Skip to contentSkip to conversation

3.12.10-stable review

PATCH In review Greg Kroah-Hartman 6 participants 136 messages Feb 4, 2014 – Feb 17, 2014

Subsystems (5)
lore
Share & tools
lore

Greg Kroah-Hartman

author maintainer
#0 lore raw

cover 3.12.10-stable review

This is the start of the stable review cycle for the 3.12.10 release. There are 133 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu Feb 6 21:07:17 UTC 2014. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

	kernel.org/pub/linux/kernel/v3.0/stable-review/patch-3.12.10-rc1.gz

and the diffstat can be found below.

thanks,

greg k-h

-------------
Pseudo-Shortlog of commits:

Greg Kroah-Hartman <[email protected]>

    Linux 3.12.10-rc1

Paul Mackerras <[email protected]>
powerpc: Make sure "cache" directory is removed when offlining cpu

Srivatsa S. Bhat <[email protected]>
powerpc: Fix the setup of CPU-to-Node mappings during CPU online

David Sterba <[email protected]>
btrfs: restrict snapshotting to own subvolumes

Wang Shilong <[email protected]>
Btrfs: handle EAGAIN case properly in btrfs_drop_snapshot()

Andy Grover <[email protected]>
target/iscsi: Fix network portal creation race

Nicholas Bellinger <[email protected]>
iscsi-target: Pre-allocate more tags to avoid ack starvation

Asias He <[email protected]>
virtio-scsi: Fix hotcpu_notifier use-after-free with virtscsi_freeze

Vijaya Mohan Guvva <[email protected]>
SCSI: bfa: Chinook quad port 16G FC HBA claim issue

Thomas Pugliese <[email protected]>
usb: core: get config and string descriptors for unauthorized devices

Mikulas Patocka <[email protected]>
hpfs: remember free space

Stephen Warren <[email protected]>
ALSA: hda/hdmi - allow PIN_OUT to be dynamically enabled

Anssi Hannula <[email protected]>
ALSA: hda - hdmi: introduce patch_nvhdmi()

Takashi Iwai <[email protected]>
ALSA: hda - Don't set indep_hp flag for old AD codecs

Mihai Caraman <[email protected]>
KVM: PPC: e500: Fix bad address type in deliver_tlb_misss()

Andreas Schwab <[email protected]>
KVM: PPC: Book3S HV: use xics_wake_cpu only when defined

Helge Deller <[email protected]>
parisc: fix cache-flushing

Mikulas Patocka <[email protected]>
alpha: fix broken network checksum

Neal Cardwell <[email protected]>
inet_diag: fix inet_diag_dump_icsk() timewait socket state logic

Duan Jiong <[email protected]>
net: gre: use icmp_hdr() to get inner ip header

Annie Li <[email protected]>
xen-netfront: fix resource leak in netfront

Holger Eitzenberger <[email protected]>
net: Fix memory leak if TPROXY used with TCP early demux

Oliver Hartkopp <[email protected]>
fib_frontend: fix possible NULL pointer dereference

Or Gerlitz <[email protected]>
net/vxlan: Share RX skb de-marking and checksum checks with ovs

Duan Jiong <[email protected]>
ip_tunnel: clear IPCB in ip_tunnel_xmit() in case dst_link_failure() is called

Heiko Carstens <[email protected]>
s390/bpf,jit: fix 32 bit divisions, use unsigned divide instructions

Eric Dumazet <[email protected]>
bpf: do not use reciprocal divide

Christoph Paasch <[email protected]>
tcp: metrics: Avoid duplicate entries with the same destination-IP

Gerald Schaefer <[email protected]>
net: rds: fix per-cpu helper usage

Richard Weinberger <[email protected]>
net,via-rhine: Fix tx_timeout handling

Hannes Frederic Sowa <[email protected]>
net: avoid reference counter overflows on fib_rules in multicast forwarding

Christian Engelmayer <[email protected]>

    ieee802154: Fix memory leak in ieee802154_add_iface()

Bjørn Mork <[email protected]>
net: usbnet: fix SG initialisation

Michal Schmidt <[email protected]>
bnx2x: fix DMA unmapping of TSO split BDs

Shane Huang <[email protected]>
i2c: piix4: Add support for AMD ML and CZ SMBus changes

Gregory CLEMENT <[email protected]>
i2c: mv64xxx: Document the newly introduced Armada XP A0 compatible

Gregory CLEMENT <[email protected]>
i2c: mv64xxx: Fix bus hang on A0 version of the Armada XP SoCs

Gregory CLEMENT <[email protected]>
ARM: mvebu: Add quirk for i2c for the OpenBlocks AX3-4 board

Gregory CLEMENT <[email protected]>
ARM: mvebu: Add support to get the ID and the revision of a SoC

Takashi Iwai <[email protected]>
hp_accel: Add a new PnP ID HPQ6007 for new HP laptops

Minchan Kim <[email protected]>
zram: fix race between reset and flushing pending work

Kent Overstreet <[email protected]>
bcache: Data corruption fix

Eric W. Biederman <[email protected]>
vfs: Is mounted should be testing mnt_ns for NULL or error.

Eric W. Biederman <[email protected]>
vfs: Remove second variable named error in __dentry_path

Theodore Ts'o <[email protected]>
ext4: avoid clearing beyond i_blocks when truncating an inline data file

Tejun Heo <[email protected]>
libata: disable LPM for some WD SATA-I devices

Simon Guinot <[email protected]>
ARM: mvebu: update the SATA compatible string for Armada 370/XP

Lior Amsalem <[email protected]>
ata: sata_mv: fix disk hotplug for Armada 370/XP SoCs

Simon Guinot <[email protected]>
ata: sata_mv: introduce compatible string "marvell, armada-370-sata"

Peter Huewe <[email protected]>

    tpm/tpm_ppi: Do not compare strcmp(a,b) == -1

Peter Huewe <[email protected]>
tpm/tpm_i2c_stm_st33: Check return code of get_burstcount

Adrien Vergé <[email protected]>
ALSA: hda - Fix silent output on MacBook Air 1,1

Dan Carpenter <[email protected]>
ALSA: bits vs bytes bug in snd_card_create()

Takashi Iwai <[email protected]>
ALSA: Enable CONFIG_ZONE_DMA for smaller PCI DMA masks

Takashi Iwai <[email protected]>
ALSA: hda - Don't create duplicated ctls for loopback paths

Takashi Iwai <[email protected]>
ALSA: hda - Correct AD1986A 3stack pin configs

Takashi Iwai <[email protected]>

    ALSA: rme9652: fix a missing comma in channel_map_9636_ds[]

Charles Keepax <[email protected]>
ASoC: wm5110: Extend SYSCLK patch file for rev D

Lars-Peter Clausen <[email protected]>

    ASoC: adau1701: Fix ADAU1701_SEROCTL_WORD_LEN_16 constant

Krzysztof Kozlowski <[email protected]>
mfd: max77686: Fix regmap resource leak on driver remove

Dongsheng Yang <[email protected]>
perf kvm: Fix kvm report without guestmount.

Chen-Yu Tsai <[email protected]>
pinctrl: sunxi: Honor GPIO output initial vaules

Stephen Warren <[email protected]>
rtc: max8907: weekday encoding fixes

Borislav Petkov <[email protected]>
rtc-cmos: Add an alarm disable quirk

Sebastian Andrzej Siewior <[email protected]>
parport: parport_pc: remove double PCI ID for NetMos

Heiko Carstens <[email protected]>
s390/uapi: fix struct statfs64 definition

Dominik Dingel <[email protected]>
KVM: s390: ioeventfd: ignore leftmost bits

Heiko Carstens <[email protected]>
KVM: s390: fix diagnose code extraction

Stephen Warren <[email protected]>
serial: 8250: enable UART_BUG_NOMSR for Tegra

Jonathan Woithe <[email protected]>
serial: 8250: Fix initialisation of Quatech cards with the AMCC PCI chip

Yegor Yefremov <[email protected]>
serial: add support for 200 v3 series Titan card

Sarah Sharp <[email protected]>
xhci: Set scatter-gather limit to avoid failed block writes.

Ben Hutchings <[email protected]>
xhci: Avoid infinite loop when sg urb requires too many trbs

Phil Pokorny <[email protected]>
hwmon: (k10temp) Add support for Kaveri CPUs

Jean-Jacques Hiblot <[email protected]>
ARM: at91: smc: bug fix in sam9_smc_cs_read()

Ludovic Desroches <[email protected]>
ARM: at91: at91sam9g45: set default mmc pinctrl-names

Marek Roszko <[email protected]>
tty/serial: at91: disable uart timer at start of shutdown

Mark Deneen <[email protected]>
tty/serial: at91: reset rx_ring when port is shutdown

Marek Roszko <[email protected]>
tty/serial: at91: fix race condition in atmel_serial_remove

Marek Roszko <[email protected]>
tty/serial: at91: Handle shutdown more safely

Tomas Winkler <[email protected]>
mei: use hbm idle state to prevent spurious resets

Malcolm Priestley <[email protected]>
staging: vt6656: CARDqGetNextTBTT correct uLowNextTBTT

Malcolm Priestley <[email protected]>
staging: vt6656: [BUG] BBvUpdatePreEDThreshold Always set sensitivity on bScanning

Larry Finger <[email protected]>
staging: r8712u: Set device type to wlan

Amir Shehata <[email protected]>
staging/lustre/ptlrpc: Fix a crash when dereferencing NULL pointer

Sujith Manoharan <[email protected]>
ath9k: Disable cross-band FCC

Sujith Manoharan <[email protected]>
ath9k: Use correct channel for RX packets

ZHAO Gang <[email protected]>
b43: fix the wrong assignment of status.freq in b43_rx()

Larry Finger <[email protected]>
b43legacy: Fix unload oops if firmware is not available

Larry Finger <[email protected]>
b43: Fix unload oops if firmware is not available

Larry Finger <[email protected]>
b43: Fix lockdep splat

Eliad Peller <[email protected]>
iwlwifi: mvm: fix missing cleanup in .start() error path

Emmanuel Grumbach <[email protected]>
iwlwifi: pcie: enable oscillator for L1 exit

Bing Zhao <[email protected]>
mwifiex: fix wrong 11ac bits setting in fw_cap_info

Amitkumar Karwar <[email protected]>
mwifiex: add missing endian conversion for fw_tsf

Larry Finger <[email protected]>
rtlwifi: rtl8188ee: Fix typo in code

Larry Finger <[email protected]>
rtlwifi: Add missing code to PWDB statics routine

Larry Finger <[email protected]>
rtlwifi: rtl8192cu: Fix some code in RF handling

Larry Finger <[email protected]>
rtlwifi: rtl8192cu: Update the power index registers

Larry Finger <[email protected]>
rtlwifi: rtl8192c: Add routines to save/restore power index registers

Larry Finger <[email protected]>
rtlwifi: Increase the RX queue length for USB drivers

Larry Finger <[email protected]>
rtlwifi: rtl8192c: Add new definitions in the dm_common header

Larry Finger <[email protected]>
rtlwifi: Set the link state

Larry Finger <[email protected]>
rtlwifi: Redo register save locations

Larry Finger <[email protected]>
rtlwifi: rtl8192cu: Add new firmware

Larry Finger <[email protected]>
rtlwifi: rtl8192c: Prevent reconnect attempts if not connected

Larry Finger <[email protected]>
rtlwifi: Update beacon statistics for USB driver

Larry Finger <[email protected]>
rtlwifi: rtl8192cu: Add new device ID

Peter Chen <[email protected]>
usb: ehci: add freescale imx28 special write register method

Alan Stern <[email protected]>
USB: fix race between hub_disconnect and recursively_mark_NOTATTACHED

Jack Pham <[email protected]>
usb: xhci: Check for XHCI_PLAT in xhci_cleanup_msix()

Mikhail Zolotaryov <[email protected]>
USB: Nokia 502 is an unusual device

Colin Leitner <[email protected]>
USB: ftdi_sio: added CS5 quirk for broken smartcard readers

Johan Hovold <[email protected]>
USB: cypress_m8: fix ring-indicator detection and reporting

Rahul Bedarkar <[email protected]>
USB: serial: add support for iBall 3.5G connect usb modem

张君 <[email protected]>
usb: option: add new zte 3g modem pids to option driver

Johan Hovold <[email protected]>
USB: pl2303: fix data corruption on termios updates

Peter Chen <[email protected]>
usb: chipidea: udc: using MultO at TD as real mult value for ISO-TX

Peter Chen <[email protected]>
usb: chipidea: need to mask INT_STATUS when write otgsc

Peter Chen <[email protected]>
usb: chipidea: imx: set CI_HDRC_IMX28_WRITE_FIX for imx28

Peter Chen <[email protected]>
usb: chipidea: add freescale imx28 special write register method

Paolo Bonzini <[email protected]>

    x86, kvm: correctly access the KVM_CPUID_FEATURES leaf at 0x40000101

Paolo Bonzini <[email protected]>
x86, kvm: cache the base of the KVM cpuid leaves

Marcelo Tosatti <[email protected]>
KVM: x86: limit PIT timer frequency

Prarit Bhargava <[email protected]>
x86: Add check for number of available vectors before CPU down

Dave Young <[email protected]>
x86/efi: Fix off-by-one bug in EFI Boot Services reservation

PaX Team <[email protected]>
x86, x32: Correct invalid use of user timespec in the kernel

David Rientjes <[email protected]>
mm/mempolicy.c: fix mempolicy printing in numa_maps

David Ertman <[email protected]>
e1000e: fix compiler warnings

Aristeu Rozanski <[email protected]>
e752x_edac: Fix pci_dev usage count

Andrea Arcangeli <[email protected]>
mm: hugetlbfs: fix hugetlbfs optimization

Ilia Mirkin <[email protected]>
drm/nouveau/bios: fix offset calculation for BMPv1 bioses

NeilBrown <[email protected]>
md/raid5: fix long-standing problem with bitmap handling on write failure.

Andrew Jones <[email protected]>
kvm: x86: fix apic_base enable check

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

Diffstat:

 Documentation/devicetree/bindings/ata/marvell.txt  |   2 +-
 .../devicetree/bindings/i2c/i2c-mv64xxx.txt        |   6 +-
 Documentation/i2c/busses/i2c-piix4                 |   2 +-
 Makefile                                           |   4 +-
 arch/alpha/lib/csum_partial_copy.c                 |   9 +-
 arch/arm/boot/dts/armada-370-xp.dtsi               |   2 +-
 arch/arm/boot/dts/at91sam9g45.dtsi                 |   2 +
 arch/arm/mach-at91/sam9_smc.c                      |   2 +-
 arch/arm/mach-mvebu/Makefile                       |   2 +-
 arch/arm/mach-mvebu/armada-370-xp.c                |  32 +++++
 arch/arm/mach-mvebu/mvebu-soc-id.c                 | 119 +++++++++++++++++
 arch/arm/mach-mvebu/mvebu-soc-id.h                 |  32 +++++
 arch/arm/net/bpf_jit_32.c                          |   6 +-
 arch/parisc/include/asm/cacheflush.h               |   2 -
 arch/parisc/include/asm/page.h                     |   3 +-
 arch/parisc/kernel/cache.c                         |  14 ++
 arch/powerpc/include/asm/topology.h                |  10 +-
 arch/powerpc/kernel/cacheinfo.c                    |   3 +
 arch/powerpc/kvm/book3s_hv.c                       |   7 +-
 arch/powerpc/kvm/e500_mmu.c                        |   2 +-
 arch/powerpc/mm/numa.c                             |  70 +++++++++-
 arch/powerpc/net/bpf_jit_comp.c                    |   7 +-
 arch/s390/include/uapi/asm/statfs.h                |  10 +-
 arch/s390/kvm/diag.c                               |   4 +-
 arch/s390/net/bpf_jit_comp.c                       |  29 +++--
 arch/sparc/net/bpf_jit_comp.c                      |  17 ++-
 arch/x86/include/asm/irq.h                         |   1 +
 arch/x86/include/asm/kvm_para.h                    |  33 ++---
 arch/x86/kernel/irq.c                              |  70 ++++++++++
 arch/x86/kernel/kvm.c                              |  32 +++++
 arch/x86/kernel/smpboot.c                          |   6 +
 arch/x86/kvm/i8254.c                               |  18 +++
 arch/x86/kvm/lapic.c                               |   5 +-
 arch/x86/kvm/x86.c                                 |   3 +
 arch/x86/kvm/x86.h                                 |   2 +
 arch/x86/net/bpf_jit_comp.c                        |  14 +-
 arch/x86/platform/efi/efi.c                        |   2 +-
 drivers/ata/libata-core.c                          |  27 ++++
 drivers/ata/libata-scsi.c                          |  18 ++-
 drivers/ata/sata_mv.c                              |  25 ++++
 drivers/char/tpm/tpm_i2c_stm_st33.c                |   7 +-
 drivers/char/tpm/tpm_ppi.c                         |   8 +-
 drivers/edac/e752x_edac.c                          |   4 +-
 drivers/gpu/drm/nouveau/core/subdev/bios/init.c    |   8 +-
 drivers/hwmon/k10temp.c                            |   1 +
 drivers/i2c/busses/Kconfig                         |   1 +
 drivers/i2c/busses/i2c-mv64xxx.c                   |   5 +
 drivers/i2c/busses/i2c-piix4.c                     |  28 +++-
 drivers/md/bcache/bset.c                           |  26 +++-
 drivers/md/raid5.c                                 |   1 +
 drivers/mfd/max77686.c                             |   2 +-
 drivers/misc/mei/hbm.c                             |  19 +++
 drivers/misc/mei/hbm.h                             |   1 +
 drivers/misc/mei/init.c                            |  12 +-
 drivers/misc/mei/interrupt.c                       |  25 ++--
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c    |  13 +-
 drivers/net/ethernet/intel/e1000e/phy.c            |  10 +-
 drivers/net/ethernet/via/via-rhine.c               |   1 +
 drivers/net/usb/usbnet.c                           |   2 +-
 drivers/net/vxlan.c                                |  20 ++-
 drivers/net/wireless/ath/ath9k/hw.c                |   7 -
 drivers/net/wireless/ath/ath9k/recv.c              |  25 ++--
 drivers/net/wireless/b43/b43.h                     |   4 +-
 drivers/net/wireless/b43/main.c                    |  27 ++--
 drivers/net/wireless/b43/xmit.c                    |   4 +-
 drivers/net/wireless/b43legacy/main.c              |   1 +
 drivers/net/wireless/iwlwifi/iwl-prph.h            |   4 +
 drivers/net/wireless/iwlwifi/mvm/ops.c             |   1 +
 drivers/net/wireless/iwlwifi/pcie/trans.c          |  22 ++++
 drivers/net/wireless/mwifiex/fw.h                  |   2 +-
 drivers/net/wireless/mwifiex/scan.c                |   4 +-
 drivers/net/wireless/rtlwifi/base.c                |   3 +-
 drivers/net/wireless/rtlwifi/core.c                |  11 ++
 drivers/net/wireless/rtlwifi/rtl8188ee/dm.c        |   2 +-
 drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c  |  36 ++++++
 drivers/net/wireless/rtlwifi/rtl8192c/dm_common.h  |  14 ++
 drivers/net/wireless/rtlwifi/rtl8192cu/dm.c        |   9 ++
 drivers/net/wireless/rtlwifi/rtl8192cu/dm.h        |   3 +
 drivers/net/wireless/rtlwifi/rtl8192cu/rf.c        |  29 +++--
 drivers/net/wireless/rtlwifi/rtl8192cu/sw.c        |  17 ++-
 drivers/net/wireless/rtlwifi/stats.c               |  14 ++
 drivers/net/wireless/rtlwifi/usb.c                 |   4 +-
 drivers/net/wireless/rtlwifi/wifi.h                |  33 ++---
 drivers/net/xen-netfront.c                         |  88 ++++---------
 drivers/parport/parport_pc.c                       |  10 +-
 drivers/pinctrl/pinctrl-sunxi.c                    |  13 +-
 drivers/platform/x86/hp_accel.c                    |   1 +
 drivers/rtc/rtc-cmos.c                             |  52 +++++++-
 drivers/rtc/rtc-max8907.c                          |  11 +-
 drivers/scsi/bfa/bfad.c                            |   6 +-
 drivers/scsi/virtio_scsi.c                         |  15 ++-
 drivers/staging/lustre/lustre/ptlrpc/niobuf.c      |   6 +-
 drivers/staging/rtl8712/usb_intf.c                 |   5 +
 drivers/staging/vt6656/baseband.c                  |   3 -
 drivers/staging/vt6656/card.c                      |   2 +-
 drivers/staging/zram/zram_drv.c                    |   4 +-
 drivers/target/iscsi/iscsi_target.c                |  34 +++--
 drivers/target/iscsi/iscsi_target_nego.c           |   2 +-
 drivers/tty/serial/8250/8250_core.c                |   4 +
 drivers/tty/serial/8250/8250_pci.c                 |   8 +-
 drivers/tty/serial/atmel_serial.c                  |  45 ++++---
 drivers/usb/chipidea/ci.h                          |  26 +++-
 drivers/usb/chipidea/ci_hdrc_imx.c                 |  32 ++++-
 drivers/usb/chipidea/core.c                        |   2 +
 drivers/usb/chipidea/host.c                        |   1 +
 drivers/usb/chipidea/otg.h                         |   6 +-
 drivers/usb/chipidea/udc.c                         |  19 ++-
 drivers/usb/core/config.c                          |   7 -
 drivers/usb/core/hub.c                             |  54 +++-----
 drivers/usb/host/ehci.h                            |  18 ++-
 drivers/usb/host/xhci-ring.c                       |   2 +-
 drivers/usb/host/xhci.c                            |   7 +-
 drivers/usb/host/xhci.h                            |   2 +-
 drivers/usb/serial/cypress_m8.h                    |   2 +-
 drivers/usb/serial/ftdi_sio.c                      |  17 ++-
 drivers/usb/serial/option.c                        |  15 +++
 drivers/usb/serial/pl2303.c                        |  34 +++--
 drivers/usb/storage/unusual_devs.h                 |   7 +
 fs/btrfs/extent-tree.c                             |   2 +-
 fs/btrfs/ioctl.c                                   |   6 +
 fs/dcache.c                                        |   1 -
 fs/ext4/inline.c                                   |   8 +-
 fs/hpfs/alloc.c                                    |  66 +++++++++-
 fs/hpfs/hpfs_fn.h                                  |   2 +-
 fs/hpfs/super.c                                    |  29 ++++-
 fs/mount.h                                         |   2 +-
 include/linux/hugetlb.h                            |   6 +
 include/linux/libata.h                             |   2 +
 include/linux/usb/chipidea.h                       |   1 +
 mm/hugetlb.c                                       |  17 +++
 mm/mempolicy.c                                     |   2 +-
 mm/swap.c                                          | 143 ++++++++++++---------
 net/compat.c                                       |   9 +-
 net/core/filter.c                                  |  30 +----
 net/ieee802154/nl-phy.c                            |   6 +-
 net/ipv4/fib_frontend.c                            |   2 +
 net/ipv4/inet_diag.c                               |   4 +-
 net/ipv4/ip_gre.c                                  |   2 +-
 net/ipv4/ip_input.c                                |   2 +-
 net/ipv4/ip_tunnel.c                               |   1 +
 net/ipv4/ipmr.c                                    |   7 +-
 net/ipv4/tcp_metrics.c                             |  51 +++++---
 net/ipv6/ip6_input.c                               |   2 +-
 net/ipv6/ip6mr.c                                   |   7 +-
 net/rds/ib_recv.c                                  |   7 +-
 sound/core/init.c                                  |   4 +-
 sound/pci/Kconfig                                  |  12 ++
 sound/pci/hda/hda_generic.c                        |   6 +-
 sound/pci/hda/patch_analog.c                       |  22 ++--
 sound/pci/hda/patch_hdmi.c                         | 100 ++++++++++----
 sound/pci/hda/patch_realtek.c                      |  19 ++-
 sound/pci/rme9652/rme9652.c                        |   2 +-
 sound/soc/codecs/adau1701.c                        |   2 +-
 sound/soc/codecs/wm5110.c                          |  48 +++++++
 tools/perf/util/session.c                          |   7 +-
 155 files changed, 1726 insertions(+), 597 deletions(-)
author maintainer
#1 lore raw

PATCH 2/133 md/raid5: fix long-standing problem with bitmap handling on write failure.

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: NeilBrown <[email protected]>

commit 9f97e4b128d2ea90a5f5063ea0ee3b0911f4c669 upstream.

Before a write starts we set a bit in the write-intent bitmap. When the write completes we clear that bit if the write was successful to all devices. However if the write wasn't fully successful we should not clear the bit. If the faulty drive is subsequently re-added, the fact that the bit is still set ensure that we will re-write the data that is missing.

This logic is mediated by the STRIPE_DEGRADED flag - we only clear the bitmap bit when this flag is not set.
Currently we correctly set the flag if a write starts when some devices are failed or missing. But we do *not* set the flag if some device failed during the write attempt.
This is wrong and can result in clearing the bit inappropriately.

So: set the flag when a write fails.

This bug has been present since bitmaps were introduces, so the fix is suitable for any -stable kernel.

---
 drivers/md/raid5.c |    1 +
 1 file changed, 1 insertion(+)
drivers/md/raid5.c +1 -0
@@ -2004,6 +2004,7 @@ static void raid5_end_write_request(stru
 			set_bit(R5_MadeGoodRepl, &sh->dev[i].flags);
 	} else {
 		if (!uptodate) {
+			set_bit(STRIPE_DEGRADED, &sh->state);
 			set_bit(WriteErrorSeen, &rdev->flags);
 			set_bit(R5_WriteError, &sh->dev[i].flags);
 			if (!test_and_set_bit(WantReplacement, &rdev->flags))
author maintainer
#2 lore raw

PATCH 3/133 drm/nouveau/bios: fix offset calculation for BMPv1 bioses

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Ilia Mirkin <[email protected]>

commit 5d2f4767c4eacab351b8450b0de4d3261fe1a957 upstream.

The only BIOS on record that needs the 14 offset has a bios major version 2 but BMP version 1.01. Another bunch of BIOSes that need the 18 offset have BMP version 2.01 or 5.01 or higher. So instead of looking at the bios major version, look at the BMP version. BIOSes with BMP version 0 do not contain a detectable script, so always return 0 for them.

See https://bugs.freedesktop.org/show_bug.cgi?id=68835

---
 drivers/gpu/drm/nouveau/core/subdev/bios/init.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/gpu/drm/nouveau/core/subdev/bios/init.c +4 -4
@@ -365,13 +365,13 @@ static u16
 init_script(struct nouveau_bios *bios, int index)
 {
 	struct nvbios_init init = { .bios = bios };
-	u16 data;
+	u16 bmp_ver = bmp_version(bios), data;
 
-	if (bmp_version(bios) && bmp_version(bios) < 0x0510) {
-		if (index > 1)
+	if (bmp_ver && bmp_ver < 0x0510) {
+		if (index > 1 || bmp_ver < 0x0100)
 			return 0x0000;
 
-		data = bios->bmp_offset + (bios->version.major < 2 ? 14 : 18);
+		data = bios->bmp_offset + (bmp_ver < 0x0200 ? 14 : 18);
 		return nv_ro16(bios, data + (index * 2));
 	}
 
author maintainer
#3 lore raw

PATCH 4/133 mm: hugetlbfs: fix hugetlbfs optimization

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Andrea Arcangeli <[email protected]>

commit 27c73ae759774e63313c1fbfeb17ba076cea64c5 upstream.

Commit 7cb2ef56e6a8 ("mm: fix aio performance regression for database caused by THP") can cause dereference of a dangling pointer if split_huge_page runs during PageHuge() if there are updates to the tail_page->private field.

Also it is repeating compound_head twice for hugetlbfs and it is running compound_head+compound_trans_head for THP when a single one is needed in both cases.

The new code within the PageSlab() check doesn't need to verify that the THP page size is never bigger than the smallest hugetlbfs page size, to avoid memory corruption.

A longstanding theoretical race condition was found while fixing the above (see the change right after the skip_unlock label, that is relevant for the compound_lock path too).

By re-establishing the _mapcount tail refcounting for all compound pages, this also fixes the below problem:

echo 0 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages

  BUG: Bad page state in process bash  pfn:59a01
  page:ffffea000139b038 count:0 mapcount:10 mapping:          (null) index:0x0
  page flags: 0x1c00000000008000(tail)
  Modules linked in:
  CPU: 6 PID: 2018 Comm: bash Not tainted 3.12.0+ #25
  Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
  Call Trace:
    dump_stack+0x55/0x76
    bad_page+0xd5/0x130
    free_pages_prepare+0x213/0x280
    __free_pages+0x36/0x80
    update_and_free_page+0xc1/0xd0
    free_pool_huge_page+0xc2/0xe0
    set_max_huge_pages.part.58+0x14c/0x220
    nr_hugepages_store_common.isra.60+0xd0/0xf0
    nr_hugepages_store+0x13/0x20
    kobj_attr_store+0xf/0x20
    sysfs_write_file+0x189/0x1e0
    vfs_write+0xc5/0x1f0
    SyS_write+0x55/0xb0
    system_call_fastpath+0x16/0x1b
Cc Pravin Shelar <[email protected]>
Cc Greg Kroah-Hartman <[email protected]>
Cc Ben Hutchings <[email protected]>
Cc Christoph Lameter <[email protected]>
Cc Johannes Weiner <[email protected]>
Cc Mel Gorman <[email protected]>
Cc Rik van Riel <[email protected]>
Cc Andi Kleen <[email protected]>
Cc Minchan Kim <[email protected]>
Cc Guillaume Morin <[email protected]>
---
 include/linux/hugetlb.h |    6 ++
 mm/hugetlb.c            |   17 +++++
 mm/swap.c               |  143 +++++++++++++++++++++++++++---------------------
 3 files changed, 106 insertions(+), 60 deletions(-)
include/linux/hugetlb.h +6 -0
@@ -31,6 +31,7 @@ struct hugepage_subpool *hugepage_new_su
 void hugepage_put_subpool(struct hugepage_subpool *spool);
 
 int PageHuge(struct page *page);
+int PageHeadHuge(struct page *page_head);
 
 void reset_vma_resv_huge_pages(struct vm_area_struct *vma);
 int hugetlb_sysctl_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *);
@@ -103,6 +104,11 @@ static inline int PageHuge(struct page *
 {
 	return 0;
 }
+
+static inline int PageHeadHuge(struct page *page_head)
+{
+	return 0;
+}
 
 static inline void reset_vma_resv_huge_pages(struct vm_area_struct *vma)
 {
mm/hugetlb.c +17 -0
@@ -736,6 +736,23 @@ int PageHuge(struct page *page)
 }
 EXPORT_SYMBOL_GPL(PageHuge);
 
+/*
+ * PageHeadHuge() only returns true for hugetlbfs head page, but not for
+ * normal or transparent huge pages.
+ */
+int PageHeadHuge(struct page *page_head)
+{
+	compound_page_dtor *dtor;
+
+	if (!PageHead(page_head))
+		return 0;
+
+	dtor = get_compound_page_dtor(page_head);
+
+	return dtor == free_huge_page;
+}
+EXPORT_SYMBOL_GPL(PageHeadHuge);
+
 pgoff_t __basepage_index(struct page *page)
 {
 	struct page *page_head = compound_head(page);
mm/swap.c +83 -60

182 lines not on this page · raw message

author maintainer
#4 lore raw

PATCH 5/133 e752x_edac: Fix pci_dev usage count

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Aristeu Rozanski <[email protected]>

commit 90ed4988b8c030d65b41b7d13140e9376dc6ec5a upstream.

In case the device 0, function 1 is not found using pci_get_device(), pci_scan_single_device() will be used but, differently than pci_get_device(), it allocates a pci_dev but doesn't does bump the usage count on the pci_dev and after few module removals and loads the pci_dev will be freed.

---
 drivers/edac/e752x_edac.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/edac/e752x_edac.c +3 -1
@@ -1182,9 +1182,11 @@ static int e752x_get_devs(struct pci_dev
 	pvt->bridge_ck = pci_get_device(PCI_VENDOR_ID_INTEL,
 				pvt->dev_info->err_dev, pvt->bridge_ck);
 
-	if (pvt->bridge_ck == NULL)
+	if (pvt->bridge_ck == NULL) {
 		pvt->bridge_ck = pci_scan_single_device(pdev->bus,
 							PCI_DEVFN(0, 1));
+		pci_dev_get(pvt->bridge_ck);
+	}
 
 	if (pvt->bridge_ck == NULL) {
 		e752x_printk(KERN_ERR, "error reporting device not found:"
author maintainer
#5 lore raw

PATCH 6/133 e1000e: fix compiler warnings

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: David Ertman <[email protected]>

commit 9e6c3b63399dd424d33a34e08b77f2cab0b84cdc upstream.

This patch is to fix a compiler warning of __bad_udelay due to a value of >999 being passed as a parameter to udelay() in the function e1000e_phy_has_link_generic(). This affects the gcc compiler when it is given a flag of -O3 and the icc compiler.

This patch is also making the change from mdelay() to msleep() in the same function, since it was determined though code inspection that this function is never called in atomic context.

---
 drivers/net/ethernet/intel/e1000e/phy.c |   10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/net/ethernet/intel/e1000e/phy.c +7 -3
@@ -1757,19 +1757,23 @@ s32 e1000e_phy_has_link_generic(struct e
 		 * it across the board.
 		 */
 		ret_val = e1e_rphy(hw, MII_BMSR, &phy_status);
-		if (ret_val)
+		if (ret_val) {
 			/* If the first read fails, another entity may have
 			 * ownership of the resources, wait and try again to
 			 * see if they have relinquished the resources yet.
 			 */
-			udelay(usec_interval);
+			if (usec_interval >= 1000)
+				msleep(usec_interval / 1000);
+			else
+				udelay(usec_interval);
+		}
 		ret_val = e1e_rphy(hw, MII_BMSR, &phy_status);
 		if (ret_val)
 			break;
 		if (phy_status & BMSR_LSTATUS)
 			break;
 		if (usec_interval >= 1000)
-			mdelay(usec_interval / 1000);
+			msleep(usec_interval / 1000);
 		else
 			udelay(usec_interval);
 	}
author maintainer
#6 lore raw

PATCH 7/133 mm/mempolicy.c: fix mempolicy printing in numa_maps

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: David Rientjes <[email protected]>

commit 8790c71a18e5d2d93532ae250bcf5eddbba729cd upstream.

As a result of commit 5606e3877ad8 ("mm: numa: Migrate on reference policy"), /proc/<pid>/numa_maps prints the mempolicy for any <pid> as "prefer:N" for the local node, N, of the process reading the file.

This should only be printed when the mempolicy of <pid> is MPOL_PREFERRED for node N.

If the process is actually only using the default mempolicy for local node allocation, make sure "default" is printed as expected.

---
 mm/mempolicy.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
mm/mempolicy.c +1 -1
@@ -2857,7 +2857,7 @@ int mpol_to_str(char *buffer, int maxlen
 	 */
 	VM_BUG_ON(maxlen < strlen("interleave") + strlen("relative") + 16);
 
-	if (!pol || pol == &default_policy)
+	if (!pol || pol == &default_policy || (pol->flags & MPOL_F_MORON))
 		mode = MPOL_DEFAULT;
 	else
 		mode = pol->mode;
author maintainer
#7 lore raw

PATCH 8/133 x86, x32: Correct invalid use of user timespec in the kernel

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: PaX Team <[email protected]>

commit 2def2ef2ae5f3990aabdbe8a755911902707d268 upstream.

The x32 case for the recvmsg() timout handling is broken:

asmlinkage long compat_sys_recvmmsg(int fd, struct compat_mmsghdr __user *mmsg,
unsigned int vlen, unsigned int flags, struct compat_timespec __user *timeout)

  {
          int datagrams;
          struct timespec ktspec;

          if (flags & MSG_CMSG_COMPAT)
                  return -EINVAL;

          if (COMPAT_USE_64BIT_TIME)
                  return __sys_recvmmsg(fd, (struct mmsghdr __user *)mmsg, vlen,
                                        flags | MSG_CMSG_COMPAT,
                                        (struct timespec *) timeout);
          ...

The timeout pointer parameter is provided by userland (hence the __user annotation) but for x32 syscalls it's simply cast to a kernel pointer and is passed to __sys_recvmmsg which will eventually directly dereference it for both reading and writing. Other callers to __sys_recvmmsg properly copy from userland to the kernel first.

The bug was introduced by commit ee4fa23c4bfc ("compat: Use COMPAT_USE_64BIT_TIME in net/compat.c") and should affect all kernels since 3.4 (and perhaps vendor kernels if they backported x32 support along with this code).

Note that CONFIG_X86_X32_ABI gets enabled at build time and only if CONFIG_X86_X32 is enabled and ld can build x32 executables.

Other uses of COMPAT_USE_64BIT_TIME seem fine.

This addresses CVE-2014-0038.

---
 net/compat.c |    9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)
net/compat.c +2 -7
@@ -780,21 +780,16 @@ asmlinkage long compat_sys_recvmmsg(int
 	if (flags & MSG_CMSG_COMPAT)
 		return -EINVAL;
 
-	if (COMPAT_USE_64BIT_TIME)
-		return __sys_recvmmsg(fd, (struct mmsghdr __user *)mmsg, vlen,
-				      flags | MSG_CMSG_COMPAT,
-				      (struct timespec *) timeout);
-
 	if (timeout == NULL)
 		return __sys_recvmmsg(fd, (struct mmsghdr __user *)mmsg, vlen,
 				      flags | MSG_CMSG_COMPAT, NULL);
 
-	if (get_compat_timespec(&ktspec, timeout))
+	if (compat_get_timespec(&ktspec, timeout))
 		return -EFAULT;
 
 	datagrams = __sys_recvmmsg(fd, (struct mmsghdr __user *)mmsg, vlen,
 				   flags | MSG_CMSG_COMPAT, &ktspec);
-	if (datagrams > 0 && put_compat_timespec(&ktspec, timeout))
+	if (datagrams > 0 && compat_put_timespec(&ktspec, timeout))
 		datagrams = -EFAULT;
 
 	return datagrams;
author maintainer
#8 lore raw

PATCH 9/133 x86/efi: Fix off-by-one bug in EFI Boot Services reservation

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Dave Young <[email protected]>

commit a7f84f03f660d93574ac88835d056c0d6468aebe upstream.

Current code check boot service region with kernel text region by:
start+size >= __pa_symbol(_text)
The end of the above region should be start + size - 1 instead.

I see this problem in ovmf + Fedora 19 grub boot:
text start: 1000000 md start: 800000 md size: 800000

---
 arch/x86/platform/efi/efi.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/platform/efi/efi.c +1 -1
@@ -440,7 +440,7 @@ void __init efi_reserve_boot_services(vo
 		 * - Not within any part of the kernel
 		 * - Not the bios reserved area
 		*/
-		if ((start+size >= __pa_symbol(_text)
+		if ((start + size > __pa_symbol(_text)
 				&& start <= __pa_symbol(_end)) ||
 			!e820_all_mapped(start, start+size, E820_RAM) ||
 			memblock_is_region_reserved(start, size)) {
author maintainer
#9 lore raw

PATCH 10/133 x86: Add check for number of available vectors before CPU down

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Prarit Bhargava <[email protected]>

commit da6139e49c7cb0f4251265cb5243b8d220adb48d upstream.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=64791

When a cpu is downed on a system, the irqs on the cpu are assigned to other cpus. It is possible, however, that when a cpu is downed there aren't enough free vectors on the remaining cpus to account for the vectors from the cpu that is being downed.

This results in an interesting "overflow" condition where irqs are "assigned" to a CPU but are not handled.

