t3517: fix section mismatch warning
arm: omap2plus: fix ads7846 pendown gpio request
arm: omap2plus: move NAND_BLOCK_SIZE out of boards
Ilya Dryomov (1):
Btrfs - use %pU to print fsid
Jamie Iles (1):
gpio/basic_mmio: add missing include of spinlock_types.h
Jan Glauber (1):
[S390] qdio: Split SBAL entry flags
Jan Schmidt (1):
Btrfs: fix extent state leak on failed nodatasum reads
Janusz Krzysztofik (1):
OMAP: fix compilation error
Jarkko Nikula (2):
omap: rx51: Set regulator V28_A always on
omap: rx51: Don't power up speaker amplifier at bootup
Jason Stubbs (1):
drm/i915: fix regression after clock gating init split
Jean-François Moine (5):
[media] gspca - ov519: Fix a regression for ovfx2 webcams
[media] gspca - ov519: Change the ovfx2 bulk transfer size
[media] gspca: Remove coarse_expo_autogain.h
[media] gspca - stv06xx: Set a lower default value of gain for
hdcs sensors
[media] gspca - ov519: Set the default frame rate to 15 fps
Jeff Layton (2):
cifs: silence printk when establishing first session on socket
cifs: trivial: add space in fsc error message
Jesper Juhl (2):
genirq: Prevent potential NULL dereference in irq_set_irq_wake()
ISDN, hfcsusb: Don't leak in hfcsusb_ph_info()
Jiri Pirko (1):
vlan: Fix the ingress VLAN_FLAG_REORDER_HDR check
Joachim Eastwood (1):
ASoC: atmel_ssc: Don't try to free ssc if request failed
Joe Perches (3):
video: Convert vmalloc/memset to vzalloc
sparc: Remove unnecessary semicolons
vsprintf: Update %pI6c to not compress a single 0
Joerg Roedel (3):
x86/amd-iommu: Fix 3 possible endless loops
x86/amd-iommu: Use only per-device dma_ops
x86/amd-iommu: Fix boot crash with hidden PCI devices
Johannes Berg (1):
mac80211: fix IBSS teardown race
John Johansen (1):
AppArmor: Fix sleep in invalid context from task_setrlimit
John Stultz (1):
rtc: Fix ioctl error path return
John W. Linville (2):
Revert "mac80211: Skip tailroom reservation for full HW-crypto devices"
Revert "mac80211: stop queues before rate control updation"
Jonghwan Choi (1):
ARM: S5PV210: Fix possible null pointer dereference
Josef Bacik (9):
Btrfs: cache bitmaps when searching for a cluster
Btrfs: noinline the cluster searching functions
Btrfs: don't commit the transaction if we dont have enough pinned bytes
Btrfs: fix bitmap regression
Btrfs: fix the allocator loop logic
Btrfs: fix duplicate checking logic
Btrfs: don't map extent buffer if path->skip_locking is set
Btrfs: unlock the trans lock properly
Btrfs: don't map extent buffer if path->skip_locking is set
Jozsef Kadlecsik (2):
netfilter: ipset: Fix return code for destroy when sets are in use
netfilter: ipset: Use the stored first cidr value instead of '1'
Julia Lawall (4):
arch/arm/mach-omap1/dma.c: Invert calls to platform_device_put
and platform_device_del
drivers/video/imxfb.c: add missing clk_put
drivers/video/pxa168fb.c: add missing clk_put
drivers/usb/host/ohci-pxa27x.c: add missing clk_put
Julian Anastasov (2):
ipvs: restore support for iptables SNAT
netfilter: nf_nat: fix crash in nf_nat_csum
Kukjin Kim (1):
ARM: S5P: Should be S3C_VA_USB_HSPHY instead of S5P_VA_XX
Kumar Gala (2):
powerpc/fsl_rio: Fix compile error when CONFIG_FSL_RIO not set
powerpc/book3e: Fix CPU feature handling on e5500 in 32-bit mode
Kyungmin Park (3):
gpio: Fix gpio-exynos4 build fails in mainline
ARM: EXYNOS4: Remove compiler warning on exynos4_pwm4_resume
ARM: SAMSUNG: Remove unused onenand plat functions
Lars-Peter Clausen (3):
ASoC: AD1836: Fix setting the PCM format
ASoC: Blackfin: bf5xx-ad1836: Fix codec device name
ASoC: snd_soc_new_{mixer,mux,pga} make sure to use right DAPM context
Laurent Pinchart (3):
[media] ivtvfb: use display information in info not in var for panning
