3.15.5-stable review
Subsystems (3)
Greg Kroah-Hartman
maintainercover 3.15.5-stable review
This is the start of the stable review cycle for the 3.15.5 release. There are 122 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 Wed Jul 9 23:57:10 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-3.15.5-rc1.gz
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 3.15.5-rc1
Rik van Riel <[email protected]>
mm/numa: Remove BUG_ON() in __handle_mm_fault()
Zhichuang SUN <[email protected]>
drivers/video/fbdev/fb-puv3.c: Add header files for function unifb_mmap
Chen Gang <[email protected]>
arch/unicore32/mm/alignment.c: include "asm/pgtable.h" to avoid compiling error
Imre Deak <[email protected]>
drm/i915: fix display power sw state reporting
Micky Ching <[email protected]>
mmc: rtsx: add R1-no-CRC mmc command type handle
Florian Westphal <[email protected]>
netfilter: ctnetlink: fix refcnt leak in dying/unconfirmed list dumper
Florian Westphal <[email protected]>
netfilter: nf_nat: fix oops on netns removal
Pablo Neira Ayuso <[email protected]>
netfilter: ctnetlink: fix dumping of dying/unconfirmed conntracks
Peter Hurley <[email protected]>
tty: Correct INPCK handling
Peter Hurley <[email protected]>
serial: Fix IGNBRK handling
Stephen Boyd <[email protected]>
clk: qcom: Fix mmcc-8974's PLL configurations
Stephen Boyd <[email protected]>
clk: qcom: Fix clk_rcg2_is_enabled() check
Wei Yang <[email protected]>
net/mlx4_core: Keep only one driver entry release mlx4_priv
Jack Morgenstein <[email protected]>
mlx4_core: Fix incorrect FLAGS1 bitmap test in mlx4_QUERY_FUNC_CAP
Vineet Gupta <[email protected]>
ARC: Fix build breakage for !CONFIG_ARC_DW2_UNWIND
Thomas Gleixner <[email protected]>
irqchip: spear_shirq: Fix interrupt offset
NeilBrown <[email protected]>
md: flush writes before starting a recovery.
Josef Bacik <[email protected]>
btrfs: only unlock block in verify_parent_transid if we locked it
Oleg Nesterov <[email protected]>
tracing/uprobes: Fix the usage of uprobe_buffer_enable() in probe_event_enable()
Oleg Nesterov <[email protected]>
tracing/uprobes: Revert "Support mix of ftrace and perf"
Steven Rostedt (Red Hat) <[email protected]>
tracing: Remove ftrace_stop/start() from reading the trace file
Suman Anna <[email protected]>
staging: tidspbridge: fix an erroneous removal of parentheses
Michal Nazarewicz <[email protected]>
tools: ffs-test: fix header values endianess
J. Bruce Fields <[email protected]>
nfsd: fix rare symlink decoding bug
Dan Carpenter <[email protected]>
staging: iio/ad7291: fix error code in ad7291_probe()
Adam Thomson <[email protected]>
iio: of_iio_channel_get_by_name() returns non-null pointers for error legs
Xiaoming Gao <[email protected]>
kvm: fix wrong address when writing Hyper-V tsc page
Paolo Bonzini <[email protected]>
KVM: x86: preserve the high 32-bits of the PAT register
Nadav Amit <[email protected]>
KVM: x86: Increase the number of fixed MTRR regs to 10
Minchan Kim <[email protected]>
zram: revalidate disk after capacity change
Michal Nazarewicz <[email protected]>
mm: page_alloc: fix CMA area initialisation when pageblock > MAX_ORDER
Jan Kara <[email protected]>
ext4: Fix hole punching for files with indirect blocks
Jan Kara <[email protected]>
ext4: Fix buffer double free in ext4_alloc_branch()
Tejun Heo <[email protected]>
blkcg: fix use-after-free in __blkg_release_rcu() by making blkcg_gq refcnt an atomic_t
Jiang Liu <[email protected]>
iommu/vt-d: fix bug in handling multiple RMRRs for the same PCI device
Keerthy <[email protected]>