For example, when downing cpus on a 1-64 logical processor system:

<snip>

[  232.021745] smpboot: CPU 61 is now offline
[  238.480275] smpboot: CPU 62 is now offline
[  245.991080] ------------[ cut here ]------------
[  245.996270] WARNING: CPU: 0 PID: 0 at net/sched/sch_generic.c:264 dev_watchdog+0x246/0x250()
[  246.005688] NETDEV WATCHDOG: p786p1 (ixgbe): transmit queue 0 timed out
[  246.013070] Modules linked in: lockd sunrpc iTCO_wdt iTCO_vendor_support sb_edac ixgbe microcode e1000e pcspkr joydev edac_core lpc_ich ioatdma ptp mdio mfd_core i2c_i801 dca pps_core i2c_core wmi acpi_cpufreq isci libsas scsi_transport_sas
[  246.037633] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.12.0+ #14
[  246.044451] Hardware name: Intel Corporation S4600LH ........../SVRBD-ROW_T, BIOS SE5C600.86B.01.08.0003.022620131521 02/26/2013
[  246.057371]  0000000000000009 ffff88081fa03d40 ffffffff8164fbf6 ffff88081fa0ee48
[  246.065728]  ffff88081fa03d90 ffff88081fa03d80 ffffffff81054ecc ffff88081fa13040
[  246.074073]  0000000000000000 ffff88200cce0000 0000000000000040 0000000000000000
[  246.082430] Call Trace:
[  246.085174]  <IRQ>  [<ffffffff8164fbf6>] dump_stack+0x46/0x58
[  246.091633]  [<ffffffff81054ecc>] warn_slowpath_common+0x8c/0xc0
[  246.098352]  [<ffffffff81054fb6>] warn_slowpath_fmt+0x46/0x50
[  246.104786]  [<ffffffff815710d6>] dev_watchdog+0x246/0x250
[  246.110923]  [<ffffffff81570e90>] ? dev_deactivate_queue.constprop.31+0x80/0x80
[  246.119097]  [<ffffffff8106092a>] call_timer_fn+0x3a/0x110
[  246.125224]  [<ffffffff8106280f>] ? update_process_times+0x6f/0x80
[  246.132137]  [<ffffffff81570e90>] ? dev_deactivate_queue.constprop.31+0x80/0x80
[  246.140308]  [<ffffffff81061db0>] run_timer_softirq+0x1f0/0x2a0
[  246.146933]  [<ffffffff81059a80>] __do_softirq+0xe0/0x220
[  246.152976]  [<ffffffff8165fedc>] call_softirq+0x1c/0x30
[  246.158920]  [<ffffffff810045f5>] do_softirq+0x55/0x90
[  246.164670]  [<ffffffff81059d35>] irq_exit+0xa5/0xb0
[  246.170227]  [<ffffffff8166062a>] smp_apic_timer_interrupt+0x4a/0x60
[  246.177324]  [<ffffffff8165f40a>] apic_timer_interrupt+0x6a/0x70
[  246.184041]  <EOI>  [<ffffffff81505a1b>] ? cpuidle_enter_state+0x5b/0xe0
[  246.191559]  [<ffffffff81505a17>] ? cpuidle_enter_state+0x57/0xe0
[  246.198374]  [<ffffffff81505b5d>] cpuidle_idle_call+0xbd/0x200
[  246.204900]  [<ffffffff8100b7ae>] arch_cpu_idle+0xe/0x30
[  246.210846]  [<ffffffff810a47b0>] cpu_startup_entry+0xd0/0x250
[  246.217371]  [<ffffffff81646b47>] rest_init+0x77/0x80
[  246.223028]  [<ffffffff81d09e8e>] start_kernel+0x3ee/0x3fb
[  246.229165]  [<ffffffff81d0989f>] ? repair_env_string+0x5e/0x5e
[  246.235787]  [<ffffffff81d095a5>] x86_64_start_reservations+0x2a/0x2c
[  246.242990]  [<ffffffff81d0969f>] x86_64_start_kernel+0xf8/0xfc
[  246.249610] ---[ end trace fb74fdef54d79039 ]---
[  246.254807] ixgbe 0000:c2:00.0 p786p1: initiating reset due to tx timeout
[  246.262489] ixgbe 0000:c2:00.0 p786p1: Reset adapter
Last login: Mon Nov 11 08:35:14 from 10.18.17.119
[root@(none) ~]# [  246.792676] ixgbe 0000:c2:00.0 p786p1: detected SFP+: 5
[  249.231598] ixgbe 0000:c2:00.0 p786p1: NIC Link is Up 10 Gbps, Flow Control: RX/TX
[  246.792676] ixgbe 0000:c2:00.0 p786p1: detected SFP+: 5
[  249.231598] ixgbe 0000:c2:00.0 p786p1: NIC Link is Up 10 Gbps, Flow Control: RX/TX

(last lines keep repeating. ixgbe driver is dead until module reload.)

If the downed cpu has more vectors than are free on the remaining cpus on the system, it is possible that some vectors are "orphaned" even though they are assigned to a cpu. In this case, since the ixgbe driver had a watchdog, the watchdog fired and notified that something was wrong.

This patch adds a function, check_vectors(), to compare the number of vectors on the CPU going down and compares it to the number of vectors available on the system. If there aren't enough vectors for the CPU to go down, an error is returned and propogated back to userspace.

v2: Do not need to look at percpu irqs
v3: Need to check affinity to prevent counting of MSIs in IOAPIC Lowest
Priority Mode
v4: Additional changes suggested by Gong Chen.
v5/v6/v7/v8: Updated comment text

---
 arch/x86/include/asm/irq.h |    1 
 arch/x86/kernel/irq.c      |   70 +++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/smpboot.c  |    6 +++
 3 files changed, 77 insertions(+)
arch/x86/include/asm/irq.h +1 -0
@@ -25,6 +25,7 @@ extern void irq_ctx_init(int cpu);
 
 #ifdef CONFIG_HOTPLUG_CPU
 #include <linux/cpumask.h>
+extern int check_irq_vectors_for_cpu_disable(void);
 extern void fixup_irqs(void);
 extern void irq_force_complete_move(int);
 #endif
arch/x86/kernel/irq.c +70 -0

76 lines not on this page · raw message

arch/x86/kernel/smpboot.c +6 -0
@@ -1310,6 +1310,12 @@ void cpu_disable_common(void)
 
 int native_cpu_disable(void)
 {
+	int ret;
+
+	ret = check_irq_vectors_for_cpu_disable();
+	if (ret)
+		return ret;
+
 	clear_local_APIC();
 
 	cpu_disable_common();
author maintainer
#12 lore raw

PATCH 11/133 KVM: x86: limit PIT timer frequency

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Marcelo Tosatti <[email protected]>

commit 9ed96e87c5748de4c2807ef17e81287c7304186c upstream.

Limit PIT timer frequency similarly to the limit applied by LAPIC timer.

---
 arch/x86/kvm/i8254.c |   18 ++++++++++++++++++
 arch/x86/kvm/lapic.c |    3 ---
 arch/x86/kvm/x86.c   |    3 +++
 arch/x86/kvm/x86.h   |    2 ++
 4 files changed, 23 insertions(+), 3 deletions(-)
arch/x86/kvm/i8254.c +18 -0
@@ -37,6 +37,7 @@
 
 #include "irq.h"
 #include "i8254.h"
+#include "x86.h"
 
 #ifndef CONFIG_X86_64
 #define mod_64(x, y) ((x) - (y) * div64_u64(x, y))
@@ -349,6 +350,23 @@ static void create_pit_timer(struct kvm
 	atomic_set(&ps->pending, 0);
 	ps->irq_ack = 1;
 
+	/*
+	 * Do not allow the guest to program periodic timers with small
+	 * interval, since the hrtimers are not throttled by the host
+	 * scheduler.
+	 */
+	if (ps->is_periodic) {
+		s64 min_period = min_timer_period_us * 1000LL;
+
+		if (ps->period < min_period) {
+			pr_info_ratelimited(
+			    "kvm: requested %lld ns "
+			    "i8254 timer period limited to %lld ns\n",
+			    ps->period, min_period);
+			ps->period = min_period;
+		}
+	}
+
 	hrtimer_start(&ps->timer, ktime_add_ns(ktime_get(), interval),
 		      HRTIMER_MODE_ABS);
 }
arch/x86/kvm/lapic.c +0 -3
@@ -71,9 +71,6 @@
 #define VEC_POS(v) ((v) & (32 - 1))
 #define REG_POS(v) (((v) >> 5) << 4)
 
-static unsigned int min_timer_period_us = 500;
-module_param(min_timer_period_us, uint, S_IRUGO | S_IWUSR);
-
 static inline void apic_set_reg(struct kvm_lapic *apic, int reg_off, u32 val)
 {
 	*((u32 *) (apic->regs + reg_off)) = val;
arch/x86/kvm/x86.c +3 -0
@@ -94,6 +94,9 @@ EXPORT_SYMBOL_GPL(kvm_x86_ops);
 static bool ignore_msrs = 0;
 module_param(ignore_msrs, bool, S_IRUGO | S_IWUSR);
 
+unsigned int min_timer_period_us = 500;
+module_param(min_timer_period_us, uint, S_IRUGO | S_IWUSR);
+
 bool kvm_has_tsc_control;
 EXPORT_SYMBOL_GPL(kvm_has_tsc_control);
 u32  kvm_max_guest_tsc_khz;
arch/x86/kvm/x86.h +2 -0
@@ -124,5 +124,7 @@ int kvm_write_guest_virt_system(struct x
 
 extern u64 host_xcr0;
 
+extern unsigned int min_timer_period_us;
+
 extern struct static_key kvm_no_apic_vcpu;
 #endif
author maintainer
#13 lore raw

PATCH 12/133 x86, kvm: cache the base of the KVM cpuid leaves

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Paolo Bonzini <[email protected]>

commit 1c300a40772dae829b91dad634999a6a522c0829 upstream.

It is unnecessary to go through hypervisor_cpuid_base every time a leaf is found (which will be every time a feature is requested after the next patch).

Fixes 1085ba7f552d84aa8ac0ae903fa8d0cc2ff9f79d
---
 arch/x86/include/asm/kvm_para.h |   22 ++++++----------------
 arch/x86/kernel/kvm.c           |   27 +++++++++++++++++++++++++++
 2 files changed, 33 insertions(+), 16 deletions(-)
arch/x86/include/asm/kvm_para.h +6 -16
@@ -85,28 +85,13 @@ static inline long kvm_hypercall4(unsign
 	return ret;
 }
 
-static inline uint32_t kvm_cpuid_base(void)
-{
-	if (boot_cpu_data.cpuid_level < 0)
-		return 0;	/* So we don't blow up on old processors */
-
-	if (cpu_has_hypervisor)
-		return hypervisor_cpuid_base("KVMKVMKVM\0\0\0", 0);
-
-	return 0;
-}
-
-static inline bool kvm_para_available(void)
-{
-	return kvm_cpuid_base() != 0;
-}
-
 static inline unsigned int kvm_arch_para_features(void)
 {
 	return cpuid_eax(KVM_CPUID_FEATURES);
 }
 
 #ifdef CONFIG_KVM_GUEST
+bool kvm_para_available(void);
 void __init kvm_guest_init(void);
 void kvm_async_pf_task_wait(u32 token);
 void kvm_async_pf_task_wake(u32 token);
@@ -126,6 +111,11 @@ static inline void kvm_spinlock_init(voi
 #define kvm_async_pf_task_wait(T) do {} while(0)
 #define kvm_async_pf_task_wake(T) do {} while(0)
 
+static inline bool kvm_para_available(void)
+{
+	return 0;
+}
+
 static inline u32 kvm_read_and_reset_pf_reason(void)
 {
 	return 0;
arch/x86/kernel/kvm.c +27 -0
@@ -500,6 +500,33 @@ void __init kvm_guest_init(void)
 #endif
 }
 
+static noinline uint32_t __kvm_cpuid_base(void)
+{
+	if (boot_cpu_data.cpuid_level < 0)
+		return 0;	/* So we don't blow up on old processors */
+
+	if (cpu_has_hypervisor)
+		return hypervisor_cpuid_base("KVMKVMKVM\0\0\0", 0);
+
+	return 0;
+}
+
+static inline uint32_t kvm_cpuid_base(void)
+{
+	static int kvm_cpuid_base = -1;
+
+	if (kvm_cpuid_base == -1)
+		kvm_cpuid_base = __kvm_cpuid_base();
+
+	return kvm_cpuid_base;
+}
+
+bool kvm_para_available(void)
+{
+	return kvm_cpuid_base() != 0;
+}
+EXPORT_SYMBOL_GPL(kvm_para_available);
+
 static uint32_t __init kvm_detect(void)
 {
 	return kvm_cpuid_base();
author maintainer
#14 lore raw

PATCH 13/133 x86, kvm: correctly access the KVM_CPUID_FEATURES leaf at 0x40000101

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Paolo Bonzini <[email protected]>

commit 77f01bdfa5e55dc19d3eb747181d2730a9bb3ca8 upstream.

When Hyper-V hypervisor leaves are present, KVM must relocate its own leaves at 0x40000100, because Windows does not look for Hyper-V leaves at indices other than 0x40000000. In this case, the KVM features are at 0x40000101, but the old code would always look at 0x40000001.

Fix by using kvm_cpuid_base(). This also requires making the function non-inline, since kvm_cpuid_base() is static.

Fixes 1085ba7f552d84aa8ac0ae903fa8d0cc2ff9f79d
---
 arch/x86/include/asm/kvm_para.h |   11 ++++++-----
 arch/x86/kernel/kvm.c           |    5 +++++
 2 files changed, 11 insertions(+), 5 deletions(-)
arch/x86/include/asm/kvm_para.h +6 -5
@@ -85,13 +85,9 @@ static inline long kvm_hypercall4(unsign
 	return ret;
 }
 
-static inline unsigned int kvm_arch_para_features(void)
-{
-	return cpuid_eax(KVM_CPUID_FEATURES);
-}
-
 #ifdef CONFIG_KVM_GUEST
 bool kvm_para_available(void);
+unsigned int kvm_arch_para_features(void);
 void __init kvm_guest_init(void);
 void kvm_async_pf_task_wait(u32 token);
 void kvm_async_pf_task_wake(u32 token);
@@ -115,6 +111,11 @@ static inline bool kvm_para_available(vo
 {
 	return 0;
 }
+
+static inline unsigned int kvm_arch_para_features(void)
+{
+	return 0;
+}
 
 static inline u32 kvm_read_and_reset_pf_reason(void)
 {
arch/x86/kernel/kvm.c +5 -0
@@ -527,6 +527,11 @@ bool kvm_para_available(void)
 }
 EXPORT_SYMBOL_GPL(kvm_para_available);
 
+unsigned int kvm_arch_para_features(void)
+{
+	return cpuid_eax(kvm_cpuid_base() | KVM_CPUID_FEATURES);
+}
+
 static uint32_t __init kvm_detect(void)
 {
 	return kvm_cpuid_base();
author maintainer
#15 lore raw

PATCH 14/133 usb: chipidea: add freescale imx28 special write register method

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Peter Chen <[email protected]>

commit ed8f8318d2ef3e5f9e4ddf79349508c116b68d7f upstream.

According to Freescale imx28 Errata, "ENGR119653 USB: ARM to USB register error issue", All USB register write operations must use the ARM SWP instruction. So, we implement special hw_write and hw_test_and_clear for imx28.

Discussion for it at below:
http://marc.info/?l=linux-usb&m=137996395529294&w=2

This patch is needed for stable tree 3.11+.

---
 drivers/usb/chipidea/ci.h    |   26 ++++++++++++++++++++++++--
 drivers/usb/chipidea/core.c  |    2 ++
 drivers/usb/chipidea/host.c  |    1 +
 include/linux/usb/chipidea.h |    1 +
 4 files changed, 28 insertions(+), 2 deletions(-)
drivers/usb/chipidea/ci.h +24 -2
@@ -135,6 +135,7 @@ struct hw_bank {
  * @id_event: indicates there is an id event, and handled at ci_otg_work
  * @b_sess_valid_event: indicates there is a vbus event, and handled
  * at ci_otg_work
+ * @imx28_write_fix: Freescale imx28 needs swp instruction for writing
  */
 struct ci_hdrc {
 	struct device			*dev;
@@ -173,6 +174,7 @@ struct ci_hdrc {
 	struct dentry			*debugfs;
 	bool				id_event;
 	bool				b_sess_valid_event;
+	bool				imx28_write_fix;
 };
 
 static inline struct ci_role_driver *ci_role(struct ci_hdrc *ci)
@@ -253,6 +255,26 @@ static inline u32 hw_read(struct ci_hdrc
 	return ioread32(ci->hw_bank.regmap[reg]) & mask;
 }
 
+#ifdef CONFIG_SOC_IMX28
+static inline void imx28_ci_writel(u32 val, volatile void __iomem *addr)
+{
+	__asm__ ("swp %0, %0, [%1]" : : "r"(val), "r"(addr));
+}
+#else
+static inline void imx28_ci_writel(u32 val, volatile void __iomem *addr)
+{
+}
+#endif
+
+static inline void __hw_write(struct ci_hdrc *ci, u32 val,
+		void __iomem *addr)
+{
+	if (ci->imx28_write_fix)
+		imx28_ci_writel(val, addr);
+	else
+		iowrite32(val, addr);
+}
+
 /**
  * hw_write: writes to a hw register
  * @reg:  register index
@@ -266,7 +288,7 @@ static inline void hw_write(struct ci_hd
 		data = (ioread32(ci->hw_bank.regmap[reg]) & ~mask)
 			| (data & mask);
 
-	iowrite32(data, ci->hw_bank.regmap[reg]);
+	__hw_write(ci, data, ci->hw_bank.regmap[reg]);
 }
 
 /**
@@ -281,7 +303,7 @@ static inline u32 hw_test_and_clear(stru
 {
 	u32 val = ioread32(ci->hw_bank.regmap[reg]) & mask;
 
-	iowrite32(val, ci->hw_bank.regmap[reg]);
+	__hw_write(ci, val, ci->hw_bank.regmap[reg]);
 	return val;
 }
 
drivers/usb/chipidea/core.c +2 -0
@@ -497,6 +497,8 @@ static int ci_hdrc_probe(struct platform
 		ci->transceiver = ci->platdata->phy;
 	else
 		ci->global_phy = true;
+	ci->imx28_write_fix = !!(ci->platdata->flags &
+		CI_HDRC_IMX28_WRITE_FIX);
 
 	ret = hw_device_init(ci, base);
 	if (ret < 0) {
drivers/usb/chipidea/host.c +1 -0
@@ -65,6 +65,7 @@ static int host_start(struct ci_hdrc *ci
 	ehci->caps = ci->hw_bank.cap;
 	ehci->has_hostpc = ci->hw_bank.lpm;
 	ehci->has_tdi_phy_lpm = ci->hw_bank.lpm;
+	ehci->imx28_write_fix = ci->imx28_write_fix;
 
 	if (ci->platdata->reg_vbus) {
 		ret = regulator_enable(ci->platdata->reg_vbus);
include/linux/usb/chipidea.h +1 -0
@@ -24,6 +24,7 @@ struct ci_hdrc_platform_data {
 	 * but otg is not supported (no register otgsc).
 	 */
 #define CI_HDRC_DUAL_ROLE_NOT_OTG	BIT(4)
+#define CI_HDRC_IMX28_WRITE_FIX		BIT(5)
 	enum usb_dr_mode	dr_mode;
 #define CI_HDRC_CONTROLLER_RESET_EVENT		0
 #define CI_HDRC_CONTROLLER_STOPPED_EVENT	1
author maintainer
#16 lore raw

PATCH 15/133 usb: chipidea: imx: set CI_HDRC_IMX28_WRITE_FIX for imx28

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Peter Chen <[email protected]>

commit 1071055e2a118a81c0b300d7f4af7eba3f7a7c82 upstream.

Due to imx28 needs ARM swp instruction for writing, we set CI_HDRC_IMX28_WRITE_FIX for imx28.

This patch is needed for stable tree 3.11+

---
 drivers/usb/chipidea/ci_hdrc_imx.c |   32 ++++++++++++++++++++++++++------
 1 file changed, 26 insertions(+), 6 deletions(-)
drivers/usb/chipidea/ci_hdrc_imx.c +26 -6
@@ -23,6 +23,26 @@
 #include "ci.h"
 #include "ci_hdrc_imx.h"
 
+#define CI_HDRC_IMX_IMX28_WRITE_FIX BIT(0)
+
+struct ci_hdrc_imx_platform_flag {
+	unsigned int flags;
+};
+
+static const struct ci_hdrc_imx_platform_flag imx27_usb_data = {
+};
+
+static const struct ci_hdrc_imx_platform_flag imx28_usb_data = {
+	.flags = CI_HDRC_IMX_IMX28_WRITE_FIX,
+};
+
+static const struct of_device_id ci_hdrc_imx_dt_ids[] = {
+	{ .compatible = "fsl,imx28-usb", .data = &imx28_usb_data},
+	{ .compatible = "fsl,imx27-usb", .data = &imx27_usb_data},
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, ci_hdrc_imx_dt_ids);
+
 struct ci_hdrc_imx_data {
 	struct usb_phy *phy;
 	struct platform_device *ci_pdev;
@@ -82,6 +102,9 @@ static int ci_hdrc_imx_probe(struct plat
 				  CI_HDRC_DISABLE_STREAMING,
 	};
 	int ret;
+	const struct of_device_id *of_id =
+			of_match_device(ci_hdrc_imx_dt_ids, &pdev->dev);
+	const struct ci_hdrc_imx_platform_flag *imx_platform_flag = of_id->data;
 
 	data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL);
 	if (!data) {
@@ -121,6 +144,9 @@ static int ci_hdrc_imx_probe(struct plat
 
 	pdata.phy = data->phy;
 
+	if (imx_platform_flag->flags & CI_HDRC_IMX_IMX28_WRITE_FIX)
+		pdata.flags |= CI_HDRC_IMX28_WRITE_FIX;
+
 	if (!pdev->dev.dma_mask)
 		pdev->dev.dma_mask = &pdev->dev.coherent_dma_mask;
 	if (!pdev->dev.coherent_dma_mask)
@@ -187,12 +213,6 @@ static int ci_hdrc_imx_remove(struct pla
 	return 0;
 }
 
-static const struct of_device_id ci_hdrc_imx_dt_ids[] = {
-	{ .compatible = "fsl,imx27-usb", },
-	{ /* sentinel */ }
-};
-MODULE_DEVICE_TABLE(of, ci_hdrc_imx_dt_ids);
-
 static struct platform_driver ci_hdrc_imx_driver = {
 	.probe = ci_hdrc_imx_probe,
 	.remove = ci_hdrc_imx_remove,
author maintainer
#17 lore raw

PATCH 16/133 usb: chipidea: need to mask INT_STATUS when write otgsc

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Peter Chen <[email protected]>

commit 5332ff1fb63c46588656e4208201bc131627c878 upstream.

For otgsc, both enable bits and status bits are in it. So we need to make sure the status bits are not be cleared when write enable bits. It can fix one bug that we plug in/out Micro AB cable fast, and sometimes, the IDIS will be cleared wrongly when handle last ID interrupt (ID 0->1), so the current interrupt will not occur.

For stable tree: 3.12+

---
 drivers/usb/chipidea/otg.h |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/usb/chipidea/otg.h +3 -3
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2013 Freescale Semiconductor, Inc.
+ * Copyright (C) 2013-2014 Freescale Semiconductor, Inc.
  *
  * Author: Peter Chen
  *
@@ -19,12 +19,12 @@ static inline void ci_clear_otg_interrup
 
 static inline void ci_enable_otg_interrupt(struct ci_hdrc *ci, u32 bits)
 {
-	hw_write(ci, OP_OTGSC, bits, bits);
+	hw_write(ci, OP_OTGSC, bits | OTGSC_INT_STATUS_BITS, bits);
 }
 
 static inline void ci_disable_otg_interrupt(struct ci_hdrc *ci, u32 bits)
 {
-	hw_write(ci, OP_OTGSC, bits, 0);
+	hw_write(ci, OP_OTGSC, bits | OTGSC_INT_STATUS_BITS, 0);
 }
 
 int ci_hdrc_otg_init(struct ci_hdrc *ci);
author maintainer
#18 lore raw

PATCH 17/133 usb: chipidea: udc: using MultO at TD as real mult value for ISO-TX

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Peter Chen <[email protected]>

commit 2fc5a7dace3c43e62402ab4e8800a8f1834ffe2a upstream.

We have met a bug that the high bandwidth ISO-TX transfer has failed at the last packet if it is less than 1024, the TD status shows it is "Transaction Error".

The root cause of this problem is: the mult value at qh is not correct for current TD's transfer length. We use TD list to queue un-transfer TDs, and change mult for new adding TDs. If new adding TDs transfer length less than 1024, but the queued un-transfer TDs transfer length is larger than 1024, the transfer error will occur, and vice versa. Usually, this problem occurs at the last packet, and the first packet for new frame.

We fixed this problem by setting Mult at QH as the largest value (3), and set MultO (Multiplier Override) at TD according to every transfer length. It can cover both hardware version less than 2.3 (the real mult is MultO if it is not 0) and 2.3+ (the real mult is min(qh.mult, td.multo)).

Since the MultO bits are only existed at TX TD, we keep the ISO-RX behavior unchanged.

For stable tree: 3.11+.

---
 drivers/usb/chipidea/udc.c |   19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)
drivers/usb/chipidea/udc.c +17 -2
@@ -394,6 +394,14 @@ static int add_td_to_list(struct ci_hw_e
 	node->ptr->token = cpu_to_le32(length << __ffs(TD_TOTAL_BYTES));
 	node->ptr->token &= cpu_to_le32(TD_TOTAL_BYTES);
 	node->ptr->token |= cpu_to_le32(TD_STATUS_ACTIVE);
+	if (hwep->type == USB_ENDPOINT_XFER_ISOC && hwep->dir == TX) {
+		u32 mul = hwreq->req.length / hwep->ep.maxpacket;
+
+		if (hwreq->req.length == 0
+				|| hwreq->req.length % hwep->ep.maxpacket)
+			mul++;
+		node->ptr->token |= mul << __ffs(TD_MULTO);
+	}
 
 	temp = (u32) (hwreq->req.dma + hwreq->req.actual);
 	if (length) {
@@ -516,10 +524,11 @@ static int _hardware_enqueue(struct ci_h
 	hwep->qh.ptr->td.token &=
 		cpu_to_le32(~(TD_STATUS_HALTED|TD_STATUS_ACTIVE));
 
-	if (hwep->type == USB_ENDPOINT_XFER_ISOC) {
+	if (hwep->type == USB_ENDPOINT_XFER_ISOC && hwep->dir == RX) {
 		u32 mul = hwreq->req.length / hwep->ep.maxpacket;
 
-		if (hwreq->req.length % hwep->ep.maxpacket)
+		if (hwreq->req.length == 0
+				|| hwreq->req.length % hwep->ep.maxpacket)
 			mul++;
 		hwep->qh.ptr->cap |= mul << __ffs(QH_MULT);
 	}
@@ -1172,6 +1181,12 @@ static int ep_enable(struct usb_ep *ep,
 	if (hwep->num)
 		cap |= QH_ZLT;
 	cap |= (hwep->ep.maxpacket << __ffs(QH_MAX_PKT)) & QH_MAX_PKT;
+	/*
+	 * For ISO-TX, we set mult at QH as the largest value, and use
+	 * MultO at TD as real mult value.
+	 */
+	if (hwep->type == USB_ENDPOINT_XFER_ISOC && hwep->dir == TX)
+		cap |= 3 << __ffs(QH_MULT);
 
 	hwep->qh.ptr->cap = cpu_to_le32(cap);
 
author maintainer
#19 lore raw

PATCH 19/133 usb: option: add new zte 3g modem pids to option driver

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: =?UTF-8?q?=E5=BC=A0=E5=90=9B?= <[email protected]>

commit 4d90b819ae4c7ea8fd5e2bb7edc68c0f334be2e4 upstream.

---
 drivers/usb/serial/option.c |   11 +++++++++++
 1 file changed, 11 insertions(+)
drivers/usb/serial/option.c +11 -0
@@ -1447,6 +1447,17 @@ static const struct usb_device_id option
 		.driver_info = (kernel_ulong_t)&net_intf3_blacklist },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x0178, 0xff, 0xff, 0xff),
 		.driver_info = (kernel_ulong_t)&net_intf3_blacklist },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xffe9, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff8b, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff8c, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff8d, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff8e, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff8f, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff90, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff91, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff92, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff93, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff94, 0xff, 0xff, 0xff) },
 
 	/* NOTE: most ZTE CDMA devices should be driven by zte_ev, not option */
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_MC2718, 0xff, 0xff, 0xff),
author maintainer
#20 lore raw

PATCH 21/133 USB: cypress_m8: fix ring-indicator detection and reporting

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Johan Hovold <[email protected]>

commit 440ebadeae9298d7de3d4d105342691841ec88d0 upstream.

Fix ring-indicator (RI) status-bit definition, which was defined as CTS, effectively preventing RI-changes from being detected while reporting false RI status.

This bug predates git.

---
 drivers/usb/serial/cypress_m8.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/serial/cypress_m8.h +1 -1
@@ -63,7 +63,7 @@
 #define UART_DSR	0x20	/* data set ready - flow control - device to host */
 #define CONTROL_RTS	0x10	/* request to send - flow control - host to device */
 #define UART_CTS	0x10	/* clear to send - flow control - device to host */
-#define UART_RI		0x10	/* ring indicator - modem - device to host */
+#define UART_RI		0x80	/* ring indicator - modem - device to host */
 #define UART_CD		0x40	/* carrier detect - modem - device to host */
 #define CYP_ERROR	0x08	/* received from input report - device to host */
 /* Note - the below has nothing to do with the "feature report" reset */
author maintainer
#21 lore raw

PATCH 22/133 USB: ftdi_sio: added CS5 quirk for broken smartcard readers

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Colin Leitner <[email protected]>

commit c1f15196ac3b541d084dc80a8fbd8a74c6a0bd44 upstream.

Genuine FTDI chips support only CS7/8. A previous fix in commit 8704211f65a2 ("USB: ftdi_sio: fixed handling of unsupported CSIZE setting") enforced this limitation and reported it back to userspace.

However, certain types of smartcard readers depend on specific driver behaviour that requests 0 data bits (not 5) to change into a different operating mode if CS5 has been set.

This patch reenables this behaviour for all FTDI devices.

Tagged to be added to stable, because it affects a lot of users of embedded systems which rely on these readers to work properly.

---
 drivers/usb/serial/ftdi_sio.c |   17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)
drivers/usb/serial/ftdi_sio.c +15 -2
@@ -2116,10 +2116,20 @@ static void ftdi_set_termios(struct tty_
 	}
 
 	/*
-	 * All FTDI UART chips are limited to CS7/8. We won't pretend to
+	 * All FTDI UART chips are limited to CS7/8. We shouldn't pretend to
 	 * support CS5/6 and revert the CSIZE setting instead.
+	 *
+	 * CS5 however is used to control some smartcard readers which abuse
+	 * this limitation to switch modes. Original FTDI chips fall back to
+	 * eight data bits.
+	 *
+	 * TODO: Implement a quirk to only allow this with mentioned
+	 *       readers. One I know of (Argolis Smartreader V1)
+	 *       returns "USB smartcard server" as iInterface string.
+	 *       The vendor didn't bother with a custom VID/PID of
+	 *       course.
 	 */
-	if ((C_CSIZE(tty) != CS8) && (C_CSIZE(tty) != CS7)) {
+	if (C_CSIZE(tty) == CS6) {
 		dev_warn(ddev, "requested CSIZE setting not supported\n");
 
 		termios->c_cflag &= ~CSIZE;
@@ -2166,6 +2176,9 @@ no_skip:
 		urb_value |= FTDI_SIO_SET_DATA_PARITY_NONE;
 	}
 	switch (cflag & CSIZE) {
+	case CS5:
+		dev_dbg(ddev, "Setting CS5 quirk\n");
+		break;
 	case CS7:
 		urb_value |= 7;
 		dev_dbg(ddev, "Setting CS7\n");
author maintainer
#22 lore raw

PATCH 23/133 USB: Nokia 502 is an unusual device

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Mikhail Zolotaryov <[email protected]>

commit 0e16114f2db4838251fb64f3b550996ad3585890 upstream.

The USB storage operation of Nokia Asha 502 Dual SIM smartphone running Asha Platform 1.1.1 is unreliable in respect of data consistency (i.e. transfered files are corrupted). A similar issue is described here:
http://discussions.nokia.com/t5/Asha-and-other-Nokia-Series-30/Nokia-301-USB-transfers-and-corrupted-files/td-p/1974170

The workaround is (MAX_SECTORS_64):
rmmod usb_storage && modprobe usb_storage quirks=0421:06aa:m

The patch adds the tested device to the unusual list permanently.

---
 drivers/usb/storage/unusual_devs.h |    7 +++++++
 1 file changed, 7 insertions(+)
drivers/usb/storage/unusual_devs.h +7 -0
@@ -234,6 +234,13 @@ UNUSUAL_DEV(  0x0421, 0x0495, 0x0370, 0x
 		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64 ),
 
+/* Patch submitted by Mikhail Zolotaryov <[email protected]> */
+UNUSUAL_DEV(  0x0421, 0x06aa, 0x1110, 0x1110,
+		"Nokia",
+		"502",
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
+		US_FL_MAX_SECTORS_64 ),
+
 #ifdef NO_SDDR09
 UNUSUAL_DEV(  0x0436, 0x0005, 0x0100, 0x0100,
 		"Microtech",
author maintainer
#23 lore raw

PATCH 24/133 usb: xhci: Check for XHCI_PLAT in xhci_cleanup_msix()

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Jack Pham <[email protected]>

commit 9005355af23856c55a5538c9024355785424821b upstream.

If CONFIG_PCI is enabled, make sure xhci_cleanup_msix() doesn't try to free a bogus PCI IRQ or dereference an invalid pci_dev when the xHCI device is actually a platform_device.

This patch should be backported to kernels as old as 3.9, that contain the commit 52fb61250a7a132b0cfb9f4a1060a1f3c49e5a25 "xhci-plat: Don't enable legacy PCI interrupts."

---
 drivers/usb/host/xhci.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/usb/host/xhci.c +3 -0
@@ -321,6 +321,9 @@ static void xhci_cleanup_msix(struct xhc
 	struct usb_hcd *hcd = xhci_to_hcd(xhci);
 	struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
 
+	if (xhci->quirks & XHCI_PLAT)
+		return;
+
 	xhci_free_irq(xhci);
 
 	if (xhci->msix_entries) {
author maintainer
#24 lore raw

PATCH 25/133 USB: fix race between hub_disconnect and recursively_mark_NOTATTACHED

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Alan Stern <[email protected]>

commit 543d7784b07ffd16cc82a9cb4e1e0323fd0040f1 upstream.

There is a race in the hub driver between hub_disconnect() and recursively_mark_NOTATTACHED(). This race can be triggered if the driver is unbound from a device at the same time as the bus's root hub is removed. When the race occurs, it can cause an oops:

BUG: unable to handle kernel NULL pointer dereference at 0000015c
IP: [<c16d5fb0>] recursively_mark_NOTATTACHED+0x20/0x60
Call Trace:
 [<c16d5fc4>] recursively_mark_NOTATTACHED+0x34/0x60
 [<c16d5fc4>] recursively_mark_NOTATTACHED+0x34/0x60
 [<c16d5fc4>] recursively_mark_NOTATTACHED+0x34/0x60
 [<c16d5fc4>] recursively_mark_NOTATTACHED+0x34/0x60
 [<c16d6082>] usb_set_device_state+0x92/0x120
 [<c16d862b>] usb_disconnect+0x2b/0x1a0
 [<c16dd4c0>] usb_remove_hcd+0xb0/0x160
 [<c19ca846>] ? _raw_spin_unlock_irqrestore+0x26/0x50
 [<c1704efc>] ehci_mid_remove+0x1c/0x30
 [<c1704f26>] ehci_mid_stop_host+0x16/0x30
 [<c16f7698>] penwell_otg_work+0xd28/0x3520
 [<c19c945b>] ? __schedule+0x39b/0x7f0
 [<c19cdb9d>] ? sub_preempt_count+0x3d/0x50
 [<c125e97d>] process_one_work+0x11d/0x3d0
 [<c19c7f4d>] ? mutex_unlock+0xd/0x10
 [<c125e0e5>] ? manage_workers.isra.24+0x1b5/0x270
 [<c125f009>] worker_thread+0xf9/0x320
 [<c19ca846>] ? _raw_spin_unlock_irqrestore+0x26/0x50
 [<c125ef10>] ? rescuer_thread+0x2b0/0x2b0
 [<c1264ac4>] kthread+0x94/0xa0
 [<c19d0f77>] ret_from_kernel_thread+0x1b/0x28
 [<c1264a30>] ? kthread_create_on_node+0xc0/0xc0

One problem is that recursively_mark_NOTATTACHED() uses the intfdata value and hub->hdev->maxchild while hub_disconnect() is clearing them. Another problem is that it uses hub->ports[i] while the port device is being released.

To fix this race, we need to hold the device_state_lock while hub_disconnect() changes the values. (Note that usb_disconnect() and hub_port_connect_change() already acquire this lock at similar critical times during a USB device's life cycle.) We also need to remove the port devices after maxchild has been set to 0, instead of before.

---
 drivers/usb/core/hub.c |   14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)
drivers/usb/core/hub.c +9 -5
@@ -1610,7 +1610,7 @@ static void hub_disconnect(struct usb_in
 {
 	struct usb_hub *hub = usb_get_intfdata(intf);
 	struct usb_device *hdev = interface_to_usbdev(intf);
-	int i;
+	int port1;
 
 	/* Take the hub off the event list and don't let it be added again */
 	spin_lock_irq(&hub_event_lock);
@@ -1625,11 +1625,15 @@ static void hub_disconnect(struct usb_in
 	hub->error = 0;
 	hub_quiesce(hub, HUB_DISCONNECT);
 
-	usb_set_intfdata (intf, NULL);
+	/* Avoid races with recursively_mark_NOTATTACHED() */
+	spin_lock_irq(&device_state_lock);
+	port1 = hdev->maxchild;
+	hdev->maxchild = 0;
+	usb_set_intfdata(intf, NULL);
+	spin_unlock_irq(&device_state_lock);
 
-	for (i = 0; i < hdev->maxchild; i++)
-		usb_hub_remove_port_device(hub, i + 1);
-	hub->hdev->maxchild = 0;
+	for (; port1 > 0; --port1)
+		usb_hub_remove_port_device(hub, port1);
 
 	if (hub->hdev->speed == USB_SPEED_HIGH)
 		highspeed_hubs--;
author maintainer
#25 lore raw

PATCH 26/133 usb: ehci: add freescale imx28 special write register method

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Peter Chen <[email protected]>

commit feffe09f510c475df082546815f9e4a573f6a233 upstream.

According to Freescale imx28 Errata, "ENGR119653 USB: ARM to USB register error issue", All USB register write operations must use the ARM SWP instruction. So, we implement a special ehci_write for imx28.

Discussion for it at below:
http://marc.info/?l=linux-usb&m=137996395529294&w=2

Without this patcheset, imx28 works unstable at high AHB bus loading. If the bus loading is not high, the imx28 usb can work well at the most of time. There is a IC errata for this problem, usually, we consider IC errata is a problem not a new feature, and this workaround is needed for that, so we need to add them to stable tree 3.11+.

---
 drivers/usb/host/ehci.h |   18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)
drivers/usb/host/ehci.h +17 -1
@@ -203,6 +203,7 @@ struct ehci_hcd {			/* one per controlle
 	unsigned		has_synopsys_hc_bug:1; /* Synopsys HC */
 	unsigned		frame_index_bug:1; /* MosChip (AKA NetMos) */
 	unsigned		need_oc_pp_cycle:1; /* MPC834X port power */
+	unsigned		imx28_write_fix:1; /* For Freescale i.MX28 */
 
 	/* required for usb32 quirk */
 	#define OHCI_CTRL_HCFS          (3 << 6)
@@ -679,6 +680,18 @@ static inline unsigned int ehci_readl(co
 #endif
 }
 
+#ifdef CONFIG_SOC_IMX28
+static inline void imx28_ehci_writel(const unsigned int val,
+		volatile __u32 __iomem *addr)
+{
+	__asm__ ("swp %0, %0, [%1]" : : "r"(val), "r"(addr));
+}
+#else
+static inline void imx28_ehci_writel(const unsigned int val,
+		volatile __u32 __iomem *addr)
+{
+}
+#endif
 static inline void ehci_writel(const struct ehci_hcd *ehci,
 		const unsigned int val, __u32 __iomem *regs)
 {
@@ -687,7 +700,10 @@ static inline void ehci_writel(const str
 		writel_be(val, regs) :
 		writel(val, regs);
 #else
-	writel(val, regs);
+	if (ehci->imx28_write_fix)
+		imx28_ehci_writel(val, regs);
+	else
+		writel(val, regs);
 #endif
 }
 
author maintainer
#26 lore raw

PATCH 27/133 rtlwifi: rtl8192cu: Add new device ID

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit f87f960b2fb802f26ee3b00c19320e57a9c583ff upstream.

---
 drivers/net/wireless/rtlwifi/rtl8192cu/sw.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/wireless/rtlwifi/rtl8192cu/sw.c +1 -0
@@ -306,6 +306,7 @@ static struct usb_device_id rtl8192c_usb
 	{RTL_USB_DEVICE(0x0bda, 0x5088, rtl92cu_hal_cfg)}, /*Thinkware-CC&C*/
 	{RTL_USB_DEVICE(0x0df6, 0x0052, rtl92cu_hal_cfg)}, /*Sitecom - Edimax*/
 	{RTL_USB_DEVICE(0x0df6, 0x005c, rtl92cu_hal_cfg)}, /*Sitecom - Edimax*/
+	{RTL_USB_DEVICE(0x0df6, 0x0077, rtl92cu_hal_cfg)}, /*Sitecom-WLA2100V2*/
 	{RTL_USB_DEVICE(0x0eb0, 0x9071, rtl92cu_hal_cfg)}, /*NO Brand - Etop*/
 	{RTL_USB_DEVICE(0x4856, 0x0091, rtl92cu_hal_cfg)}, /*NetweeN - Feixun*/
 	/* HP - Lite-On ,8188CUS Slim Combo */
author maintainer
#27 lore raw

PATCH 28/133 rtlwifi: Update beacon statistics for USB driver

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 65b9cc97c6852fae19dc5c7745e9abc8dd380aad upstream.

The USB drivers were not updating the beacon statistics, which led to false beacon loss indications.

---
 drivers/net/wireless/rtlwifi/usb.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/net/wireless/rtlwifi/usb.c +2 -0
@@ -484,6 +484,8 @@ static void _rtl_usb_rx_process_agg(stru
 			if (unicast)
 				rtlpriv->link_info.num_rx_inperiod++;
 		}
+		/* static bcn for roaming */
+		rtl_beacon_statistic(hw, skb);
 	}
 }
 
author maintainer
#28 lore raw

PATCH 29/133 rtlwifi: rtl8192c: Prevent reconnect attempts if not connected

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 8fd77aec1a9d6f4328fc0244f21932114e066df3 upstream.

This driver has a watchdog timer that attempts to reconnect when beacon frames are not seen for 6 seconds. This patch disables that reconnect whenever the device has never been connected.

---
 drivers/net/wireless/rtlwifi/base.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/wireless/rtlwifi/base.c +2 -1
@@ -1437,7 +1437,8 @@ void rtl_watchdog_wq_callback(void *data
 			/* if we can't recv beacon for 6s, we should
 			 * reconnect this AP
 			 */
-			if (rtlpriv->link_info.roam_times >= 3) {
+			if ((rtlpriv->link_info.roam_times >= 3) &&
+			    !is_zero_ether_addr(rtlpriv->mac80211.bssid)) {
 				RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG,
 					 "AP off, try to reconnect now\n");
 				rtlpriv->link_info.roam_times = 0;
author maintainer
#29 lore raw

PATCH 30/133 rtlwifi: rtl8192cu: Add new firmware

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 62009b7f12793c932aaba0df946b04cb4a77d022 upstream.