[media] v4l: Fix media_entity_to_video_device macro argument name
[media] media: Fix media device minor registration
Li Zefan (2):
Btrfs: avoid stack bloat in btrfs_ioctl_fs_info()
Btrfs: use join_transaction in btrfs_evict_inode()
Libor Pechacek (1):
USB: core: Tolerate protocol stall during hub and port status read
Linus Torvalds (4):
tty_buffer: get rid of 'seen_tail' logic in flush_to_ldisc
selinux: simplify and clean up inode_has_perm()
vfs: reorganize 'struct inode' layout a bit
Linux 3.0-rc3
Linus Walleij (2):
genirq: Fix descriptor init on non-sparse IRQs
gpio/nomadik: fix sleepmode for elder Nomadik
Luciano Coelho (1):
nl80211: fix overflow in ssid_len
Maarten Lankhorst (2):
xhci: Add defines for hardcoded slot states
xhci: Do not issue device reset when device is not setup
Marcelo Tosatti (1):
KVM: x86: use proper port value when checking io instruction permission
Marcus Meissner (1):
net/ipv6: check for mistakenly passed in non-AF_INET6 sockaddrs
Marek Olšák (1):
drm/radeon/kms: add missing Evergreen texture formats to the CS parser
Mark Brown (8):
rtc: Staticize non-exported __rtc_set_alarm()
genirq: Ensure we locate the passed IRQ in irq_alloc_descs()
ASoC: Only update SYSCLK_ENA when pausing WM8915 SYSCLK
ASoC: Add missing break in WM8915 FLL source selection
ASoC: Check for NULL register bank in snd_soc_get_cache_val()
USB: s3c-hsotg: Tone down debugging
ASoC: Fix WM8962 headphone volume update for use of advanced caches
ASoC: WM8804 does not support sample rates below 32kHz
Martin Schwidefsky (2):
[S390] use generic RCU page-table freeing code
[S390] fix kvm defines for 31 bit compile
Mathias Krause (1):
exec: delay address limit change until point of no return
Matt Evans (2):
xhci: Bigendian fix for skip_isoc_td()
xhci: Bigendian fix for xhci_check_bandwidth()
Mauro Carvalho Chehab (2):
[media] uvc_entity: initialize return value
[media] soc_camera: preserve const attribute
Michael Chang (1):
staging: gma500: get control from firmware framebuffer if conflicts
Michael Hennerich (1):
staging: iio: industrialio-trigger: set iio_poll_func private_data
Michal Marek (5):
kbuild: Fix KERNELVERSION for empty SUBLEVEL or PATCHLEVEL
kbuild: Fix <linux/version.h> for empty SUBLEVEL or PATCHLEVEL
kbuild: Move depmod call to a separate script
kbuild: Hack for depmod not handling X.Y versions
perf: Use make kernelversion instead of parsing the Makefile
Mika Westerberg (5):
ep93xx: set DMA masks for the ep93xx_eth
net: ep93xx_eth: pass struct device to DMA API functions
net: ep93xx_eth: allocate buffers using kmalloc()
net: ep93xx_eth: drop GFP_DMA from call to dma_alloc_coherent()
net: ep93xx_eth: fix DMA API violations
Mike Frysinger (1):
musb: fix prefetch build failure
Mike McCormack (2):
rtlwifi: Fix logic in rx_interrupt
rtlwifi: Avoid modifying skbs that are resubmitted
Mike Waychison (1):
KVM: Initialize kvm before registering the mmu notifier
Neil Horman (1):
bonding: reset queue mapping prior to transmission to physical device (v5)
Nick Kossifidis (1):
ath5k: Disable fast channel switching by default
Nicolas Kaiser (2):
drm: i915: correct return status in intel_hdmi_mode_valid()
staging: rts_pstor: use bitwise operator instead of logical one
OGAWA Hirofumi (1):
fat: Fix corrupt inode flags when remove ATTR_SYS flag
Omar Ramirez Luna (1):
OMAP: iovmm: fix SW flags passed by user
Pablo Neira Ayuso (1):
netfilter: nf_conntrack: fix ct refcount leak in l4proto->error()
Padmavathi Venna (1):
ARM: S5P64X0: Fix SPI platform device name
Patrik Jakobsson (1):
staging: gma500: Skip bogus LVDS VBT mode and check for LVDS
before adding backlight
Paul Mundt (1):
fbdev: sh_mobile_lcdcfb: Fix up fallout from MERAM changes.