regulator: tps65218: Correct the the config register for LDO1
Keerthy <[email protected]>
regulator: tps65218: Add the missing of_node assignment in probe
Steve French <[email protected]>
CIFS: fix mount failure with broken pathnames when smb3 mount with mapchars option
Björn Baumbach <[email protected]>
fs/cifs: fix regression in cifs_create_mf_symlink()
Rafał Miłecki <[email protected]>
b43: fix frequency reported on G-PHY with /new/ firmware
Maxime Ripard <[email protected]>
net: allwinner: emac: Add missing free_irq
ChiaHao <[email protected]>
arm64: Bug fix in stack alignment exception
David R. Piegdon <[email protected]>
ARM: OMAP2+: Fix parser-bug in platform muxing code
Steve Capper <[email protected]>
arm64: mm: Make icache synchronisation logic huge page aware
Gregory CLEMENT <[email protected]>
ARM: mvebu: Fix the improper use of the compatible string armada38x using a wildcard
Andrew Morton <[email protected]>
ia64: arch/ia64/include/uapi/asm/fcntl.h needs personality.h
David Rientjes <[email protected]>
mm, pcp: allow restoring percpu_pagelist_fraction default
Naoya Horiguchi <[email protected]>
hugetlb: fix copy_hugetlb_page_range() to handle migration/hwpoisoned entry
Steven Miao <[email protected]>
mm: nommu: per-thread vma cache fix
Emmanuel Grumbach <[email protected]>
iwlwifi: pcie: try to get ownership several times
Felix Fietkau <[email protected]>
mac80211: fix a memory leak on sta rate selection table
Arik Nemtsov <[email protected]>
mac80211: don't check netdev state for debugfs read/write
Krzysztof Hałasa <[email protected]>
mac80211: fix IBSS join by initializing last_scan_completed
Paul Kocialkowski <[email protected]>
twl4030-madc: Request processed values in twl4030_get_madc_conversion
Doug Smythies <[email protected]>
intel_pstate: Correct rounding in busy calculation
Fabio Baltieri <[email protected]>
hwmon: (ina2xx) Cast to s16 on shunt and current regs
Ilya Dryomov <[email protected]>
rbd: handle parent_overlap on writes correctly
Alex Elder <[email protected]>
rbd: use reference counts for image requests
Lukas Czerner <[email protected]>
dm thin: update discard_granularity to reflect the thin-pool blocksize
Joe Thornber <[email protected]>
dm era: check for a non-NULL metadata object before closing it
Marcin Kraglak <[email protected]>
Bluetooth: Allow change security level on ATT_CID in slave role
Johan Hedberg <[email protected]>
Bluetooth: Fix locking of hdev when calling into SMP code
Jukka Taimisto <[email protected]>
Bluetooth: Fix deadlock in l2cap_conn_del()
Johan Hedberg <[email protected]>
Bluetooth: Fix setting correct authentication information for SMP STK
Johan Hedberg <[email protected]>
Bluetooth: Reuse hci_stop_discovery function when cleaning up HCI state
Johan Hedberg <[email protected]>
Bluetooth: Refactor discovery stopping into its own function
Johan Hedberg <[email protected]>
Bluetooth: Fix indicating discovery state when canceling inquiry
Johan Hedberg <[email protected]>
Bluetooth: Fix check for connection encryption
Johan Hedberg <[email protected]>
Bluetooth: Fix SSP acceptor just-works confirmation without MITM
Johan Hedberg <[email protected]>
Bluetooth: Fix incorrectly overriding conn->src_type
Guido Martínez <[email protected]>
drm/i2c: tda998x: move drm_i2c_encoder_destroy call
Thomas Hellstrom <[email protected]>
drm/vmwgfx: Fix incorrect write to read-only register v2:
Mengdong Lin <[email protected]>
ALSA: hda - restore BCLK M/N value as per CDCLK for HSW/BDW display HDA controller
Jani Nikula <[email protected]>
drm/i915: provide interface for audio driver to query cdclk