Vendor driver rtl8188C_8192C_8192D_usb_linux_v3.4.2_3727.20120404 introduced new firmware for these chips. The code try for the new file, and fall back to the original firmware if the new file is not available.

---
 drivers/net/wireless/rtlwifi/core.c         |   10 ++++++++++
 drivers/net/wireless/rtlwifi/rtl8192cu/sw.c |   16 +++++++++++++---
 drivers/net/wireless/rtlwifi/wifi.h         |    1 +
 3 files changed, 24 insertions(+), 3 deletions(-)
drivers/net/wireless/rtlwifi/core.c +10 -0
@@ -46,10 +46,20 @@ void rtl_fw_cb(const struct firmware *fi
 			 "Firmware callback routine entered!\n");
 	complete(&rtlpriv->firmware_loading_complete);
 	if (!firmware) {
+		if (rtlpriv->cfg->alt_fw_name) {
+			err = request_firmware(&firmware,
+					       rtlpriv->cfg->alt_fw_name,
+					       rtlpriv->io.dev);
+			pr_info("Loading alternative firmware %s\n",
+				rtlpriv->cfg->alt_fw_name);
+			if (!err)
+				goto found_alt;
+		}
 		pr_err("Firmware %s not available\n", rtlpriv->cfg->fw_name);
 		rtlpriv->max_fw_size = 0;
 		return;
 	}
+found_alt:
 	if (firmware->size > rtlpriv->max_fw_size) {
 		RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG,
 			 "Firmware is too big!\n");
drivers/net/wireless/rtlwifi/rtl8192cu/sw.c +13 -3
@@ -49,6 +49,9 @@ MODULE_AUTHOR("Larry Finger	<Larry.Finge
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Realtek 8192C/8188C 802.11n USB wireless");
 MODULE_FIRMWARE("rtlwifi/rtl8192cufw.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8192cufw_A.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8192cufw_B.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8192cufw_TMSC.bin");
 
 static int rtl92cu_init_sw_vars(struct ieee80211_hw *hw)
 {
@@ -68,14 +71,21 @@ static int rtl92cu_init_sw_vars(struct i
 			 "Can't alloc buffer for fw\n");
 		return 1;
 	}
-
+	if (IS_VENDOR_UMC_A_CUT(rtlpriv->rtlhal.version) &&
+	    !IS_92C_SERIAL(rtlpriv->rtlhal.version)) {
+		rtlpriv->cfg->fw_name = "rtlwifi/rtl8192cufw_A.bin";
+	} else if (IS_81xxC_VENDOR_UMC_B_CUT(rtlpriv->rtlhal.version)) {
+		rtlpriv->cfg->fw_name = "rtlwifi/rtl8192cufw_B.bin";
+	} else {
+		rtlpriv->cfg->fw_name = "rtlwifi/rtl8192cufw_TMSC.bin";
+	}
+	/* provide name of alternative file */
+	rtlpriv->cfg->alt_fw_name = "rtlwifi/rtl8192cufw.bin";
 	pr_info("Loading firmware %s\n", rtlpriv->cfg->fw_name);
 	rtlpriv->max_fw_size = 0x4000;
 	err = request_firmware_nowait(THIS_MODULE, 1,
 				      rtlpriv->cfg->fw_name, rtlpriv->io.dev,
 				      GFP_KERNEL, hw, rtl_fw_cb);
-
-
 	return err;
 }
 
drivers/net/wireless/rtlwifi/wifi.h +1 -0
@@ -1806,6 +1806,7 @@ struct rtl_hal_cfg {
 	bool write_readback;
 	char *name;
 	char *fw_name;
+	char *alt_fw_name;
 	struct rtl_hal_ops *ops;
 	struct rtl_mod_params *mod_params;
 	struct rtl_hal_usbint_cfg *usb_interface_cfg;
author maintainer
#30 lore raw

PATCH 31/133 rtlwifi: Redo register save locations

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit b9a758a8c905fc59e783ae91ad645452d877ea88 upstream.

The initial USB driver did not use some register save locations in the private data storage. To save some memory, a union was used to overlay these variables with USB I/O components. In an update of the gain-control code, these register save locations are now needed for USB drivers.

---
 drivers/net/wireless/rtlwifi/wifi.h |   32 ++++++++++++++++----------------
 1 file changed, 16 insertions(+), 16 deletions(-)
drivers/net/wireless/rtlwifi/wifi.h +16 -16
@@ -1035,6 +1035,7 @@ struct rtl_ht_agg {
 
 struct rssi_sta {
 	long undec_sm_pwdb;
+	long undec_sm_cck;
 };
 
 struct rtl_tid_data {
@@ -1325,8 +1326,10 @@ struct fast_ant_training {
 struct rtl_dm {
 	/*PHY status for Dynamic Management */
 	long entry_min_undec_sm_pwdb;
+	long undec_sm_cck;
 	long undec_sm_pwdb;	/*out dm */
 	long entry_max_undec_sm_pwdb;
+	s32 ofdm_pkt_cnt;
 	bool dm_initialgain_enable;
 	bool dynamic_txpower_enable;
 	bool current_turbo_edca;
@@ -1341,6 +1344,7 @@ struct rtl_dm {
 	bool inform_fw_driverctrldm;
 	bool current_mrc_switch;
 	u8 txpowercount;
+	u8 powerindex_backup[6];
 
 	u8 thermalvalue_rxgain;
 	u8 thermalvalue_iqk;
@@ -1352,7 +1356,9 @@ struct rtl_dm {
 	bool done_txpower;
 	u8 dynamic_txhighpower_lvl;	/*Tx high power level */
 	u8 dm_flag;		/*Indicate each dynamic mechanism's status. */
+	u8 dm_flag_tmp;
 	u8 dm_type;
+	u8 dm_rssi_sel;
 	u8 txpower_track_control;
 	bool interrupt_migration;
 	bool disable_tx_int;
@@ -1951,6 +1957,7 @@ struct dig_t {
 	u8 pre_ccastate;
 	u8 cur_ccasate;
 	u8 large_fa_hit;
+	u8 dig_dynamic_min;
 	u8 forbidden_igi;
 	u8 dig_state;
 	u8 dig_highpwrstate;
@@ -2031,22 +2038,15 @@ struct rtl_priv {
 	struct dig_t dm_digtable;
 	struct ps_t dm_pstable;
 
-	/* section shared by individual drivers */
-	union {
-		struct {	/* data buffer pointer for USB reads */
-			__le32 *usb_data;
-			int usb_data_index;
-			bool initialized;
-		};
-		struct {	/* section for 8723ae */
-			bool reg_init;	/* true if regs saved */
-			u32 reg_874;
-			u32 reg_c70;
-			u32 reg_85c;
-			u32 reg_a74;
-			bool bt_operation_on;
-		};
-	};
+	u32 reg_874;
+	u32 reg_c70;
+	u32 reg_85c;
+	u32 reg_a74;
+	bool reg_init;	/* true if regs saved */
+	bool bt_operation_on;
+	__le32 *usb_data;
+	int usb_data_index;
+	bool initialized;
 	bool enter_ps;	/* true when entering PS */
 	u8 rate_mask[5];
 
author maintainer
#31 lore raw

PATCH 32/133 rtlwifi: Set the link state

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 619ce76f8bb850b57032501a39f26aa6c6731c70 upstream.

The present code fails to set the linked state when an interface is added.

---
 drivers/net/wireless/rtlwifi/core.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/wireless/rtlwifi/core.c +1 -0
@@ -194,6 +194,7 @@ static int rtl_op_add_interface(struct i
 					rtlpriv->cfg->maps
 					[RTL_IBSS_INT_MASKS]);
 		}
+		mac->link_state = MAC80211_LINKED;
 		break;
 	case NL80211_IFTYPE_ADHOC:
 		RT_TRACE(rtlpriv, COMP_MAC80211, DBG_LOUD,
author maintainer
#32 lore raw

PATCH 33/133 rtlwifi: rtl8192c: Add new definitions in the dm_common header

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit c908c74e005de780fddbe8cb6fcd44803f5d4b74 upstream.

Changes in the gain-control mechanism will require some changes in the header.

---
 drivers/net/wireless/rtlwifi/rtl8192c/dm_common.h |   14 ++++++++++++++
 1 file changed, 14 insertions(+)
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.h +14 -0
@@ -91,6 +91,17 @@
 #define TX_POWER_NEAR_FIELD_THRESH_LVL2		74
 #define TX_POWER_NEAR_FIELD_THRESH_LVL1		67
 
+#define DYNAMIC_FUNC_DISABLE			0x0
+#define DYNAMIC_FUNC_DIG			BIT(0)
+#define DYNAMIC_FUNC_HP				BIT(1)
+#define DYNAMIC_FUNC_SS				BIT(2) /*Tx Power Tracking*/
+#define DYNAMIC_FUNC_BT				BIT(3)
+#define DYNAMIC_FUNC_ANT_DIV			BIT(4)
+
+#define	RSSI_CCK				0
+#define	RSSI_OFDM				1
+#define	RSSI_DEFAULT				2
+
 struct swat_t {
 	u8 failure_cnt;
 	u8 try_flag;
@@ -167,5 +178,8 @@ void rtl92c_phy_lc_calibrate(struct ieee
 void rtl92c_phy_iq_calibrate(struct ieee80211_hw *hw, bool recovery);
 void rtl92c_dm_dynamic_txpower(struct ieee80211_hw *hw);
 void rtl92c_dm_bt_coexist(struct ieee80211_hw *hw);
+void dm_savepowerindex(struct ieee80211_hw *hw);
+void dm_writepowerindex(struct ieee80211_hw *hw, u8 value);
+void dm_restorepowerindex(struct ieee80211_hw *hw);
 
 #endif
author maintainer
#33 lore raw

PATCH 34/133 rtlwifi: Increase the RX queue length for USB drivers

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit dc6405712268fe514d3dd89aa936c4397b0871b9 upstream.

The current number of RX buffers queued is 32, which is too small under heavy load. That number is doubled.

---
 drivers/net/wireless/rtlwifi/usb.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/rtlwifi/usb.c +1 -1
@@ -557,7 +557,7 @@ static void _rtl_rx_pre_process(struct i
 	}
 }
 
-#define __RX_SKB_MAX_QUEUED	32
+#define __RX_SKB_MAX_QUEUED	64
 
 static void _rtl_rx_work(unsigned long param)
 {
author maintainer
#34 lore raw

PATCH 35/133 rtlwifi: rtl8192c: Add routines to save/restore power index registers

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 97204e93f01868eeba6ae5c4f3270f32905bb418 upstream.

---
 drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c |   36 ++++++++++++++++++++++
 1 file changed, 36 insertions(+)
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c +36 -0
@@ -158,6 +158,42 @@ static const u8 cckswing_table_ch14[CCK_
 	{0x09, 0x08, 0x07, 0x04, 0x00, 0x00, 0x00, 0x00}
 };
 
+static u32 power_index_reg[6] = {0xc90, 0xc91, 0xc92, 0xc98, 0xc99, 0xc9a};
+
+void dm_restorepowerindex(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u8	index;
+
+	for (index = 0; index < 6; index++)
+		rtl_write_byte(rtlpriv, power_index_reg[index],
+			       rtlpriv->dm.powerindex_backup[index]);
+}
+EXPORT_SYMBOL_GPL(dm_restorepowerindex);
+
+void dm_writepowerindex(struct ieee80211_hw *hw, u8 value)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u8 index;
+
+	for (index = 0; index < 6; index++)
+		rtl_write_byte(rtlpriv, power_index_reg[index], value);
+}
+EXPORT_SYMBOL_GPL(dm_writepowerindex);
+
+void dm_savepowerindex(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u8 index;
+	u8 tmp;
+
+	for (index = 0; index < 6; index++) {
+		tmp = rtl_read_byte(rtlpriv, power_index_reg[index]);
+		rtlpriv->dm.powerindex_backup[index] = tmp;
+	}
+}
+EXPORT_SYMBOL_GPL(dm_savepowerindex);
+
 static void rtl92c_dm_diginit(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
author maintainer
#35 lore raw

PATCH 36/133 rtlwifi: rtl8192cu: Update the power index registers

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 9806eacf5de27ab01d680c5d75c92a3a89734e4f upstream.

This patch uses the newly introduced power index register routines.

---
 drivers/net/wireless/rtlwifi/rtl8192cu/dm.c |    9 +++++++++
 drivers/net/wireless/rtlwifi/rtl8192cu/dm.h |    3 +++
 2 files changed, 12 insertions(+)
drivers/net/wireless/rtlwifi/rtl8192cu/dm.c +9 -0
@@ -101,6 +101,15 @@ void rtl92cu_dm_dynamic_txpower(struct i
 			 "PHY_SetTxPowerLevel8192S() Channel = %d\n",
 			 rtlphy->current_channel);
 		rtl92c_phy_set_txpower_level(hw, rtlphy->current_channel);
+		if (rtlpriv->dm.dynamic_txhighpower_lvl ==
+		    TXHIGHPWRLEVEL_NORMAL)
+			dm_restorepowerindex(hw);
+		else if (rtlpriv->dm.dynamic_txhighpower_lvl ==
+			 TXHIGHPWRLEVEL_LEVEL1)
+			dm_writepowerindex(hw, 0x14);
+		else if (rtlpriv->dm.dynamic_txhighpower_lvl ==
+			 TXHIGHPWRLEVEL_LEVEL2)
+			dm_writepowerindex(hw, 0x10);
 	}
 
 	rtlpriv->dm.last_dtp_lvl = rtlpriv->dm.dynamic_txhighpower_lvl;
drivers/net/wireless/rtlwifi/rtl8192cu/dm.h +3 -0
@@ -30,3 +30,6 @@
 #include "../rtl8192ce/dm.h"
 
 void rtl92cu_dm_dynamic_txpower(struct ieee80211_hw *hw);
+void dm_savepowerindex(struct ieee80211_hw *hw);
+void dm_writepowerindex(struct ieee80211_hw *hw, u8 value);
+void dm_restorepowerindex(struct ieee80211_hw *hw);
author maintainer
#36 lore raw

PATCH 37/133 rtlwifi: rtl8192cu: Fix some code in RF handling

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit e9b0784bb9de3152e787ee779868c626b137fb3b upstream.

---
 drivers/net/wireless/rtlwifi/rtl8192cu/rf.c |   29 +++++++++++++++-------------
 1 file changed, 16 insertions(+), 13 deletions(-)
drivers/net/wireless/rtlwifi/rtl8192cu/rf.c +16 -13
@@ -85,17 +85,15 @@ void rtl92cu_phy_rf6052_set_cck_txpower(
 	if (mac->act_scanning) {
 		tx_agc[RF90_PATH_A] = 0x3f3f3f3f;
 		tx_agc[RF90_PATH_B] = 0x3f3f3f3f;
-		if (turbo_scanoff) {
-			for (idx1 = RF90_PATH_A; idx1 <= RF90_PATH_B; idx1++) {
-				tx_agc[idx1] = ppowerlevel[idx1] |
-				    (ppowerlevel[idx1] << 8) |
-				    (ppowerlevel[idx1] << 16) |
-				    (ppowerlevel[idx1] << 24);
-				if (rtlhal->interface == INTF_USB) {
-					if (tx_agc[idx1] > 0x20 &&
-					    rtlefuse->external_pa)
-						tx_agc[idx1] = 0x20;
-				}
+		for (idx1 = RF90_PATH_A; idx1 <= RF90_PATH_B; idx1++) {
+			tx_agc[idx1] = ppowerlevel[idx1] |
+			    (ppowerlevel[idx1] << 8) |
+			    (ppowerlevel[idx1] << 16) |
+			    (ppowerlevel[idx1] << 24);
+			if (rtlhal->interface == INTF_USB) {
+				if (tx_agc[idx1] > 0x20 &&
+				    rtlefuse->external_pa)
+					tx_agc[idx1] = 0x20;
 			}
 		}
 	} else {
@@ -107,7 +105,7 @@ void rtl92cu_phy_rf6052_set_cck_txpower(
 			   TXHIGHPWRLEVEL_LEVEL2) {
 			tx_agc[RF90_PATH_A] = 0x00000000;
 			tx_agc[RF90_PATH_B] = 0x00000000;
-		} else{
+		} else {
 			for (idx1 = RF90_PATH_A; idx1 <= RF90_PATH_B; idx1++) {
 				tx_agc[idx1] = ppowerlevel[idx1] |
 				    (ppowerlevel[idx1] << 8) |
@@ -373,7 +371,12 @@ static void _rtl92c_write_ofdm_power_reg
 			    regoffset == RTXAGC_B_MCS07_MCS04)
 				regoffset = 0xc98;
 			for (i = 0; i < 3; i++) {
-				writeVal = (writeVal > 6) ? (writeVal - 6) : 0;
+				if (i != 2)
+					writeVal = (writeVal > 8) ?
+						   (writeVal - 8) : 0;
+				else
+					writeVal = (writeVal > 6) ?
+						   (writeVal - 6) : 0;
 				rtl_write_byte(rtlpriv, (u32)(regoffset + i),
 					      (u8)writeVal);
 			}
author maintainer
#37 lore raw

PATCH 38/133 rtlwifi: Add missing code to PWDB statics routine

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit d82403a9f407217b6aed5260aa92a120e8e98310 upstream.

---
 drivers/net/wireless/rtlwifi/stats.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)
drivers/net/wireless/rtlwifi/stats.c +14 -0
@@ -176,6 +176,7 @@ static void rtl_process_pwdb(struct ieee
 	struct rtl_sta_info *drv_priv = NULL;
 	struct ieee80211_sta *sta = NULL;
 	long undec_sm_pwdb;
+	long undec_sm_cck;
 
 	rcu_read_lock();
 	if (rtlpriv->mac80211.opmode != NL80211_IFTYPE_STATION)
@@ -185,12 +186,16 @@ static void rtl_process_pwdb(struct ieee
 	if (sta) {
 		drv_priv = (struct rtl_sta_info *) sta->drv_priv;
 		undec_sm_pwdb = drv_priv->rssi_stat.undec_sm_pwdb;
+		undec_sm_cck = drv_priv->rssi_stat.undec_sm_cck;
 	} else {
 		undec_sm_pwdb = rtlpriv->dm.undec_sm_pwdb;
+		undec_sm_cck = rtlpriv->dm.undec_sm_cck;
 	}
 
 	if (undec_sm_pwdb < 0)
 		undec_sm_pwdb = pstatus->rx_pwdb_all;
+	if (undec_sm_cck < 0)
+		undec_sm_cck = pstatus->rx_pwdb_all;
 	if (pstatus->rx_pwdb_all > (u32) undec_sm_pwdb) {
 		undec_sm_pwdb = (((undec_sm_pwdb) *
 		      (RX_SMOOTH_FACTOR - 1)) +
@@ -200,6 +205,15 @@ static void rtl_process_pwdb(struct ieee
 		undec_sm_pwdb = (((undec_sm_pwdb) * (RX_SMOOTH_FACTOR - 1)) +
 		     (pstatus->rx_pwdb_all)) / (RX_SMOOTH_FACTOR);
 	}
+	if (pstatus->rx_pwdb_all > (u32) undec_sm_cck) {
+		undec_sm_cck = (((undec_sm_pwdb) *
+		      (RX_SMOOTH_FACTOR - 1)) +
+		     (pstatus->rx_pwdb_all)) / (RX_SMOOTH_FACTOR);
+		undec_sm_cck = undec_sm_cck + 1;
+	} else {
+		undec_sm_pwdb = (((undec_sm_cck) * (RX_SMOOTH_FACTOR - 1)) +
+		     (pstatus->rx_pwdb_all)) / (RX_SMOOTH_FACTOR);
+	}
 
 	if (sta) {
 		drv_priv->rssi_stat.undec_sm_pwdb = undec_sm_pwdb;
author maintainer
#38 lore raw

PATCH 39/133 rtlwifi: rtl8188ee: Fix typo in code

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit f699273d6a624266ebc9198774f06ee64a3847a1 upstream.

The static analyser "cppcheck" shows the following typo:
drivers/net/wireless/rtlwifi/rtl8188ee/dm.c:1081]: (style) Same expression on both sides of '!='.

---
 drivers/net/wireless/rtlwifi/rtl8188ee/dm.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/rtlwifi/rtl8188ee/dm.c +1 -1
@@ -1078,7 +1078,7 @@ static void rtl88e_dm_txpower_tracking_c
 				rtldm->swing_flag_ofdm = true;
 			}
 
-			if (rtldm->swing_idx_cck != rtldm->swing_idx_cck) {
+			if (rtldm->swing_idx_cck_cur != rtldm->swing_idx_cck) {
 				rtldm->swing_idx_cck_cur = rtldm->swing_idx_cck;
 				rtldm->swing_flag_cck = true;
 			}
author maintainer
#39 lore raw

PATCH 40/133 mwifiex: add missing endian conversion for fw_tsf

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Amitkumar Karwar <[email protected]>

commit 9795229752c31da0c5f8a7dc4c827665327b52f9 upstream.

It is u64 data received from firmware. Little endian to cpu conversion is required here.

---
 drivers/net/wireless/mwifiex/scan.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/wireless/mwifiex/scan.c +2 -2
@@ -1681,7 +1681,7 @@ int mwifiex_ret_802_11_scan(struct mwifi
 		const u8 *ie_buf;
 		size_t ie_len;
 		u16 channel = 0;
-		u64 fw_tsf = 0;
+		__le64 fw_tsf = 0;
 		u16 beacon_size = 0;
 		u32 curr_bcn_bytes;
 		u32 freq;
@@ -1815,7 +1815,7 @@ int mwifiex_ret_802_11_scan(struct mwifi
 					      ie_buf, ie_len, rssi, GFP_KERNEL);
 				bss_priv = (struct mwifiex_bss_priv *)bss->priv;
 				bss_priv->band = band;
-				bss_priv->fw_tsf = fw_tsf;
+				bss_priv->fw_tsf = le64_to_cpu(fw_tsf);
 				if (priv->media_connected &&
 				    !memcmp(bssid,
 					    priv->curr_bss_params.bss_descriptor
author maintainer
#40 lore raw

PATCH 41/133 mwifiex: fix wrong 11ac bits setting in fw_cap_info

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Bing Zhao <[email protected]>

commit 1e202242ee1432d68a8bea4919b2ae0ef19d9e06 upstream.

bit 14 is actually reserved and bit 12 & 13 should be used for 11ac capability in fw_cap_info.

---
 drivers/net/wireless/mwifiex/fw.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mwifiex/fw.h +1 -1
@@ -226,7 +226,7 @@ enum MWIFIEX_802_11_PRIVACY_FILTER {
 
 /* HW_SPEC fw_cap_info */
 
-#define ISSUPP_11ACENABLED(fw_cap_info) (fw_cap_info & (BIT(13)|BIT(14)))
+#define ISSUPP_11ACENABLED(fw_cap_info) (fw_cap_info & (BIT(12)|BIT(13)))
 
 #define GET_VHTCAP_CHWDSET(vht_cap_info)    ((vht_cap_info >> 2) & 0x3)
 #define GET_VHTNSSMCS(mcs_mapset, nss) ((mcs_mapset >> (2 * (nss - 1))) & 0x3)
author maintainer
#41 lore raw

PATCH 42/133 iwlwifi: pcie: enable oscillator for L1 exit

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Emmanuel Grumbach <[email protected]>

commit 2d93aee152b1758a94a18fe15d72153ba73b5679 upstream.

Enabling the oscillator consumes slightly more power (100uA) but allows to make sure that we exit from L1 on time.

Not doing so might lead to a PCIe specification violation since we might wake up from L1 at the wrong time. This issue has been identified on 3160 and 7260 only. On older NICs L1 off is not enabled, on newer NICs (7265), the issue is fixed.

When the bug occurs the user sees that the NIC has disappeared from the PCI bridge, any access to the device returns 0xff.

This fixes:

	https://bugzilla.kernel.org/show_bug.cgi?id=64541

and has been extensively discussed here:

	http://markmail.org/thread/mfmpzqt3r333n4bo
Fixes 99cd47142399 ("iwlwifi: add 7000 series device configuration")
Reported-and-tested-by wzyboy <[email protected]>
---
 drivers/net/wireless/iwlwifi/iwl-prph.h   |    4 ++++
 drivers/net/wireless/iwlwifi/pcie/trans.c |   22 ++++++++++++++++++++++
 2 files changed, 26 insertions(+)
drivers/net/wireless/iwlwifi/iwl-prph.h +4 -0
@@ -272,4 +272,8 @@ static inline unsigned int SCD_QUEUE_STA
 
 /*********************** END TX SCHEDULER *************************************/
 
+/* Oscillator clock */
+#define OSC_CLK				(0xa04068)
+#define OSC_CLK_FORCE_CONTROL		(0x8)
+
 #endif				/* __iwl_prph_h__ */
drivers/net/wireless/iwlwifi/pcie/trans.c +22 -0
@@ -206,6 +206,28 @@ static int iwl_pcie_apm_init(struct iwl_
 		goto out;
 	}
 
+	if (trans->cfg->host_interrupt_operation_mode) {
+		/*
+		 * This is a bit of an abuse - This is needed for 7260 / 3160
+		 * only check host_interrupt_operation_mode even if this is
+		 * not related to host_interrupt_operation_mode.
+		 *
+		 * Enable the oscillator to count wake up time for L1 exit. This
+		 * consumes slightly more power (100uA) - but allows to be sure
+		 * that we wake up from L1 on time.
+		 *
+		 * This looks weird: read twice the same register, discard the
+		 * value, set a bit, and yet again, read that same register
+		 * just to discard the value. But that's the way the hardware
+		 * seems to like it.
+		 */
+		iwl_read_prph(trans, OSC_CLK);
+		iwl_read_prph(trans, OSC_CLK);
+		iwl_set_bits_prph(trans, OSC_CLK, OSC_CLK_FORCE_CONTROL);
+		iwl_read_prph(trans, OSC_CLK);
+		iwl_read_prph(trans, OSC_CLK);
+	}
+
 	/*
 	 * Enable DMA clock and wait for it to stabilize.
 	 *
author maintainer
#42 lore raw

PATCH 43/133 iwlwifi: mvm: fix missing cleanup in .start() error path

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Eliad Peller <[email protected]>

commit 91b0d1198417cf4fd9a7bd4138b6909f0b359099 upstream.

Cleanup of iwl_mvm_leds was missing in case of error, resulting in the following warning:

WARNING: at lib/kobject.c:196 kobject_add_internal+0x1f4/0x210()

kobject_add_internal failed for phy0-led with -EEXIST, don't try to register things with the same name in the same directory.

which prevents further reloads of the driver.

---
 drivers/net/wireless/iwlwifi/mvm/ops.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/wireless/iwlwifi/mvm/ops.c +1 -0
@@ -446,6 +446,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *
 
  out_unregister:
 	ieee80211_unregister_hw(mvm->hw);
+	iwl_mvm_leds_exit(mvm);
  out_free:
 	iwl_phy_db_free(mvm->phy_db);
 	kfree(mvm->scan_cmd);
author maintainer
#43 lore raw

PATCH 44/133 b43: Fix lockdep splat

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 09164043f63c947a49797750a09ca1cd7c31108e upstream.

In https://bugzilla.kernel.org/show_bug.cgi?id=67561, a locking dependency is reported when b43 is used with hostapd, and rfkill is used to kill the radio output.

The lockdep splat (in part) is as follows:

======================================================
[ INFO: possible circular locking dependency detected ] 3.12.0 #1 Not tainted
-------------------------------------------------------
rfkill/10040 is trying to acquire lock:
(rtnl_mutex){+.+.+.}, at: [<ffffffff8146f282>] rtnl_lock+0x12/0x20

but task is already holding lock:
(rfkill_global_mutex){+.+.+.}, at: [<ffffffffa04832ca>] rfkill_fop_write+0x6a/0x170 [rfkill]

--snip--

Chain exists of:
  rtnl_mutex --> misc_mtx --> rfkill_global_mutex

The fix is to move the initialization of the hardware random number generator outside the code range covered by the rtnl_mutex.

---
 drivers/net/wireless/b43/main.c |   17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)
drivers/net/wireless/b43/main.c +11 -6
@@ -2424,6 +2424,7 @@ error:
 
 static int b43_one_core_attach(struct b43_bus_dev *dev, struct b43_wl *wl);
 static void b43_one_core_detach(struct b43_bus_dev *dev);
+static int b43_rng_init(struct b43_wl *wl);
 
 static void b43_request_firmware(struct work_struct *work)
 {
@@ -2475,6 +2476,10 @@ start_ieee80211:
 		goto err_one_core_detach;
 	wl->hw_registred = true;
 	b43_leds_register(wl->current_dev);
+
+	/* Register HW RNG driver */
+	b43_rng_init(wl);
+
 	goto out;
 
 err_one_core_detach:
@@ -4636,9 +4641,6 @@ static void b43_wireless_core_exit(struc
 	if (!dev || b43_status(dev) != B43_STAT_INITIALIZED)
 		return;
 
-	/* Unregister HW RNG driver */
-	b43_rng_exit(dev->wl);
-
 	b43_set_status(dev, B43_STAT_UNINIT);
 
 	/* Stop the microcode PSM. */
@@ -4795,9 +4797,6 @@ static int b43_wireless_core_init(struct
 
 	b43_set_status(dev, B43_STAT_INITIALIZED);
 
-	/* Register HW RNG driver */
-	b43_rng_init(dev->wl);
-
 out:
 	return err;
 
@@ -5464,6 +5463,9 @@ static void b43_bcma_remove(struct bcma_
 
 	b43_one_core_detach(wldev->dev);
 
+	/* Unregister HW RNG driver */
+	b43_rng_exit(wl);
+
 	b43_leds_unregister(wl);
 
 	ieee80211_free_hw(wl->hw);
@@ -5541,6 +5543,9 @@ static void b43_ssb_remove(struct ssb_de
 
 	b43_one_core_detach(dev);
 
+	/* Unregister HW RNG driver */
+	b43_rng_exit(wl);
+
 	if (list_empty(&wl->devlist)) {
 		b43_leds_unregister(wl);
 		/* Last core on the chip unregistered.
author maintainer
#44 lore raw

PATCH 45/133 b43: Fix unload oops if firmware is not available

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 0673effd41dba323d6a280ef37b5ef29f3f5a653 upstream.

The asyncronous firmware load uses a completion struct to hold firmware processing until the user-space routines are up and running. There is. however, a problem in that the waiter is nevered canceled during teardown. As a result, unloading the driver when firmware is not available causes an oops.

To be able to access the completion structure at teardown, it had to be moved into the b43_wldev structure.

This patch also fixes a typo in a comment.

---
 drivers/net/wireless/b43/b43.h  |    4 ++--
 drivers/net/wireless/b43/main.c |   10 +++++-----
 2 files changed, 7 insertions(+), 7 deletions(-)
drivers/net/wireless/b43/b43.h +2 -2
@@ -731,8 +731,6 @@ enum b43_firmware_file_type {
 struct b43_request_fw_context {
 	/* The device we are requesting the fw for. */
 	struct b43_wldev *dev;
-	/* a completion event structure needed if this call is asynchronous */
-	struct completion fw_load_complete;
 	/* a pointer to the firmware object */
 	const struct firmware *blob;
 	/* The type of firmware to request. */
@@ -809,6 +807,8 @@ enum {
 struct b43_wldev {
 	struct b43_bus_dev *dev;
 	struct b43_wl *wl;
+	/* a completion event structure needed if this call is asynchronous */
+	struct completion fw_load_complete;
 
 	/* The device initialization status.
 	 * Use b43_status() to query. */
drivers/net/wireless/b43/main.c +5 -5
@@ -2070,6 +2070,7 @@ void b43_do_release_fw(struct b43_firmwa
 
 static void b43_release_firmware(struct b43_wldev *dev)
 {
+	complete(&dev->fw_load_complete);
 	b43_do_release_fw(&dev->fw.ucode);
 	b43_do_release_fw(&dev->fw.pcm);
 	b43_do_release_fw(&dev->fw.initvals);
@@ -2095,7 +2096,7 @@ static void b43_fw_cb(const struct firmw
 	struct b43_request_fw_context *ctx = context;
 
 	ctx->blob = firmware;
-	complete(&ctx->fw_load_complete);
+	complete(&ctx->dev->fw_load_complete);
 }
 
 int b43_do_request_fw(struct b43_request_fw_context *ctx,
@@ -2142,7 +2143,7 @@ int b43_do_request_fw(struct b43_request
 	}
 	if (async) {
 		/* do this part asynchronously */
-		init_completion(&ctx->fw_load_complete);
+		init_completion(&ctx->dev->fw_load_complete);
 		err = request_firmware_nowait(THIS_MODULE, 1, ctx->fwname,
 					      ctx->dev->dev->dev, GFP_KERNEL,
 					      ctx, b43_fw_cb);
@@ -2150,12 +2151,11 @@ int b43_do_request_fw(struct b43_request
 			pr_err("Unable to load firmware\n");
 			return err;
 		}
-		/* stall here until fw ready */
-		wait_for_completion(&ctx->fw_load_complete);
+		wait_for_completion(&ctx->dev->fw_load_complete);
 		if (ctx->blob)
 			goto fw_ready;
 	/* On some ARM systems, the async request will fail, but the next sync
-	 * request works. For this reason, we dall through here
+	 * request works. For this reason, we fall through here
 	 */
 	}
 	err = request_firmware(&ctx->blob, ctx->fwname,
author maintainer
#45 lore raw

PATCH 46/133 b43legacy: Fix unload oops if firmware is not available

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 452028665312672c6ba9e16a19248ee00ead9400 upstream.

The asyncronous firmware load uses a completion struct to hold firmware processing until the user-space routines are up and running. There is. however, a problem in that the waiter is nevered canceled during teardown. As a result, unloading the driver when firmware is not available causes an oops.

---
 drivers/net/wireless/b43legacy/main.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/wireless/b43legacy/main.c +1 -0
@@ -3919,6 +3919,7 @@ static void b43legacy_remove(struct ssb_
 	 * as the ieee80211 unreg will destroy the workqueue. */
 	cancel_work_sync(&wldev->restart_work);
 	cancel_work_sync(&wl->firmware_load);
+	complete(&wldev->fw_load_complete);
 
 	B43legacy_WARN_ON(!wl);
 	if (!wldev->fw.ucode)
author maintainer
#46 lore raw

PATCH 47/133 b43: fix the wrong assignment of status.freq in b43_rx()

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: ZHAO Gang <[email protected]>

commit 64e5acb09ca6b50c97299cff9ef51299470b29f2 upstream.

Use the right function to update frequency value.

If rx skb is probe response or beacon, the wrong frequency value can cause problem that bss info can't be updated when it should be.

Fixes 8318d78a44d4 ("cfg80211 API for channels/bitrates, mac80211 and driver conversion")
---
 drivers/net/wireless/b43/xmit.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/wireless/b43/xmit.c +2 -2
@@ -821,10 +821,10 @@ void b43_rx(struct b43_wldev *dev, struc
 		 * channel number in b43. */
 		if (chanstat & B43_RX_CHAN_5GHZ) {
 			status.band = IEEE80211_BAND_5GHZ;
-			status.freq = b43_freq_to_channel_5ghz(chanid);
+			status.freq = b43_channel_to_freq_5ghz(chanid);
 		} else {
 			status.band = IEEE80211_BAND_2GHZ;
-			status.freq = b43_freq_to_channel_2ghz(chanid);
+			status.freq = b43_channel_to_freq_2ghz(chanid);
 		}
 		break;
 	default:
author maintainer
#47 lore raw

PATCH 48/133 ath9k: Use correct channel for RX packets

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Sujith Manoharan <[email protected]>

commit ff9a93f2ebb88ac7aab9568de80b64b92078e96d upstream.

Accessing the current channel definition in mac80211 when processing RX packets is problematic because it could have been updated when a scan is issued. Since a channel change involves flushing the existing packets in the RX queue before a chip-reset is done, they would be processed using the wrong band/channel information.

To avoid this, use the current channel information maintained in the driver.

---
 drivers/net/wireless/ath/ath9k/recv.c |   25 ++++++++++++++-----------
 1 file changed, 14 insertions(+), 11 deletions(-)
drivers/net/wireless/ath/ath9k/recv.c +14 -11
@@ -848,20 +848,15 @@ static int ath9k_process_rate(struct ath
 	enum ieee80211_band band;
 	unsigned int i = 0;
 	struct ath_softc __maybe_unused *sc = common->priv;
+	struct ath_hw *ah = sc->sc_ah;
 
-	band = hw->conf.chandef.chan->band;
+	band = ah->curchan->chan->band;
 	sband = hw->wiphy->bands[band];
 
-	switch (hw->conf.chandef.width) {
-	case NL80211_CHAN_WIDTH_5:
+	if (IS_CHAN_QUARTER_RATE(ah->curchan))
 		rxs->flag |= RX_FLAG_5MHZ;
-		break;
-	case NL80211_CHAN_WIDTH_10:
+	else if (IS_CHAN_HALF_RATE(ah->curchan))
 		rxs->flag |= RX_FLAG_10MHZ;
-		break;
-	default:
-		break;
-	}
 
 	if (rx_stats->rs_rate & 0x80) {
 		/* HT rate */
@@ -1175,6 +1170,14 @@ static int ath9k_rx_skb_preprocess(struc
 		ath_start_rx_poll(sc, 3);
 	}
 
+	/*
+	 * This shouldn't happen, but have a safety check anyway.
+	 */
+	if (WARN_ON(!ah->curchan)) {
+		ret = -EINVAL;
+		goto exit;
+	}
+
 	if (ath9k_process_rate(common, hw, rx_stats, rx_status)) {
 		ret =-EINVAL;
 		goto exit;
@@ -1182,8 +1185,8 @@ static int ath9k_rx_skb_preprocess(struc
 
 	ath9k_process_rssi(common, hw, rx_stats, rx_status);
 
-	rx_status->band = hw->conf.chandef.chan->band;
-	rx_status->freq = hw->conf.chandef.chan->center_freq;
+	rx_status->band = ah->curchan->chan->band;
+	rx_status->freq = ah->curchan->chan->center_freq;
 	rx_status->antenna = rx_stats->rs_antenna;
 	rx_status->flag |= RX_FLAG_MACTIME_END;
 
author maintainer
#48 lore raw

PATCH 49/133 ath9k: Disable cross-band FCC

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Sujith Manoharan <[email protected]>

commit 1e2f9295f4c657500111514f92a3d3894d0e05b4 upstream.

Fast Channel Change across bands was enabled for AR9462 recently, but this is causing baseband issues. Disable it until this feature is tested well. Also, remove the feature bit for AR9565 since it is a single-band card and doesn't support this feature.

---
 drivers/net/wireless/ath/ath9k/hw.c |    7 -------
 1 file changed, 7 deletions(-)
drivers/net/wireless/ath/ath9k/hw.c +0 -7
@@ -2608,13 +2608,6 @@ int ath9k_hw_fill_cap_info(struct ath_hw
 	    ah->eep_ops->get_eeprom(ah, EEP_PAPRD))
 			pCap->hw_caps |= ATH9K_HW_CAP_PAPRD;
 
-	/*
-	 * Fast channel change across bands is available
-	 * only for AR9462 and AR9565.
-	 */
-	if (AR_SREV_9462(ah) || AR_SREV_9565(ah))
-		pCap->hw_caps |= ATH9K_HW_CAP_FCC_BAND_SWITCH;
-
 	return 0;
 }
 
author maintainer
#49 lore raw

PATCH 50/133 staging/lustre/ptlrpc: Fix a crash when dereferencing NULL pointer

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Amir Shehata <[email protected]>

commit 3c92a0bf4d72737035a16c4fe357ccd439c9b7d2 upstream.

When a system runs out of memory and the function ptlrpc_register_bulk() is called from ptl_send_rpc() the call to LNetMEAttach() fails due to failure to allocate memory. This forces the code into an error path, which most probably previously went untested. The error path:

if (rc != 0) {
        CERROR("%s: LNetMEAttach failed x"LPU64"/%d: rc = %dn",
                desc->bd_export->exp_obd->obd_name, xid,
                posted_md, rc);
        break;
}

This print assumes that desc->bd_export is not NULL. However, it is. In fact it is expected to be NULL. desc->bd_import is the correct structure to access in this case.

Lustre-change: http://review.whamcloud.com/7121
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3585

---
 drivers/staging/lustre/lustre/ptlrpc/niobuf.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/staging/lustre/lustre/ptlrpc/niobuf.c +3 -3
@@ -179,7 +179,7 @@ int ptlrpc_register_bulk(struct ptlrpc_r
 				  LNET_UNLINK, LNET_INS_AFTER, &me_h);
 		if (rc != 0) {
 			CERROR("%s: LNetMEAttach failed x"LPU64"/%d: rc = %d\n",
-			       desc->bd_export->exp_obd->obd_name, xid,
+			       desc->bd_import->imp_obd->obd_name, xid,
 			       posted_md, rc);
 			break;
 		}
@@ -189,7 +189,7 @@ int ptlrpc_register_bulk(struct ptlrpc_r
 				  &desc->bd_mds[posted_md]);
 		if (rc != 0) {
 			CERROR("%s: LNetMDAttach failed x"LPU64"/%d: rc = %d\n",
-			       desc->bd_export->exp_obd->obd_name, xid,
+			       desc->bd_import->imp_obd->obd_name, xid,
 			       posted_md, rc);
 			rc2 = LNetMEUnlink(me_h);
 			LASSERT(rc2 == 0);
@@ -219,7 +219,7 @@ int ptlrpc_register_bulk(struct ptlrpc_r
 	/* Holler if peer manages to touch buffers before he knows the xid */
 	if (desc->bd_md_count != total_md)
 		CWARN("%s: Peer %s touched %d buffers while I registered\n",
-		      desc->bd_export->exp_obd->obd_name, libcfs_id2str(peer),
+		      desc->bd_import->imp_obd->obd_name, libcfs_id2str(peer),
 		      total_md - desc->bd_md_count);
 	spin_unlock(&desc->bd_lock);
 
author maintainer
#50 lore raw

PATCH 51/133 staging: r8712u: Set device type to wlan

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Larry Finger <[email protected]>

commit 3a21f00a5002b14e4aab52aef59d33ed28468a13 upstream.