Peter Foley (1):
kbuild: silence Nothing to be done for 'all' message
Peter Zijlstra (5):
sched: Fix cross-cpu clock sync on remote wakeups
sched: Fix schedstat.nr_wakeups_migrate
perf, cgroups: Fix up for new API
lockdep: Fix lock_is_held() on recursion
sched: Fix/clarify set_task_cpu() locking rules
Rabin Vincent (1):
OMAP4: fix return value of omap4_l3_init
Rafael J. Wysocki (2):
PM / Intel IOMMU: Fix init_iommu_pm_ops() for CONFIG_PM unset
PM / Runtime: Fix loops in pm_runtime_clk_notify()
Rafał Miłecki (1):
ssb: fix PCI(e) driver regression causing oops on PCI cards
Randy Dunlap (2):
staging: fix olpc_dcon build, needs BACKLIGHT_CLASS_DEVICE
staging: fix ath6kl build when CFG80211 is not enabled
Robert Richter (4):
oprofile, x86: Add comments to IBS LVT offset initialization
oprofile: Free potentially owned tasks in case of errors
oprofile: Fix locking dependency in sync_start()
oprofile, dcookies: Fix possible circular locking dependency
Roland Stigge (1):
staging: iio: max517: Fix iio_info changes
Roland Vossen (1):
staging: brcm80211: fix for 'multiple definition of
wl_msg_level' build err
Russell King (1):
GPIO: OMAP: fix section mismatch warnings
Russell King - ARM Linux (3):
NET: am79c961: ensure asm() statements are marked volatile
NET: am79c961: ensure multicast filter is correctly set at open
NET: am79c961: fix assembler warnings
Ryusuke Konishi (3):
nilfs2: fix incorrect block address termination in node concatenation
nilfs2: fix missing block address termination in btree node shrinking
nilfs2: fix problem in setting checkpoint interval
Sage Weil (8):
usb: remove bad dput after dentry_unhash
usb: remove bad dput after dentry_unhash
ceph: use ihold when we already have an inode ref
ceph: fix sync vs canceled write
ceph: fix short sync reads from the OSD
ceph: fix ENOENT logic in striped_read
ceph: unwind canceled flock state
Btrfs: clear current->journal_info on async transaction commit
Sangbeom Kim (1):
ASoC: SAMSUNG: Fix the incorrect referencing of I2SCON register
Sanjeev Premi (1):
[media] omap3isp: fix compiler warning
Santosh Shilimkar (2):
OMAP2+: Fix 9 section mismatch(es) warnings from mach-omap2/built-in.o
OMAP: SRAM: Fix warning: format '%08lx' expects type 'long unsigned int'
Sarah Sharp (1):
xhci: Disable MSI for some Fresco Logic hosts.