Zhaowei Yuan <[email protected]>
drm: fix NULL pointer access by wrong ioctl
Chris Wilson <[email protected]>
drm/i95: Initialize active ring->pid to -1
Jani Nikula <[email protected]>
drm/i915: default to having backlight if VBT not available
Chris Wilson <[email protected]>
drm/i915: Hold the table lock whilst walking the file's idr and counting the objects in debugfs
Takashi Iwai <[email protected]>
drm/i915, HD-audio: Don't continue probing when nomodeset is given
Jani Nikula <[email protected]>
drm/i915: set backlight duty cycle after backlight enable for gen4
Ville Syrjälä <[email protected]>
drm/i915: Avoid div-by-zero when pixel_multiplier is zero
Chris Wilson <[email protected]>
drm/i915: Disable FBC by default also on Haswell and later
Chris Wilson <[email protected]>
drm/i915: Reorder semaphore deadlock check
Ben Widawsky <[email protected]>
drm/i915/bdw: Only use 2g GGTT for 32b platforms
Ilia Mirkin <[email protected]>
drm/gk208/gr: add missing registers to grctx init
Mario Kleiner <[email protected]>
drm/nv50-/mc: fix kms pageflip events by reordering irq handling order.
Alex Deucher <[email protected]>
drm/radeon/cik: fix typo in EOP packet
Alex Deucher <[email protected]>
drm/radeon/dpm: fix vddci setup typo on cayman
Alex Deucher <[email protected]>
drm/radeon/dpm: fix typo in vddci setup for eg/btc
Alex Deucher <[email protected]>
drm/radeon: add missing vce init case for hawaii
Alex Deucher <[email protected]>
drm/radeon/atom: fix dithering on certain panels
Alex Deucher <[email protected]>
drm/radeon/dp: fix lane/clock setup for dp 1.2 capable devices
Alex Deucher <[email protected]>
drm/radeon: fix typo in radeon_connector_is_dp12_capable()
Alex Deucher <[email protected]>
drm/radeon: only apply hdmi bpc pll flags when encoder mode is hdmi
Mario Kleiner <[email protected]>
drm/nouveau/kms: reference vblank for crtc during pageflip.
Mario Kleiner <[email protected]>
drm/nouveau/disp/nv04-nv40: abort scanoutpos query on vga analog.
Mario Kleiner <[email protected]>
drm/nouveau/kms/nv04-nv40: fix pageflip events via special case.
Alex Deucher <[email protected]>
vgaswitcheroo: switch the mux to the igp on power down when runpm is enabled
Thomas Petazzoni <[email protected]>
mtd: pxa3xx_nand: make the driver work on big-endian systems
pekon gupta <[email protected]>
mtd: nand: omap: fix BCHx ecc.correct to return detected bit-flips in erased-page
Pekon Gupta <[email protected]>
mtd: eLBC NAND: fix subpage write support
Stanislaw Gruszka <[email protected]>
rt2x00: fix rfkill regression on rt2500pci
Stanislaw Gruszka <[email protected]>
rt2x00: disable TKIP on USB
Michal Nazarewicz <[email protected]>
usb: gadget: f_fs: fix NULL pointer dereference when there are no strings
Michal Nazarewicz <[email protected]>
usb: gadget: f_fs: resurect usb_functionfs_descs_head structure
Johan Hovold <[email protected]>
USB: ftdi_sio: fix null deref at port probe
Peter Chen <[email protected]>
usb: chipidea: udc: delete td from req's td list at ep_dequeue
Ezequiel Garcia <[email protected]>
usb: musb: Fix panic upon musb_am335x module removal
Thomas Gleixner <[email protected]>
usb: musb: Ensure that cppi41 timer gets armed on premature DMA TX irq
Linus Walleij <[email protected]>
usb: musb: ux500: don't propagate the OF node
Bjørn Mork <[email protected]>
usb: option: add/modify Olivetti Olicard modems
Oliver Neukum <[email protected]>
USB: option: add device ID for SpeedUp SU9800 usb 3g modem
Wang, Yu <[email protected]>
xhci: Fix runtime suspended xhci from blocking system suspend.