The latest version of NetworkManager does not recognize the device as wireless without this change.

---
 drivers/staging/rtl8712/usb_intf.c |    5 +++++
 1 file changed, 5 insertions(+)
drivers/staging/rtl8712/usb_intf.c +5 -0
@@ -358,6 +358,10 @@ static u8 key_2char2num(u8 hch, u8 lch)
 	return (hex_to_bin(hch) << 4) | hex_to_bin(lch);
 }
 
+static const struct device_type wlan_type = {
+	.name = "wlan",
+};
+
 /*
  * drv_init() - a device potentially for us
  *
@@ -393,6 +397,7 @@ static int r871xu_drv_init(struct usb_in
 	padapter->pusb_intf = pusb_intf;
 	usb_set_intfdata(pusb_intf, pnetdev);
 	SET_NETDEV_DEV(pnetdev, &pusb_intf->dev);
+	pnetdev->dev.type = &wlan_type;
 	/* step 2. */
 	padapter->dvobj_init = &r8712_usb_dvobj_init;
 	padapter->dvobj_deinit = &r8712_usb_dvobj_deinit;
author maintainer
#51 lore raw

PATCH 52/133 staging: vt6656: [BUG] BBvUpdatePreEDThreshold Always set sensitivity on bScanning

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Malcolm Priestley <[email protected]>

commit 8f248dae133668bfb8e9379b4b3f0571c858b24a upstream.

byBBPreEDIndex value is initially 0, this means that from cold BBvUpdatePreEDThreshold is never set.

This means that sensitivity may be in an ambiguous state, failing to scan any wireless points or at least distant ones.

---
 drivers/staging/vt6656/baseband.c |    3 ---
 1 file changed, 3 deletions(-)
drivers/staging/vt6656/baseband.c +0 -3
@@ -1464,7 +1464,6 @@ void BBvUpdatePreEDThreshold(struct vnt_
 
             if( bScanning )
             {   // need Max sensitivity //RSSI -69, -70,....
-                if(pDevice->byBBPreEDIndex == 0) break;
                 pDevice->byBBPreEDIndex = 0;
                 ControlvWriteByte(pDevice, MESSAGE_REQUEST_BBREG, 0xC9, 0x00); //CR201(0xC9)
                 ControlvWriteByte(pDevice, MESSAGE_REQUEST_BBREG, 0xCE, 0x30); //CR206(0xCE)
@@ -1607,7 +1606,6 @@ void BBvUpdatePreEDThreshold(struct vnt_
 
             if( bScanning )
             {   // need Max sensitivity  //RSSI -69, -70, ...
-                if(pDevice->byBBPreEDIndex == 0) break;
                 pDevice->byBBPreEDIndex = 0;
                 ControlvWriteByte(pDevice, MESSAGE_REQUEST_BBREG, 0xC9, 0x00); //CR201(0xC9)
                 ControlvWriteByte(pDevice, MESSAGE_REQUEST_BBREG, 0xCE, 0x24); //CR206(0xCE)
@@ -1759,7 +1757,6 @@ void BBvUpdatePreEDThreshold(struct vnt_
         case RF_VT3342A0: //RobertYu:20060627, testing table
             if( bScanning )
             {   // need Max sensitivity  //RSSI -67, -68, ...
-                if(pDevice->byBBPreEDIndex == 0) break;
                 pDevice->byBBPreEDIndex = 0;
                 ControlvWriteByte(pDevice, MESSAGE_REQUEST_BBREG, 0xC9, 0x00); //CR201(0xC9)
                 ControlvWriteByte(pDevice, MESSAGE_REQUEST_BBREG, 0xCE, 0x38); //CR206(0xCE)
author maintainer
#52 lore raw

PATCH 53/133 staging: vt6656: CARDqGetNextTBTT correct uLowNextTBTT

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Malcolm Priestley <[email protected]>

commit 9acec059c0cef0bf086c738f4c0b1f4447782a48 upstream.

value uLowNextTBTT yields wrong value.

ULL is needed with qwTSF

---
 drivers/staging/vt6656/card.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/vt6656/card.c +1 -1
@@ -731,7 +731,7 @@ u64 CARDqGetNextTBTT(u64 qwTSF, u16 wBea
 
     uBeaconInterval = wBeaconInterval * 1024;
     // Next TBTT = ((local_current_TSF / beacon_interval) + 1 ) * beacon_interval
-	uLowNextTBTT = ((qwTSF & 0xffffffffU) >> 10) << 10;
+	uLowNextTBTT = ((qwTSF & 0xffffffffULL) >> 10) << 10;
 	uLowRemain = (uLowNextTBTT) % uBeaconInterval;
 	uHighRemain = ((0x80000000 % uBeaconInterval) * 2 * (u32)(qwTSF >> 32))
 		% uBeaconInterval;
author maintainer
#53 lore raw

PATCH 54/133 mei: use hbm idle state to prevent spurious resets

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Tomas Winkler <[email protected]>

commit 66ae460b13c31a176b41550259683c841a62af3e upstream.

When reset is caused by hbm protocol mismatch or timeout we might end up in an endless reset loop and hbm protocol will never sync

---
 drivers/misc/mei/hbm.c       |   19 +++++++++++++++++++
 drivers/misc/mei/hbm.h       |    1 +
 drivers/misc/mei/init.c      |   12 ++++++++----
 drivers/misc/mei/interrupt.c |   25 +++++++++++++++----------
 4 files changed, 43 insertions(+), 14 deletions(-)
drivers/misc/mei/hbm.c +19 -0
@@ -128,6 +128,17 @@ static bool is_treat_specially_client(st
 	return false;
 }
 
+/**
+ * mei_hbm_idle - set hbm to idle state
+ *
+ * @dev: the device structure
+ */
+void mei_hbm_idle(struct mei_device *dev)
+{
+	dev->init_clients_timer = 0;
+	dev->hbm_state = MEI_HBM_IDLE;
+}
+
 int mei_hbm_start_wait(struct mei_device *dev)
 {
 	int ret;
@@ -577,6 +588,14 @@ void mei_hbm_dispatch(struct mei_device
 	mei_read_slots(dev, dev->rd_msg_buf, hdr->length);
 	mei_msg = (struct mei_bus_message *)dev->rd_msg_buf;
 
+	/* ignore spurious message and prevent reset nesting
+	 * hbm is put to idle during system reset
+	 */
+	if (dev->hbm_state == MEI_HBM_IDLE) {
+		dev_dbg(&dev->pdev->dev, "hbm: state is idle ignore spurious messages\n");
+		return 0;
+	}
+
 	switch (mei_msg->hbm_cmd) {
 	case HOST_START_RES_CMD:
 		version_res = (struct hbm_host_version_response *)mei_msg;
drivers/misc/mei/hbm.h +1 -0
@@ -49,6 +49,7 @@ static inline void mei_hbm_hdr(struct me
 	hdr->reserved = 0;
 }
 
+void mei_hbm_idle(struct mei_device *dev);
 int mei_hbm_start_req(struct mei_device *dev);
 int mei_hbm_start_wait(struct mei_device *dev);
 int mei_hbm_cl_flow_control_req(struct mei_device *dev, struct mei_cl *cl);
drivers/misc/mei/init.c +8 -4
@@ -139,14 +139,19 @@ void mei_reset(struct mei_device *dev, i
 			dev->dev_state != MEI_DEV_POWER_DOWN &&
 			dev->dev_state != MEI_DEV_POWER_UP);
 
+	/* we're already in reset, cancel the init timer
+	 * if the reset was called due the hbm protocol error
+	 * we need to call it before hw start
+	 * so the hbm watchdog won't kick in
+	 */
+	mei_hbm_idle(dev);
+
 	ret = mei_hw_reset(dev, interrupts_enabled);
 	if (ret) {
 		dev_err(&dev->pdev->dev, "hw reset failed disabling the device\n");
 		interrupts_enabled = false;
-		dev->dev_state = MEI_DEV_DISABLED;
 	}
 
-	dev->hbm_state = MEI_HBM_IDLE;
 
 	if (dev->dev_state != MEI_DEV_INITIALIZING &&
 	    dev->dev_state != MEI_DEV_POWER_UP) {
@@ -175,8 +180,6 @@ void mei_reset(struct mei_device *dev, i
 		memset(&dev->wr_ext_msg, 0, sizeof(dev->wr_ext_msg));
 	}
 
-	/* we're already in reset, cancel the init timer */
-	dev->init_clients_timer = 0;
 
 	dev->me_clients_num = 0;
 	dev->rd_msg_hdr = 0;
@@ -188,6 +191,7 @@ void mei_reset(struct mei_device *dev, i
 
 	if (!interrupts_enabled) {
 		dev_dbg(&dev->pdev->dev, "intr not enabled end of reset\n");
+		dev->dev_state = MEI_DEV_DISABLED;
 		return;
 	}
 
drivers/misc/mei/interrupt.c +15 -10
@@ -536,7 +536,6 @@ EXPORT_SYMBOL_GPL(mei_irq_write_handler)
  *
  * @work: pointer to the work_struct structure
  *
- * NOTE: This function is called by timer interrupt work
  */
 void mei_timer(struct work_struct *work)
 {
@@ -551,18 +550,24 @@ void mei_timer(struct work_struct *work)
 
 
 	mutex_lock(&dev->device_lock);
-	if (dev->dev_state != MEI_DEV_ENABLED) {
-		if (dev->dev_state == MEI_DEV_INIT_CLIENTS) {
-			if (dev->init_clients_timer) {
-				if (--dev->init_clients_timer == 0) {
-					dev_err(&dev->pdev->dev, "reset: init clients timeout hbm_state = %d.\n",
-						dev->hbm_state);
-					mei_reset(dev, 1);
-				}
+
+	/* Catch interrupt stalls during HBM init handshake */
+	if (dev->dev_state == MEI_DEV_INIT_CLIENTS &&
+	    dev->hbm_state != MEI_HBM_IDLE) {
+
+		if (dev->init_clients_timer) {
+			if (--dev->init_clients_timer == 0) {
+				dev_err(&dev->pdev->dev, "timer: init clients timeout hbm_state = %d.\n",
+					dev->hbm_state);
+				mei_reset(dev, 1);
+				goto out;
 			}
 		}
-		goto out;
 	}
+
+	if (dev->dev_state != MEI_DEV_ENABLED)
+		goto out;
+
 	/*** connect/disconnect timeouts ***/
 	list_for_each_entry_safe(cl_pos, cl_next, &dev->file_list, link) {
 		if (cl_pos->timer_count) {
author maintainer
#54 lore raw

PATCH 55/133 tty/serial: at91: Handle shutdown more safely

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Marek Roszko <[email protected]>

commit 0cc7c6c7916b1b6f34350ff1473b80b9f7e459c0 upstream.

Interrupts were being cleaned up late in the shutdown handler, it is possible that an interrupt can occur and schedule a tasklet that runs after the port is cleaned up. There is a null dereference due to this race condition with the following stacktrace:

[<c02092b0>] (atmel_tasklet_func+0x514/0x814) from [<c001fd34>] (tasklet_action+0x70/0xa8) [<c001fd34>] (tasklet_action+0x70/0xa8) from [<c001f60c>] (__do_softirq+0x90/0x144) [<c001f60c>] (__do_softirq+0x90/0x144) from [<c001fa18>] (irq_exit+0x40/0x4c) [<c001fa18>] (irq_exit+0x40/0x4c) from [<c000e298>] (handle_IRQ+0x64/0x84) [<c000e298>] (handle_IRQ+0x64/0x84) from [<c000d6c0>] (__irq_svc+0x40/0x50) [<c000d6c0>] (__irq_svc+0x40/0x50) from [<c0208060>] (atmel_rx_dma_release+0x88/0xb8) [<c0208060>] (atmel_rx_dma_release+0x88/0xb8) from [<c0209740>] (atmel_shutdown+0x104/0x160) [<c0209740>] (atmel_shutdown+0x104/0x160) from [<c0205e8c>] (uart_port_shutdown+0x2c/0x38)

---
 drivers/tty/serial/atmel_serial.c |   20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)
drivers/tty/serial/atmel_serial.c +13 -7
@@ -1633,12 +1633,24 @@ static int atmel_startup(struct uart_por
 static void atmel_shutdown(struct uart_port *port)
 {
 	struct atmel_uart_port *atmel_port = to_atmel_uart_port(port);
+
+	/*
+	 * Clear out any scheduled tasklets before
+	 * we destroy the buffers
+	 */
+	tasklet_kill(&atmel_port->tasklet);
+
 	/*
-	 * Ensure everything is stopped.
+	 * Ensure everything is stopped and
+	 * disable all interrupts, port and break condition.
 	 */
 	atmel_stop_rx(port);
 	atmel_stop_tx(port);
 
+	UART_PUT_CR(port, ATMEL_US_RSTSTA);
+	UART_PUT_IDR(port, -1);
+
+
 	/*
 	 * Shut-down the DMA.
 	 */
@@ -1648,12 +1660,6 @@ static void atmel_shutdown(struct uart_p
 		atmel_port->release_tx(port);
 
 	/*
-	 * Disable all interrupts, port and break condition.
-	 */
-	UART_PUT_CR(port, ATMEL_US_RSTSTA);
-	UART_PUT_IDR(port, -1);
-
-	/*
 	 * Free the interrupt
 	 */
 	free_irq(port->irq, port);
author maintainer
#55 lore raw

PATCH 56/133 tty/serial: at91: fix race condition in atmel_serial_remove

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Marek Roszko <[email protected]>

commit f50c995f9ebf064cea1368bf361c4e29679415b4 upstream.

The _remove callback could be called when a tasklet is scheduled. tasklet_kill was called inside the function in order to free up any scheduled tasklets. However it was called after uart_remove_one_port which destroys tty references needed in the port for atmel_tasklet_func.
Simply putting the tasklet_kill at the start of the function will prevent this conflict.

---
 drivers/tty/serial/atmel_serial.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/tty/serial/atmel_serial.c +2 -1
@@ -2430,11 +2430,12 @@ static int atmel_serial_remove(struct pl
 	struct atmel_uart_port *atmel_port = to_atmel_uart_port(port);
 	int ret = 0;
 
+	tasklet_kill(&atmel_port->tasklet);
+
 	device_init_wakeup(&pdev->dev, 0);
 
 	ret = uart_remove_one_port(&atmel_uart, port);
 
-	tasklet_kill(&atmel_port->tasklet);
 	kfree(atmel_port->rx_ring.buf);
 
 	/* "port" is allocated statically, so we shouldn't free it */
author maintainer
#56 lore raw

PATCH 57/133 tty/serial: at91: reset rx_ring when port is shutdown

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Mark Deneen <[email protected]>

commit bb7e73c598fb226c75f7625088a8f6a45a0fc892 upstream.

When using RX DMA, the driver won't pass any data to the uart layer until the buffer is flipped. When the port is shutdown, the dma buffers are unmapped, but the head and tail of the ring buffer are not reseted. Since the serial console will keep the port open, this will only present itself when the uart is not shared.

To reproduce the issue, with an unpatched driver, run a getty on /dev/ttyS0 with no serial console and exit. Getty will exit, and when the new one returns you will be unable to log in. If you hold down a key long enough to fill the DMA buffer and flip it, you can then log in.

---
 drivers/tty/serial/atmel_serial.c |    6 ++++++
 1 file changed, 6 insertions(+)
drivers/tty/serial/atmel_serial.c +6 -0
@@ -1660,6 +1660,12 @@ static void atmel_shutdown(struct uart_p
 		atmel_port->release_tx(port);
 
 	/*
+	 * Reset ring buffer pointers
+	 */
+	atmel_port->rx_ring.head = 0;
+	atmel_port->rx_ring.tail = 0;
+
+	/*
 	 * Free the interrupt
 	 */
 	free_irq(port->irq, port);
author maintainer
#57 lore raw

PATCH 58/133 tty/serial: at91: disable uart timer at start of shutdown

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Marek Roszko <[email protected]>

commit 8bc661bfc0c2d221e209f4205bdaaf574d50100c upstream.

The uart timer will schedule a tasklet when it fires. It is possible that it can fire inside _shutdown before it is killed in the dma and pdc cleanup routines. This causes a tasklet that exists after the port is shutdown, so when the kernel finally executes it, it panics as the tty port is NULL.

This is a somewhat rare condition but its possible if a program keeps on opening/closing the port. It has been observed in particular with systemd boot messages that were causing a kernel panic because of this behavior.

Moving the timer deletion to the beginning of the function stops a tasklet from being scheduled unexpectedly.

[[email protected]: modify commit message, call setup_timer() in any case]
---
 drivers/tty/serial/atmel_serial.c |   22 ++++++++++------------
 1 file changed, 10 insertions(+), 12 deletions(-)
drivers/tty/serial/atmel_serial.c +10 -12
@@ -824,9 +824,6 @@ static void atmel_release_rx_dma(struct
 	atmel_port->desc_rx = NULL;
 	atmel_port->chan_rx = NULL;
 	atmel_port->cookie_rx = -EINVAL;
-
-	if (!atmel_port->is_usart)
-		del_timer_sync(&atmel_port->uart_timer);
 }
 
 static void atmel_rx_from_dma(struct uart_port *port)
@@ -1228,9 +1225,6 @@ static void atmel_release_rx_pdc(struct
 				 DMA_FROM_DEVICE);
 		kfree(pdc->buf);
 	}
-
-	if (!atmel_port->is_usart)
-		del_timer_sync(&atmel_port->uart_timer);
 }
 
 static void atmel_rx_from_pdc(struct uart_port *port)
@@ -1587,12 +1581,13 @@ static int atmel_startup(struct uart_por
 	/* enable xmit & rcvr */
 	UART_PUT_CR(port, ATMEL_US_TXEN | ATMEL_US_RXEN);
 
+	setup_timer(&atmel_port->uart_timer,
+			atmel_uart_timer_callback,
+			(unsigned long)port);
+
 	if (atmel_use_pdc_rx(port)) {
 		/* set UART timeout */
 		if (!atmel_port->is_usart) {
-			setup_timer(&atmel_port->uart_timer,
-					atmel_uart_timer_callback,
-					(unsigned long)port);
 			mod_timer(&atmel_port->uart_timer,
 					jiffies + uart_poll_timeout(port));
 		/* set USART timeout */
@@ -1607,9 +1602,6 @@ static int atmel_startup(struct uart_por
 	} else if (atmel_use_dma_rx(port)) {
 		/* set UART timeout */
 		if (!atmel_port->is_usart) {
-			setup_timer(&atmel_port->uart_timer,
-					atmel_uart_timer_callback,
-					(unsigned long)port);
 			mod_timer(&atmel_port->uart_timer,
 					jiffies + uart_poll_timeout(port));
 		/* set USART timeout */
@@ -1635,6 +1627,12 @@ static void atmel_shutdown(struct uart_p
 	struct atmel_uart_port *atmel_port = to_atmel_uart_port(port);
 
 	/*
+	 * Prevent any tasklets being scheduled during
+	 * cleanup
+	 */
+	del_timer_sync(&atmel_port->uart_timer);
+
+	/*
 	 * Clear out any scheduled tasklets before
 	 * we destroy the buffers
 	 */
author maintainer
#58 lore raw

PATCH 59/133 ARM: at91: at91sam9g45: set default mmc pinctrl-names

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Ludovic Desroches <[email protected]>

commit 0645b93f6c223b594c0dca348e2ae0a23bccf6e3 upstream.

pinctrl-names property was missing from mmc nodes.

---
 arch/arm/boot/dts/at91sam9g45.dtsi |    2 ++
 1 file changed, 2 insertions(+)
arch/arm/boot/dts/at91sam9g45.dtsi +2 -0
@@ -618,6 +618,7 @@
 				compatible = "atmel,hsmci";
 				reg = <0xfff80000 0x600>;
 				interrupts = <11 IRQ_TYPE_LEVEL_HIGH 0>;
+				pinctrl-names = "default";
 				dmas = <&dma 1 AT91_DMA_CFG_PER_ID(0)>;
 				dma-names = "rxtx";
 				#address-cells = <1>;
@@ -629,6 +630,7 @@
 				compatible = "atmel,hsmci";
 				reg = <0xfffd0000 0x600>;
 				interrupts = <29 IRQ_TYPE_LEVEL_HIGH 0>;
+				pinctrl-names = "default";
 				dmas = <&dma 1 AT91_DMA_CFG_PER_ID(13)>;
 				dma-names = "rxtx";
 				#address-cells = <1>;
author maintainer
#59 lore raw

PATCH 60/133 ARM: at91: smc: bug fix in sam9_smc_cs_read()

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Jean-Jacques Hiblot <[email protected]>

commit 1588c51cf6d782e63a8719681d905ef0ac22ee62 upstream.

There was a copy/paste error when reading the nwe_pulse value.

---
 arch/arm/mach-at91/sam9_smc.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/mach-at91/sam9_smc.c +1 -1
@@ -101,7 +101,7 @@ static void sam9_smc_cs_read(void __iome
 	/* Pulse register */
 	val = __raw_readl(base + AT91_SMC_PULSE);
 
-	config->nwe_setup = val & AT91_SMC_NWEPULSE;
+	config->nwe_pulse = val & AT91_SMC_NWEPULSE;
 	config->ncs_write_pulse = (val & AT91_SMC_NCS_WRPULSE) >> 8;
 	config->nrd_pulse = (val & AT91_SMC_NRDPULSE) >> 16;
 	config->ncs_read_pulse = (val & AT91_SMC_NCS_RDPULSE) >> 24;
author maintainer
#60 lore raw

PATCH 61/133 hwmon: (k10temp) Add support for Kaveri CPUs

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Phil Pokorny <[email protected]>

commit d303b1b5fbb688282bbf72a534b9dfed7af9fe4f upstream.

Add new PCI ID to support new model "Kaveri" family.

---
 drivers/hwmon/k10temp.c |    1 +
 1 file changed, 1 insertion(+)
drivers/hwmon/k10temp.c +1 -0
@@ -210,6 +210,7 @@ static DEFINE_PCI_DEVICE_TABLE(k10temp_i
 	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_CNB17H_F3) },
 	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_15H_NB_F3) },
 	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_15H_M10H_F3) },
+	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_15H_M30H_NB_F3) },
 	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_16H_NB_F3) },
 	{}
 };
author maintainer
#61 lore raw

PATCH 62/133 xhci: Avoid infinite loop when sg urb requires too many trbs

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Ben Hutchings <[email protected]>

commit d6c9ea9069af684358efedcaf2f2f687f51c58ee upstream.

Currently prepare_ring() returns -ENOMEM if the urb won't fit into a single ring segment. usb_sg_wait() treats this error as a temporary condition and will keep retrying until something else goes wrong.

The number of retries should be limited in usb_sg_wait(), but also prepare_ring() should not return an error code that suggests it might be worth retrying. Change it to -EINVAL.

Reported-by [email protected]

References: http://bugs.debian.org/733907

Fixes 35773dac5f86 ('usb: xhci: Link TRB must not occur within a USB payload burst')
---
 drivers/usb/host/xhci-ring.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/host/xhci-ring.c +1 -1
@@ -2964,7 +2964,7 @@ static int prepare_ring(struct xhci_hcd
 			if (num_trbs >= TRBS_PER_SEGMENT) {
 				xhci_err(xhci, "Too many fragments %d, max %d\n",
 						num_trbs, TRBS_PER_SEGMENT - 1);
-				return -ENOMEM;
+				return -EINVAL;
 			}
 
 			nop_cmd = cpu_to_le32(TRB_TYPE(TRB_TR_NOOP) |
author maintainer
#62 lore raw

PATCH 63/133 xhci: Set scatter-gather limit to avoid failed block writes.

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Sarah Sharp <[email protected]>

commit f2d9b991c549f159dc9ae81f77d8206c790cbfee upstream.

Commit 35773dac5f862cb1c82ea151eba3e2f6de51ec3e "usb: xhci: Link TRB must not occur within a USB payload burst" attempted to fix an issue found with USB ethernet adapters, and inadvertently broke USB storage devices. The patch attempts to ensure that transfers never span a segment, and rejects transfers that have more than 63 entries (or possibly less, if some entries cross 64KB boundaries).

usb-storage limits the maximum transfer size to 120K, and we had assumed the block layer would pass a scatter-gather list of 4K entries, resulting in no more than 31 sglist entries:

http://marc.info/?l=linux-usb&m=138498190419312&w=2

That assumption was wrong, since we've seen the driver reject a write that was 218 sectors long (of probably 512 bytes each):

Jan  1 07:04:49 jidanni5 kernel: [  559.624704] xhci_hcd 0000:00:14.0: Too many fragments 79, max 63
...
Jan  1 07:04:58 jidanni5 kernel: [  568.622583] Write(10): 2a 00 00 06 85 0e 00 00 da 00

Limit the number of scatter-gather entries to half a ring segment. That should be margin enough in case some entries cross 64KB boundaries. Increase the number of TRBs per segment from 64 to 256, which should result in ring segments fitting on a 4K page.

References: http://bugs.debian.org/733907

Fixes 35773dac5f86 ('usb: xhci: Link TRB must not occur within a USB payload burst')
---
 drivers/usb/host/xhci.c |    4 ++--
 drivers/usb/host/xhci.h |    2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)
drivers/usb/host/xhci.c +2 -2
@@ -4727,8 +4727,8 @@ int xhci_gen_setup(struct usb_hcd *hcd,
 	struct device		*dev = hcd->self.controller;
 	int			retval;
 
-	/* Accept arbitrarily long scatter-gather lists */
-	hcd->self.sg_tablesize = ~0;
+	/* Limit the block layer scatter-gather lists to half a segment. */
+	hcd->self.sg_tablesize = TRBS_PER_SEGMENT / 2;
 
 	/* support to build packet from discontinuous buffers */
 	hcd->self.no_sg_constraint = 1;
drivers/usb/host/xhci.h +1 -1
@@ -1262,7 +1262,7 @@ union xhci_trb {
  * since the command ring is 64-byte aligned.
  * It must also be greater than 16.
  */
-#define TRBS_PER_SEGMENT	64
+#define TRBS_PER_SEGMENT	256
 /* Allow two commands + a link TRB, along with any reserved command TRBs */
 #define MAX_RSVD_CMD_TRBS	(TRBS_PER_SEGMENT - 3)
 #define TRB_SEGMENT_SIZE	(TRBS_PER_SEGMENT*16)
author maintainer
#63 lore raw

PATCH 64/133 serial: add support for 200 v3 series Titan card

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Yegor Yefremov <[email protected]>

commit 48c0247d7b7bf58abb85a39021099529df365c4d upstream.

---
 drivers/tty/serial/8250/8250_pci.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/tty/serial/8250/8250_pci.c +4 -0
@@ -1545,6 +1545,7 @@ pci_wch_ch353_setup(struct serial_privat
 #define PCI_DEVICE_ID_TITAN_800E	0xA014
 #define PCI_DEVICE_ID_TITAN_200EI	0xA016
 #define PCI_DEVICE_ID_TITAN_200EISI	0xA017
+#define PCI_DEVICE_ID_TITAN_200V3	0xA306
 #define PCI_DEVICE_ID_TITAN_400V3	0xA310
 #define PCI_DEVICE_ID_TITAN_410V3	0xA312
 #define PCI_DEVICE_ID_TITAN_800V3	0xA314
@@ -4140,6 +4141,9 @@ static struct pci_device_id serial_pci_t
 	{	PCI_VENDOR_ID_TITAN, PCI_DEVICE_ID_TITAN_200EISI,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_oxsemi_2_4000000 },
+	{	PCI_VENDOR_ID_TITAN, PCI_DEVICE_ID_TITAN_200V3,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
+		pbn_b0_bt_2_921600 },
 	{	PCI_VENDOR_ID_TITAN, PCI_DEVICE_ID_TITAN_400V3,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b0_4_921600 },
author maintainer
#64 lore raw

PATCH 65/133 serial: 8250: Fix initialisation of Quatech cards with the AMCC PCI chip

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Jonathan Woithe <[email protected]>

commit 9c5320f8d7d9a2cf623e65d50e1113f34d9b9eb1 upstream.

Fix the initialisation of older Quatech serial cards which are fitted with the AMCC PCI Matchmaker interface chip.

---
 drivers/tty/serial/8250/8250_pci.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/tty/serial/8250/8250_pci.c +2 -2
@@ -1260,10 +1260,10 @@ static int pci_quatech_init(struct pci_d
 		unsigned long base = pci_resource_start(dev, 0);
 		if (base) {
 			u32 tmp;
-			outl(inl(base + 0x38), base + 0x38);
+			outl(inl(base + 0x38) | 0x00002000, base + 0x38);
 			tmp = inl(base + 0x3c);
 			outl(tmp | 0x01000000, base + 0x3c);
-			outl(tmp, base + 0x3c);
+			outl(tmp &= ~0x01000000, base + 0x3c);
 		}
 	}
 	return 0;
author maintainer
#65 lore raw

PATCH 66/133 serial: 8250: enable UART_BUG_NOMSR for Tegra

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Stephen Warren <[email protected]>

commit 3685f19e07802ec4207b52465c408f185b66490e upstream.

Tegra chips have 4 or 5 identical UART modules embedded. UARTs C..E have their MODEM-control signals tied off to a static state. However UARTs A and B can optionally route those signals to/from package pins, depending on the exact pinmux configuration.

When these signals are not routed to package pins, false interrupts may trigger either temporarily, or permanently, all while not showing up in the IIR; it will read as NO_INT. This will eventually lead to the UART IRQ being disabled due to unhandled interrupts. When this happens, the kernel may print e.g.:

irq 68: nobody cared (try booting with the "irqpoll" option)

In order to prevent this, enable UART_BUG_NOMSR. This prevents UART_IER_MSI from being enabled, which prevents the false interrupts from triggering.

In practice, this is not needed under any of the following conditions:

* On Tegra chips after Tegra30, since the HW bug has apparently been fixed.

* On UARTs C..E since their MODEM control signals are tied to the correct static state which doesn't trigger the issue.

* On UARTs A..B if the MODEM control signals are routed out to package pins, since they will then carry valid signals.

However, we ignore these exceptions for now, since they are only relevant if a board actually hooks up more than a 4-wire UART, and no currently supported board does this. If we ever support a board that does, we can refine the algorithm that enables UART_BUG_NOMSR to take those exceptions into account, and/or read a flag from DT/... that indicates that the board has hooked up and pinmux'd more than a 4-wire UART.

---
 drivers/tty/serial/8250/8250_core.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/tty/serial/8250/8250_core.c +4 -0
@@ -2670,6 +2670,10 @@ static void serial8250_config_port(struc
 	if (port->type == PORT_16550A && port->iotype == UPIO_AU)
 		up->bugs |= UART_BUG_NOMSR;
 
+	/* HW bugs may trigger IRQ while IIR == NO_INT */
+	if (port->type == PORT_TEGRA)
+		up->bugs |= UART_BUG_NOMSR;
+
 	if (port->type != PORT_UNKNOWN && flags & UART_CONFIG_IRQ)
 		autoconfig_irq(up);
 
author maintainer
#66 lore raw

PATCH 67/133 KVM: s390: fix diagnose code extraction

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Heiko Carstens <[email protected]>

commit 743db27c526e0f31cc507959d662e97e2048a86f upstream.

The diagnose code to be used is the contents of the base register (if not zero), plus the displacement. The current code ignores the base register contents. So let's fix that...

---
 arch/s390/kvm/diag.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/s390/kvm/diag.c +1 -1
@@ -139,7 +139,7 @@ static int __diag_virtio_hypercall(struc
 
 int kvm_s390_handle_diag(struct kvm_vcpu *vcpu)
 {
-	int code = (vcpu->arch.sie_block->ipb & 0xfff0000) >> 16;
+	int code = kvm_s390_get_base_disp_rs(vcpu) & 0xffff;
 
 	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
 		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
author maintainer
#67 lore raw

PATCH 68/133 KVM: s390: ioeventfd: ignore leftmost bits

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Dominik Dingel <[email protected]>

commit ff1f3cb4b3ac5d039f02679f34cb1498d110d241 upstream.

The diagnose 500 subcode 3 contains the 32 bit subchannel id in bits 32-63 (counting from the left). As for other I/O instructions, bits 0-31 should be ignored and thus not be passed to kvm_io_bus_write_cookie().

This fixes a bug where the guest passed non-zero bits 0-31 which the host tried to interpret, leading to ioeventfd notification failures.

---
 arch/s390/kvm/diag.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/s390/kvm/diag.c +1 -1
@@ -122,7 +122,7 @@ static int __diag_virtio_hypercall(struc
 	 * - gpr 4 contains the index on the bus (optionally)
 	 */
 	ret = kvm_io_bus_write_cookie(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
-				      vcpu->run->s.regs.gprs[2],
+				      vcpu->run->s.regs.gprs[2] & 0xffffffff,
 				      8, &vcpu->run->s.regs.gprs[3],
 				      vcpu->run->s.regs.gprs[4]);
 	srcu_read_unlock(&vcpu->kvm->srcu, idx);
author maintainer
#68 lore raw

PATCH 69/133 s390/uapi: fix struct statfs64 definition

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Heiko Carstens <[email protected]>

commit 4e078146dff728f4865270a47710d57797e81eb6 upstream.

With b8668fd0a7e1b59f "s390/uapi: change struct statfs[64] member types to unsigned values" the size of a couple of struct statfs64 member got incorrectly changed from 64 to 32 bit for 32 bit builds.

Fix this by changing the type of couple of struct statfs64 members from unsigned long to unsigned long long.
The definition of struct compat_statfs64 was correct however.

---
 arch/s390/include/uapi/asm/statfs.h |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
arch/s390/include/uapi/asm/statfs.h +5 -5
@@ -35,11 +35,11 @@ struct statfs {
 struct statfs64 {
 	unsigned int	f_type;
 	unsigned int	f_bsize;
-	unsigned long	f_blocks;
-	unsigned long	f_bfree;
-	unsigned long	f_bavail;
-	unsigned long	f_files;
-	unsigned long	f_ffree;
+	unsigned long long f_blocks;
+	unsigned long long f_bfree;
+	unsigned long long f_bavail;
+	unsigned long long f_files;
+	unsigned long long f_ffree;
 	__kernel_fsid_t f_fsid;
 	unsigned int	f_namelen;
 	unsigned int	f_frsize;
author maintainer
#69 lore raw

PATCH 70/133 parport: parport_pc: remove double PCI ID for NetMos

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Sebastian Andrzej Siewior <[email protected]>

commit d6a484520c5572a4170fa915109ccfc0c38f5008 upstream.

In commit 85747f ("PATCH] parport: add NetMOS 9805 support") Max added the PCI ID for NetMOS 9805 based on a Debian bug report from 2k4 which was at the v2.4.26 time frame. The patch made into 2.6.14. Shortly before that patch akpm merged commit 296d3c783b ("[PATCH] Support NetMOS based PCI cards providing serial and parallel ports") which made into v2.6.9-rc1.
Now we have two different entries for the same PCI id.
I have here the NetMos 9805 which claims to support SPP/EPP/ECP mode. This patch takes Max's entry for titan_1284p1 (base != -1 specifies the ioport for ECP mode) and replaces akpm's entry for netmos_9805 which specified -1 (=none). Both share the same PCI-ID (my card has subsystem 0x1000 / 0x0020 so it should match PCI_ANY).

While here I also drop the entry for titan_1284p2 which is the same as netmos_9815.

---
 drivers/parport/parport_pc.c |   10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)
drivers/parport/parport_pc.c +2 -8
@@ -2596,8 +2596,6 @@ enum parport_pc_pci_cards {
 	syba_2p_epp,
 	syba_1p_ecp,
 	titan_010l,
-	titan_1284p1,
-	titan_1284p2,
 	avlab_1p,
 	avlab_2p,
 	oxsemi_952,
@@ -2656,8 +2654,6 @@ static struct parport_pc_pci {
 	/* syba_2p_epp AP138B */	{ 2, { { 0, 0x078 }, { 0, 0x178 }, } },
 	/* syba_1p_ecp W83787 */	{ 1, { { 0, 0x078 }, } },
 	/* titan_010l */		{ 1, { { 3, -1 }, } },
-	/* titan_1284p1 */              { 1, { { 0, 1 }, } },
-	/* titan_1284p2 */		{ 2, { { 0, 1 }, { 2, 3 }, } },
 	/* avlab_1p		*/	{ 1, { { 0, 1}, } },
 	/* avlab_2p		*/	{ 2, { { 0, 1}, { 2, 3 },} },
 	/* The Oxford Semi cards are unusual: 954 doesn't support ECP,
@@ -2673,8 +2669,8 @@ static struct parport_pc_pci {
 	/* netmos_9705 */               { 1, { { 0, -1 }, } },
 	/* netmos_9715 */               { 2, { { 0, 1 }, { 2, 3 },} },
 	/* netmos_9755 */               { 2, { { 0, 1 }, { 2, 3 },} },
-	/* netmos_9805 */               { 1, { { 0, -1 }, } },
-	/* netmos_9815 */               { 2, { { 0, -1 }, { 2, -1 }, } },
+	/* netmos_9805 */		{ 1, { { 0, 1 }, } },
+	/* netmos_9815 */		{ 2, { { 0, 1 }, { 2, 3 }, } },
 	/* netmos_9901 */               { 1, { { 0, -1 }, } },
 	/* netmos_9865 */               { 1, { { 0, -1 }, } },
 	/* quatech_sppxp100 */		{ 1, { { 0, 1 }, } },
@@ -2718,8 +2714,6 @@ static const struct pci_device_id parpor
 	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, syba_1p_ecp },
 	{ PCI_VENDOR_ID_TITAN, PCI_DEVICE_ID_TITAN_010L,
 	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, titan_010l },
-	{ 0x9710, 0x9805, 0x1000, 0x0010, 0, 0, titan_1284p1 },
-	{ 0x9710, 0x9815, 0x1000, 0x0020, 0, 0, titan_1284p2 },
 	/* PCI_VENDOR_ID_AVLAB/Intek21 has another bunch of cards ...*/
 	/* AFAVLAB_TK9902 */
 	{ 0x14db, 0x2120, PCI_ANY_ID, PCI_ANY_ID, 0, 0, avlab_1p},
author maintainer
#70 lore raw

PATCH 71/133 rtc-cmos: Add an alarm disable quirk

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Borislav Petkov <[email protected]>

commit d5a1c7e3fc38d9c7d629e1e47f32f863acbdec3d upstream.

41c7f7424259f ("rtc: Disable the alarm in the hardware (v2)") added the functionality to disable the RTC wake alarm when shutting down the box.

However, there are at least two b0rked BIOSes we know about:

https://bugzilla.novell.com/show_bug.cgi?id=812592
https://bugzilla.novell.com/show_bug.cgi?id=805740

where, when wakeup alarm is enabled in the BIOS, the machine reboots automatically right after shutdown, regardless of what wakeup time is programmed.

Bisecting the issue lead to this patch so disable its functionality with a DMI quirk only for those boxes.

Cc Brecht Machiels <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc John Stultz <[email protected]>
Cc Rabin Vincent <[email protected]>
[jstultz: Changed variable name for clarity, added extra dmi entry]
---
 drivers/rtc/rtc-cmos.c |   52 ++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 51 insertions(+), 1 deletion(-)
drivers/rtc/rtc-cmos.c +51 -1
@@ -34,11 +34,11 @@
 #include <linux/interrupt.h>
 #include <linux/spinlock.h>
 #include <linux/platform_device.h>
-#include <linux/mod_devicetable.h>
 #include <linux/log2.h>
 #include <linux/pm.h>
 #include <linux/of.h>
 #include <linux/of_platform.h>
+#include <linux/dmi.h>
 
 /* this is for "generic access to PC-style RTC" using CMOS_READ/CMOS_WRITE */
 #include <asm-generic/rtc.h>
@@ -377,6 +377,51 @@ static int cmos_set_alarm(struct device
 	return 0;
 }
 
+/*
+ * Do not disable RTC alarm on shutdown - workaround for b0rked BIOSes.
+ */
+static bool alarm_disable_quirk;
+
+static int __init set_alarm_disable_quirk(const struct dmi_system_id *id)
+{
+	alarm_disable_quirk = true;
+	pr_info("rtc-cmos: BIOS has alarm-disable quirk. ");
+	pr_info("RTC alarms disabled\n");
+	return 0;
+}
+
+static const struct dmi_system_id rtc_quirks[] __initconst = {
+	/* https://bugzilla.novell.com/show_bug.cgi?id=805740 */
+	{
+		.callback = set_alarm_disable_quirk,
+		.ident    = "IBM Truman",
+		.matches  = {
+			DMI_MATCH(DMI_SYS_VENDOR, "TOSHIBA"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "4852570"),
+		},
+	},
+	/* https://bugzilla.novell.com/show_bug.cgi?id=812592 */
+	{
+		.callback = set_alarm_disable_quirk,
+		.ident    = "Gigabyte GA-990XA-UD3",
+		.matches  = {
+			DMI_MATCH(DMI_SYS_VENDOR,
+					"Gigabyte Technology Co., Ltd."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "GA-990XA-UD3"),
+		},
+	},
+	/* http://permalink.gmane.org/gmane.linux.kernel/1604474 */
+	{
+		.callback = set_alarm_disable_quirk,
+		.ident    = "Toshiba Satellite L300",
+		.matches  = {
+			DMI_MATCH(DMI_SYS_VENDOR, "TOSHIBA"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Satellite L300"),
+		},
+	},
+	{}
+};
+
 static int cmos_alarm_irq_enable(struct device *dev, unsigned int enabled)
 {
 	struct cmos_rtc	*cmos = dev_get_drvdata(dev);
@@ -385,6 +430,9 @@ static int cmos_alarm_irq_enable(struct
 	if (!is_valid_irq(cmos->irq))
 		return -EINVAL;
 
+	if (alarm_disable_quirk)
+		return 0;
+
 	spin_lock_irqsave(&rtc_lock, flags);
 
 	if (enabled)
@@ -1158,6 +1206,8 @@ static int __init cmos_init(void)
 			platform_driver_registered = true;
 	}
 
+	dmi_check_system(rtc_quirks);
+
 	if (retval == 0)
 		return 0;
 
author maintainer
#71 lore raw

PATCH 72/133 rtc: max8907: weekday encoding fixes

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Stephen Warren <[email protected]>

commit 75ea799df4cb07e505c91b4abaa87bc28aad3e66 upstream.