Sebastian Andrzej Siewior (7):
genirq: Print threaded handler in spurious debug output
irq: Handle spurios irq detection for threaded irqs
timers: Consider slack value in mod_timer()
usb/mv_udc_core: fix compile
usb/pxa25x_udc: cleanup the LUBBOCK err path
usb/s3c-hsudc: fix error path
usb/renesas_usbhs: free uep on removal
Shaohui Xie (1):
powerpc/85xx: fix race bug of calling request_irq after enable
elbc interrupts
Shubhrajyoti D (1):
omap4: hwmod: Enable the keypad
Stanislaw Gruszka (5):
iwlagn: fix channel switch locking
iwlagn: use cts-to-self protection on 5000 adapters series
rt2x00: fix rmmod crash
iwl4965: set tx power after rxon_assoc
iwlegacy: fix channel switch locking
Stefano Stabellini (1):
xen: use the trigger info we already have to choose the irq handler
Steffen Klassert (2):
xfrm: Fix off by one in the replay advance functions
ipv4: Fix packet size calculation for raw IPsec packets in
__ip_append_data
Steffen Sledz (1):
USB: serial: add another 4N-GALAXY.DE PID to ftdi_sio driver
Steven Miao (1):
fbdev: bf537-lq035: add missing blacklight properties type
Steven Rostedt (2):
kprobes/trace: Fix kprobe selftest for gcc 4.6
ftrace: Revert 8ab2b7efd ftrace: Remove unnecessary disabling of irqs
Sucheta Chakraborty (1):
qlcnic: Avoid double free of skb in tx path
Suleiman Souhlal (1):
SLAB: Record actual last user of freed objects.
Suresh Jayaraman (2):
cifs: fix the kernel release version in the default security
warning message
cifs: make CIFS depend on CRYPTO_ECB
Takashi Iwai (3):
ALSA: hda - Fix invalid unsol tag for some alc262 model quirks
ALSA: hda - Fix initialization of hp pins with master_mute in Realtek
ALSA: Use %pV for snd_printk()
Tasslehoff Kjappfot (1):
omap3: Free Beagle rev gpios when they are read, so others can
read them later
Tejun Heo (1):
libata: fix unexpectedly frozen port after ata_eh_reset()
Thadeu Lima de Souza Cascardo (1):
mac80211: call dev_alloc_name before copying name to sdata
Theodore Ts'o (1):
vfs: make unlink() and rmdir() return ENOENT in preference to EROFS
Thomas Abraham (1):
gpio/exynos4: Fix incorrect mapping of gpio pull-up macro to
register setting
Thomas Gleixner (2):
clockevents: Handle empty cpumask gracefully
x86: cpu-hotplug: Prevent softirq wakeup on wrong CPU
Timur Tabi (1):
ASoC: fsl: fix initialization of DMA buffers
Toby Gray (1):
USB: cdc-acm: Adding second ACM channel support for Nokia E7 and C7
Todd Poynor (1):
ARM: OMAP2: Add missing iounmap in omap4430_phy_init
Tormod Volden (1):
savagefb: Use panel CVT mode as default
Torsten Hilbrich (1):
USB: option Add blacklist for ZTE K3765-Z (19d2:2002)
Tushar Behera (1):
ARM: S5P: Fix compilation error for exynos4_defconfig
Vasanthakumar Thiagarajan (1):
staging: ath6kl: Fix a kernel panic during suspend/resume
Vasily Averin (1):
possible memory corruption in cifs_parse_mount_options()
Vince Weaver (2):
perf: Comment /proc/sys/kernel/perf_event_paranoid to be part of user ABI
perf: Fix comments in include/linux/perf_event.h
Virupax Sadashivpetimath (1):
spi-pl022: Add missing return value update
WANG Cong (1):
netpoll: prevent netpoll setup on slave devices
Wey-Yi Guy (1):
iwlagn: send tx power command if defer cause by RXON not match
Williams, Mitch A (1):
igb: fix i350 SR-IOV failture
Wolfram Sang (1):
dtc/powerpc: remove obsolete .gitignore entries
Wu Jiajun-B06378 (1):
gianfar:localized filer table
Yegor Yefremov (1):
ethtool.h: fix typos
oftedal (3):
Restructure sun4d_build_device_irq so that timer interrupts can
be allocated
Do not skip interrupt sources in sun4d interrupt handler and
acknowledge interrupts correctly
Add support for allocating irqs for bootbus devices
richard kennedy (1):
btrfs: remove 64bit alignment padding to allow extent_buffer to
fit into one fewer cacheline
On Mon, 13 Jun 2011, Linus Torvalds wrote:
> I'm including the shortlog, which is fairly self-explanatory: small
> random changes, nothing hugely interesting. It will matter to people
> only if it happens to fix that one small nagging issue you had.