Mathias Nyman <[email protected]>
xhci: correct burst count field for isoc transfers on 1.0 xhci hosts
Mathias Nyman <[email protected]>
xhci: Use correct SLOT ID when handling a reset device command
Paolo Bonzini <[email protected]>
virtio-scsi: fix various bad behavior on aborted requests
Ulrich Obergfell <[email protected]>
scsi_error: fix invalid setting of host byte
Paolo Bonzini <[email protected]>
virtio-scsi: avoid cancelling uninitialized work items
Brian King <[email protected]>
ibmvscsi: Add memory barriers for send / receive
Brian King <[email protected]>
ibmvscsi: Abort init sequence during error recovery
Alan Stern <[email protected]>
usb-storage/SCSI: Add broken_fua blacklist flag
Martin K. Petersen <[email protected]>
SCSI: use the scsi data buffer length to extract transfer size
-------------
Diffstat:
.../devicetree/bindings/arm/armada-38x.txt | 14 ++- Documentation/sysctl/vm.txt | 3 +- Makefile | 4 +- arch/arc/kernel/ctx_sw_asm.S | 2 +- arch/arm/boot/dts/armada-380.dtsi | 2 +- arch/arm/boot/dts/armada-385-db.dts | 2 +- arch/arm/boot/dts/armada-385-rd.dts | 2 +- arch/arm/boot/dts/armada-385.dtsi | 2 +- arch/arm/boot/dts/armada-38x.dtsi | 2 +- arch/arm/mach-omap2/mux.c | 6 +- arch/arm64/kernel/entry.S | 1 - arch/arm64/mm/flush.c | 3 +- arch/ia64/include/uapi/asm/fcntl.h | 1 + arch/unicore32/mm/alignment.c | 1 + arch/x86/include/asm/kvm_host.h | 4 +- arch/x86/kvm/x86.c | 2 +- block/blk-cgroup.c | 7 +- block/blk-cgroup.h | 17 ++-- drivers/block/rbd.c | 19 +++- drivers/block/zram/zram_drv.c | 5 +- drivers/clk/qcom/clk-rcg2.c | 2 +- drivers/clk/qcom/mmcc-msm8974.c | 13 ++- drivers/cpufreq/intel_pstate.c | 5 +- drivers/gpu/drm/drm_drv.c | 3 +- drivers/gpu/drm/i2c/tda998x_drv.c | 2 +- drivers/gpu/drm/i915/i915_debugfs.c | 2 + drivers/gpu/drm/i915/i915_gem_gtt.c | 7 ++ drivers/gpu/drm/i915/i915_gpu_error.c | 3 +- drivers/gpu/drm/i915/i915_irq.c | 18 +++- drivers/gpu/drm/i915/intel_bios.c | 6 +- drivers/gpu/drm/i915/intel_panel.c | 5 +- drivers/gpu/drm/i915/intel_pm.c | 55 +++++++++-- drivers/gpu/drm/i915/intel_ringbuffer.h | 2 +- drivers/gpu/drm/i915/intel_sdvo.c | 4 +- drivers/gpu/drm/nouveau/core/engine/disp/nv04.c | 8 ++ .../gpu/drm/nouveau/core/engine/graph/ctxnv108.c | 2 + drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c | 2 +- drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c | 2 +- drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c | 2 +- drivers/gpu/drm/nouveau/nouveau_display.c | 17 +++- drivers/gpu/drm/radeon/atombios_crtc.c | 48 +++++----- drivers/gpu/drm/radeon/atombios_dp.c | 17 +++- drivers/gpu/drm/radeon/atombios_encoders.c | 5 +- drivers/gpu/drm/radeon/cikd.h | 2 +- drivers/gpu/drm/radeon/cypress_dpm.c | 