The current MAX8907 driver has two issues related to weekday value handling:

1)

The HW WEEKDAY register has range 0..6 rather than 1..7 as documented. Note that I validated the actual HW range by observing the HW register roll from 6->0 rather than 6->7->1 as would otherwise be expected.

This matches Linux's tm_wday range of 0..6.

When the CMOS RAM content is lost, the date returned from the device is

2007-01-01 00:00:00, which is a Monday.  The WEEKDAY register reads 1 in

this case. This matches the numbering in Linux's tm_wday field.

Hence we should write Linux's tm_wday value to the register without modifying it. Hence, remove the +1/-1 calculations for WEEKDAY/tm_wday.

2)

There's no need to make alarms match on the WEEKDAY register, since the other fields together uniquely define the alarm date/time. Ignoring the WEEKDAY value in the match isolates the driver from any incorrect value in the current time copy of the WEEKDAY register.

Each change individually, or both together, solves an issue that I observed; "hwclock -r" would time out waiting for its alarm to fire if the CMOS RAM content had been lost, and hence the WEEKDAY register value mismatched what the driver expected it to be. "hwclock -w" would solve this by over-writing the HW default WEEKDAY register value with what the driver expected.

---
 drivers/rtc/rtc-max8907.c |   11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/rtc/rtc-max8907.c +5 -6
@@ -51,7 +51,7 @@ static irqreturn_t max8907_irq_handler(i
 {
 	struct max8907_rtc *rtc = data;
 
-	regmap_update_bits(rtc->regmap, MAX8907_REG_ALARM0_CNTL, 0x7f, 0);
+	regmap_write(rtc->regmap, MAX8907_REG_ALARM0_CNTL, 0);
 
 	rtc_update_irq(rtc->rtc_dev, 1, RTC_IRQF | RTC_AF);
 
@@ -64,7 +64,7 @@ static void regs_to_tm(u8 *regs, struct
 		bcd2bin(regs[RTC_YEAR1]) - 1900;
 	tm->tm_mon = bcd2bin(regs[RTC_MONTH] & 0x1f) - 1;
 	tm->tm_mday = bcd2bin(regs[RTC_DATE] & 0x3f);
-	tm->tm_wday = (regs[RTC_WEEKDAY] & 0x07) - 1;
+	tm->tm_wday = (regs[RTC_WEEKDAY] & 0x07);
 	if (regs[RTC_HOUR] & HOUR_12) {
 		tm->tm_hour = bcd2bin(regs[RTC_HOUR] & 0x01f);
 		if (tm->tm_hour == 12)
@@ -88,7 +88,7 @@ static void tm_to_regs(struct rtc_time *
 	regs[RTC_YEAR1] = bin2bcd(low);
 	regs[RTC_MONTH] = bin2bcd(tm->tm_mon + 1);
 	regs[RTC_DATE] = bin2bcd(tm->tm_mday);
-	regs[RTC_WEEKDAY] = tm->tm_wday + 1;
+	regs[RTC_WEEKDAY] = tm->tm_wday;
 	regs[RTC_HOUR] = bin2bcd(tm->tm_hour);
 	regs[RTC_MIN] = bin2bcd(tm->tm_min);
 	regs[RTC_SEC] = bin2bcd(tm->tm_sec);
@@ -153,7 +153,7 @@ static int max8907_rtc_set_alarm(struct
 	tm_to_regs(&alrm->time, regs);
 
 	/* Disable alarm while we update the target time */
-	ret = regmap_update_bits(rtc->regmap, MAX8907_REG_ALARM0_CNTL, 0x7f, 0);
+	ret = regmap_write(rtc->regmap, MAX8907_REG_ALARM0_CNTL, 0);
 	if (ret < 0)
 		return ret;
 
@@ -163,8 +163,7 @@ static int max8907_rtc_set_alarm(struct
 		return ret;
 
 	if (alrm->enabled)
-		ret = regmap_update_bits(rtc->regmap, MAX8907_REG_ALARM0_CNTL,
-					 0x7f, 0x7f);
+		ret = regmap_write(rtc->regmap, MAX8907_REG_ALARM0_CNTL, 0x77);
 
 	return ret;
 }
author maintainer
#72 lore raw

PATCH 73/133 pinctrl: sunxi: Honor GPIO output initial vaules

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Chen-Yu Tsai <[email protected]>

commit fa8cf57c923e86a693a85aff1df579245a27cbb3 upstream.

Some GPIO users, such as fixed-regulator, request GPIO output with initial value of 1. This was ignored by sunxi driver.

---
 drivers/pinctrl/pinctrl-sunxi.c |   13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)
drivers/pinctrl/pinctrl-sunxi.c +7 -6
@@ -469,12 +469,6 @@ static int sunxi_pinctrl_gpio_get(struct
 	return val;
 }
 
-static int sunxi_pinctrl_gpio_direction_output(struct gpio_chip *chip,
-					unsigned offset, int value)
-{
-	return pinctrl_gpio_direction_output(chip->base + offset);
-}
-
 static void sunxi_pinctrl_gpio_set(struct gpio_chip *chip,
 				unsigned offset, int value)
 {
@@ -498,6 +492,13 @@ static void sunxi_pinctrl_gpio_set(struc
 	spin_unlock_irqrestore(&pctl->lock, flags);
 }
 
+static int sunxi_pinctrl_gpio_direction_output(struct gpio_chip *chip,
+					unsigned offset, int value)
+{
+	sunxi_pinctrl_gpio_set(chip, offset, value);
+	return pinctrl_gpio_direction_output(chip->base + offset);
+}
+
 static int sunxi_pinctrl_gpio_of_xlate(struct gpio_chip *gc,
 				const struct of_phandle_args *gpiospec,
 				u32 *flags)
author maintainer
#73 lore raw

PATCH 74/133 perf kvm: Fix kvm report without guestmount.

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Dongsheng Yang <[email protected]>

commit ad85ace07a05062ef6b59c35a5e80b6eaee1eee6 upstream.

Currently, if we use perf kvm --guestkallsyms --guestmodules report, we can not get the perf information from perf data file. All sample are shown as unknown.

Reproducing steps:

	# perf kvm --guestkallsyms /tmp/kallsyms --guestmodules /tmp/modules record -a sleep 1
	[ perf record: Woken up 1 times to write data ]
	[ perf record: Captured and wrote 0.624 MB perf.data.guest (~27260 samples) ]
	# perf kvm --guestkallsyms /tmp/kallsyms --guestmodules /tmp/modules report |grep %
	   100.00%  [guest/6471]  [unknown]         [g] 0xffffffff8164f330

This bug was introduced by 207b57926 (perf kvm: Fix regression with guest machine creation). In original code, it uses perf_session__find_machine(), it means we deliver symbol to machine which has the same pid, if no machine found, deliver it to *default* guest. But if we use perf_session__findnew_machine() here, if no machine was found, new machine with pid will be built and added. Then the default guest which with pid == 0 will never get a symbol.

And because the new machine initialized here has no kernel map created, the symbol delivered to it will be marked as "unknown".

This patch here is to revert commit 207b57926 and fix the SEGFAULT bug in another way.

Verification steps:

	# ./perf kvm --guestkallsyms /home/kallsyms --guestmodules /home/modules record -a sleep 1
	[ perf record: Woken up 1 times to write data ]
	[ perf record: Captured and wrote 0.651 MB perf.data.guest (~28437 samples) ]
	# ./perf kvm --guestkallsyms /home/kallsyms --guestmodules /home/modules report |grep %
	    22.64%    :6471  [guest.kernel.kallsyms]  [g] update_rq_clock.part.70
	    19.99%    :6471  [guest.kernel.kallsyms]  [g] d_free
	    18.46%    :6471  [guest.kernel.kallsyms]  [g] bio_phys_segments
	    16.25%    :6471  [guest.kernel.kallsyms]  [g] dequeue_task
	    12.78%    :6471  [guest.kernel.kallsyms]  [g] __switch_to
	     7.91%    :6471  [guest.kernel.kallsyms]  [g] scheduler_tick
	     1.75%    :6471  [guest.kernel.kallsyms]  [g] native_apic_mem_write
	     0.21%    :6471  [guest.kernel.kallsyms]  [g] apic_timer_interrupt
---
 tools/perf/util/session.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
tools/perf/util/session.c +6 -1
@@ -870,6 +870,7 @@ static struct machine *
 					       struct perf_sample *sample)
 {
 	const u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct machine *machine;
 
 	if (perf_guest &&
 	    ((cpumode == PERF_RECORD_MISC_GUEST_KERNEL) ||
@@ -882,7 +883,11 @@ static struct machine *
 		else
 			pid = sample->pid;
 
-		return perf_session__findnew_machine(session, pid);
+		machine = perf_session__find_machine(session, pid);
+		if (!machine)
+			machine = perf_session__findnew_machine(session,
+						DEFAULT_GUEST_KERNEL_ID);
+		return machine;
 	}
 
 	return &session->machines.host;
author maintainer
#74 lore raw

PATCH 75/133 mfd: max77686: Fix regmap resource leak on driver remove

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Krzysztof Kozlowski <[email protected]>

commit 74142ffc0b52cfe6f9d2f6f34a5f3eedbfe3ce51 upstream.

The regmap used by max77686 MFD driver was not freed with regmap_exit() on driver exit. This lead to leak of resources.

Replace regmap_init_i2c() call in driver probe with initialization of managed register map so the regmap will be properly freed by the device management code.

---
 drivers/mfd/max77686.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mfd/max77686.c +1 -1
@@ -103,7 +103,7 @@ static int max77686_i2c_probe(struct i2c
 	max77686->irq_gpio = pdata->irq_gpio;
 	max77686->irq = i2c->irq;
 
-	max77686->regmap = regmap_init_i2c(i2c, &max77686_regmap_config);
+	max77686->regmap = devm_regmap_init_i2c(i2c, &max77686_regmap_config);
 	if (IS_ERR(max77686->regmap)) {
 		ret = PTR_ERR(max77686->regmap);
 		dev_err(max77686->dev, "Failed to allocate register map: %d\n",
author maintainer
#75 lore raw

PATCH 76/133 ASoC: adau1701: Fix ADAU1701_SEROCTL_WORD_LEN_16 constant

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Lars-Peter Clausen <[email protected]>

commit e20970ada3f699c113fe64b04492af083d11a7d8 upstream.

The driver defines ADAU1701_SEROCTL_WORD_LEN_16 as 0x10 while it should be b10, so 0x2. This patch fixes it.

---
 sound/soc/codecs/adau1701.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/codecs/adau1701.c +1 -1
@@ -71,7 +71,7 @@
 
 #define ADAU1701_SEROCTL_WORD_LEN_24	0x0000
 #define ADAU1701_SEROCTL_WORD_LEN_20	0x0001
-#define ADAU1701_SEROCTL_WORD_LEN_16	0x0010
+#define ADAU1701_SEROCTL_WORD_LEN_16	0x0002
 #define ADAU1701_SEROCTL_WORD_LEN_MASK	0x0003
 
 #define ADAU1701_AUXNPOW_VBPD		0x40
author maintainer
#76 lore raw

PATCH 77/133 ASoC: wm5110: Extend SYSCLK patch file for rev D

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Charles Keepax <[email protected]>

commit 34354792432b6e0a3b156819a1a19716c50d3473 upstream.

Latest evaluation of the the device has given some patch file additions for improved performance.

---
 sound/soc/codecs/wm5110.c |   48 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 48 insertions(+)
sound/soc/codecs/wm5110.c +48 -0
@@ -43,6 +43,54 @@ static const struct reg_default wm5110_s
 	{ 0x3133, 0x1201 },
 	{ 0x3183, 0x1501 },
 	{ 0x31D3, 0x1401 },
+	{ 0x0049, 0x01ea },
+	{ 0x004a, 0x01f2 },
+	{ 0x0057, 0x01e7 },
+	{ 0x0058, 0x01fb },
+	{ 0x33ce, 0xc4f5 },
+	{ 0x33cf, 0x1361 },
+	{ 0x33d0, 0x0402 },
+	{ 0x33d1, 0x4700 },
+	{ 0x33d2, 0x026d },
+	{ 0x33d3, 0xff00 },
+	{ 0x33d4, 0x026d },
+	{ 0x33d5, 0x0101 },
+	{ 0x33d6, 0xc4f5 },
+	{ 0x33d7, 0x0361 },
+	{ 0x33d8, 0x0402 },
+	{ 0x33d9, 0x6701 },
+	{ 0x33da, 0xc4f5 },
+	{ 0x33db, 0x136f },
+	{ 0x33dc, 0xc4f5 },
+	{ 0x33dd, 0x134f },
+	{ 0x33de, 0xc4f5 },
+	{ 0x33df, 0x131f },
+	{ 0x33e0, 0x026d },
+	{ 0x33e1, 0x4f01 },
+	{ 0x33e2, 0x026d },
+	{ 0x33e3, 0xf100 },
+	{ 0x33e4, 0x026d },
+	{ 0x33e5, 0x0001 },
+	{ 0x33e6, 0xc4f5 },
+	{ 0x33e7, 0x0361 },
+	{ 0x33e8, 0x0402 },
+	{ 0x33e9, 0x6601 },
+	{ 0x33ea, 0xc4f5 },
+	{ 0x33eb, 0x136f },
+	{ 0x33ec, 0xc4f5 },
+	{ 0x33ed, 0x134f },
+	{ 0x33ee, 0xc4f5 },
+	{ 0x33ef, 0x131f },
+	{ 0x33f0, 0x026d },
+	{ 0x33f1, 0x4e01 },
+	{ 0x33f2, 0x026d },
+	{ 0x33f3, 0xf000 },
+	{ 0x33f6, 0xc4f5 },
+	{ 0x33f7, 0x1361 },
+	{ 0x33f8, 0x0402 },
+	{ 0x33f9, 0x4600 },
+	{ 0x33fa, 0x026d },
+	{ 0x33fb, 0xfe00 },
 };
 
 static int wm5110_sysclk_ev(struct snd_soc_dapm_widget *w,
author maintainer
#77 lore raw

PATCH 78/133 ALSA: rme9652: fix a missing comma in channel_map_9636_ds[]

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Takashi Iwai <[email protected]>

commit 770bd4bf2e664939a9dacd3d26ec9ff7a3933210 upstream.

The lack of comma leads to the wrong channel for an SPDIF channel. Unfortunately this wasn't caught by compiler because it's still a valid expression.

---
 sound/pci/rme9652/rme9652.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/pci/rme9652/rme9652.c +1 -1
@@ -285,7 +285,7 @@ static char channel_map_9636_ds[26] = {
 	/* ADAT channels are remapped */
 	1, 3, 5, 7, 9, 11, 13, 15,
 	/* channels 8 and 9 are S/PDIF */
-	24, 25
+	24, 25,
 	/* others don't exist */
 	-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1
 };
author maintainer
#78 lore raw

PATCH 79/133 ALSA: hda - Correct AD1986A 3stack pin configs

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Takashi Iwai <[email protected]>

commit ed0e0d0617a8dc3d8b82c6e54827f269f2247b07 upstream.

The 3stack pin configs for AD1986A codec had incorrect values that resulted in broken mic and line-in.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=66621

---
 sound/pci/hda/patch_analog.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
sound/pci/hda/patch_analog.c +3 -3
@@ -264,11 +264,11 @@ static const struct hda_fixup ad1986a_fi
 		.v.pins = (const struct hda_pintbl[]) {
 			{ 0x1a, 0x02214021 }, /* headphone */
 			{ 0x1b, 0x01014011 }, /* front */
-			{ 0x1c, 0x01013012 }, /* surround */
-			{ 0x1d, 0x01019015 }, /* clfe */
+			{ 0x1c, 0x01813030 }, /* line-in */
+			{ 0x1d, 0x01a19020 }, /* rear mic */
 			{ 0x1e, 0x411111f0 }, /* N/A */
 			{ 0x1f, 0x02a190f0 }, /* mic */
-			{ 0x20, 0x018130f0 }, /* line-in */
+			{ 0x20, 0x411111f0 }, /* N/A */
 			{}
 		},
 	},
author maintainer
#79 lore raw

PATCH 80/133 ALSA: hda - Dont create duplicated ctls for loopback paths

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Takashi Iwai <[email protected]>

commit 43a8e50a46a4e1dd1451e4a4ffa1f7695fb7d287 upstream.

AD1986A mic pins (0x1d and 0x1f) share the same widget for controlling the loopback volume/mute, but the generic parser didn't check it. This ended up with the duplicated controls for the same effect.

This patch adds the check of the duplication for avoiding it.

After this fix, there will be only one control although it affects both paths; this remaining issue should be fixed later in a different patch.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=66621

---
 sound/pci/hda/hda_generic.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
sound/pci/hda/hda_generic.c +4 -2
@@ -2857,9 +2857,11 @@ static bool look_for_mix_leaf_ctls(struc
 	if (num_conns < idx)
 		return false;
 	nid = list[idx];
-	if (!*mix_val && nid_has_volume(codec, nid, HDA_OUTPUT))
+	if (!*mix_val && nid_has_volume(codec, nid, HDA_OUTPUT) &&
+	    !is_ctl_associated(codec, nid, HDA_OUTPUT, 0, NID_PATH_VOL_CTL))
 		*mix_val = HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT);
-	if (!*mute_val && nid_has_mute(codec, nid, HDA_OUTPUT))
+	if (!*mute_val && nid_has_mute(codec, nid, HDA_OUTPUT) &&
+	    !is_ctl_associated(codec, nid, HDA_OUTPUT, 0, NID_PATH_MUTE_CTL))
 		*mute_val = HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT);
 
 	return *mix_val || *mute_val;
author maintainer
#80 lore raw

PATCH 81/133 ALSA: Enable CONFIG_ZONE_DMA for smaller PCI DMA masks

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Takashi Iwai <[email protected]>

commit 80ab8eae70e51d578ebbeb228e0f7a562471b8b7 upstream.

The PCI devices with DMA masks smaller than 32bit should enable CONFIG_ZONE_DMA. Since the recent change of page allocator, page allocations via dma_alloc_coherent() with the limited DMA mask bits may fail more frequently, ended up with no available buffers, when CONFIG_ZONE_DMA isn't enabled. With CONFIG_ZONE_DMA, the system has much more chance to obtain such pages.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=68221

---
 sound/pci/Kconfig |   12 ++++++++++++
 1 file changed, 12 insertions(+)
sound/pci/Kconfig +12 -0
@@ -25,6 +25,7 @@ config SND_ALS300
 	select SND_PCM
 	select SND_AC97_CODEC
 	select SND_OPL3_LIB
+	select ZONE_DMA
 	help
 	  Say 'Y' or 'M' to include support for Avance Logic ALS300/ALS300+
 
@@ -49,6 +50,7 @@ config SND_ALI5451
 	tristate "ALi M5451 PCI Audio Controller"
 	select SND_MPU401_UART
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y here to include support for the integrated AC97 sound
 	  device on motherboards using the ALi M5451 Audio Controller
@@ -153,6 +155,7 @@ config SND_AZT3328
 	select SND_PCM
 	select SND_RAWMIDI
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y here to include support for Aztech AZF3328 (PCI168)
 	  soundcards.
@@ -458,6 +461,7 @@ config SND_EMU10K1
 	select SND_HWDEP
 	select SND_RAWMIDI
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y to include support for Sound Blaster PCI 512, Live!,
 	  Audigy and E-mu APS (partially supported) soundcards.
@@ -473,6 +477,7 @@ config SND_EMU10K1X
 	tristate "Emu10k1X (Dell OEM Version)"
 	select SND_AC97_CODEC
 	select SND_RAWMIDI
+	select ZONE_DMA
 	help
 	  Say Y here to include support for the Dell OEM version of the
 	  Sound Blaster Live!.
@@ -506,6 +511,7 @@ config SND_ES1938
 	select SND_OPL3_LIB
 	select SND_MPU401_UART
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y here to include support for soundcards based on ESS Solo-1
 	  (ES1938, ES1946, ES1969) chips.
@@ -517,6 +523,7 @@ config SND_ES1968
 	tristate "ESS ES1968/1978 (Maestro-1/2/2E)"
 	select SND_MPU401_UART
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y here to include support for soundcards based on ESS Maestro
 	  1/2/2E chips.
@@ -605,6 +612,7 @@ config SND_ICE1712
 	select SND_MPU401_UART
 	select SND_AC97_CODEC
 	select BITREVERSE
+	select ZONE_DMA
 	help
 	  Say Y here to include support for soundcards based on the
 	  ICE1712 (Envy24) chip.
@@ -692,6 +700,7 @@ config SND_LX6464ES
 config SND_MAESTRO3
 	tristate "ESS Allegro/Maestro3"
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y here to include support for soundcards based on ESS Maestro 3
 	  (Allegro) chips.
@@ -788,6 +797,7 @@ config SND_SIS7019
 	tristate "SiS 7019 Audio Accelerator"
 	depends on X86 && !X86_64
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y here to include support for the SiS 7019 Audio Accelerator.
 
@@ -799,6 +809,7 @@ config SND_SONICVIBES
 	select SND_OPL3_LIB
 	select SND_MPU401_UART
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y here to include support for soundcards based on the S3
 	  SonicVibes chip.
@@ -810,6 +821,7 @@ config SND_TRIDENT
 	tristate "Trident 4D-Wave DX/NX; SiS 7018"
 	select SND_MPU401_UART
 	select SND_AC97_CODEC
+	select ZONE_DMA
 	help
 	  Say Y here to include support for soundcards based on Trident
 	  4D-Wave DX/NX or SiS 7018 chips.
author maintainer
#81 lore raw

PATCH 82/133 ALSA: bits vs bytes bug in snd_card_create()

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Dan Carpenter <[email protected]>

commit 4c3773eda49c872a3034382f8ec3080002e715bf upstream.

The test here is intended intended to prevent shift wrapping bugs when we do "1U << idx2". We should consider the number of bits in a u32 instead of the number of bytes.

[fix another chunk similarly by tiwai]

Fixes 7bb2491b35a2 ('ALSA: Add kconfig to specify the max card numbers')
---
 sound/core/init.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
sound/core/init.c +2 -2
@@ -170,7 +170,7 @@ int snd_card_create(int idx, const char
 	if (idx < 0) {
 		for (idx2 = 0; idx2 < SNDRV_CARDS; idx2++) {
 			/* idx == -1 == 0xffff means: take any free slot */
-			if (idx2 < sizeof(int) && !(idx & (1U << idx2)))
+			if (idx2 < 32 && !(idx & (1U << idx2)))
 				continue;
 			if (!test_bit(idx2, snd_cards_lock)) {
 				if (module_slot_match(module, idx2)) {
@@ -183,7 +183,7 @@ int snd_card_create(int idx, const char
 	if (idx < 0) {
 		for (idx2 = 0; idx2 < SNDRV_CARDS; idx2++) {
 			/* idx == -1 == 0xffff means: take any free slot */
-			if (idx2 < sizeof(int) && !(idx & (1U << idx2)))
+			if (idx2 < 32 && !(idx & (1U << idx2)))
 				continue;
 			if (!test_bit(idx2, snd_cards_lock)) {
 				if (!slots[idx2] || !*slots[idx2]) {
author maintainer
#82 lore raw

PATCH 84/133 tpm/tpm_i2c_stm_st33: Check return code of get_burstcount

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Peter Huewe <[email protected]>

commit 85c5e0d451125c6ddb78663972e40af810b83644 upstream.

The 'get_burstcount' function can in some circumstances 'return -EBUSY' which in tpm_stm_i2c_send is stored in an 'u32 burstcnt'
thus converting the signed value into an unsigned value, resulting in 'burstcnt' being huge.
Changing the type to u32 only does not solve the problem as the signed value is converted to an unsigned in I2C_WRITE_DATA, resulting in the same effect.

Thus
-> Change type of burstcnt to u32 (the return type of get_burstcount) -> Add a check for the return value of 'get_burstcount' and propagate a potential error.

This makes also sense in the 'I2C_READ_DATA' case, where the there is no signed/unsigned conversion.

found by coverity

---
 drivers/char/tpm/tpm_i2c_stm_st33.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/char/tpm/tpm_i2c_stm_st33.c +6 -1
@@ -410,6 +410,8 @@ static int recv_data(struct tpm_chip *ch
 			     &chip->vendor.read_queue)
 	       == 0) {
 		burstcnt = get_burstcount(chip);
+		if (burstcnt < 0)
+			return burstcnt;
 		len = min_t(int, burstcnt, count - size);
 		I2C_READ_DATA(client, TPM_DATA_FIFO, buf + size, len);
 		size += len;
@@ -451,7 +453,8 @@ static irqreturn_t tpm_ioserirq_handler(
 static int tpm_stm_i2c_send(struct tpm_chip *chip, unsigned char *buf,
 			    size_t len)
 {
-	u32 status, burstcnt = 0, i, size;
+	u32 status, i, size;
+	int burstcnt = 0;
 	int ret;
 	u8 data;
 	struct i2c_client *client;
@@ -482,6 +485,8 @@ static int tpm_stm_i2c_send(struct tpm_c
 
 	for (i = 0; i < len - 1;) {
 		burstcnt = get_burstcount(chip);
+		if (burstcnt < 0)
+			return burstcnt;
 		size = min_t(int, len - i - 1, burstcnt);
 		ret = I2C_WRITE_DATA(client, TPM_DATA_FIFO, buf, size);
 		if (ret < 0)
author maintainer
#83 lore raw

PATCH 85/133 tpm/tpm_ppi: Do not compare strcmp(a,b) == -1

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Peter Huewe <[email protected]>

commit 747d35bd9bb4ae6bd74b19baa5bbe32f3e0cee11 upstream.

Depending on the implementation strcmp might return the difference between two strings not only -1,0,1 consequently
if (strcmp (a,b) == -1)
might lead to taking the wrong branch

-> compare with < 0 instead,
which in any case is more canonical.

---
 drivers/char/tpm/tpm_ppi.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/char/tpm/tpm_ppi.c +4 -4
@@ -172,7 +172,7 @@ static ssize_t tpm_store_ppi_request(str
 	 * is updated with function index from SUBREQ to SUBREQ2 since PPI
 	 * version 1.1
 	 */
-	if (strcmp(version, "1.1") == -1)
+	if (strcmp(version, "1.1") < 0)
 		params[2].integer.value = TPM_PPI_FN_SUBREQ;
 	else
 		params[2].integer.value = TPM_PPI_FN_SUBREQ2;
@@ -182,7 +182,7 @@ static ssize_t tpm_store_ppi_request(str
 	 * string/package type. For PPI version 1.0 and 1.1, use buffer type
 	 * for compatibility, and use package type since 1.2 according to spec.
 	 */
-	if (strcmp(version, "1.2") == -1) {
+	if (strcmp(version, "1.2") < 0) {
 		params[3].type = ACPI_TYPE_BUFFER;
 		params[3].buffer.length = sizeof(req);
 		sscanf(buf, "%d", &req);
@@ -248,7 +248,7 @@ static ssize_t tpm_show_ppi_transition_a
 	 * (e.g. Capella with PPI 1.0) need integer/string/buffer type, so for
 	 * compatibility, define params[3].type as buffer, if PPI version < 1.2
 	 */
-	if (strcmp(version, "1.2") == -1) {
+	if (strcmp(version, "1.2") < 0) {
 		params[3].type = ACPI_TYPE_BUFFER;
 		params[3].buffer.length =  0;
 		params[3].buffer.pointer = NULL;
@@ -390,7 +390,7 @@ static ssize_t show_ppi_operations(char
 	kfree(output.pointer);
 	output.length = ACPI_ALLOCATE_BUFFER;
 	output.pointer = NULL;
-	if (strcmp(version, "1.2") == -1)
+	if (strcmp(version, "1.2") < 0)
 		return -EPERM;
 
 	params[2].integer.value = TPM_PPI_FN_GETOPR;
author maintainer
#84 lore raw

PATCH 86/133 ata: sata_mv: introduce compatible string "marvell, armada-370-sata"

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Simon Guinot <[email protected]>

commit b1f5c73bd5a4752efb7d7af019034044b08aafe9 upstream.

The sata_mv driver supports the SATA IP found in several Marvell SoCs. As some new SATA registers have been introduced with the Armada 370/XP SoCs, a way to identify them is needed.

This patch introduces a new compatible string for the SATA IP found in Armada 370/XP SoCs.

Cc Thomas Petazzoni <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Andrew Lunn <[email protected]>
Cc Gregory Clement <[email protected]>
Cc Sebastian Hesselbarth <[email protected]>
Cc Lior Amsalem <[email protected]>
---
 Documentation/devicetree/bindings/ata/marvell.txt |    2 +-
 drivers/ata/sata_mv.c                             |    1 +
 2 files changed, 2 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/ata/marvell.txt +1 -1
@@ -1,7 +1,7 @@
 * Marvell Orion SATA
 
 Required Properties:
-- compatibility : "marvell,orion-sata"
+- compatibility : "marvell,orion-sata" or "marvell,armada-370-sata"
 - reg           : Address range of controller
 - interrupts    : Interrupt controller is using
 - nr-ports      : Number of SATA ports in use.
drivers/ata/sata_mv.c +1 -0
@@ -4209,6 +4209,7 @@ static int mv_platform_resume(struct pla
 
 #ifdef CONFIG_OF
 static struct of_device_id mv_sata_dt_ids[] = {
+	{ .compatible = "marvell,armada-370-sata", },
 	{ .compatible = "marvell,orion-sata", },
 	{},
 };
author maintainer
#85 lore raw

PATCH 87/133 ata: sata_mv: fix disk hotplug for Armada 370/XP SoCs

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Lior Amsalem <[email protected]>

commit 9013d64e661fc2a37a1742670202171c27fef4b5 upstream.

On Armada 370/XP SoCs, once a disk is removed from a SATA port, then the re-plug events are not detected by the sata_mv driver. This patch fixes the issue by updating the PHY speed in the LP_PHY_CTL register (0x58) according to the SControl speed.

Note that this fix is only applied if the compatible string "marvell,armada-370-sata" is found in the SATA DT node.

Fixes 9ae6f740b49f ("arm: mach-mvebu: add support for Armada 370 and Armada XP with DT")
Cc Thomas Petazzoni <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Andrew Lunn <[email protected]>
Cc Gregory Clement <[email protected]>
Cc Sebastian Hesselbarth <[email protected]>
---
 drivers/ata/sata_mv.c |   24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
drivers/ata/sata_mv.c +24 -0
@@ -304,6 +304,7 @@ enum {
 	MV5_LTMODE		= 0x30,
 	MV5_PHY_CTL		= 0x0C,
 	SATA_IFCFG		= 0x050,
+	LP_PHY_CTL		= 0x058,
 
 	MV_M2_PREAMP_MASK	= 0x7e0,
 
@@ -431,6 +432,7 @@ enum {
 	MV_HP_CUT_THROUGH	= (1 << 10),	/* can use EDMA cut-through */
 	MV_HP_FLAG_SOC		= (1 << 11),	/* SystemOnChip, no PCI */
 	MV_HP_QUIRK_LED_BLINK_EN = (1 << 12),	/* is led blinking enabled? */
+	MV_HP_FIX_LP_PHY_CTL	= (1 << 13),	/* fix speed in LP_PHY_CTL ? */
 
 	/* Port private flags (pp_flags) */
 	MV_PP_FLAG_EDMA_EN	= (1 << 0),	/* is EDMA engine enabled? */
@@ -1358,6 +1360,7 @@ static int mv_scr_write(struct ata_link
 
 	if (ofs != 0xffffffffU) {
 		void __iomem *addr = mv_ap_base(link->ap) + ofs;
+		struct mv_host_priv *hpriv = link->ap->host->private_data;
 		if (sc_reg_in == SCR_CONTROL) {
 			/*
 			 * Workaround for 88SX60x1 FEr SATA#26:
@@ -1374,6 +1377,18 @@ static int mv_scr_write(struct ata_link
 			 */
 			if ((val & 0xf) == 1 || (readl(addr) & 0xf) == 1)
 				val |= 0xf000;
+
+			if (hpriv->hp_flags & MV_HP_FIX_LP_PHY_CTL) {
+				void __iomem *lp_phy_addr =
+					mv_ap_base(link->ap) + LP_PHY_CTL;
+				/*
+				 * Set PHY speed according to SControl speed.
+				 */
+				if ((val & 0xf0) == 0x10)
+					writelfl(0x7, lp_phy_addr);
+				else
+					writelfl(0x227, lp_phy_addr);
+			}
 		}
 		writelfl(val, addr);
 		return 0;
@@ -4110,6 +4125,15 @@ static int mv_platform_probe(struct plat
 	if (rc)
 		goto err;
 
+	/*
+	 * To allow disk hotplug on Armada 370/XP SoCs, the PHY speed must be
+	 * updated in the LP_PHY_CTL register.
+	 */
+	if (pdev->dev.of_node &&
+		of_device_is_compatible(pdev->dev.of_node,
+					"marvell,armada-370-sata"))
+		hpriv->hp_flags |= MV_HP_FIX_LP_PHY_CTL;
+
 	/* initialize adapter */
 	rc = mv_init_host(host);
 	if (rc)
author maintainer
#86 lore raw

PATCH 88/133 ARM: mvebu: update the SATA compatible string for Armada 370/XP

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Simon Guinot <[email protected]>

commit a96cc303e42ad7830dde929aad0046e448a05505 upstream.

This patch updates the Armada 370/XP SATA node with the new compatible string "marvell,armada-370-sata".

Cc Thomas Petazzoni <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Andrew Lunn <[email protected]>
Cc Gregory Clement <[email protected]>
Cc Sebastian Hesselbarth <[email protected]>
Cc Lior Amsalem <[email protected]>
---
 arch/arm/boot/dts/armada-370-xp.dtsi |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/armada-370-xp.dtsi +1 -1
@@ -143,7 +143,7 @@
 			};
 
 			sata@a0000 {
-				compatible = "marvell,orion-sata";
+				compatible = "marvell,armada-370-sata";
 				reg = <0xa0000 0x5000>;
 				interrupts = <55>;
 				clocks = <&gateclk 15>, <&gateclk 30>;
author maintainer
#87 lore raw

PATCH 89/133 libata: disable LPM for some WD SATA-I devices

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Tejun Heo <[email protected]>

commit ecd75ad514d73efc1bbcc5f10a13566c3ace5f53 upstream.

For some reason, some early WD drives spin up and down drives erratically when the link is put into slumber mode which can reduce the life expectancy of the device significantly. Unfortunately, we don't have full list of devices and given the nature of the issue it'd be better to err on the side of false positives than the other way around. Let's disable LPM on all WD devices which match one of the known problematic model prefixes and are SATA-I.

As horkage list doesn't support matching SATA capabilities, this is implemented as two horkages - WD_BROKEN_LPM and NOLPM. The former is set for the known prefixes and sets the latter if the matched device is SATA-I.

Note that this isn't optimal as this disables all LPM operations and partial link power state reportedly works fine on these; however, the way LPM is implemented in libata makes it difficult to precisely map libata LPM setting to specific link power state. Well, these devices are already fairly outdated. Let's just disable whole LPM for now.

References: https://bugzilla.kernel.org/show_bug.cgi?id=57211

---
 drivers/ata/libata-core.c |   27 +++++++++++++++++++++++++++
 drivers/ata/libata-scsi.c |   18 +++++++++++++++---
 include/linux/libata.h    |    2 ++
 3 files changed, 44 insertions(+), 3 deletions(-)
drivers/ata/libata-core.c +27 -0
@@ -2222,6 +2222,16 @@ int ata_dev_configure(struct ata_device
 	if (rc)
 		return rc;
 
+	/* some WD SATA-1 drives have issues with LPM, turn on NOLPM for them */
+	if ((dev->horkage & ATA_HORKAGE_WD_BROKEN_LPM) &&
+	    (id[ATA_ID_SATA_CAPABILITY] & 0xe) == 0x2)
+		dev->horkage |= ATA_HORKAGE_NOLPM;
+
+	if (dev->horkage & ATA_HORKAGE_NOLPM) {
+		ata_dev_warn(dev, "LPM support broken, forcing max_power\n");
+		dev->link->ap->target_lpm_policy = ATA_LPM_MAX_POWER;
+	}
+
 	/* let ACPI work its magic */
 	rc = ata_acpi_on_devcfg(dev);
 	if (rc)
@@ -4216,6 +4226,23 @@ static const struct ata_blacklist_entry
 	{ "Micron_M500*",		NULL,	ATA_HORKAGE_NO_NCQ_TRIM, },
 	{ "Crucial_CT???M500SSD1",	NULL,	ATA_HORKAGE_NO_NCQ_TRIM, },
 
+	/*
+	 * Some WD SATA-I drives spin up and down erratically when the link
+	 * is put into the slumber mode.  We don't have full list of the
+	 * affected devices.  Disable LPM if the device matches one of the
+	 * known prefixes and is SATA-1.  As a side effect LPM partial is
+	 * lost too.
+	 *
+	 * https://bugzilla.kernel.org/show_bug.cgi?id=57211
+	 */
+	{ "WDC WD800JD-*",		NULL,	ATA_HORKAGE_WD_BROKEN_LPM },
+	{ "WDC WD1200JD-*",		NULL,	ATA_HORKAGE_WD_BROKEN_LPM },
+	{ "WDC WD1600JD-*",		NULL,	ATA_HORKAGE_WD_BROKEN_LPM },
+	{ "WDC WD2000JD-*",		NULL,	ATA_HORKAGE_WD_BROKEN_LPM },
+	{ "WDC WD2500JD-*",		NULL,	ATA_HORKAGE_WD_BROKEN_LPM },
+	{ "WDC WD3000JD-*",		NULL,	ATA_HORKAGE_WD_BROKEN_LPM },
+	{ "WDC WD3200JD-*",		NULL,	ATA_HORKAGE_WD_BROKEN_LPM },
+
 	/* End Marker */
 	{ }
 };
drivers/ata/libata-scsi.c +15 -3
@@ -111,12 +111,14 @@ static const char *ata_lpm_policy_names[
 	[ATA_LPM_MIN_POWER]	= "min_power",
 };
 
-static ssize_t ata_scsi_lpm_store(struct device *dev,
+static ssize_t ata_scsi_lpm_store(struct device *device,
 				  struct device_attribute *attr,
 				  const char *buf, size_t count)
 {
-	struct Scsi_Host *shost = class_to_shost(dev);
+	struct Scsi_Host *shost = class_to_shost(device);
 	struct ata_port *ap = ata_shost_to_port(shost);
+	struct ata_link *link;
+	struct ata_device *dev;
 	enum ata_lpm_policy policy;
 	unsigned long flags;
 
@@ -132,10 +134,20 @@ static ssize_t ata_scsi_lpm_store(struct
 		return -EINVAL;
 
 	spin_lock_irqsave(ap->lock, flags);
+
+	ata_for_each_link(link, ap, EDGE) {
+		ata_for_each_dev(dev, &ap->link, ENABLED) {
+			if (dev->horkage & ATA_HORKAGE_NOLPM) {
+				count = -EOPNOTSUPP;
+				goto out_unlock;
+			}
+		}
+	}
+
 	ap->target_lpm_policy = policy;
 	ata_port_schedule_eh(ap);
+out_unlock:
 	spin_unlock_irqrestore(ap->lock, flags);
-
 	return count;
 }
 
include/linux/libata.h +2 -0
@@ -419,6 +419,8 @@ enum {
 	ATA_HORKAGE_MAX_SEC_LBA48 = (1 << 17),	/* Set max sects to 65535 */
 	ATA_HORKAGE_ATAPI_DMADIR = (1 << 18),	/* device requires dmadir */
 	ATA_HORKAGE_NO_NCQ_TRIM	= (1 << 19),	/* don't use queued TRIM */
+	ATA_HORKAGE_NOLPM	= (1 << 20),	/* don't use LPM */
+	ATA_HORKAGE_WD_BROKEN_LPM = (1 << 21),	/* some WDs have broken LPM */
 
 	 /* DMA mask for user DMA control: User visible values; DO NOT
 	    renumber */
author maintainer
#88 lore raw

PATCH 90/133 ext4: avoid clearing beyond i_blocks when truncating an inline data file

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Theodore Ts'o <[email protected]>

commit 09c455aaa8f47a94d5bafaa23d58365768210507 upstream.

A missing cast means that when we are truncating a file which is less than 60 bytes, we don't clear the correct area of memory, and in fact we can end up truncating the next inode in the inode table, or worse yet, some other kernel data structure.

Addresses-Coverity-Id: #751987

---
 fs/ext4/inline.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
fs/ext4/inline.c +5 -3
@@ -1925,9 +1925,11 @@ void ext4_inline_data_truncate(struct in
 		}
 
 		/* Clear the content within i_blocks. */
-		if (i_size < EXT4_MIN_INLINE_DATA_SIZE)
-			memset(ext4_raw_inode(&is.iloc)->i_block + i_size, 0,
-					EXT4_MIN_INLINE_DATA_SIZE - i_size);
+		if (i_size < EXT4_MIN_INLINE_DATA_SIZE) {
+			void *p = (void *) ext4_raw_inode(&is.iloc)->i_block;
+			memset(p + i_size, 0,
+			       EXT4_MIN_INLINE_DATA_SIZE - i_size);
+		}
 
 		EXT4_I(inode)->i_inline_size = i_size <
 					EXT4_MIN_INLINE_DATA_SIZE ?
author maintainer
#89 lore raw

PATCH 91/133 vfs: Remove second variable named error in __dentry_path

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: "Eric W. Biederman" <[email protected]>

commit a8323da0366d3398eda62741d2ac1130c8a172ed upstream.