>
> But as usual, I'll ask people to give it a good testing,
>
> Linus
Breaks modules_install, as far as I can see it is creating the right
folders but not putting the actual files into them.
INSTALL arch/x86/crypto/aes-i586.ko
cp: cannot stat `arch/x86/crypto/aes-i586.ko': No such file or directory
INSTALL arch/x86/crypto/crc32c-intel.ko
cp: cannot stat `arch/x86/crypto/crc32c-intel.ko': No such file or
directory
INSTALL arch/x86/crypto/salsa20-i586.ko
etc ...
ls -l /lib/modules/3.0.0-rc3/kernel/arch/x86/
total 0
drwxr-xr-x 1 root root 0 Jun 14 13:05 crypto
drwxr-xr-x 1 root root 0 Jun 14 13:05 kernel
Gerhard
--
Gerhard Mack
[email protected]
<>< As a computer, I find your faith in technology amusing.
On Tue, Jun 14, 2011 at 1:01 AM, Linus Torvalds
<[email protected]> wrote:
> .. and this time even with a timely tar-ball, since I'm not traveling any more.
>
> What do we have in it? More than in -rc2. I'm clearly not the only one
> who was in Japan for LinuxCon, or something else just made people wake
> up.
Linus, I know I run a risk of being not the first person to ask this,
but anyway.
I've got a patch for my project to fix parsing of kernel version which
has only two numbers. Basically,
- scanf(ver, "%u.%u.%u", &a, &b, &c)
+ sscanf(ver, "%u.%u", &a, &b)
I can take it, but it made me thinking: how many other projects
will be similarly affected? Must be hundreds, even thousands.
I propose to still use three digits. I mean, if you want to use
3.0, 3.1, 3.2 numbering for your releases and leave third digit
for stable series, just number them 3.0.0, 3.1.0, 3.2.0 instead.
This way, many userspace projects will need less patching
in order to work with 3.x. In many cases, they can
avoid patching altogether.
--
vda
Op 14-06-11 14:15, Denys Vlasenko schreef:
> On Tue, Jun 14, 2011 at 1:01 AM, Linus Torvalds
> <[email protected]> wrote:
>> .. and this time even with a timely tar-ball, since I'm not traveling any more.
>>
>> What do we have in it? More than in -rc2. I'm clearly not the only one
>> who was in Japan for LinuxCon, or something else just made people wake
>> up.
> Linus, I know I run a risk of being not the first person to ask this,
> but anyway.
>
> I've got a patch for my project to fix parsing of kernel version which
> has only two numbers. Basically,
>
> - scanf(ver, "%u.%u.%u", &a, &b, &c)
> + sscanf(ver, "%u.%u", &a, &b)
>
> I can take it, but it made me thinking: how many other projects
> will be similarly affected? Must be hundreds, even thousands.
>
> I propose to still use three digits. I mean, if you want to use
> 3.0, 3.1, 3.2 numbering for your releases and leave third digit
> for stable series, just number them 3.0.0, 3.1.0, 3.2.0 instead.
>
> This way, many userspace projects will need less patching
> in order to work with 3.x. In many cases, they can
> avoid patching altogether.
>
~$ uname -r
3.0.0-rc3-patser+
~Maarten
On Tue, Jun 14, 2011 at 3:15 PM, Denys Vlasenko
<[email protected]> wrote:
> I've got a patch for my project to fix parsing of kernel version which
> has only two numbers. Basically,
Why does your project care about kernel version?