In commit 232d2d60aa5469bb097f55728f65146bd49c1d25

Author: Waiman Long <[email protected]>
Date:   Mon Sep 9 12:18:13 2013 -0400

dcache: Translating dentry into pathname without taking rename_lock

The __dentry_path locking was changed and the variable error was intended to be moved outside of the loop. Unfortunately the inner declaration of error was not removed. Resulting in a version of __dentry_path that will never return an error.

Remove the problematic inner declaration of error and allow __dentry_path to return errors once again.

---
 fs/dcache.c |    1 -
 1 file changed, 1 deletion(-)
fs/dcache.c +0 -1
@@ -3140,7 +3140,6 @@ restart:
 	read_seqbegin_or_lock(&rename_lock, &seq);
 	while (!IS_ROOT(dentry)) {
 		struct dentry *parent = dentry->d_parent;
-		int error;
 
 		prefetch(parent);
 		error = prepend_name(&end, &len, &dentry->d_name);
author maintainer
#90 lore raw

PATCH 92/133 vfs: Is mounted should be testing mnt_ns for NULL or error.

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: "Eric W. Biederman" <[email protected]>

commit 260a459d2e39761fbd39803497205ce1690bc7b1 upstream.

A bug was introduced with the is_mounted helper function in commit f7a99c5b7c8bd3d3f533c8b38274e33f3da9096e

Author: Al Viro <[email protected]>
Date:   Sat Jun 9 00:59:08 2012 -0400
    get rid of ->mnt_longterm

    it's enough to set ->mnt_ns of internal vfsmounts to something
    distinct from all struct mnt_namespace out there; then we can
    just use the check for ->mnt_ns != NULL in the fast path of
    mntput_no_expire()

    Signed-off-by: Al Viro <[email protected]>

The intent was to test if the real_mount(vfsmount)->mnt_ns was NULL_OR_ERR but the code is actually testing real_mount(vfsmount) and always returning true.

The result is d_absolute_path returning paths it should be hiding.

---
 fs/mount.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/mount.h +1 -1
@@ -73,7 +73,7 @@ static inline int mnt_has_parent(struct
 static inline int is_mounted(struct vfsmount *mnt)
 {
 	/* neither detached nor internal? */
-	return !IS_ERR_OR_NULL(real_mount(mnt));
+	return !IS_ERR_OR_NULL(real_mount(mnt)->mnt_ns);
 }
 
 extern struct mount *__lookup_mnt(struct vfsmount *, struct dentry *, int);
author maintainer
#91 lore raw

PATCH 93/133 bcache: Data corruption fix

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Kent Overstreet <[email protected]>

commit ef71ec00002d92a08eb27e9d036e3d48835b6597 upstream.

The code that handles overlapping extents that we've just read back in from disk was depending on the behaviour of the code that handles overlapping extents as we're inserting into a btree node in the case of an insert that forced an existing extent to be split: on insert, if we had to split we'd also insert a new extent to represent the top part of the old extent - and then that new extent would get written out.

The code that read the extents back in thus not bother with splitting extents - if it saw an extent that ovelapped in the middle of an older extent, it would trim the old extent to only represent the bottom part, assuming that the original insert would've inserted a new extent to represent the top part.

I still haven't figured out _how_ it can happen, but I'm now pretty convinced (and testing has confirmed) that there's some kind of an obscure corner case (probably involving extent merging, and multiple overwrites in different sets) that breaks this. The fix is to change the mergesort fixup code to split extents itself when required.

---
 drivers/md/bcache/bset.c |   26 ++++++++++++++++++++++----
 1 file changed, 22 insertions(+), 4 deletions(-)
drivers/md/bcache/bset.c +22 -4
@@ -935,7 +935,7 @@ static void sort_key_next(struct btree_i
 		*i = iter->data[--iter->used];
 }
 
-static void btree_sort_fixup(struct btree_iter *iter)
+static struct bkey *btree_sort_fixup(struct btree_iter *iter, struct bkey *tmp)
 {
 	while (iter->used > 1) {
 		struct btree_iter_set *top = iter->data, *i = top + 1;
@@ -963,9 +963,22 @@ static void btree_sort_fixup(struct btre
 		} else {
 			/* can't happen because of comparison func */
 			BUG_ON(!bkey_cmp(&START_KEY(top->k), &START_KEY(i->k)));
-			bch_cut_back(&START_KEY(i->k), top->k);
+
+			if (bkey_cmp(i->k, top->k) < 0) {
+				bkey_copy(tmp, top->k);
+
+				bch_cut_back(&START_KEY(i->k), tmp);
+				bch_cut_front(i->k, top->k);
+				heap_sift(iter, 0, btree_iter_cmp);
+
+				return tmp;
+			} else {
+				bch_cut_back(&START_KEY(i->k), top->k);
+			}
 		}
 	}
+
+	return NULL;
 }
 
 static void btree_mergesort(struct btree *b, struct bset *out,
@@ -973,15 +986,20 @@ static void btree_mergesort(struct btree
 			    bool fixup, bool remove_stale)
 {
 	struct bkey *k, *last = NULL;
+	BKEY_PADDED(k) tmp;
 	bool (*bad)(struct btree *, const struct bkey *) = remove_stale
 		? bch_ptr_bad
 		: bch_ptr_invalid;
 
 	while (!btree_iter_end(iter)) {
 		if (fixup && !b->level)
-			btree_sort_fixup(iter);
+			k = btree_sort_fixup(iter, &tmp.k);
+		else
+			k = NULL;
+
+		if (!k)
+			k = bch_btree_iter_next(iter);
 
-		k = bch_btree_iter_next(iter);
 		if (bad(b, k))
 			continue;
 
author maintainer
#92 lore raw

PATCH 94/133 zram: fix race between reset and flushing pending work

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Minchan Kim <[email protected]>

commit da4a04126baa3be03bc566d4a2ee0944c5e783d0 upstream.

Dan and Sergey reported that there is a racy between reset and flushing of pending work so that it could make oops by freeing zram->meta in reset while zram_slot_free can access zram->meta if new request is adding during the race window.

This patch moves flush after taking init_lock so it prevents new request so that it closes the race.

---
 drivers/staging/zram/zram_drv.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/staging/zram/zram_drv.c +2 -2
@@ -552,14 +552,14 @@ static void zram_reset_device(struct zra
 	size_t index;
 	struct zram_meta *meta;
 
-	flush_work(&zram->free_work);
-
 	down_write(&zram->init_lock);
 	if (!zram->init_done) {
 		up_write(&zram->init_lock);
 		return;
 	}
 
+	flush_work(&zram->free_work);
+
 	meta = zram->meta;
 	zram->init_done = 0;
 
author maintainer
#93 lore raw

PATCH 96/133 ARM: mvebu: Add support to get the ID and the revision of a SoC

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Gregory CLEMENT <[email protected]>

commit af8d1c63afcbf36eea06789c92e22d4af118d2fb upstream.

All the mvebu SoCs have information related to their variant and revision that can be read from the PCI control register.

This patch adds support for Armada XP and Armada 370. This reading of the revision and the ID are done before the PCI initialization to avoid any conflicts. Once these data are retrieved, the resources are freed to let the PCI subsystem use it.

Fixes 930ab3d403ae (i2c: mv64xxx: Add I2C Transaction Generator support)
---
 arch/arm/mach-mvebu/Makefile       |    2 
 arch/arm/mach-mvebu/mvebu-soc-id.c |  119 +++++++++++++++++++++++++++++++++++++
 arch/arm/mach-mvebu/mvebu-soc-id.h |   32 +++++++++
 3 files changed, 152 insertions(+), 1 deletion(-)
arch/arm/mach-mvebu/Makefile +1 -1
@@ -3,7 +3,7 @@ ccflags-$(CONFIG_ARCH_MULTIPLATFORM) :=
 
 AFLAGS_coherency_ll.o		:= -Wa,-march=armv7-a
 
-obj-y				 += system-controller.o
+obj-y				 += system-controller.o mvebu-soc-id.o
 obj-$(CONFIG_MACH_ARMADA_370_XP) += armada-370-xp.o
 obj-$(CONFIG_ARCH_MVEBU)	 += coherency.o coherency_ll.o pmsu.o
 obj-$(CONFIG_SMP)                += platsmp.o headsmp.o
arch/arm/mach-mvebu/mvebu-soc-id.c new file +119 -0

119 lines not on this page · raw message

arch/arm/mach-mvebu/mvebu-soc-id.h new file +32 -0
@@ -0,0 +1,32 @@
+/*
+ * Marvell EBU SoC ID and revision definitions.
+ *
+ * Copyright (C) 2014 Marvell Semiconductor
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __LINUX_MVEBU_SOC_ID_H
+#define __LINUX_MVEBU_SOC_ID_H
+
+/* Armada XP ID */
+#define MV78230_DEV_ID	    0x7823
+#define MV78260_DEV_ID	    0x7826
+#define MV78460_DEV_ID	    0x7846
+
+/* Armada XP Revision */
+#define MV78XX0_A0_REV	    0x1
+#define MV78XX0_B0_REV	    0x2
+
+#ifdef CONFIG_ARCH_MVEBU
+int mvebu_get_soc_id(u32 *dev, u32 *rev);
+#else
+static inline int mvebu_get_soc_id(u32 *dev, u32 *rev)
+{
+	return -1;
+}
+#endif
+
+#endif /* __LINUX_MVEBU_SOC_ID_H */
author maintainer
#94 lore raw

PATCH 97/133 ARM: mvebu: Add quirk for i2c for the OpenBlocks AX3-4 board

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Gregory CLEMENT <[email protected]>

commit 85e618a1be2b2092318178d1d66bdad49cbbeeeb upstream.

The first variants of Armada XP SoCs (A0 stepping) have issues related to the i2c controller which prevent to use the offload mechanism and lead to a kernel hang during boot.

This commit add quirk in the mvebu platform code to check the SoC version and then update the compatible string for the i2c controller according to the revision of the SoC. Currently only some OpenBlocks AX3-4 boards are known to use an A0 revision so the check is done only for these boards.

Fixes 930ab3d403ae (i2c: mv64xxx: Add I2C Transaction Generator support)
---
 arch/arm/mach-mvebu/armada-370-xp.c |   32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)
arch/arm/mach-mvebu/armada-370-xp.c +32 -0
@@ -21,6 +21,7 @@
 #include <linux/clocksource.h>
 #include <linux/dma-mapping.h>
 #include <linux/mbus.h>
+#include <linux/slab.h>
 #include <asm/hardware/cache-l2x0.h>
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
@@ -28,6 +29,7 @@
 #include "armada-370-xp.h"
 #include "common.h"
 #include "coherency.h"
+#include "mvebu-soc-id.h"
 
 static void __init armada_370_xp_map_io(void)
 {
@@ -45,8 +47,38 @@ static void __init armada_370_xp_timer_a
 #endif
 }
 
+static void __init i2c_quirk(void)
+{
+	struct device_node *np;
+	u32 dev, rev;
+
+	/*
+	 * Only revisons more recent than A0 support the offload
+	 * mechanism. We can exit only if we are sure that we can
+	 * get the SoC revision and it is more recent than A0.
+	 */
+	if (mvebu_get_soc_id(&rev, &dev) == 0 && dev > MV78XX0_A0_REV)
+		return;
+
+	for_each_compatible_node(np, NULL, "marvell,mv78230-i2c") {
+		struct property *new_compat;
+
+		new_compat = kzalloc(sizeof(*new_compat), GFP_KERNEL);
+
+		new_compat->name = kstrdup("compatible", GFP_KERNEL);
+		new_compat->length = sizeof("marvell,mv78230-a0-i2c");
+		new_compat->value = kstrdup("marvell,mv78230-a0-i2c",
+						GFP_KERNEL);
+
+		of_update_property(np, new_compat);
+	}
+	return;
+}
+
 static void __init armada_370_xp_dt_init(void)
 {
+	if (of_machine_is_compatible("plathome,openblocks-ax3-4"))
+		i2c_quirk();
 	of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL);
 }
 
author maintainer
#95 lore raw

PATCH 98/133 i2c: mv64xxx: Fix bus hang on A0 version of the Armada XP SoCs

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Gregory CLEMENT <[email protected]>

commit 6cf70ae928bae17077efc0d528dec49bc380438b upstream.

The first variants of Armada XP SoCs (A0 stepping) have issues related to the i2c controller which prevent to use the offload mechanism and lead to a kernel hang during boot.

The commit introduces a new the compatible string marvell,mv78230-a0-i2c for the i2c controller. When this compatible string is used the driver disables the offload mechanism and the kernel no more hangs on these SoCs.

Fixes 930ab3d403ae (i2c: mv64xxx: Add I2C Transaction Generator support)
---
 drivers/i2c/busses/i2c-mv64xxx.c |    5 +++++
 1 file changed, 5 insertions(+)
drivers/i2c/busses/i2c-mv64xxx.c +5 -0
@@ -692,6 +692,7 @@ static const struct of_device_id mv64xxx
 	{ .compatible = "allwinner,sun4i-i2c", .data = &mv64xxx_i2c_regs_sun4i},
 	{ .compatible = "marvell,mv64xxx-i2c", .data = &mv64xxx_i2c_regs_mv64xxx},
 	{ .compatible = "marvell,mv78230-i2c", .data = &mv64xxx_i2c_regs_mv64xxx},
+	{ .compatible = "marvell,mv78230-a0-i2c", .data = &mv64xxx_i2c_regs_mv64xxx},
 	{}
 };
 MODULE_DEVICE_TABLE(of, mv64xxx_i2c_of_match_table);
@@ -783,6 +784,10 @@ mv64xxx_of_config(struct mv64xxx_i2c_dat
 		drv_data->errata_delay = true;
 	}
 
+	if (of_device_is_compatible(np, "marvell,mv78230-a0-i2c")) {
+		drv_data->offload_enabled = false;
+		drv_data->errata_delay = true;
+	}
 out:
 	return rc;
 #endif
author maintainer
#96 lore raw

PATCH 99/133 i2c: mv64xxx: Document the newly introduced Armada XP A0 compatible

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Gregory CLEMENT <gregory.clement-wi1+55ScJUtKEb57/[email protected]>

commit f8b94beb7e6a374cb0de531b72377c49857b35ca upstream.

The first variants of Armada XP SoCs (A0 stepping) have issues related to the i2c controller which prevent to use the offload mechanism and lead to a kernel hang during boot.

The commit introduces a new the compatible string
marvell,mv78230-a0-i2c for the i2c controller.

---
 Documentation/devicetree/bindings/i2c/i2c-mv64xxx.txt |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/i2c/i2c-mv64xxx.txt +5 -1
@@ -5,7 +5,11 @@ Required properties :
 
  - reg             : Offset and length of the register set for the device
  - compatible      : Should be "marvell,mv64xxx-i2c" or "allwinner,sun4i-i2c"
-                     or "marvell,mv78230-i2c"
+                     or "marvell,mv78230-i2c" or "marvell,mv78230-a0-i2c"
+                     Note: Only use "marvell,mv78230-a0-i2c" for a very rare,
+                     initial version of the SoC which had broken offload
+                     support.  Linux auto-detects this and sets it
+                     appropriately.
  - interrupts      : The interrupt number
 
 Optional properties :

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

author maintainer
#97 lore raw

PATCH 100/133 i2c: piix4: Add support for AMD ML and CZ SMBus changes

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Shane Huang <[email protected]>

commit 032f708bc4f6da868ec49dac48ddf3670d8035d3 upstream.

The locations of SMBus register base address and enablement bit are changed from AMD ML, which need this patch to be supported.

---
 Documentation/i2c/busses/i2c-piix4 |    2 +-
 drivers/i2c/busses/Kconfig         |    1 +
 drivers/i2c/busses/i2c-piix4.c     |   28 +++++++++++++++++++++++-----
 3 files changed, 25 insertions(+), 6 deletions(-)
Documentation/i2c/busses/i2c-piix4 +1 -1
@@ -13,7 +13,7 @@ Supported adapters:
   * AMD SP5100 (SB700 derivative found on some server mainboards)
     Datasheet: Publicly available at the AMD website
     http://support.amd.com/us/Embedded_TechDocs/44413.pdf
-  * AMD Hudson-2, CZ
+  * AMD Hudson-2, ML, CZ
     Datasheet: Not publicly available
   * Standard Microsystems (SMSC) SLC90E66 (Victory66) southbridge
     Datasheet: Publicly available at the SMSC website http://www.smsc.com
drivers/i2c/busses/Kconfig +1 -0
@@ -151,6 +151,7 @@ config I2C_PIIX4
 	    ATI SB700/SP5100
 	    ATI SB800
 	    AMD Hudson-2
+	    AMD ML
 	    AMD CZ
 	    Serverworks OSB4
 	    Serverworks CSB5
drivers/i2c/busses/i2c-piix4.c +23 -5
@@ -22,7 +22,7 @@
 	Intel PIIX4, 440MX
 	Serverworks OSB4, CSB5, CSB6, HT-1000, HT-1100
 	ATI IXP200, IXP300, IXP400, SB600, SB700/SP5100, SB800
-	AMD Hudson-2, CZ
+	AMD Hudson-2, ML, CZ
 	SMSC Victory66
 
    Note: we assume there can only be one device, with one or more
@@ -235,7 +235,8 @@ static int piix4_setup_sb800(struct pci_
 {
 	unsigned short piix4_smba;
 	unsigned short smba_idx = 0xcd6;
-	u8 smba_en_lo, smba_en_hi, i2ccfg, i2ccfg_offset = 0x10, smb_en;
+	u8 smba_en_lo, smba_en_hi, smb_en, smb_en_status;
+	u8 i2ccfg, i2ccfg_offset = 0x10;
 
 	/* SB800 and later SMBus does not support forcing address */
 	if (force || force_addr) {
@@ -245,7 +246,15 @@ static int piix4_setup_sb800(struct pci_
 	}
 
 	/* Determine the address of the SMBus areas */
-	smb_en = (aux) ? 0x28 : 0x2c;
+	if ((PIIX4_dev->vendor == PCI_VENDOR_ID_AMD &&
+	     PIIX4_dev->device == PCI_DEVICE_ID_AMD_HUDSON2_SMBUS &&
+	     PIIX4_dev->revision >= 0x41) ||
+	    (PIIX4_dev->vendor == PCI_VENDOR_ID_AMD &&
+	     PIIX4_dev->device == 0x790b &&
+	     PIIX4_dev->revision >= 0x49))
+		smb_en = 0x00;
+	else
+		smb_en = (aux) ? 0x28 : 0x2c;
 
 	if (!request_region(smba_idx, 2, "smba_idx")) {
 		dev_err(&PIIX4_dev->dev, "SMBus base address index region "
@@ -258,13 +267,22 @@ static int piix4_setup_sb800(struct pci_
 	smba_en_hi = inb_p(smba_idx + 1);
 	release_region(smba_idx, 2);
 
-	if ((smba_en_lo & 1) == 0) {
+	if (!smb_en) {
+		smb_en_status = smba_en_lo & 0x10;
+		piix4_smba = smba_en_hi << 8;
+		if (aux)
+			piix4_smba |= 0x20;
+	} else {
+		smb_en_status = smba_en_lo & 0x01;
+		piix4_smba = ((smba_en_hi << 8) | smba_en_lo) & 0xffe0;
+	}
+
+	if (!smb_en_status) {
 		dev_err(&PIIX4_dev->dev,
 			"Host SMBus controller not enabled!\n");
 		return -ENODEV;
 	}
 
-	piix4_smba = ((smba_en_hi << 8) | smba_en_lo) & 0xffe0;
 	if (acpi_check_region(piix4_smba, SMBIOSIZE, piix4_driver.name))
 		return -ENODEV;
 
author maintainer
#98 lore raw

PATCH 101/133 bnx2x: fix DMA unmapping of TSO split BDs

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Michal Schmidt <[email protected]>

[ Upstream commit 95e92fd40c967c363ad66b2fd1ce4dcd68132e54 ]

bnx2x triggers warnings with CONFIG_DMA_API_DEBUG=y:

  WARNING: CPU: 0 PID: 2253 at lib/dma-debug.c:887 check_unmap+0xf8/0x920()
  bnx2x 0000:28:00.0: DMA-API: device driver frees DMA memory with
  different size [device address=0x00000000da2b389e] [map size=1490 bytes]
  [unmap size=66 bytes]

The reason is that bnx2x splits a TSO BD into two BDs (headers + data) using one DMA mapping for both, but it uses only the length of the first BD when unmapping.

This patch fixes the bug by unmapping the whole length of the two BDs.

---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c |   13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +9 -4
@@ -160,6 +160,7 @@ static u16 bnx2x_free_tx_pkt(struct bnx2
 	struct sk_buff *skb = tx_buf->skb;
 	u16 bd_idx = TX_BD(tx_buf->first_bd), new_cons;
 	int nbd;
+	u16 split_bd_len = 0;
 
 	/* prefetch skb end pointer to speedup dev_kfree_skb() */
 	prefetch(&skb->end);
@@ -167,10 +168,7 @@ static u16 bnx2x_free_tx_pkt(struct bnx2
 	DP(NETIF_MSG_TX_DONE, "fp[%d]: pkt_idx %d  buff @(%p)->skb %p\n",
 	   txdata->txq_index, idx, tx_buf, skb);
 
-	/* unmap first bd */
 	tx_start_bd = &txdata->tx_desc_ring[bd_idx].start_bd;
-	dma_unmap_single(&bp->pdev->dev, BD_UNMAP_ADDR(tx_start_bd),
-			 BD_UNMAP_LEN(tx_start_bd), DMA_TO_DEVICE);
 
 	nbd = le16_to_cpu(tx_start_bd->nbd) - 1;
 #ifdef BNX2X_STOP_ON_ERROR
@@ -188,12 +186,19 @@ static u16 bnx2x_free_tx_pkt(struct bnx2
 	--nbd;
 	bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
 
-	/* ...and the TSO split header bd since they have no mapping */
+	/* TSO headers+data bds share a common mapping. See bnx2x_tx_split() */
 	if (tx_buf->flags & BNX2X_TSO_SPLIT_BD) {
+		tx_data_bd = &txdata->tx_desc_ring[bd_idx].reg_bd;
+		split_bd_len = BD_UNMAP_LEN(tx_data_bd);
 		--nbd;
 		bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
 	}
 
+	/* unmap first bd */
+	dma_unmap_single(&bp->pdev->dev, BD_UNMAP_ADDR(tx_start_bd),
+			 BD_UNMAP_LEN(tx_start_bd) + split_bd_len,
+			 DMA_TO_DEVICE);
+
 	/* now free frags */
 	while (nbd > 0) {
 
author maintainer
#99 lore raw

PATCH 103/133 ieee802154: Fix memory leak in ieee802154_add_iface()

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Christian Engelmayer <[email protected]>

[ Upstream commit 267d29a69c6af39445f36102a832b25ed483f299 ]

Fix a memory leak in the ieee802154_add_iface() error handling path. Detected by Coverity: CID 710490.

---
 net/ieee802154/nl-phy.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/ieee802154/nl-phy.c +4 -2
@@ -224,8 +224,10 @@ static int ieee802154_add_iface(struct s
 
 	if (info->attrs[IEEE802154_ATTR_DEV_TYPE]) {
 		type = nla_get_u8(info->attrs[IEEE802154_ATTR_DEV_TYPE]);
-		if (type >= __IEEE802154_DEV_MAX)
-			return -EINVAL;
+		if (type >= __IEEE802154_DEV_MAX) {
+			rc = -EINVAL;
+			goto nla_put_failure;
+		}
 	}
 
 	dev = phy->add_iface(phy, devname, type);
author maintainer
#100 lore raw

PATCH 104/133 net: avoid reference counter overflows on fib_rules in multicast forwarding

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Hannes Frederic Sowa <[email protected]>

[ Upstream commit 95f4a45de1a0f172b35451fc52283290adb21f6e ]

Bob Falken reported that after 4G packets, multicast forwarding stopped working. This was because of a rule reference counter overflow which freed the rule as soon as the overflow happend.

This patch solves this by adding the FIB_LOOKUP_NOREF flag to fib_rules_lookup calls. This is safe even from non-rcu locked sections as in this case the flag only implies not taking a reference to the rule, which we don't need at all.

Rules only hold references to the namespace, which are guaranteed to be available during the call of the non-rcu protected function reg_vif_xmit because of the interface reference which itself holds a reference to the net namespace.

Fixes f0ad0860d01e47 ("ipv4: ipmr: support multiple tables")
Fixes d1db275dd3f6e4 ("ipv6: ip6mr: support multiple tables")
Cc Patrick McHardy <[email protected]>
Cc Thomas Graf <[email protected]>
Cc Julian Anastasov <[email protected]>
Cc Eric Dumazet <[email protected]>
---
 net/ipv4/ipmr.c  |    7 +++++--
 net/ipv6/ip6mr.c |    7 +++++--
 2 files changed, 10 insertions(+), 4 deletions(-)
net/ipv4/ipmr.c +5 -2
@@ -157,9 +157,12 @@ static struct mr_table *ipmr_get_table(s
 static int ipmr_fib_lookup(struct net *net, struct flowi4 *flp4,
 			   struct mr_table **mrt)
 {
-	struct ipmr_result res;
-	struct fib_lookup_arg arg = { .result = &res, };
 	int err;
+	struct ipmr_result res;
+	struct fib_lookup_arg arg = {
+		.result = &res,
+		.flags = FIB_LOOKUP_NOREF,
+	};
 
 	err = fib_rules_lookup(net->ipv4.mr_rules_ops,
 			       flowi4_to_flowi(flp4), 0, &arg);
net/ipv6/ip6mr.c +5 -2
@@ -141,9 +141,12 @@ static struct mr6_table *ip6mr_get_table
 static int ip6mr_fib_lookup(struct net *net, struct flowi6 *flp6,
 			    struct mr6_table **mrt)
 {
-	struct ip6mr_result res;
-	struct fib_lookup_arg arg = { .result = &res, };
 	int err;
+	struct ip6mr_result res;
+	struct fib_lookup_arg arg = {
+		.result = &res,
+		.flags = FIB_LOOKUP_NOREF,
+	};
 
 	err = fib_rules_lookup(net->ipv6.mr6_rules_ops,
 			       flowi6_to_flowi(flp6), 0, &arg);
author maintainer
#101 lore raw

PATCH 105/133 net,via-rhine: Fix tx_timeout handling

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Richard Weinberger <[email protected]>

[ Upstream commit a926592f5e4e900f3fa903298c4619a131e60963 ]

rhine_reset_task() misses to disable the tx scheduler upon reset, this can lead to a crash if work is still scheduled while we're resetting the tx queue.

Fixes:

[   93.591707] BUG: unable to handle kernel NULL pointer dereference at 0000004c
[   93.595514] IP: [<c119d10d>] rhine_napipoll+0x491/0x6
---
 drivers/net/ethernet/via/via-rhine.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/via/via-rhine.c +1 -0
@@ -1615,6 +1615,7 @@ static void rhine_reset_task(struct work
 		goto out_unlock;
 
 	napi_disable(&rp->napi);
+	netif_tx_disable(dev);
 	spin_lock_bh(&rp->lock);
 
 	/* clear all descriptors */
author maintainer
#102 lore raw

PATCH 106/133 net: rds: fix per-cpu helper usage

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Gerald Schaefer <[email protected]>

[ Upstream commit c196403b79aa241c3fefb3ee5bb328aa7c5cc860 ]

commit ae4b46e9d "net: rds: use this_cpu_* per-cpu helper" broke per-cpu handling for rds. chpfirst is the result of __this_cpu_read(), so it is an absolute pointer and not __percpu. Therefore, __this_cpu_write() should not operate on chpfirst, but rather on cache->percpu->first, just like __this_cpu_read() did before.

Signed-off-byd Gerald Schaefer <[email protected]>

---
 net/rds/ib_recv.c |    7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
net/rds/ib_recv.c +3 -4
@@ -421,8 +421,7 @@ static void rds_ib_recv_cache_put(struct
 				 struct rds_ib_refill_cache *cache)
 {
 	unsigned long flags;
-	struct list_head *old;
-	struct list_head __percpu *chpfirst;
+	struct list_head *old, *chpfirst;
 
 	local_irq_save(flags);
 
@@ -432,7 +431,7 @@ static void rds_ib_recv_cache_put(struct
 	else /* put on front */
 		list_add_tail(new_item, chpfirst);
 
-	__this_cpu_write(chpfirst, new_item);
+	__this_cpu_write(cache->percpu->first, new_item);
 	__this_cpu_inc(cache->percpu->count);
 
 	if (__this_cpu_read(cache->percpu->count) < RDS_IB_RECYCLE_BATCH_COUNT)
@@ -452,7 +451,7 @@ static void rds_ib_recv_cache_put(struct
 	} while (old);
 
 
-	__this_cpu_write(chpfirst, NULL);
+	__this_cpu_write(cache->percpu->first, NULL);
 	__this_cpu_write(cache->percpu->count, 0);
 end:
 	local_irq_restore(flags);
author maintainer
#103 lore raw

PATCH 107/133 tcp: metrics: Avoid duplicate entries with the same destination-IP

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Christoph Paasch <[email protected]>

[ Upstream commit 77f99ad16a07aa062c2d30fae57b1fee456f6ef6 ]

Because the tcp-metrics is an RCU-list, it may be that two soft-interrupts are inside __tcp_get_metrics() for the same destination-IP at the same time. If this destination-IP is not yet part of the tcp-metrics, both soft-interrupts will end up in tcpm_new and create a new entry for this IP.
So, we will have two tcp-metrics with the same destination-IP in the list.

This patch checks twice __tcp_get_metrics(). First without holding the lock, then while holding the lock. The second one is there to confirm that the entry has not been added by another soft-irq while waiting for the spin-lock.

Fixes 51c5d0c4b169b (tcp: Maintain dynamic metrics in local cache.)
---
 net/ipv4/tcp_metrics.c |   51 ++++++++++++++++++++++++++++++-------------------
 1 file changed, 32 insertions(+), 19 deletions(-)
net/ipv4/tcp_metrics.c +32 -19
@@ -22,6 +22,9 @@
 
 int sysctl_tcp_nometrics_save __read_mostly;
 
+static struct tcp_metrics_block *__tcp_get_metrics(const struct inetpeer_addr *addr,
+						   struct net *net, unsigned int hash);
+
 struct tcp_fastopen_metrics {
 	u16	mss;
 	u16	syn_loss:10;		/* Recurring Fast Open SYN losses */
@@ -130,16 +133,41 @@ static void tcpm_suck_dst(struct tcp_met
 	}
 }
 
+#define TCP_METRICS_TIMEOUT		(60 * 60 * HZ)
+
+static void tcpm_check_stamp(struct tcp_metrics_block *tm, struct dst_entry *dst)
+{
+	if (tm && unlikely(time_after(jiffies, tm->tcpm_stamp + TCP_METRICS_TIMEOUT)))
+		tcpm_suck_dst(tm, dst, false);
+}
+
+#define TCP_METRICS_RECLAIM_DEPTH	5
+#define TCP_METRICS_RECLAIM_PTR		(struct tcp_metrics_block *) 0x1UL
+
 static struct tcp_metrics_block *tcpm_new(struct dst_entry *dst,
 					  struct inetpeer_addr *addr,
-					  unsigned int hash,
-					  bool reclaim)
+					  unsigned int hash)
 {
 	struct tcp_metrics_block *tm;
 	struct net *net;
+	bool reclaim = false;
 
 	spin_lock_bh(&tcp_metrics_lock);
 	net = dev_net(dst->dev);
+
+	/* While waiting for the spin-lock the cache might have been populated
+	 * with this entry and so we have to check again.
+	 */
+	tm = __tcp_get_metrics(addr, net, hash);
+	if (tm == TCP_METRICS_RECLAIM_PTR) {
+		reclaim = true;
+		tm = NULL;
+	}
+	if (tm) {
+		tcpm_check_stamp(tm, dst);
+		goto out_unlock;
+	}
+
 	if (unlikely(reclaim)) {
 		struct tcp_metrics_block *oldest;
 
@@ -169,17 +197,6 @@ out_unlock:
 	return tm;
 }
 
-#define TCP_METRICS_TIMEOUT		(60 * 60 * HZ)
-
-static void tcpm_check_stamp(struct tcp_metrics_block *tm, struct dst_entry *dst)
-{
-	if (tm && unlikely(time_after(jiffies, tm->tcpm_stamp + TCP_METRICS_TIMEOUT)))
-		tcpm_suck_dst(tm, dst, false);
-}
-
-#define TCP_METRICS_RECLAIM_DEPTH	5
-#define TCP_METRICS_RECLAIM_PTR		(struct tcp_metrics_block *) 0x1UL
-
 static struct tcp_metrics_block *tcp_get_encode(struct tcp_metrics_block *tm, int depth)
 {
 	if (tm)
@@ -280,7 +297,6 @@ static struct tcp_metrics_block *tcp_get
 	struct inetpeer_addr addr;
 	unsigned int hash;
 	struct net *net;
-	bool reclaim;
 
 	addr.family = sk->sk_family;
 	switch (addr.family) {
@@ -300,13 +316,10 @@ static struct tcp_metrics_block *tcp_get
 	hash = hash_32(hash, net->ipv4.tcp_metrics_hash_log);
 
 	tm = __tcp_get_metrics(&addr, net, hash);
-	reclaim = false;
-	if (tm == TCP_METRICS_RECLAIM_PTR) {
-		reclaim = true;
+	if (tm == TCP_METRICS_RECLAIM_PTR)
 		tm = NULL;
-	}
 	if (!tm && create)
-		tm = tcpm_new(dst, &addr, hash, reclaim);
+		tm = tcpm_new(dst, &addr, hash);
 	else
 		tcpm_check_stamp(tm, dst);
 
author maintainer
#104 lore raw

PATCH 108/133 bpf: do not use reciprocal divide

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Eric Dumazet <[email protected]>

[ Upstream commit aee636c4809fa54848ff07a899b326eb1f9987a2 ]

At first Jakub Zawadzki noticed that some divisions by reciprocal_divide were not correct. (off by one in some cases)
http://www.wireshark.org/~darkjames/reciprocal-buggy.c

He could also show this with BPF:
http://www.wireshark.org/~darkjames/set-and-dump-filter-k-bug.c

The reciprocal divide in linux kernel is not generic enough, lets remove its use in BPF, as it is not worth the pain with current cpus.

Cc Mircea Gherzan <[email protected]>
Cc Daniel Borkmann <[email protected]>
Cc Hannes Frederic Sowa <[email protected]>
Cc Matt Evans <[email protected]>
Cc Martin Schwidefsky <[email protected]>
Cc Heiko Carstens <[email protected]>
Cc David S. Miller <[email protected]>
---
 arch/arm/net/bpf_jit_32.c       |    6 +++---
 arch/powerpc/net/bpf_jit_comp.c |    7 ++++---
 arch/s390/net/bpf_jit_comp.c    |   17 ++++++++++++-----
 arch/sparc/net/bpf_jit_comp.c   |   17 ++++++++++++++---
 arch/x86/net/bpf_jit_comp.c     |   14 ++++++++++----
 net/core/filter.c               |   30 ++----------------------------
 6 files changed, 45 insertions(+), 46 deletions(-)
arch/arm/net/bpf_jit_32.c +3 -3
@@ -637,10 +637,10 @@ load_ind:
 			emit(ARM_MUL(r_A, r_A, r_X), ctx);
 			break;
 		case BPF_S_ALU_DIV_K:
-			/* current k == reciprocal_value(userspace k) */
+			if (k == 1)
+				break;
 			emit_mov_i(r_scratch, k, ctx);
-			/* A = top 32 bits of the product */
-			emit(ARM_UMULL(r_scratch, r_A, r_A, r_scratch), ctx);
+			emit_udiv(r_A, r_A, r_scratch, ctx);
 			break;
 		case BPF_S_ALU_DIV_X:
 			update_on_xread(ctx);
arch/powerpc/net/bpf_jit_comp.c +4 -3
@@ -209,10 +209,11 @@ static int bpf_jit_build_body(struct sk_
 			}
 			PPC_DIVWU(r_A, r_A, r_X);
 			break;
-		case BPF_S_ALU_DIV_K: /* A = reciprocal_divide(A, K); */
+		case BPF_S_ALU_DIV_K: /* A /= K */
+			if (K == 1)
+				break;
 			PPC_LI32(r_scratch1, K);
-			/* Top 32 bits of 64bit result -> A */
-			PPC_MULHWU(r_A, r_A, r_scratch1);
+			PPC_DIVWU(r_A, r_A, r_scratch1);
 			break;
 		case BPF_S_ALU_AND_X:
 			ctx->seen |= SEEN_XREG;
arch/s390/net/bpf_jit_comp.c +12 -5
@@ -371,11 +371,13 @@ static int bpf_jit_insn(struct bpf_jit *
 		/* dr %r4,%r12 */
 		EMIT2(0x1d4c);
 		break;
-	case BPF_S_ALU_DIV_K: /* A = reciprocal_divide(A, K) */
-		/* m %r4,<d(K)>(%r13) */
-		EMIT4_DISP(0x5c40d000, EMIT_CONST(K));
-		/* lr %r5,%r4 */
-		EMIT2(0x1854);
+	case BPF_S_ALU_DIV_K: /* A /= K */
+		if (K == 1)
+			break;
+		/* lhi %r4,0 */
+		EMIT4(0xa7480000);
+		/* d %r4,<d(K)>(%r13) */
+		EMIT4_DISP(0x5d40d000, EMIT_CONST(K));
 		break;
 	case BPF_S_ALU_MOD_X: /* A %= X */
 		jit->seen |= SEEN_XREG | SEEN_RET0;
@@ -391,6 +393,11 @@ static int bpf_jit_insn(struct bpf_jit *
 		EMIT2(0x1854);
 		break;
 	case BPF_S_ALU_MOD_K: /* A %= K */
+		if (K == 1) {
+			/* lhi %r5,0 */
+			EMIT4(0xa7580000);
+			break;
+		}
 		/* lhi %r4,0 */
 		EMIT4(0xa7480000);
 		/* d %r4,<d(K)>(%r13) */
arch/sparc/net/bpf_jit_comp.c +14 -3
@@ -497,9 +497,20 @@ void bpf_jit_compile(struct sk_filter *f
 			case BPF_S_ALU_MUL_K:	/* A *= K */
 				emit_alu_K(MUL, K);
 				break;
-			case BPF_S_ALU_DIV_K:	/* A /= K */
-				emit_alu_K(MUL, K);
-				emit_read_y(r_A);
+			case BPF_S_ALU_DIV_K:	/* A /= K with K != 0*/
+				if (K == 1)
+					break;
+				emit_write_y(G0);
+#ifdef CONFIG_SPARC32
+				/* The Sparc v8 architecture requires
+				 * three instructions between a %y
+				 * register write and the first use.
+				 */
+				emit_nop();
+				emit_nop();
+				emit_nop();
+#endif
+				emit_alu_K(DIV, K);
 				break;
 			case BPF_S_ALU_DIV_X:	/* A /= X; */
 				emit_cmpi(r_X, 0);
arch/x86/net/bpf_jit_comp.c +10 -4
@@ -359,15 +359,21 @@ void bpf_jit_compile(struct sk_filter *f
 				EMIT2(0x89, 0xd0);	/* mov %edx,%eax */
 				break;
 			case BPF_S_ALU_MOD_K: /* A %= K; */
+				if (K == 1) {
+					CLEAR_A();
+					break;
+				}
 				EMIT2(0x31, 0xd2);	/* xor %edx,%edx */
 				EMIT1(0xb9);EMIT(K, 4);	/* mov imm32,%ecx */
 				EMIT2(0xf7, 0xf1);	/* div %ecx */
 				EMIT2(0x89, 0xd0);	/* mov %edx,%eax */
 				break;
-			case BPF_S_ALU_DIV_K: /* A = reciprocal_divide(A, K); */
-				EMIT3(0x48, 0x69, 0xc0); /* imul imm32,%rax,%rax */
-				EMIT(K, 4);
-				EMIT4(0x48, 0xc1, 0xe8, 0x20); /* shr $0x20,%rax */
+			case BPF_S_ALU_DIV_K: /* A /= K */
+				if (K == 1)
+					break;
+				EMIT2(0x31, 0xd2);	/* xor %edx,%edx */
+				EMIT1(0xb9);EMIT(K, 4);	/* mov imm32,%ecx */
+				EMIT2(0xf7, 0xf1);	/* div %ecx */
 				break;
 			case BPF_S_ALU_AND_X:
 				seen |= SEEN_XREG;
net/core/filter.c +2 -28
@@ -36,7 +36,6 @@
 #include <asm/uaccess.h>
 #include <asm/unaligned.h>
 #include <linux/filter.h>
-#include <linux/reciprocal_div.h>
 #include <linux/ratelimit.h>
 #include <linux/seccomp.h>
 #include <linux/if_vlan.h>
@@ -166,7 +165,7 @@ unsigned int sk_run_filter(const struct
 			A /= X;
 			continue;
 		case BPF_S_ALU_DIV_K:
-			A = reciprocal_divide(A, K);
+			A /= K;
 			continue;
 		case BPF_S_ALU_MOD_X:
 			if (X == 0)
@@ -553,11 +552,6 @@ int sk_chk_filter(struct sock_filter *fi
 		/* Some instructions need special checks */
 		switch (code) {
 		case BPF_S_ALU_DIV_K:
-			/* check for division by zero */
-			if (ftest->k == 0)
-				return -EINVAL;
-			ftest->k = reciprocal_value(ftest->k);
-			break;
 		case BPF_S_ALU_MOD_K:
 			/* check for division by zero */
 			if (ftest->k == 0)
@@ -853,27 +847,7 @@ void sk_decode_filter(struct sock_filter
 	to->code = decodes[code];
 	to->jt = filt->jt;
 	to->jf = filt->jf;
-
-	if (code == BPF_S_ALU_DIV_K) {
-		/*
-		 * When loaded this rule user gave us X, which was
-		 * translated into R = r(X). Now we calculate the
-		 * RR = r(R) and report it back. If next time this
-		 * value is loaded and RRR = r(RR) is calculated
-		 * then the R == RRR will be true.
-		 *
-		 * One exception. X == 1 translates into R == 0 and
-		 * we can't calculate RR out of it with r().
-		 */
-
-		if (filt->k == 0)
-			to->k = 1;
-		else
-			to->k = reciprocal_value(filt->k);
-
-		BUG_ON(reciprocal_value(to->k) != filt->k);
-	} else
-		to->k = filt->k;
+	to->k = filt->k;
 }
 
 int sk_get_filter(struct sock *sk, struct sock_filter __user *ubuf, unsigned int len)
author maintainer
#105 lore raw

PATCH 109/133 s390/bpf,jit: fix 32 bit divisions, use unsigned divide instructions

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Heiko Carstens <[email protected]>

[ Upstream commit 3af57f78c38131b7a66e2b01e06fdacae01992a3 ]

The s390 bpf jit compiler emits the signed divide instructions "dr" and "d" for unsigned divisions.
This can cause problems: the dividend will be zero extended to a 64 bit value and the divisor is the 32 bit signed value as specified A or X accumulator, even though A and X are supposed to be treated as unsigned values.