On Tue, Jun 14, 2011 at 2:55 PM, Alexey Dobriyan <[email protected]> wrote:
> On Tue, Jun 14, 2011 at 3:15 PM, Denys Vlasenko
> <[email protected]> wrote:
>> I've got a patch for my project to fix parsing of kernel version which
>> has only two numbers. Basically,
>
> Why does your project care about kernel version?
It's depmod. It was assuming that if argument starts with three digits,
it's a kernel version:
/* If a version is provided, then that kernel version's module directory
* is used, rather than the current kernel version (as returned by
* "uname -r"). */
if (*argv && sscanf(*argv, "%u.%u.%u", &tmp, &tmp, &tmp) == 3) {
version = *argv++;
} else {
uname(&uts);
version = uts.release;
}
--
vda
On Tue, Jun 14, 2011 at 5:15 AM, Denys Vlasenko
<[email protected]> wrote:
>
> I've got a patch for my project to fix parsing of kernel version which
> has only two numbers. Basically,
>
> - scanf(ver, "%u.%u.%u", &a, &b, &c)
> + sscanf(ver, "%u.%u", &a, &b)
Please just fix it.
The projects that care about kernel version are buggy in so many ways
that it's not funny.
It's not just the "two versus three digits" bug either. The *bigger*
bug is usually that the tests are done the wrong way entirely, namely
to fail when you cannot parse the version. And that's just f*%!ing
stupid!
If you cannot parse the version, then rather than fail, a project
should have gone "uh, it's some future version that I don't recognize
or some other OS entirely, so I'll just do whatever the most modern
thing is". Having an "assert()" or returning an error is just the mark
of incompetence.
So please just fix things. Preferably by removing the version check
entirely, but if you really feel that you cannot do that, then AT
LEAST realize that the reason you're looking at the version is to
support *old* versions, not new ones, and make the logic work that way
(so that next time around, when we change the version numbers to be
sanscrit characters and you fail parsing them *again*, you don't
actually fail, you just go "oh, this is modern" and go on with your
life).
In other words: don't be stupid.
Ok?
Linus
On Mon, 13 Jun 2011 16:01:26 -0700 Linus Torvalds wrote:
> .. and this time even with a timely tar-ball, since I'm not traveling any more.
ERROR: "pm_idle" [arch/x86/kernel/apm.ko] undefined!
ERROR: "default_idle" [arch/x86/kernel/apm.ko] undefined!
Fix was posted by Andy Whitcroft on June 8 and Ben Hutchings on June 9.
https://lkml.org/lkml/2011/6/8/170
https://lkml.org/lkml/2011/6/9/679
Below is Andy's patch since it was posted first.
Please merge some fix.
---
The commit below removed the export for pm_idle/default_idle unless the
apm module was modularised and CONFIG_APM_CPU_IDLE was set. The apm
module uses pm_idle/default_idle unconditionally, CONFIG_APM_CPU_IDLE
only affects the bios idle threshold. Adjust the export accordingly.
commit 06ae40ce073daf233607a3c54a489f2c1e44683e
Author: Len Brown <[email protected]>
Date: Fri Apr 1 15:28:09 2011 -0400
x86 idle: EXPORT_SYMBOL(default_idle, pm_idle) only when APM demands it
Signed-off-by: Andy Whitcroft <[email protected]>
---
arch/x86/kernel/process.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 2e4928d..b8265d0 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -337,7 +337,7 @@ EXPORT_SYMBOL(boot_option_idle_override);
* Powermanagement idle function, if any..
*/
void (*pm_idle)(void);
-#if defined(CONFIG_APM_MODULE) && defined(CONFIG_APM_CPU_IDLE)
+#if defined(CONFIG_APM_MODULE)
EXPORT_SYMBOL(pm_idle);
#endif
@@ -399,7 +399,7 @@ void default_idle(void)
cpu_relax();
}
}
-#if defined(CONFIG_APM_MODULE) && defined(CONFIG_APM_CPU_IDLE)
+#if defined(CONFIG_APM_MODULE)
EXPORT_SYMBOL(default_idle);
#endif
--
On Tue, Jun 14, 2011 at 10:59:41AM -0700, Linus Torvalds wrote:
> On Tue, Jun 14, 2011 at 5:15 AM, Denys Vlasenko
> <[email protected]> wrote:
> >
> > I've got a patch for my project to fix parsing of kernel version which
> > has only two numbers. Basically,
> >
> > - scanf(ver, "%u.%u.%u", &a, &b, &c)
> > + sscanf(ver, "%u.%u", &a, &b)
>
> Please just fix it.