The divide instrunctions will generate an exception if the result cannot be expressed with a 32 bit signed value.
This is the case if e.g. the dividend is 0xffffffff and the divisor either 1 or also 0xffffffff (signed: -1).

To avoid all these issues simply use unsigned divide instructions.

---
 arch/s390/net/bpf_jit_comp.c |   16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
arch/s390/net/bpf_jit_comp.c +8 -8
@@ -368,16 +368,16 @@ static int bpf_jit_insn(struct bpf_jit *
 		EMIT4_PCREL(0xa7840000, (jit->ret0_ip - jit->prg));
 		/* lhi %r4,0 */
 		EMIT4(0xa7480000);
-		/* dr %r4,%r12 */
-		EMIT2(0x1d4c);
+		/* dlr %r4,%r12 */
+		EMIT4(0xb997004c);
 		break;
 	case BPF_S_ALU_DIV_K: /* A /= K */
 		if (K == 1)
 			break;
 		/* lhi %r4,0 */
 		EMIT4(0xa7480000);
-		/* d %r4,<d(K)>(%r13) */
-		EMIT4_DISP(0x5d40d000, EMIT_CONST(K));
+		/* dl %r4,<d(K)>(%r13) */
+		EMIT6_DISP(0xe340d000, 0x0097, EMIT_CONST(K));
 		break;
 	case BPF_S_ALU_MOD_X: /* A %= X */
 		jit->seen |= SEEN_XREG | SEEN_RET0;
@@ -387,8 +387,8 @@ static int bpf_jit_insn(struct bpf_jit *
 		EMIT4_PCREL(0xa7840000, (jit->ret0_ip - jit->prg));
 		/* lhi %r4,0 */
 		EMIT4(0xa7480000);
-		/* dr %r4,%r12 */
-		EMIT2(0x1d4c);
+		/* dlr %r4,%r12 */
+		EMIT4(0xb997004c);
 		/* lr %r5,%r4 */
 		EMIT2(0x1854);
 		break;
@@ -400,8 +400,8 @@ static int bpf_jit_insn(struct bpf_jit *
 		}
 		/* lhi %r4,0 */
 		EMIT4(0xa7480000);
-		/* d %r4,<d(K)>(%r13) */
-		EMIT4_DISP(0x5d40d000, EMIT_CONST(K));
+		/* dl %r4,<d(K)>(%r13) */
+		EMIT6_DISP(0xe340d000, 0x0097, EMIT_CONST(K));
 		/* lr %r5,%r4 */
 		EMIT2(0x1854);
 		break;
author maintainer
#106 lore raw

PATCH 110/133 ip_tunnel: clear IPCB in ip_tunnel_xmit() in case dst_link_failure() is called

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Duan Jiong <[email protected]>

[ Upstream commit 11c21a307d79ea5f6b6fc0d3dfdeda271e5e65f6 ]

commit a622260254ee48("ip_tunnel: fix kernel panic with icmp_dest_unreach") clear IPCB in ip_tunnel_xmit() , or else skb->cb[] may contain garbage from GSO segmentation layer.

But commit 0e6fbc5b6c621("ip_tunnels: extend iptunnel_xmit()") refactor codes, and it clear IPCB behind the dst_link_failure().

So clear IPCB in ip_tunnel_xmit() just like commti a622260254ee48("ip_tunnel:
fix kernel panic with icmp_dest_unreach").

---
 net/ipv4/ip_tunnel.c |    1 +
 1 file changed, 1 insertion(+)
net/ipv4/ip_tunnel.c +1 -0
@@ -618,6 +618,7 @@ void ip_tunnel_xmit(struct sk_buff *skb,
 				tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
 			tunnel->err_count--;
 
+			memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
 			dst_link_failure(skb);
 		} else
 			tunnel->err_count = 0;
author maintainer
#107 lore raw

PATCH 111/133 net/vxlan: Share RX skb de-marking and checksum checks with ovs

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Or Gerlitz <[email protected]>

[ Upstream commit d0bc65557ad09a57b4db176e9e3ccddb26971453 ]

Make sure the practice set by commit 0afb166 "vxlan: Add capability of Rx checksum offload for inner packet" is applied when the skb goes through the portion of the RX code which is shared between vxlan netdevices and ovs vxlan port instances.

---
 drivers/net/vxlan.c |   20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)
drivers/net/vxlan.c +9 -11
@@ -1051,6 +1051,15 @@ static int vxlan_udp_encap_recv(struct s
 	if (!vs)
 		goto drop;
 
+	/* If the NIC driver gave us an encapsulated packet
+	 * with the encapsulation mark, the device checksummed it
+	 * for us. Otherwise force the upper layers to verify it.
+	 */
+	if (skb->ip_summed != CHECKSUM_UNNECESSARY || !skb->encapsulation)
+		skb->ip_summed = CHECKSUM_NONE;
+
+	skb->encapsulation = 0;
+
 	vs->rcv(vs, skb, vxh->vx_vni);
 	return 0;
 
@@ -1109,17 +1118,6 @@ static void vxlan_rcv(struct vxlan_sock
 
 	skb_reset_network_header(skb);
 
-	/* If the NIC driver gave us an encapsulated packet with
-	 * CHECKSUM_UNNECESSARY and Rx checksum feature is enabled,
-	 * leave the CHECKSUM_UNNECESSARY, the device checksummed it
-	 * for us. Otherwise force the upper layers to verify it.
-	 */
-	if (skb->ip_summed != CHECKSUM_UNNECESSARY || !skb->encapsulation ||
-	    !(vxlan->dev->features & NETIF_F_RXCSUM))
-		skb->ip_summed = CHECKSUM_NONE;
-
-	skb->encapsulation = 0;
-
 	if (oip6)
 		err = IP6_ECN_decapsulate(oip6, skb);
 	if (oip)
author maintainer
#108 lore raw

PATCH 112/133 fib_frontend: fix possible NULL pointer dereference

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Oliver Hartkopp <[email protected]>

[ Upstream commit a0065f266a9b5d51575535a25c15ccbeed9a9966 ]

The two commits 0115e8e30d (net: remove delay at device dismantle) and 748e2d9396a (net: reinstate rtnl in call_netdevice_notifiers()) silently removed a NULL pointer check for in_dev since Linux 3.7.

This patch re-introduces this check as it causes crashing the kernel when setting small mtu values on non-ip capable netdevices.

---
 net/ipv4/fib_frontend.c |    2 ++
 1 file changed, 2 insertions(+)
net/ipv4/fib_frontend.c +2 -0
@@ -1049,6 +1049,8 @@ static int fib_netdev_event(struct notif
 	}
 
 	in_dev = __in_dev_get_rtnl(dev);
+	if (!in_dev)
+		return NOTIFY_DONE;
 
 	switch (event) {
 	case NETDEV_UP:
author maintainer
#109 lore raw

PATCH 113/133 net: Fix memory leak if TPROXY used with TCP early demux

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Holger Eitzenberger <[email protected]>

[ Upstream commit a452ce345d63ddf92cd101e4196569f8718ad319 ]

I see a memory leak when using a transparent HTTP proxy using TPROXY together with TCP early demux and Kernel v3.8.13.15 (Ubuntu stable):

unreferenced object 0xffff88008cba4a40 (size 1696):
comm "softirq", pid 0, jiffies 4294944115 (age 8907.520s) hex dump (first 32 bytes):

    0a e0 20 6a 40 04 1b 37 92 be 32 e2 e8 b4 00 00  .. [email protected].....
    02 00 07 01 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<ffffffff810b710a>] kmem_cache_alloc+0xad/0xb9
    [<ffffffff81270185>] sk_prot_alloc+0x29/0xc5
    [<ffffffff812702cf>] sk_clone_lock+0x14/0x283
    [<ffffffff812aaf3a>] inet_csk_clone_lock+0xf/0x7b
    [<ffffffff8129a893>] netlink_broadcast+0x14/0x16
    [<ffffffff812c1573>] tcp_create_openreq_child+0x1b/0x4c3
    [<ffffffff812c033e>] tcp_v4_syn_recv_sock+0x38/0x25d
    [<ffffffff812c13e4>] tcp_check_req+0x25c/0x3d0
    [<ffffffff812bf87a>] tcp_v4_do_rcv+0x287/0x40e
    [<ffffffff812a08a7>] ip_route_input_noref+0x843/0xa55
    [<ffffffff812bfeca>] tcp_v4_rcv+0x4c9/0x725
    [<ffffffff812a26f4>] ip_local_deliver_finish+0xe9/0x154
    [<ffffffff8127a927>] __netif_receive_skb+0x4b2/0x514
    [<ffffffff8127aa77>] process_backlog+0xee/0x1c5
    [<ffffffff8127c949>] net_rx_action+0xa7/0x200
    [<ffffffff81209d86>] add_interrupt_randomness+0x39/0x157

But there are many more, resulting in the machine going OOM after some days.

Quoted 1 line
Quoted 1 line

From looking at the TPROXY code, and with help from Florian, I see

that the memory leak is introduced in tcp_v4_early_demux():

  void tcp_v4_early_demux(struct sk_buff *skb)
  {
    /* ... */
    iph = ip_hdr(skb);
    th = tcp_hdr(skb);

    if (th->doff < sizeof(struct tcphdr) / 4)
        return;

    sk = __inet_lookup_established(dev_net(skb->dev), &tcp_hashinfo,
                       iph->saddr, th->source,
                       iph->daddr, ntohs(th->dest),
                       skb->skb_iif);
    if (sk) {
        skb->sk = sk;

where the socket is assigned unconditionally to skb->sk, also bumping the refcnt on it. This is problematic, because in our case the skb has already a socket assigned in the TPROXY target. This then results in the leak I see.

The very same issue seems to be with IPv6, but haven't tested.

---
 net/ipv4/ip_input.c  |    2 +-
 net/ipv6/ip6_input.c |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
net/ipv4/ip_input.c +1 -1
@@ -314,7 +314,7 @@ static int ip_rcv_finish(struct sk_buff
 	const struct iphdr *iph = ip_hdr(skb);
 	struct rtable *rt;
 
-	if (sysctl_ip_early_demux && !skb_dst(skb)) {
+	if (sysctl_ip_early_demux && !skb_dst(skb) && skb->sk == NULL) {
 		const struct net_protocol *ipprot;
 		int protocol = iph->protocol;
 
net/ipv6/ip6_input.c +1 -1
@@ -49,7 +49,7 @@
 
 int ip6_rcv_finish(struct sk_buff *skb)
 {
-	if (sysctl_ip_early_demux && !skb_dst(skb)) {
+	if (sysctl_ip_early_demux && !skb_dst(skb) && skb->sk == NULL) {
 		const struct inet6_protocol *ipprot;
 
 		ipprot = rcu_dereference(inet6_protos[ipv6_hdr(skb)->nexthdr]);
author maintainer
#110 lore raw

PATCH 114/133 xen-netfront: fix resource leak in netfront

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Annie Li <[email protected]>

[ Upstream commit cefe0078eea52af17411eb1248946a94afb84ca5 ]

This patch removes grant transfer releasing code from netfront, and uses gnttab_end_foreign_access to end grant access since gnttab_end_foreign_access_ref may fail when the grant entry is currently used for reading or writing.

* clean up grant transfer code kept from old netfront(2.6.18) which grants pages for access/map and transfer. But grant transfer is deprecated in current netfront, so remove corresponding release code for transfer.

* fix resource leak, release grant access (through gnttab_end_foreign_access) and skb for tx/rx path, use get_page to ensure page is released when grant access is completed successfully.

Xen-blkfront/xen-tpmfront/xen-pcifront also have similar issue, but patches for them will be created separately.

V6: Correct subject line and commit message.

V5: Remove unecessary change in xennet_end_access.

V4: Revert put_page in gnttab_end_foreign_access, and keep netfront change in single patch.

V3: Changes as suggestion from David Vrabel, ensure pages are not freed untill grant acess is ended.

V2: Improve patch comments.

---
 drivers/net/xen-netfront.c |   88 +++++++++++++--------------------------------
 1 file changed, 26 insertions(+), 62 deletions(-)
drivers/net/xen-netfront.c +26 -62
@@ -117,6 +117,7 @@ struct netfront_info {
 	} tx_skbs[NET_TX_RING_SIZE];
 	grant_ref_t gref_tx_head;
 	grant_ref_t grant_tx_ref[NET_TX_RING_SIZE];
+	struct page *grant_tx_page[NET_TX_RING_SIZE];
 	unsigned tx_skb_freelist;
 
 	spinlock_t   rx_lock ____cacheline_aligned_in_smp;
@@ -395,6 +396,7 @@ static void xennet_tx_buf_gc(struct net_
 			gnttab_release_grant_reference(
 				&np->gref_tx_head, np->grant_tx_ref[id]);
 			np->grant_tx_ref[id] = GRANT_INVALID_REF;
+			np->grant_tx_page[id] = NULL;
 			add_id_to_freelist(&np->tx_skb_freelist, np->tx_skbs, id);
 			dev_kfree_skb_irq(skb);
 		}
@@ -451,6 +453,7 @@ static void xennet_make_frags(struct sk_
 		gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id,
 						mfn, GNTMAP_readonly);
 
+		np->grant_tx_page[id] = virt_to_page(data);
 		tx->gref = np->grant_tx_ref[id] = ref;
 		tx->offset = offset;
 		tx->size = len;
@@ -496,6 +499,7 @@ static void xennet_make_frags(struct sk_
 							np->xbdev->otherend_id,
 							mfn, GNTMAP_readonly);
 
+			np->grant_tx_page[id] = page;
 			tx->gref = np->grant_tx_ref[id] = ref;
 			tx->offset = offset;
 			tx->size = bytes;
@@ -595,6 +599,7 @@ static int xennet_start_xmit(struct sk_b
 	mfn = virt_to_mfn(data);
 	gnttab_grant_foreign_access_ref(
 		ref, np->xbdev->otherend_id, mfn, GNTMAP_readonly);
+	np->grant_tx_page[id] = virt_to_page(data);
 	tx->gref = np->grant_tx_ref[id] = ref;
 	tx->offset = offset;
 	tx->size = len;
@@ -1119,10 +1124,11 @@ static void xennet_release_tx_bufs(struc
 			continue;
 
 		skb = np->tx_skbs[i].skb;
-		gnttab_end_foreign_access_ref(np->grant_tx_ref[i],
-					      GNTMAP_readonly);
-		gnttab_release_grant_reference(&np->gref_tx_head,
-					       np->grant_tx_ref[i]);
+		get_page(np->grant_tx_page[i]);
+		gnttab_end_foreign_access(np->grant_tx_ref[i],
+					  GNTMAP_readonly,
+					  (unsigned long)page_address(np->grant_tx_page[i]));
+		np->grant_tx_page[i] = NULL;
 		np->grant_tx_ref[i] = GRANT_INVALID_REF;
 		add_id_to_freelist(&np->tx_skb_freelist, np->tx_skbs, i);
 		dev_kfree_skb_irq(skb);
@@ -1131,78 +1137,35 @@ static void xennet_release_tx_bufs(struc
 
 static void xennet_release_rx_bufs(struct netfront_info *np)
 {
-	struct mmu_update      *mmu = np->rx_mmu;
-	struct multicall_entry *mcl = np->rx_mcl;
-	struct sk_buff_head free_list;
-	struct sk_buff *skb;
-	unsigned long mfn;
-	int xfer = 0, noxfer = 0, unused = 0;
 	int id, ref;
 
-	dev_warn(&np->netdev->dev, "%s: fix me for copying receiver.\n",
-			 __func__);
-	return;
-
-	skb_queue_head_init(&free_list);
-
 	spin_lock_bh(&np->rx_lock);
 
 	for (id = 0; id < NET_RX_RING_SIZE; id++) {
-		ref = np->grant_rx_ref[id];
-		if (ref == GRANT_INVALID_REF) {
-			unused++;
-			continue;
-		}
+		struct sk_buff *skb;
+		struct page *page;
 
 		skb = np->rx_skbs[id];
-		mfn = gnttab_end_foreign_transfer_ref(ref);
-		gnttab_release_grant_reference(&np->gref_rx_head, ref);
-		np->grant_rx_ref[id] = GRANT_INVALID_REF;
-
-		if (0 == mfn) {
-			skb_shinfo(skb)->nr_frags = 0;
-			dev_kfree_skb(skb);
-			noxfer++;
+		if (!skb)
 			continue;
-		}
 
-		if (!xen_feature(XENFEAT_auto_translated_physmap)) {
-			/* Remap the page. */
-			const struct page *page =
-				skb_frag_page(&skb_shinfo(skb)->frags[0]);
-			unsigned long pfn = page_to_pfn(page);
-			void *vaddr = page_address(page);
+		ref = np->grant_rx_ref[id];
+		if (ref == GRANT_INVALID_REF)
+			continue;
 
-			MULTI_update_va_mapping(mcl, (unsigned long)vaddr,
-						mfn_pte(mfn, PAGE_KERNEL),
-						0);
-			mcl++;
-			mmu->ptr = ((u64)mfn << PAGE_SHIFT)
-				| MMU_MACHPHYS_UPDATE;
-			mmu->val = pfn;
-			mmu++;
+		page = skb_frag_page(&skb_shinfo(skb)->frags[0]);
 
-			set_phys_to_machine(pfn, mfn);
-		}
-		__skb_queue_tail(&free_list, skb);
-		xfer++;
-	}
-
-	dev_info(&np->netdev->dev, "%s: %d xfer, %d noxfer, %d unused\n",
-		 __func__, xfer, noxfer, unused);
+		/* gnttab_end_foreign_access() needs a page ref until
+		 * foreign access is ended (which may be deferred).
+		 */
+		get_page(page);
+		gnttab_end_foreign_access(ref, 0,
+					  (unsigned long)page_address(page));
+		np->grant_rx_ref[id] = GRANT_INVALID_REF;
 
-	if (xfer) {
-		if (!xen_feature(XENFEAT_auto_translated_physmap)) {
-			/* Do all the remapping work and M2P updates. */
-			MULTI_mmu_update(mcl, np->rx_mmu, mmu - np->rx_mmu,
-					 NULL, DOMID_SELF);
-			mcl++;
-			HYPERVISOR_multicall(np->rx_mcl, mcl - np->rx_mcl);
-		}
+		kfree_skb(skb);
 	}
 
-	__skb_queue_purge(&free_list);
-
 	spin_unlock_bh(&np->rx_lock);
 }
 
@@ -1349,6 +1312,7 @@ static struct net_device *xennet_create_
 	for (i = 0; i < NET_RX_RING_SIZE; i++) {
 		np->rx_skbs[i] = NULL;
 		np->grant_rx_ref[i] = GRANT_INVALID_REF;
+		np->grant_tx_page[i] = NULL;
 	}
 
 	/* A grant for every tx ring slot */
author maintainer
#111 lore raw

PATCH 115/133 net: gre: use icmp_hdr() to get inner ip header

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Duan Jiong <[email protected]>

[ Upstream commit c0c0c50ff7c3e331c90bab316d21f724fb9e1994 ]

When dealing with icmp messages, the skb->data points the ip header that triggered the sending of the icmp message.

In gre_cisco_err(), the parse_gre_header() is called, and the iptunnel_pull_header() is called to pull the skb at the end of the parse_gre_header(), so the skb->data doesn't point the inner ip header.

Unfortunately, the ipgre_err still needs those ip addresses in inner ip header to look up tunnel by ip_tunnel_lookup().

So just use icmp_hdr() to get inner ip header instead of skb->data.

---
 net/ipv4/ip_gre.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/ip_gre.c +1 -1
@@ -178,7 +178,7 @@ static int ipgre_err(struct sk_buff *skb
 	else
 		itn = net_generic(net, ipgre_net_id);
 
-	iph = (const struct iphdr *)skb->data;
+	iph = (const struct iphdr *)(icmp_hdr(skb) + 1);
 	t = ip_tunnel_lookup(itn, skb->dev->ifindex, tpi->flags,
 			     iph->daddr, iph->saddr, tpi->key);
 
author maintainer
#112 lore raw

PATCH 116/133 inet_diag: fix inet_diag_dump_icsk() timewait socket state logic

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Neal Cardwell <[email protected]>

[ Based upon upstream commit 70315d22d3c7383f9a508d0aab21e2eb35b2303a ]

Fix inet_diag_dump_icsk() to reflect the fact that both TIME_WAIT and FIN_WAIT2 connections are represented by inet_timewait_sock (not just TIME_WAIT). Thus:

(a) We need to iterate through the time_wait buckets if the user wants either TIME_WAIT or FIN_WAIT2. (Before fixing this, "ss -nemoi state fin-wait-2" would not return any sockets, even if there were some in FIN_WAIT2.)

(b) We need to check tw_substate to see if the user wants to dump sockets in the particular substate (TIME_WAIT or FIN_WAIT2) that a given connection is in. (Before fixing this, "ss -nemoi state time-wait" would actually return sockets in state FIN_WAIT2.)

An analogous fix is in v3.13: 70315d22d3c7383f9a508d0aab21e2eb35b2303a ("inet_diag: fix inet_diag_dump_icsk() to use correct state for timewait sockets") but that patch is quite different because 3.13 code is very different in this area due to the unification of TCP hash tables in 05dbc7b ("tcp/dccp: remove twchain") in v3.13-rc1.

I tested that this applies cleanly between v3.3 and v3.12, and tested that it works in both 3.3 and 3.12. It does not apply cleanly to 3.2 and earlier (though it makes semantic sense), and semantically is not the right fix for 3.13 and beyond (as mentioned above).

---
 net/ipv4/inet_diag.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
net/ipv4/inet_diag.c +3 -1
@@ -961,7 +961,7 @@ next_normal:
 			++num;
 		}
 
-		if (r->idiag_states & TCPF_TIME_WAIT) {
+		if (r->idiag_states & (TCPF_TIME_WAIT | TCPF_FIN_WAIT2)) {
 			struct inet_timewait_sock *tw;
 
 			inet_twsk_for_each(tw, node,
@@ -971,6 +971,8 @@ next_normal:
 
 				if (num < s_num)
 					goto next_dying;
+				if (!(r->idiag_states & (1 << tw->tw_substate)))
+					goto next_dying;
 				if (r->sdiag_family != AF_UNSPEC &&
 						tw->tw_family != r->sdiag_family)
 					goto next_dying;
author maintainer
#113 lore raw

PATCH 117/133 alpha: fix broken network checksum

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Mikulas Patocka <[email protected]>

commit 0ef38d70d4118b2ce1a538d14357be5ff9dc2bbd upstream.

The patch 3ddc5b46a8e90f3c9251338b60191d0a804b0d92 breaks networking on alpha (there is a follow-up fix 5cfe8f1ba5eebe6f4b6e5858cdb1a5be4f3272a6, but networking is still broken even with the second patch).

The patch 3ddc5b46a8e90f3c9251338b60191d0a804b0d92 makes csum_partial_copy_from_user check the pointer with access_ok. However, csum_partial_copy_from_user is called also from csum_partial_copy_nocheck and csum_partial_copy_nocheck is called on kernel pointers and it is supposed not to check pointer validity.

This bug results in ssh session hangs if the system is loaded and bulk data are printed to ssh terminal.

This patch fixes csum_partial_copy_nocheck to call set_fs(KERNEL_DS), so that access_ok in csum_partial_copy_from_user accepts kernel-space addresses.

---
 arch/alpha/lib/csum_partial_copy.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
arch/alpha/lib/csum_partial_copy.c +7 -2
@@ -378,6 +378,11 @@ csum_partial_copy_from_user(const void _
 __wsum
 csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum)
 {
-	return csum_partial_copy_from_user((__force const void __user *)src,
-			dst, len, sum, NULL);
+	__wsum checksum;
+	mm_segment_t oldfs = get_fs();
+	set_fs(KERNEL_DS);
+	checksum = csum_partial_copy_from_user((__force const void __user *)src,
+						dst, len, sum, NULL);
+	set_fs(oldfs);
+	return checksum;
 }
author maintainer
#117 lore raw

PATCH 118/133 parisc: fix cache-flushing

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Helge Deller <[email protected]>

commit 57737c49dd72c96cfbcd4f66559f3ffc399aeb4f upstream.

This commit:
f8dae00684d678afa13041ef170cecfd1297ed40: parisc: Ensure full cache coherency for kmap/kunmap caused negative caching side-effects, e.g. hanging processes with expect and too many inequivalent alias messages from flush_dcache_page() on Debian 5 systems.

This patch now partly reverts it and has been in production use on our debian buildd makeservers since a week without any major problems.

---
 arch/parisc/include/asm/cacheflush.h |    2 --
 arch/parisc/include/asm/page.h       |    3 ++-
 arch/parisc/kernel/cache.c           |   14 ++++++++++++++
 3 files changed, 16 insertions(+), 3 deletions(-)
arch/parisc/include/asm/cacheflush.h +0 -2
@@ -132,7 +132,6 @@ void mark_rodata_ro(void);
 static inline void *kmap(struct page *page)
 {
 	might_sleep();
-	flush_dcache_page(page);
 	return page_address(page);
 }
 
@@ -144,7 +143,6 @@ static inline void kunmap(struct page *p
 static inline void *kmap_atomic(struct page *page)
 {
 	pagefault_disable();
-	flush_dcache_page(page);
 	return page_address(page);
 }
 
arch/parisc/include/asm/page.h +2 -1
@@ -29,7 +29,8 @@ struct page;
 void clear_page_asm(void *page);
 void copy_page_asm(void *to, void *from);
 #define clear_user_page(vto, vaddr, page) clear_page_asm(vto)
-#define copy_user_page(vto, vfrom, vaddr, page) copy_page_asm(vto, vfrom)
+void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,
+			struct page *pg);
 
 /* #define CONFIG_PARISC_TMPALIAS */
 
arch/parisc/kernel/cache.c +14 -0
@@ -388,6 +388,20 @@ void flush_kernel_dcache_page_addr(void
 }
 EXPORT_SYMBOL(flush_kernel_dcache_page_addr);
 
+void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,
+	struct page *pg)
+{
+       /* Copy using kernel mapping.  No coherency is needed (all in
+	  kunmap) for the `to' page.  However, the `from' page needs to
+	  be flushed through a mapping equivalent to the user mapping
+	  before it can be accessed through the kernel mapping. */
+	preempt_disable();
+	flush_dcache_page_asm(__pa(vfrom), vaddr);
+	preempt_enable();
+	copy_page_asm(vto, vfrom);
+}
+EXPORT_SYMBOL(copy_user_page);
+
 void purge_tlb_entries(struct mm_struct *mm, unsigned long addr)
 {
 	unsigned long flags;
author maintainer
#118 lore raw

PATCH 119/133 KVM: PPC: Book3S HV: use xics_wake_cpu only when defined

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Andreas Schwab <[email protected]>

commit 48eaef0518a565d3852e301c860e1af6a6db5a84 upstream.

---
 arch/powerpc/kvm/book3s_hv.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
arch/powerpc/kvm/book3s_hv.c +6 -1
@@ -82,10 +82,13 @@ void kvmppc_fast_vcpu_kick(struct kvm_vc
 
 	/* CPU points to the first thread of the core */
 	if (cpu != me && cpu >= 0 && cpu < nr_cpu_ids) {
+#ifdef CONFIG_KVM_XICS
 		int real_cpu = cpu + vcpu->arch.ptid;
 		if (paca[real_cpu].kvm_hstate.xics_phys)
 			xics_wake_cpu(real_cpu);
-		else if (cpu_online(cpu))
+		else
+#endif
+		if (cpu_online(cpu))
 			smp_send_reschedule(cpu);
 	}
 	put_cpu();
@@ -1089,7 +1092,9 @@ static void kvmppc_start_thread(struct k
 	smp_wmb();
 #if defined(CONFIG_PPC_ICP_NATIVE) && defined(CONFIG_SMP)
 	if (vcpu->arch.ptid) {
+#ifdef CONFIG_KVM_XICS
 		xics_wake_cpu(cpu);
+#endif
 		++vc->n_woken;
 	}
 #endif
author maintainer
#119 lore raw

PATCH 120/133 KVM: PPC: e500: Fix bad address type in deliver_tlb_misss()

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Mihai Caraman <[email protected]>

commit 70713fe315ed14cd1bb07d1a7f33e973d136ae3d upstream.

Use gva_t instead of unsigned int for eaddr in deliver_tlb_miss().

---
 arch/powerpc/kvm/e500_mmu.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/kvm/e500_mmu.c +1 -1
@@ -127,7 +127,7 @@ static int kvmppc_e500_tlb_index(struct
 }
 
 static inline void kvmppc_e500_deliver_tlb_miss(struct kvm_vcpu *vcpu,
-		unsigned int eaddr, int as)
+		gva_t eaddr, int as)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 	unsigned int victim, tsized;
author maintainer
#120 lore raw

PATCH 121/133 ALSA: hda - Dont set indep_hp flag for old AD codecs

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Takashi Iwai <[email protected]>

commit cbd209f41ea5f39394de5c1fe2dd9aa54a9c5744 upstream.

Some old AD codecs don't like the independent HP handling, either it contains a single DAC (AD1981) or it mandates the mixer routing (AD1986A). This patch removes the indep_hp flag for such codecs.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=68081

---
 sound/pci/hda/patch_analog.c |   16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
sound/pci/hda/patch_analog.c +8 -8
@@ -171,7 +171,7 @@ static const struct hda_codec_ops ad198x
 };
 
 
-static int ad198x_parse_auto_config(struct hda_codec *codec)
+static int ad198x_parse_auto_config(struct hda_codec *codec, bool indep_hp)
 {
 	struct ad198x_spec *spec = codec->spec;
 	struct auto_pin_cfg *cfg = &spec->gen.autocfg;
@@ -181,7 +181,7 @@ static int ad198x_parse_auto_config(stru
 	codec->no_trigger_sense = 1;
 	codec->no_sticky_stream = 1;
 
-	spec->gen.indep_hp = 1;
+	spec->gen.indep_hp = indep_hp;
 
 	err = snd_hda_parse_pin_defcfg(codec, cfg, NULL, 0);
 	if (err < 0)
@@ -362,7 +362,7 @@ static int patch_ad1986a(struct hda_code
 			   ad1986a_fixups);
 	snd_hda_apply_fixup(codec, HDA_FIXUP_ACT_PRE_PROBE);
 
-	err = ad198x_parse_auto_config(codec);
+	err = ad198x_parse_auto_config(codec, false);
 	if (err < 0) {
 		snd_hda_gen_free(codec);
 		return err;
@@ -464,7 +464,7 @@ static int patch_ad1983(struct hda_codec
 
 	spec->gen.beep_nid = 0x10;
 	set_beep_amp(spec, 0x10, 0, HDA_OUTPUT);
-	err = ad198x_parse_auto_config(codec);
+	err = ad198x_parse_auto_config(codec, false);
 	if (err < 0)
 		goto error;
 	err = ad1983_add_spdif_mux_ctl(codec);
@@ -564,7 +564,7 @@ static int patch_ad1981(struct hda_codec
 	snd_hda_pick_fixup(codec, NULL, ad1981_fixup_tbl, ad1981_fixups);
 	snd_hda_apply_fixup(codec, HDA_FIXUP_ACT_PRE_PROBE);
 
-	err = ad198x_parse_auto_config(codec);
+	err = ad198x_parse_auto_config(codec, false);
 	if (err < 0)
 		goto error;
 	err = ad1983_add_spdif_mux_ctl(codec);
@@ -890,7 +890,7 @@ static int patch_ad1988(struct hda_codec
 	snd_hda_pick_fixup(codec, ad1988_fixup_models, NULL, ad1988_fixups);
 	snd_hda_apply_fixup(codec, HDA_FIXUP_ACT_PRE_PROBE);
 
-	err = ad198x_parse_auto_config(codec);
+	err = ad198x_parse_auto_config(codec, true);
 	if (err < 0)
 		goto error;
 	err = ad1988_add_spdif_mux_ctl(codec);
@@ -1064,7 +1064,7 @@ static int patch_ad1884(struct hda_codec
 	snd_hda_pick_fixup(codec, NULL, ad1884_fixup_tbl, ad1884_fixups);
 	snd_hda_apply_fixup(codec, HDA_FIXUP_ACT_PRE_PROBE);
 
-	err = ad198x_parse_auto_config(codec);
+	err = ad198x_parse_auto_config(codec, true);
 	if (err < 0)
 		goto error;
 	err = ad1983_add_spdif_mux_ctl(codec);
@@ -1106,7 +1106,7 @@ static int patch_ad1882(struct hda_codec
 	spec->gen.mixer_merge_nid = 0x21;
 	spec->gen.beep_nid = 0x10;
 	set_beep_amp(spec, 0x10, 0, HDA_OUTPUT);
-	err = ad198x_parse_auto_config(codec);
+	err = ad198x_parse_auto_config(codec, true);
 	if (err < 0)
 		goto error;
 	err = ad1988_add_spdif_mux_ctl(codec);
author maintainer
#121 lore raw

PATCH 122/133 ALSA: hda - hdmi: introduce patch_nvhdmi()

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Anssi Hannula <[email protected]>

(This is a backport of *part* of upstream 611885bc963a "ALSA: hda -
hdmi: Disallow unsupported 2ch remapping on NVIDIA codecs" to stable 3.10 through 3.12. Later stable already contain all of the original patch.)

Mainline commit 611885bc963a "ALSA: hda - hdmi: Disallow unsupported 2ch remapping on NVIDIA codecs" introduces function patch_nvhdmi(). That function is edited by 75fae117a5db "ALSA: hda/hdmi - allow PIN_OUT to be dynamically enabled". In order to backport the PIN_OUT patch, I am first back-porting just the addition of function patch_nvhdmi(), so that the conflicts applying the PIN_OUT patch are simplified.

Ideally, one might backport all of 611885bc963a. However, that commit doesn't apply to stable kernels, since it relies on a chain of other patches which implement new features.

[swarren, extracted just a small part of the original patch]
---
 sound/pci/hda/patch_hdmi.c |   60 +++++++++++++++++++++++++++------------------
 1 file changed, 37 insertions(+), 23 deletions(-)
sound/pci/hda/patch_hdmi.c +37 -23
@@ -2560,6 +2560,20 @@ static int patch_nvhdmi_8ch_7x(struct hd
 	return 0;
 }
 
+static int patch_nvhdmi(struct hda_codec *codec)
+{
+	struct hdmi_spec *spec;
+	int err;
+
+	err = patch_generic_hdmi(codec);
+	if (err)
+		return err;
+
+	spec = codec->spec;
+
+	return 0;
+}
+
 /*
  * ATI-specific implementations
  *
@@ -2632,30 +2646,30 @@ static const struct hda_codec_preset snd
 { .id = 0x10de0005, .name = "MCP77/78 HDMI",	.patch = patch_nvhdmi_8ch_7x },
 { .id = 0x10de0006, .name = "MCP77/78 HDMI",	.patch = patch_nvhdmi_8ch_7x },
 { .id = 0x10de0007, .name = "MCP79/7A HDMI",	.patch = patch_nvhdmi_8ch_7x },
-{ .id = 0x10de000a, .name = "GPU 0a HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de000b, .name = "GPU 0b HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de000c, .name = "MCP89 HDMI",	.patch = patch_generic_hdmi },
-{ .id = 0x10de000d, .name = "GPU 0d HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0010, .name = "GPU 10 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0011, .name = "GPU 11 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0012, .name = "GPU 12 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0013, .name = "GPU 13 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0014, .name = "GPU 14 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0015, .name = "GPU 15 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0016, .name = "GPU 16 HDMI/DP",	.patch = patch_generic_hdmi },
+{ .id = 0x10de000a, .name = "GPU 0a HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de000b, .name = "GPU 0b HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de000c, .name = "MCP89 HDMI",	.patch = patch_nvhdmi },
+{ .id = 0x10de000d, .name = "GPU 0d HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0010, .name = "GPU 10 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0011, .name = "GPU 11 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0012, .name = "GPU 12 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0013, .name = "GPU 13 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0014, .name = "GPU 14 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0015, .name = "GPU 15 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0016, .name = "GPU 16 HDMI/DP",	.patch = patch_nvhdmi },
 /* 17 is known to be absent */
-{ .id = 0x10de0018, .name = "GPU 18 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0019, .name = "GPU 19 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de001a, .name = "GPU 1a HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de001b, .name = "GPU 1b HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de001c, .name = "GPU 1c HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0040, .name = "GPU 40 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0041, .name = "GPU 41 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0042, .name = "GPU 42 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0043, .name = "GPU 43 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0044, .name = "GPU 44 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0051, .name = "GPU 51 HDMI/DP",	.patch = patch_generic_hdmi },
-{ .id = 0x10de0060, .name = "GPU 60 HDMI/DP",	.patch = patch_generic_hdmi },
+{ .id = 0x10de0018, .name = "GPU 18 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0019, .name = "GPU 19 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de001a, .name = "GPU 1a HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de001b, .name = "GPU 1b HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de001c, .name = "GPU 1c HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0040, .name = "GPU 40 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0041, .name = "GPU 41 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0042, .name = "GPU 42 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0043, .name = "GPU 43 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0044, .name = "GPU 44 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0051, .name = "GPU 51 HDMI/DP",	.patch = patch_nvhdmi },
+{ .id = 0x10de0060, .name = "GPU 60 HDMI/DP",	.patch = patch_nvhdmi },
 { .id = 0x10de0067, .name = "MCP67 HDMI",	.patch = patch_nvhdmi_2ch },
 { .id = 0x10de8001, .name = "MCP73 HDMI",	.patch = patch_nvhdmi_2ch },
 { .id = 0x11069f80, .name = "VX900 HDMI/DP",	.patch = patch_via_hdmi },
author maintainer
#122 lore raw

PATCH 123/133 ALSA: hda/hdmi - allow PIN_OUT to be dynamically enabled

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Stephen Warren <[email protected]>

(This is upstream 75fae117a5db "ALSA: hda/hdmi - allow PIN_OUT to be dynamically enabled", backported to stable 3.10 through 3.12. 3.13 and later can take the original patch.)

Commit 384a48d71520 "ALSA: hda: HDMI: Support codecs with fewer cvts than pins" dynamically enabled each pin widget's PIN_OUT only when the pin was actively in use. This was required on certain NVIDIA CODECs for correct operation. Specifically, if multiple pin widgets each had their mux input select the same audio converter widget and each pin widget had PIN_OUT enabled, then only one of the pin widgets would actually receive the audio, and often not the one the user wanted!

However, this apparently broke some Intel systems, and commit 6169b673618b "ALSA: hda - Always turn on pins for HDMI/DP" reverted the dynamic setting of PIN_OUT. This in turn broke the afore-mentioned NVIDIA CODECs.

This change supports either dynamic or static handling of PIN_OUT, selected by a flag set up during CODEC initialization. This flag is enabled for all recent NVIDIA GPUs.