>
> The projects that care about kernel version are buggy in so many ways
> that it's not funny.
His problem is recognizing "kernel version" vs. "filename" as a
parameter.
depmod [ -e ] [ -FSystem.map ] [ -n ] [ -v ] [ version ] [ filename... ]
Oh course, that's a stupid interface, but probably a hard one to
change.
OG.
On 06/15/2011 03:15 AM, Olivier Galibert wrote:
> On Tue, Jun 14, 2011 at 10:59:41AM -0700, Linus Torvalds wrote:
>> On Tue, Jun 14, 2011 at 5:15 AM, Denys Vlasenko
>> <[email protected]> wrote:
>>>
>>> I've got a patch for my project to fix parsing of kernel version which
>>> has only two numbers. Basically,
>>>
>>> - scanf(ver, "%u.%u.%u",&a,&b,&c)
>>> + sscanf(ver, "%u.%u",&a,&b)
>>
>> Please just fix it.
>>
>> The projects that care about kernel version are buggy in so many ways
>> that it's not funny.
>
> His problem is recognizing "kernel version" vs. "filename" as a
> parameter.
> depmod [ -e ] [ -FSystem.map ] [ -n ] [ -v ] [ version ] [ filename... ]
>
> Oh course, that's a stupid interface, but probably a hard one to
> change.
Well, the man page says that the version parameter is used to find
/lib/modules/version. So why not check if /lib/modules/version exists
and, if not, try the filename?
/me is still annoyed that some expensive commercial software over here
won't start if you have funny characters like '+' in your kernel version.
--Andy
On Wed, Jun 15, 2011 at 12:29 PM, Andy Lutomirski <[email protected]> wrote:
>
> Well, the man page says that the version parameter is used to find
> /lib/modules/version. ?So why not check if /lib/modules/version exists and,
> if not, try the filename?
Yeah, that's the only sane way to see if something is a filename: look
up the file. If you can find it, it's a filename.
It's what 'git' uses to decide between the ambiguity of a filename vs
a commit name, for example. It works really well. And sometimes, when
the ambiguity is not resolvable (ie you really meant a filename, but
there's a version that has the exact same format), then you are
fundamentally screwed - you need some other way to tell the difference
(which could be a simple priority decision - "filename overrides
version" - or it can be a command line thing).
So stop trying to parse the version number, and just see if
'/lib/modules/%s' exists for that (unparsed) version string. Or, if
you want to prioritize it the other way, just check for existence of
the file of the (again, unparsed) version string with a simple
"access(%s, F_OK)" or something.
I really don't see why you should ever parse a version number.
Linus
On Wed, Jun 15, 2011 at 12:44 PM, Linus Torvalds
<[email protected]> wrote:
>
> I really don't see why you should ever parse a version number.
Let me modulate that: I do see why you would ever say "is this version
xyz" - but then you don't have to be very smart about parsing things,
because a failure to parse just means "no". So there is never any case
where a failure to parse something means that you cannot answer the
question, and failure to parse it never becomes a failure for the
program itself.
So doing
if (!strncmp(version, "v2.4", 4))
workaround_for_old_version = true;
is valid. But only if you really have special code for those specific
old versions that you know, and have normal code for everything else.
So if you then confinue with
else if (!strncmp(version, "v2.6", 4))
error("unknown version");
then you're just being an ass.
Why? At that point you're no longer correcting for known bugs or
mis-features, at that point you're just being stupid and parsing the
version string for no good reason.
So "checking for _particular_ old versions that you know about" is
fine. Checking for versions you _don't_ know about or understand is
stupid.
Linus