---
 sound/pci/hda/patch_hdmi.c |   40 ++++++++++++++++++++++++++++++++++++----
 1 file changed, 36 insertions(+), 4 deletions(-)
sound/pci/hda/patch_hdmi.c +36 -4
@@ -88,6 +88,9 @@ struct hdmi_spec {
 	unsigned int channels_max; /* max over all cvts */
 
 	struct hdmi_eld temp_eld;
+
+	bool dyn_pin_out;
+
 	/*
 	 * Non-generic ATI/NVIDIA specific
 	 */
@@ -452,15 +455,25 @@ static void hdmi_write_dip_byte(struct h
 
 static void hdmi_init_pin(struct hda_codec *codec, hda_nid_t pin_nid)
 {
+	struct hdmi_spec *spec = codec->spec;
+	int pin_out;
+
 	/* Unmute */
 	if (get_wcaps(codec, pin_nid) & AC_WCAP_OUT_AMP)
 		snd_hda_codec_write(codec, pin_nid, 0,
 				AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE);
-	/* Enable pin out: some machines with GM965 gets broken output when
-	 * the pin is disabled or changed while using with HDMI
-	 */
+
+	if (spec->dyn_pin_out)
+		/* Disable pin out until stream is active */
+		pin_out = 0;
+	else
+		/* Enable pin out: some machines with GM965 gets broken output
+		 * when the pin is disabled or changed while using with HDMI
+		 */
+		pin_out = PIN_OUT;
+
 	snd_hda_codec_write(codec, pin_nid, 0,
-			    AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT);
+			    AC_VERB_SET_PIN_WIDGET_CONTROL, pin_out);
 }
 
 static int hdmi_get_channel_count(struct hda_codec *codec, hda_nid_t cvt_nid)
@@ -1535,6 +1548,7 @@ static int generic_hdmi_playback_pcm_pre
 	struct hdmi_spec_per_pin *per_pin = get_pin(spec, pin_idx);
 	hda_nid_t pin_nid = per_pin->pin_nid;
 	bool non_pcm;
+	int pinctl;
 
 	non_pcm = check_non_pcm_per_cvt(codec, cvt_nid);
 	per_pin->channels = substream->runtime->channels;
@@ -1544,6 +1558,14 @@ static int generic_hdmi_playback_pcm_pre
 
 	hdmi_setup_audio_infoframe(codec, per_pin, non_pcm);
 
+	if (spec->dyn_pin_out) {
+		pinctl = snd_hda_codec_read(codec, pin_nid, 0,
+					    AC_VERB_GET_PIN_WIDGET_CONTROL, 0);
+		snd_hda_codec_write(codec, pin_nid, 0,
+				    AC_VERB_SET_PIN_WIDGET_CONTROL,
+				    pinctl | PIN_OUT);
+	}
+
 	return hdmi_setup_stream(codec, cvt_nid, pin_nid, stream_tag, format);
 }
 
@@ -1563,6 +1585,7 @@ static int hdmi_pcm_close(struct hda_pcm
 	int cvt_idx, pin_idx;
 	struct hdmi_spec_per_cvt *per_cvt;
 	struct hdmi_spec_per_pin *per_pin;
+	int pinctl;
 
 	if (hinfo->nid) {
 		cvt_idx = cvt_nid_to_cvt_index(spec, hinfo->nid);
@@ -1579,6 +1602,14 @@ static int hdmi_pcm_close(struct hda_pcm
 			return -EINVAL;
 		per_pin = get_pin(spec, pin_idx);
 
+		if (spec->dyn_pin_out) {
+			pinctl = snd_hda_codec_read(codec, per_pin->pin_nid, 0,
+					AC_VERB_GET_PIN_WIDGET_CONTROL, 0);
+			snd_hda_codec_write(codec, per_pin->pin_nid, 0,
+					    AC_VERB_SET_PIN_WIDGET_CONTROL,
+					    pinctl & ~PIN_OUT);
+		}
+
 		snd_hda_spdif_ctls_unassign(codec, pin_idx);
 		per_pin->chmap_set = false;
 		memset(per_pin->chmap, 0, sizeof(per_pin->chmap));
@@ -2570,6 +2601,7 @@ static int patch_nvhdmi(struct hda_codec
 		return err;
 
 	spec = codec->spec;
+	spec->dyn_pin_out = true;
 
 	return 0;
 }
author maintainer
#123 lore raw

PATCH 124/133 hpfs: remember free space

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Mikulas Patocka <[email protected]>

commit 2cbe5c76fc5e38e9af4b709593146e4b8272b69e upstream.

Previously, hpfs scanned all bitmaps each time the user asked for free space using statfs. This patch changes it so that hpfs scans the bitmaps only once, remembes the free space and on next invocation of statfs it returns the value instantly.

New versions of wine are hammering on the statfs syscall very heavily, making some games unplayable when they're stored on hpfs, with load times in minutes.

This should be backported to the stable kernels because it fixes user-visible problem (excessive level load times in wine).

---
 fs/hpfs/alloc.c   |   66 ++++++++++++++++++++++++++++++++++++++++++++++++++++--
 fs/hpfs/hpfs_fn.h |    2 -
 fs/hpfs/super.c   |   29 ++++++++++++++++++-----
 3 files changed, 87 insertions(+), 10 deletions(-)
fs/hpfs/alloc.c +64 -2
@@ -8,6 +8,58 @@
 
 #include "hpfs_fn.h"
 
+static void hpfs_claim_alloc(struct super_block *s, secno sec)
+{
+	struct hpfs_sb_info *sbi = hpfs_sb(s);
+	if (sbi->sb_n_free != (unsigned)-1) {
+		if (unlikely(!sbi->sb_n_free)) {
+			hpfs_error(s, "free count underflow, allocating sector %08x", sec);
+			sbi->sb_n_free = -1;
+			return;
+		}
+		sbi->sb_n_free--;
+	}
+}
+
+static void hpfs_claim_free(struct super_block *s, secno sec)
+{
+	struct hpfs_sb_info *sbi = hpfs_sb(s);
+	if (sbi->sb_n_free != (unsigned)-1) {
+		if (unlikely(sbi->sb_n_free >= sbi->sb_fs_size)) {
+			hpfs_error(s, "free count overflow, freeing sector %08x", sec);
+			sbi->sb_n_free = -1;
+			return;
+		}
+		sbi->sb_n_free++;
+	}
+}
+
+static void hpfs_claim_dirband_alloc(struct super_block *s, secno sec)
+{
+	struct hpfs_sb_info *sbi = hpfs_sb(s);
+	if (sbi->sb_n_free_dnodes != (unsigned)-1) {
+		if (unlikely(!sbi->sb_n_free_dnodes)) {
+			hpfs_error(s, "dirband free count underflow, allocating sector %08x", sec);
+			sbi->sb_n_free_dnodes = -1;
+			return;
+		}
+		sbi->sb_n_free_dnodes--;
+	}
+}
+
+static void hpfs_claim_dirband_free(struct super_block *s, secno sec)
+{
+	struct hpfs_sb_info *sbi = hpfs_sb(s);
+	if (sbi->sb_n_free_dnodes != (unsigned)-1) {
+		if (unlikely(sbi->sb_n_free_dnodes >= sbi->sb_dirband_size / 4)) {
+			hpfs_error(s, "dirband free count overflow, freeing sector %08x", sec);
+			sbi->sb_n_free_dnodes = -1;
+			return;
+		}
+		sbi->sb_n_free_dnodes++;
+	}
+}
+
 /*
  * Check if a sector is allocated in bitmap
  * This is really slow. Turned on only if chk==2
@@ -203,9 +255,15 @@ secno hpfs_alloc_sector(struct super_blo
 	}
 	sec = 0;
 	ret:
+	if (sec) {
+		i = 0;
+		do
+			hpfs_claim_alloc(s, sec + i);
+		while (unlikely(++i < n));
+	}
 	if (sec && f_p) {
 		for (i = 0; i < forward; i++) {
-			if (!hpfs_alloc_if_possible(s, sec + i + 1)) {
+			if (!hpfs_alloc_if_possible(s, sec + n + i)) {
 				hpfs_error(s, "Prealloc doesn't work! Wanted %d, allocated at %08x, can't allocate %d", forward, sec, i);
 				sec = 0;
 				break;
@@ -228,6 +286,7 @@ static secno alloc_in_dirband(struct sup
 	nr >>= 2;
 	sec = alloc_in_bmp(s, (~0x3fff) | nr, 1, 0);
 	if (!sec) return 0;
+	hpfs_claim_dirband_alloc(s, sec);
 	return ((sec & 0x3fff) << 2) + sbi->sb_dirband_start;
 }
 
@@ -242,6 +301,7 @@ int hpfs_alloc_if_possible(struct super_
 		bmp[(sec & 0x3fff) >> 5] &= cpu_to_le32(~(1 << (sec & 0x1f)));
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
+		hpfs_claim_alloc(s, sec);
 		return 1;
 	}
 	hpfs_brelse4(&qbh);
@@ -275,6 +335,7 @@ void hpfs_free_sectors(struct super_bloc
 		return;
 	}
 	bmp[(sec & 0x3fff) >> 5] |= cpu_to_le32(1 << (sec & 0x1f));
+	hpfs_claim_free(s, sec);
 	if (!--n) {
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
@@ -359,6 +420,7 @@ void hpfs_free_dnode(struct super_block
 		bmp[ssec >> 5] |= cpu_to_le32(1 << (ssec & 0x1f));
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
+		hpfs_claim_dirband_free(s, dno);
 	}
 }
 
@@ -366,7 +428,7 @@ struct dnode *hpfs_alloc_dnode(struct su
 			 dnode_secno *dno, struct quad_buffer_head *qbh)
 {
 	struct dnode *d;
-	if (hpfs_count_one_bitmap(s, hpfs_sb(s)->sb_dmap) > FREE_DNODES_ADD) {
+	if (hpfs_get_free_dnodes(s) > FREE_DNODES_ADD) {
 		if (!(*dno = alloc_in_dirband(s, near)))
 			if (!(*dno = hpfs_alloc_sector(s, near, 4, 0))) return NULL;
 	} else {
fs/hpfs/hpfs_fn.h +1 -1
@@ -311,7 +311,7 @@ static inline struct hpfs_sb_info *hpfs_
 __printf(2, 3)
 void hpfs_error(struct super_block *, const char *, ...);
 int hpfs_stop_cycles(struct super_block *, int, int *, int *, char *);
-unsigned hpfs_count_one_bitmap(struct super_block *, secno);
+unsigned hpfs_get_free_dnodes(struct super_block *);
 
 /*
  * local time (HPFS) to GMT (Unix)
fs/hpfs/super.c +22 -7
@@ -115,7 +115,7 @@ static void hpfs_put_super(struct super_
 	kfree(sbi);
 }
 
-unsigned hpfs_count_one_bitmap(struct super_block *s, secno secno)
+static unsigned hpfs_count_one_bitmap(struct super_block *s, secno secno)
 {
 	struct quad_buffer_head qbh;
 	unsigned long *bits;
@@ -123,7 +123,7 @@ unsigned hpfs_count_one_bitmap(struct su
 
 	bits = hpfs_map_4sectors(s, secno, &qbh, 0);
 	if (!bits)
-		return 0;
+		return (unsigned)-1;
 	count = bitmap_weight(bits, 2048 * BITS_PER_BYTE);
 	hpfs_brelse4(&qbh);
 	return count;
@@ -138,30 +138,45 @@ static unsigned count_bitmaps(struct sup
 		hpfs_prefetch_bitmap(s, n);
 	}
 	for (n = 0; n < n_bands; n++) {
+		unsigned c;
 		hpfs_prefetch_bitmap(s, n + COUNT_RD_AHEAD);
-		count += hpfs_count_one_bitmap(s, le32_to_cpu(hpfs_sb(s)->sb_bmp_dir[n]));
+		c = hpfs_count_one_bitmap(s, le32_to_cpu(hpfs_sb(s)->sb_bmp_dir[n]));
+		if (c != (unsigned)-1)
+			count += c;
 	}
 	return count;
 }
 
+unsigned hpfs_get_free_dnodes(struct super_block *s)
+{
+	struct hpfs_sb_info *sbi = hpfs_sb(s);
+	if (sbi->sb_n_free_dnodes == (unsigned)-1) {
+		unsigned c = hpfs_count_one_bitmap(s, sbi->sb_dmap);
+		if (c == (unsigned)-1)
+			return 0;
+		sbi->sb_n_free_dnodes = c;
+	}
+	return sbi->sb_n_free_dnodes;
+}
+
 static int hpfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct super_block *s = dentry->d_sb;
 	struct hpfs_sb_info *sbi = hpfs_sb(s);
 	u64 id = huge_encode_dev(s->s_bdev->bd_dev);
+
 	hpfs_lock(s);
 
-	/*if (sbi->sb_n_free == -1) {*/
+	if (sbi->sb_n_free == (unsigned)-1)
 		sbi->sb_n_free = count_bitmaps(s);
-		sbi->sb_n_free_dnodes = hpfs_count_one_bitmap(s, sbi->sb_dmap);
-	/*}*/
+
 	buf->f_type = s->s_magic;
 	buf->f_bsize = 512;
 	buf->f_blocks = sbi->sb_fs_size;
 	buf->f_bfree = sbi->sb_n_free;
 	buf->f_bavail = sbi->sb_n_free;
 	buf->f_files = sbi->sb_dirband_size / 4;
-	buf->f_ffree = sbi->sb_n_free_dnodes;
+	buf->f_ffree = hpfs_get_free_dnodes(s);
 	buf->f_fsid.val[0] = (u32)id;
 	buf->f_fsid.val[1] = (u32)(id >> 32);
 	buf->f_namelen = 254;
author maintainer
#124 lore raw

PATCH 125/133 usb: core: get config and string descriptors for unauthorized devices

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Thomas Pugliese <[email protected]>

commit 83e83ecb79a8225e79bc8e54e9aff3e0e27658a2 upstream.

There is no need to skip querying the config and string descriptors for unauthorized WUSB devices when usb_new_device is called. It is allowed by WUSB spec. The only action that needs to be delayed until authorization time is the set config. This change allows user mode tools to see the config and string descriptors earlier in enumeration which is needed for some WUSB devices to function properly on Android systems. It also reduces the amount of divergent code paths needed for WUSB devices.

---
 drivers/usb/core/config.c |    7 -------
 drivers/usb/core/hub.c    |   40 +++++++---------------------------------
 2 files changed, 7 insertions(+), 40 deletions(-)
drivers/usb/core/config.c +0 -7
@@ -651,10 +651,6 @@ void usb_destroy_configuration(struct us
  *
  * hub-only!! ... and only in reset path, or usb_new_device()
  * (used by real hubs and virtual root hubs)
- *
- * NOTE: if this is a WUSB device and is not authorized, we skip the
- *       whole thing. A non-authorized USB device has no
- *       configurations.
  */
 int usb_get_configuration(struct usb_device *dev)
 {
@@ -666,8 +662,6 @@ int usb_get_configuration(struct usb_dev
 	struct usb_config_descriptor *desc;
 
 	cfgno = 0;
-	if (dev->authorized == 0)	/* Not really an error */
-		goto out_not_authorized;
 	result = -ENOMEM;
 	if (ncfg > USB_MAXCONFIG) {
 		dev_warn(ddev, "too many configurations: %d, "
@@ -751,7 +745,6 @@ int usb_get_configuration(struct usb_dev
 
 err:
 	kfree(desc);
-out_not_authorized:
 	dev->descriptor.bNumConfigurations = cfgno;
 err2:
 	if (result == -ENOMEM)
drivers/usb/core/hub.c +7 -33
@@ -2242,18 +2242,13 @@ static int usb_enumerate_device(struct u
 			return err;
 		}
 	}
-	if (udev->wusb == 1 && udev->authorized == 0) {
-		udev->product = kstrdup("n/a (unauthorized)", GFP_KERNEL);
-		udev->manufacturer = kstrdup("n/a (unauthorized)", GFP_KERNEL);
-		udev->serial = kstrdup("n/a (unauthorized)", GFP_KERNEL);
-	}
-	else {
-		/* read the standard strings and cache them if present */
-		udev->product = usb_cache_string(udev, udev->descriptor.iProduct);
-		udev->manufacturer = usb_cache_string(udev,
-						      udev->descriptor.iManufacturer);
-		udev->serial = usb_cache_string(udev, udev->descriptor.iSerialNumber);
-	}
+
+	/* read the standard strings and cache them if present */
+	udev->product = usb_cache_string(udev, udev->descriptor.iProduct);
+	udev->manufacturer = usb_cache_string(udev,
+					      udev->descriptor.iManufacturer);
+	udev->serial = usb_cache_string(udev, udev->descriptor.iSerialNumber);
+
 	err = usb_enumerate_device_otg(udev);
 	if (err < 0)
 		return err;
@@ -2435,16 +2430,6 @@ int usb_deauthorize_device(struct usb_de
 	usb_dev->authorized = 0;
 	usb_set_configuration(usb_dev, -1);
 
-	kfree(usb_dev->product);
-	usb_dev->product = kstrdup("n/a (unauthorized)", GFP_KERNEL);
-	kfree(usb_dev->manufacturer);
-	usb_dev->manufacturer = kstrdup("n/a (unauthorized)", GFP_KERNEL);
-	kfree(usb_dev->serial);
-	usb_dev->serial = kstrdup("n/a (unauthorized)", GFP_KERNEL);
-
-	usb_destroy_configuration(usb_dev);
-	usb_dev->descriptor.bNumConfigurations = 0;
-
 out_unauthorized:
 	usb_unlock_device(usb_dev);
 	return 0;
@@ -2472,17 +2457,7 @@ int usb_authorize_device(struct usb_devi
 		goto error_device_descriptor;
 	}
 
-	kfree(usb_dev->product);
-	usb_dev->product = NULL;
-	kfree(usb_dev->manufacturer);
-	usb_dev->manufacturer = NULL;
-	kfree(usb_dev->serial);
-	usb_dev->serial = NULL;
-
 	usb_dev->authorized = 1;
-	result = usb_enumerate_device(usb_dev);
-	if (result < 0)
-		goto error_enumerate;
 	/* Choose and set the configuration.  This registers the interfaces
 	 * with the driver core and lets interface drivers bind to them.
 	 */
@@ -2498,7 +2473,6 @@ int usb_authorize_device(struct usb_devi
 	}
 	dev_info(&usb_dev->dev, "authorized to connect\n");
 
-error_enumerate:
 error_device_descriptor:
 	usb_autosuspend_device(usb_dev);
 error_autoresume:
author maintainer
#125 lore raw

PATCH 126/133 SCSI: bfa: Chinook quad port 16G FC HBA claim issue

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Vijaya Mohan Guvva <[email protected]>

commit dcaf9aed995c2b2a49fb86bbbcfa2f92c797ab5d upstream.

Bfa driver crash is observed while pushing the firmware on to chinook quad port card due to uninitialized bfi_image_ct2 access which gets initialized only for CT2 ASIC based cards after request_firmware(). For quard port chinook (CT2 ASIC based), bfi_image_ct2 is not getting initialized as there is no check for chinook PCI device ID before request_firmware and instead bfi_image_cb is initialized as it is the default case for card type check.

This patch includes changes to read the right firmware for quad port chinook.

---
 drivers/scsi/bfa/bfad.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/scsi/bfa/bfad.c +4 -2
@@ -1832,7 +1832,7 @@ out:
 static u32 *
 bfad_load_fwimg(struct pci_dev *pdev)
 {
-	if (pdev->device == BFA_PCI_DEVICE_ID_CT2) {
+	if (bfa_asic_id_ct2(pdev->device)) {
 		if (bfi_image_ct2_size == 0)
 			bfad_read_firmware(pdev, &bfi_image_ct2,
 				&bfi_image_ct2_size, BFAD_FW_FILE_CT2);
@@ -1842,12 +1842,14 @@ bfad_load_fwimg(struct pci_dev *pdev)
 			bfad_read_firmware(pdev, &bfi_image_ct,
 				&bfi_image_ct_size, BFAD_FW_FILE_CT);
 		return bfi_image_ct;
-	} else {
+	} else if (bfa_asic_id_cb(pdev->device)) {
 		if (bfi_image_cb_size == 0)
 			bfad_read_firmware(pdev, &bfi_image_cb,
 				&bfi_image_cb_size, BFAD_FW_FILE_CB);
 		return bfi_image_cb;
 	}
+
+	return NULL;
 }
 
 static void
author maintainer
#126 lore raw

PATCH 127/133 virtio-scsi: Fix hotcpu_notifier use-after-free with virtscsi_freeze

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Asias He <[email protected]>

commit f466f75385369a181409e46da272db3de6f5c5cb upstream.

vqs are freed in virtscsi_freeze but the hotcpu_notifier is not unregistered. We will have a use-after-free usage when the notifier callback is called after virtscsi_freeze.

Fixes 285e71ea6f3583a85e27cb2b9a7d8c35d4c0d558

("virtio-scsi: reset virtqueue affinity when doing cpu hotplug")

---
 drivers/scsi/virtio_scsi.c |   15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)
drivers/scsi/virtio_scsi.c +14 -1
@@ -957,6 +957,10 @@ static void virtscsi_remove(struct virti
 #ifdef CONFIG_PM
 static int virtscsi_freeze(struct virtio_device *vdev)
 {
+	struct Scsi_Host *sh = virtio_scsi_host(vdev);
+	struct virtio_scsi *vscsi = shost_priv(sh);
+
+	unregister_hotcpu_notifier(&vscsi->nb);
 	virtscsi_remove_vqs(vdev);
 	return 0;
 }
@@ -965,8 +969,17 @@ static int virtscsi_restore(struct virti
 {
 	struct Scsi_Host *sh = virtio_scsi_host(vdev);
 	struct virtio_scsi *vscsi = shost_priv(sh);
+	int err;
+
+	err = virtscsi_init(vdev, vscsi);
+	if (err)
+		return err;
+
+	err = register_hotcpu_notifier(&vscsi->nb);
+	if (err)
+		vdev->config->del_vqs(vdev);
 
-	return virtscsi_init(vdev, vscsi);
+	return err;
 }
 #endif
 
author maintainer
#127 lore raw

PATCH 128/133 iscsi-target: Pre-allocate more tags to avoid ack starvation

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Nicholas Bellinger <[email protected]>

commit 4a4caa29f1abcb14377e05d57c0793d338fb945d upstream.

This patch addresses an traditional iscsi-target fabric ack starvation issue where iscsit_allocate_cmd() -> percpu_ida_alloc_state() ends up hitting slow path percpu-ida code, because iscsit_ack_from_expstatsn() is expected to free ack'ed tags after tag allocation.

This is done to take into account the tags waiting to be acknowledged and released in iscsit_ack_from_expstatsn(), but who's number are not directly limited by the CmdSN Window queue_depth being enforced by the target.

So that said, this patch bumps up the pre-allocated number of per session tags to:

(max(queue_depth, ISCSIT_MIN_TAGS) * 2) + ISCSIT_EXTRA_TAGS

for good measure to avoid the percpu_ida_alloc_state() slow path.

---
 drivers/target/iscsi/iscsi_target_nego.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/target/iscsi/iscsi_target_nego.c +1 -1
@@ -1192,7 +1192,7 @@ get_target:
 	 */
 alloc_tags:
 	tag_num = max_t(u32, ISCSIT_MIN_TAGS, queue_depth);
-	tag_num += (tag_num / 2) + ISCSIT_EXTRA_TAGS;
+	tag_num = (tag_num * 2) + ISCSIT_EXTRA_TAGS;
 	tag_size = sizeof(struct iscsi_cmd) + conn->conn_transport->priv_size;
 
 	ret = transport_alloc_session_tags(sess->se_sess, tag_num, tag_size);
author maintainer
#128 lore raw

PATCH 129/133 target/iscsi: Fix network portal creation race

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Andy Grover <[email protected]>

commit ee291e63293146db64668e8d65eb35c97e8324f4 upstream.

When creating network portals rapidly, such as when restoring a configuration, LIO's code to reuse existing portals can return a false negative if the thread hasn't run yet and set np_thread_state to ISCSI_NP_THREAD_ACTIVE. This causes an error in the network stack when attempting to bind to the same address/port.

This patch sets NP_THREAD_ACTIVE before the np is placed on g_np_list, so even if the thread hasn't run yet, iscsit_get_np will return the existing np.

Also, convert np_lock -> np_mutex + hold across adding new net portal to g_np_list to prevent a race where two threads may attempt to create the same network portal, resulting in one of them failing.

(nab: Add missing mutex_unlocks in iscsit_add_np failure paths) (DanC: Fix incorrect spin_unlock -> spin_unlock_bh)

---
 drivers/target/iscsi/iscsi_target.c |   34 +++++++++++++++++++++-------------
 1 file changed, 21 insertions(+), 13 deletions(-)
drivers/target/iscsi/iscsi_target.c +21 -13
@@ -52,7 +52,7 @@
 static LIST_HEAD(g_tiqn_list);
 static LIST_HEAD(g_np_list);
 static DEFINE_SPINLOCK(tiqn_lock);
-static DEFINE_SPINLOCK(np_lock);
+static DEFINE_MUTEX(np_lock);
 
 static struct idr tiqn_idr;
 struct idr sess_idr;
@@ -307,6 +307,9 @@ bool iscsit_check_np_match(
 	return false;
 }
 
+/*
+ * Called with mutex np_lock held
+ */
 static struct iscsi_np *iscsit_get_np(
 	struct __kernel_sockaddr_storage *sockaddr,
 	int network_transport)
@@ -314,11 +317,10 @@ static struct iscsi_np *iscsit_get_np(
 	struct iscsi_np *np;
 	bool match;
 
-	spin_lock_bh(&np_lock);
 	list_for_each_entry(np, &g_np_list, np_list) {
-		spin_lock(&np->np_thread_lock);
+		spin_lock_bh(&np->np_thread_lock);
 		if (np->np_thread_state != ISCSI_NP_THREAD_ACTIVE) {
-			spin_unlock(&np->np_thread_lock);
+			spin_unlock_bh(&np->np_thread_lock);
 			continue;
 		}
 
@@ -330,13 +332,11 @@ static struct iscsi_np *iscsit_get_np(
 			 * while iscsi_tpg_add_network_portal() is called.
 			 */
 			np->np_exports++;
-			spin_unlock(&np->np_thread_lock);
-			spin_unlock_bh(&np_lock);
+			spin_unlock_bh(&np->np_thread_lock);
 			return np;
 		}
-		spin_unlock(&np->np_thread_lock);
+		spin_unlock_bh(&np->np_thread_lock);
 	}
-	spin_unlock_bh(&np_lock);
 
 	return NULL;
 }
@@ -350,16 +350,22 @@ struct iscsi_np *iscsit_add_np(
 	struct sockaddr_in6 *sock_in6;
 	struct iscsi_np *np;
 	int ret;
+
+	mutex_lock(&np_lock);
+
 	/*
 	 * Locate the existing struct iscsi_np if already active..
 	 */
 	np = iscsit_get_np(sockaddr, network_transport);
-	if (np)
+	if (np) {
+		mutex_unlock(&np_lock);
 		return np;
+	}
 
 	np = kzalloc(sizeof(struct iscsi_np), GFP_KERNEL);
 	if (!np) {
 		pr_err("Unable to allocate memory for struct iscsi_np\n");
+		mutex_unlock(&np_lock);
 		return ERR_PTR(-ENOMEM);
 	}
 
@@ -382,6 +388,7 @@ struct iscsi_np *iscsit_add_np(
 	ret = iscsi_target_setup_login_socket(np, sockaddr);
 	if (ret != 0) {
 		kfree(np);
+		mutex_unlock(&np_lock);
 		return ERR_PTR(ret);
 	}
 
@@ -390,6 +397,7 @@ struct iscsi_np *iscsit_add_np(
 		pr_err("Unable to create kthread: iscsi_np\n");
 		ret = PTR_ERR(np->np_thread);
 		kfree(np);
+		mutex_unlock(&np_lock);
 		return ERR_PTR(ret);
 	}
 	/*
@@ -400,10 +408,10 @@ struct iscsi_np *iscsit_add_np(
 	 * point because iscsi_np has not been added to g_np_list yet.
 	 */
 	np->np_exports = 1;
+	np->np_thread_state = ISCSI_NP_THREAD_ACTIVE;
 
-	spin_lock_bh(&np_lock);
 	list_add_tail(&np->np_list, &g_np_list);
-	spin_unlock_bh(&np_lock);
+	mutex_unlock(&np_lock);
 
 	pr_debug("CORE[0] - Added Network Portal: %s:%hu on %s\n",
 		np->np_ip, np->np_port, np->np_transport->name);
@@ -470,9 +478,9 @@ int iscsit_del_np(struct iscsi_np *np)
 
 	np->np_transport->iscsit_free_np(np);
 
-	spin_lock_bh(&np_lock);
+	mutex_lock(&np_lock);
 	list_del(&np->np_list);
-	spin_unlock_bh(&np_lock);
+	mutex_unlock(&np_lock);
 
 	pr_debug("CORE[0] - Removed Network Portal: %s:%hu on %s\n",
 		np->np_ip, np->np_port, np->np_transport->name);
author maintainer
#129 lore raw

PATCH 130/133 Btrfs: handle EAGAIN case properly in btrfs_drop_snapshot()

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Wang Shilong <[email protected]>

commit 90515e7f5d7d24cbb2a4038a3f1b5cfa2921aa17 upstream.

We may return early in btrfs_drop_snapshot(), we shouldn't call btrfs_std_err() for this case, fix it.

---
 fs/btrfs/extent-tree.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/extent-tree.c +1 -1
@@ -7718,7 +7718,7 @@ out:
 	 */
 	if (!for_reloc && root_dropped == false)
 		btrfs_add_dead_root(root);
-	if (err)
+	if (err && err != -EAGAIN)
 		btrfs_std_error(root->fs_info, err);
 	return err;
 }
author maintainer
#130 lore raw

PATCH 131/133 btrfs: restrict snapshotting to own subvolumes

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: David Sterba <[email protected]>

commit d024206133ce21936b3d5780359afc00247655b7 upstream.

Currently, any user can snapshot any subvolume if the path is accessible and thus indirectly create and keep files he does not own under his direcotries. This is not possible with traditional directories.

In security context, a user can snapshot root filesystem and pin any potentially buggy binaries, even if the updates are applied.

All the snapshots are visible to the administrator, so it's possible to verify if there are suspicious snapshots.

Another more practical problem is that any user can pin the space used by eg. root and cause ENOSPC.

Original report:
https://bugs.launchpad.net/ubuntu/+source/apparmor/+bug/484786

---
 fs/btrfs/ioctl.c |    6 ++++++
 1 file changed, 6 insertions(+)
fs/btrfs/ioctl.c +6 -0
@@ -1548,6 +1548,12 @@ static noinline int btrfs_ioctl_snap_cre
 			printk(KERN_INFO "btrfs: Snapshot src from "
 			       "another FS\n");
 			ret = -EINVAL;
+		} else if (!inode_owner_or_capable(src_inode)) {
+			/*
+			 * Subvolume creation is not restricted, but snapshots
+			 * are limited to own subvolumes only
+			 */
+			ret = -EPERM;
 		} else {
 			ret = btrfs_mksubvol(&file->f_path, name, namelen,
 					     BTRFS_I(src_inode)->root,
author maintainer
#131 lore raw

PATCH 132/133 powerpc: Fix the setup of CPU-to-Node mappings during CPU online

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: "Srivatsa S. Bhat" <[email protected]>

commit d4edc5b6c480a0917e61d93d55531d7efa6230be upstream.

On POWER platforms, the hypervisor can notify the guest kernel about dynamic changes in the cpu-numa associativity (VPHN topology update). Hence the cpu-to-node mappings that we got from the firmware during boot, may no longer be valid after such updates. This is handled using the arch_update_cpu_topology() hook in the scheduler, and the sched-domains are rebuilt according to the new mappings.

But unfortunately, at the moment, CPU hotplug ignores these updated mappings and instead queries the firmware for the cpu-to-numa relationships and uses them during CPU online. So the kernel can end up assigning wrong NUMA nodes to CPUs during subsequent CPU hotplug online operations (after booting).

Further, a particularly problematic scenario can result from this bug:
On POWER platforms, the SMT mode can be switched between 1, 2, 4 (and even 8) threads per core. The switch to Single-Threaded (ST) mode is performed by offlining all except the first CPU thread in each core. Switching back to SMT mode involves onlining those other threads back, in each core.

Now consider this scenario:

1. During boot, the kernel gets the cpu-to-node mappings from the firmware and assigns the CPUs to NUMA nodes appropriately, during CPU online.

2. Later on, the hypervisor updates the cpu-to-node mappings dynamically and communicates this update to the kernel. The kernel in turn updates its cpu-to-node associations and rebuilds its sched domains. Everything is fine so far.

3. Now, the user switches the machine from SMT to ST mode (say, by running ppc64_cpu --smt=1). This involves offlining all except 1 thread in each core.

4. The user then tries to switch back from ST to SMT mode (say, by running ppc64_cpu --smt=4), and this involves onlining those threads back. Since CPU hotplug ignores the new mappings, it queries the firmware and tries to associate the newly onlined sibling threads to the old NUMA nodes. This results in sibling threads within the same core getting associated with different NUMA nodes, which is incorrect.

The scheduler's build-sched-domains code gets thoroughly confused with this and enters an infinite loop and causes soft-lockups, as explained in detail in commit 3be7db6ab (powerpc: VPHN topology change updates all siblings).

So to fix this, use the numa_cpu_lookup_table to remember the updated cpu-to-node mappings, and use them during CPU hotplug online operations. Further, we also need to ensure that all threads in a core are assigned to a common NUMA node, irrespective of whether all those threads were online during the topology update. To achieve this, we take care not to use cpu_sibling_mask() since it is not hotplug invariant. Instead, we use cpu_first_sibling_thread() and set up the mappings manually using the 'threads_per_core' value for that particular platform. This helps us ensure that we don't hit this bug with any combination of CPU hotplug and SMT mode switching.

---
 arch/powerpc/include/asm/topology.h |   10 ++++-
 arch/powerpc/mm/numa.c              |   70 ++++++++++++++++++++++++++++++++++--
 2 files changed, 76 insertions(+), 4 deletions(-)
arch/powerpc/include/asm/topology.h +9 -1
@@ -22,7 +22,15 @@ struct device_node;
 
 static inline int cpu_to_node(int cpu)
 {
-	return numa_cpu_lookup_table[cpu];
+	int nid;
+
+	nid = numa_cpu_lookup_table[cpu];
+
+	/*
+	 * During early boot, the numa-cpu lookup table might not have been
+	 * setup for all CPUs yet. In such cases, default to node 0.
+	 */
+	return (nid < 0) ? 0 : nid;
 }
 
 #define parent_node(node)	(node)
arch/powerpc/mm/numa.c +67 -3

111 lines not on this page · raw message

author maintainer
#132 lore raw

PATCH 133/133 powerpc: Make sure "cache" directory is removed when offlining cpu

In reply to Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Paul Mackerras <[email protected]>

commit 91b973f90c1220d71923e7efe1e61f5329806380 upstream.

The code in remove_cache_dir() is supposed to remove the "cache" subdirectory from the sysfs directory for a CPU when that CPU is being offlined. It tries to do this by calling kobject_put() on the kobject for the subdirectory. However, the subdirectory only gets removed once the last reference goes away, and the reference being put here may well not be the last reference. That means that the "cache" subdirectory may still exist when the offlining operation has finished. If the same CPU subsequently gets onlined, the code tries to add a new "cache" subdirectory. If the old subdirectory has not yet been removed, we get a WARN_ON in the sysfs code, with stack trace, and an error message printed on the console. Further, we ultimately end up with an online cpu with no "cache" subdirectory.

This fixes it by doing an explicit kobject_del() at the point where we want the subdirectory to go away. kobject_del() removes the sysfs directory even though the object still exists in memory. The object will get freed at some point in the future. A subsequent onlining operation can create a new sysfs directory, even if the old object still exists in memory, without causing any problems.

---
 arch/powerpc/kernel/cacheinfo.c |    3 +++
 1 file changed, 3 insertions(+)
arch/powerpc/kernel/cacheinfo.c +3 -0
@@ -794,6 +794,9 @@ static void remove_cache_dir(struct cach
 {
 	remove_index_dirs(cache_dir);
 
+	/* Remove cache dir from sysfs */
+	kobject_del(cache_dir->kobj);
+
 	kobject_put(cache_dir->kobj);
 
 	kfree(cache_dir);
#10 lore raw

Re: [PATCH 3.12 010/133] x86: Add check for number of available vectors before CPU down

In reply to Greg Kroah-Hartman
Quoted 10 lines · Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Prarit Bhargava <[email protected]>

commit da6139e49c7cb0f4251265cb5243b8d220adb48d upstream.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=64791

Greg, there is a follow up to this patch that removes declarations from the stack and makes them global within the file. This was done to avoid a -Wframe-larger-than warning that was found in with 0day kernel build service, kbuild.

So I don't think this should go into -stable yet. I've pushed a patch onto LKML to resolve the kbuild warning, and hopefully on the x86 maintainers pick it up.

P.

author maintainer
#11 lore raw

Re: [PATCH 3.12 010/133] x86: Add check for number of available vectors before CPU down

Quoted 14 lines · Prarit Bhargava
Quoted 10 lines · Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Prarit Bhargava <[email protected]>

commit da6139e49c7cb0f4251265cb5243b8d220adb48d upstream.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=64791

Quoted 8 lines

Greg, there is a follow up to this patch that removes declarations from the stack and makes them global within the file. This was done to avoid a -Wframe-larger-than warning that was found in with 0day kernel build service, kbuild.

So I don't think this should go into -stable yet. I've pushed a patch onto LKML to resolve the kbuild warning, and hopefully on the x86 maintainers pick it up.

Ok, I've dropped it from the 3.10, 3.12, and 3.13-stable queues. Please let [email protected] know when the fix is in Linus's tree, so that this, and the fix, can be added to the stable trees.

thanks,

greg k-h

#133 lore raw
In reply to Greg Kroah-Hartman
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 3.12.10 release. There are 133 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu Feb 6 21:07:17 UTC 2014. Anything received after that time might be too late.

Version tested: v3.12.9-132

Build results:

    total: 126 pass: 121 skipped: 5 fail: 0

qemu tests all passed.

Results are as expected.

Details are available at http://server.roeck-us.net:8010/builders.

Guenter

#134 lore raw
Quoted 22 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 3.12.10 release. There are 133 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu Feb 6 21:07:17 UTC 2014. Anything received after that time might be too late.

Version tested: v3.12.9-132

Build results:

     total: 126 pass: 121 skipped: 5 fail: 0

qemu tests all passed.

Results are as expected.

Details are available at http://server.roeck-us.net:8010/builders.

Quoted 10 lines

Guenter

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Patch applied cleanly. boot tests passed on all test systems. No dmesg regressions: emerg, crit, alert, err are clean. No regressions in warn.

Minor Note: git apply --index complained about extra white space at the end of patch file. patch -p1 worked just fine.

-- Shuah

Signature · 4 lines
Shuah Khan
Senior Linux Kernel Developer - Open Source Group
Samsung Research America(Silicon Valley)
[email protected] | (970) 672-0658
#135 lore raw
Quoted 34 lines · Shuah Khan
Quoted 22 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 3.12.10 release. There are 133 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu Feb 6 21:07:17 UTC 2014. Anything received after that time might be too late.

Version tested: v3.12.9-132

Build results:

    total: 126 pass: 121 skipped: 5 fail: 0

qemu tests all passed.

Results are as expected.

Details are available at http://server.roeck-us.net:8010/builders.

Quoted 10 lines

Guenter

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Quoted 7 lines

Patch applied cleanly. boot tests passed on all test systems. No dmesg regressions: emerg, crit, alert, err are clean. No regressions in warn.

Minor Note: git apply --index complained about extra white space at the end of patch file. patch -p1 worked just fine.

I noticed. Usually that is a problem with the original patch, and Greg (rightfully) doesn't modify it just to avoid such messages.

Guenter

#114 lore raw

Re: [PATCH 3.12 117/133] alpha: fix broken network checksum

In reply to Greg Kroah-Hartman
Quoted 11 lines · Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Mikulas Patocka <[email protected]>

commit 0ef38d70d4118b2ce1a538d14357be5ff9dc2bbd upstream.

The patch 3ddc5b46a8e90f3c9251338b60191d0a804b0d92 breaks networking on alpha (there is a follow-up fix 5cfe8f1ba5eebe6f4b6e5858cdb1a5be4f3272a6, but networking is still broken even with the second patch).

I wonder whether that follow-up (partial) fix mentioned (5cfe8f1ba5eebe6f4b6e5858cdb1a5be4f3272a6) should be pulled into 3.12 stable too?

Cheers
Michael.

#115 lore raw

Re: [PATCH 3.12 117/133] alpha: fix broken network checksum

Quoted 13 lines · Michael Cree
Quoted 11 lines · Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Mikulas Patocka <[email protected]>

commit 0ef38d70d4118b2ce1a538d14357be5ff9dc2bbd upstream.

The patch 3ddc5b46a8e90f3c9251338b60191d0a804b0d92 breaks networking on alpha (there is a follow-up fix 5cfe8f1ba5eebe6f4b6e5858cdb1a5be4f3272a6, but networking is still broken even with the second patch).

Quoted 6 lines

I wonder whether that follow-up (partial) fix mentioned (5cfe8f1ba5eebe6f4b6e5858cdb1a5be4f3272a6) should be pulled into 3.12 stable too?

Cheers
Michael.

Hi

It is quite confusing - csum_partial_copy_from_user is only called from csum_and_copy_from_user in include/net/checksum.h.

csum_and_copy_from_user already verifies the source memory range with access_ok. Despite this, people started to add access_ok to many architecture-specific versions of csum_partial_copy_from_user (for example 3ddc5b46a8e90f3c9251338b60191d0a804b0d92).

It seems that the best thing is to revert 0ef38d70d4118b2ce1a538d14357be5ff9dc2bbd, 5cfe8f1ba5eebe6f4b6e5858cdb1a5be4f3272a6 and csum_partial_copy_from_user change from 3ddc5b46a8e90f3c9251338b60191d0a804b0d92.

David, you are maintainer of networking - does it have any deeper sense to perform access_ok in csum_partial_copy_from_user if the caller performs this check already? Should this be just removed?

Also - the x86 and alpha implementation of csum_and_copy_from_user zero the destination buffer on userspace-access error - but csum_and_copy_from_user (that calls these function) doesn't zero the destination buffer on error. Should the destination buffer be zeroed on error or not?

Mikulas

#116 lore raw

Re: [PATCH 3.12 117/133] alpha: fix broken network checksum

In reply to Michael Cree
Quoted 13 lines · Michael Cree
Quoted 11 lines · Greg Kroah-Hartman

3.12-stable review patch. If anyone has any objections, please let me know.

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

From: Mikulas Patocka <[email protected]>

commit 0ef38d70d4118b2ce1a538d14357be5ff9dc2bbd upstream.

The patch 3ddc5b46a8e90f3c9251338b60191d0a804b0d92 breaks networking on alpha (there is a follow-up fix 5cfe8f1ba5eebe6f4b6e5858cdb1a5be4f3272a6, but networking is still broken even with the second patch).

Quoted 6 lines

I wonder whether that follow-up (partial) fix mentioned (5cfe8f1ba5eebe6f4b6e5858cdb1a5be4f3272a6) should be pulled into 3.12 stable too?

Cheers
Michael.

I think it doesn't have to be backported.

Mikulas