2 +- drivers/gpu/drm/radeon/ni_dpm.c | 2 +- drivers/gpu/drm/radeon/radeon_connectors.c | 2 +- drivers/gpu/drm/radeon/radeon_vce.c | 1 + drivers/gpu/drm/vmwgfx/vmwgfx_fb.c | 1 - drivers/gpu/vga/vga_switcheroo.c | 3 +- drivers/hwmon/ina2xx.c | 7 +- drivers/iio/adc/twl4030-madc.c | 1 + drivers/iio/inkern.c | 6 +- drivers/iommu/intel-iommu.c | 9 +- drivers/irqchip/spear-shirq.c | 2 +- drivers/md/dm-era-target.c | 3 +- drivers/md/dm-thin.c | 3 +- drivers/md/md.c | 13 +++ drivers/mmc/host/rtsx_pci_sdmmc.c | 3 + drivers/mtd/nand/fsl_elbc_nand.c | 14 +++ drivers/mtd/nand/omap2.c | 2 +- drivers/mtd/nand/pxa3xx_nand.c | 4 +- drivers/net/ethernet/allwinner/sun4i-emac.c | 1 + drivers/net/ethernet/mellanox/mlx4/fw.c | 2 +- drivers/net/ethernet/mellanox/mlx4/main.c | 2 +- drivers/net/wireless/b43/xmit.c | 10 +- drivers/net/wireless/iwlwifi/pcie/trans.c | 26 ++++-- drivers/net/wireless/rt2x00/rt2500pci.c | 7 +- drivers/net/wireless/rt2x00/rt2x00.h | 1 + drivers/net/wireless/rt2x00/rt2x00dev.c | 24 ++++- drivers/net/wireless/rt2x00/rt2x00mac.c | 2 + drivers/regulator/tps65218-regulator.c | 3 +- drivers/scsi/ibmvscsi/ibmvscsi.c | 13 ++- drivers/scsi/scsi_error.c | 8 +- drivers/scsi/sd.c | 5 +- drivers/scsi/virtio_scsi.c | 26 +++++- drivers/staging/iio/adc/ad7291.c | 4 +- drivers/staging/tidspbridge/core/tiomap3430.c | 6 +- drivers/tty/n_tty.c | 19 ++-- drivers/tty/serial/8250/8250_core.c | 2 +- drivers/tty/serial/altera_uart.c | 6 ++ drivers/tty/serial/amba-pl010.c | 2 +- drivers/tty/serial/amba-pl011.c | 2 +- drivers/tty/serial/atmel_serial.c | 2 +- drivers/tty/serial/bcm63xx_uart.c | 2 +- drivers/tty/serial/bfin_uart.c | 2 +- drivers/tty/serial/dz.c | 2 +- drivers/tty/serial/efm32-uart.c | 2 +- drivers/tty/serial/fsl_lpuart.c | 2 +- drivers/tty/serial/ip22zilog.c | 2 +- drivers/tty/serial/m32r_sio.c | 2 +- drivers/tty/serial/max310x.c | 2 +- drivers/tty/serial/mcf.c | 6 ++ drivers/tty/serial/mfd.c | 2 +- drivers/tty/serial/mpsc.c | 2 +- drivers/tty/serial/msm_serial.c | 2 +- drivers/tty/serial/mxs-auart.c | 2 +- drivers/tty/serial/netx-serial.c | 2 +- drivers/tty/serial/pmac_zilog.c | 2 +- drivers/tty/serial/pnx8xxx_uart.c | 2 +- drivers/tty/serial/pxa.c | 2 +- drivers/tty/serial/sb1250-duart.c | 2 +- drivers/tty/serial/sccnxp.c | 2 +- drivers/tty/serial/serial_ks8695.c | 2 +- drivers/tty/serial/serial_txx9.c | 2 +- drivers/tty/serial/sirfsoc_uart.c | 2 +- drivers/tty/serial/st-asc.c | 2 +- drivers/tty/serial/sunsab.c | 2 +- drivers/tty/serial/sunsu.c | 2 +- drivers/tty/serial/sunzilog.c | 2 +- drivers/tty/serial/ucc_uart.c | 2 +- drivers/tty/serial/vr41xx_siu.c | 2 +- drivers/tty/serial/zs.c | 2 +- drivers/usb/chipidea/udc.c | 7 ++ drivers/usb/gadget/f_fs.c | 12 ++- drivers/usb/host/xhci-ring.c | 9 +- drivers/usb/host/xhci.c | 10 +- drivers/usb/musb/musb_am335x.c | 23 ++--- drivers/usb/musb/musb_cppi41.c | 2 +- drivers/usb/musb/ux500.c | 1 - drivers/usb/serial/ftdi_sio.c | 7 +- drivers/usb/serial/option.c | 26 ++++-- drivers/usb/storage/scsiglue.c | 4 + drivers/usb/storage/unusual_devs.h | 7 ++ drivers/video/fbdev/fb-puv3.c | 2 + fs/btrfs/disk-io.c | 3 +- fs/cifs/cifs_unicode.c | 7 +- fs/cifs/link.c | 2 +- fs/ext4/indirect.c | 20 +++- fs/nfsd/nfs4proc.c | 9 -- fs/nfsd/nfs4xdr.c | 13 ++- include/drm/i915_powerwell.h | 5 +- include/linux/usb_usual.h | 4 +- include/scsi/scsi_cmnd.h | 2 +- include/scsi/scsi_device.h | 1 + include/uapi/linux/usb/functionfs.h | 7 ++ kernel/sysctl.c | 3 +- kernel/trace/trace.c | 2 - kernel/trace/trace_uprobe.c | 40 +++++--- mm/hugetlb.c | 71 ++++++++------ mm/memory.c | 3 - mm/nommu.c | 2 +- mm/page_alloc.c | 56 ++++++++--- net/bluetooth/hci_conn.c | 7 +- net/bluetooth/hci_event.c | 11 ++- net/bluetooth/l2cap_core.c | 8 +- net/bluetooth/l2cap_sock.c | 5 - net/bluetooth/mgmt.c | 104 +++++++++++---------- net/bluetooth/smp.c | 9 +- net/mac80211/debugfs_netdev.c | 6 +- net/mac80211/ibss.c | 1 + net/mac80211/sta_info.c | 1 + net/netfilter/nf_conntrack_netlink.c | 14 +-- net/netfilter/nf_nat_core.c | 35 ++++++- sound/pci/hda/hda_i915.c | 67 +++++++++++-- sound/pci/hda/hda_i915.h | 6 +- sound/pci/hda/hda_intel.c | 57 +++-------- tools/usb/ffs-test.c | 4 +- 158 files changed, 894 insertions(+), 445 deletions(-)
Greg Kroah-Hartman
maintainerPATCH 1/122 SCSI: use the scsi data buffer length to extract transfer size
3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Martin K. Petersen" <[email protected]>
commit 5616b0a46ed82eb9a093f752fc4d7bd3cc688583 upstream.
Commit 8846bab180fa introduced a helper that can be used to query the wire transfer size for a SCSI command taking protection information into account.
However, some commands do not have a 1:1 mapping between the block range they work on and the payload size (discard, write same). After the scatterlist has been set up these requests use __data_len to store the number of bytes to report completion on. This means that callers of scsi_transfer_length() would get the wrong byte count for these types of requests.
To overcome this we make scsi_transfer_length() use the scatterlist length in the scsi_data_buffer as basis for the wire transfer calculation instead of __data_len.
--- include/scsi/scsi_cmnd.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/scsi/scsi_cmnd.h +1 -1
@@ -309,7 +309,7 @@ static inline void set_driver_byte(strucstatic inline unsigned scsi_transfer_length (struct scsi_cmnd *scmd) { -unsigned int xfer_len =blk_rq_bytes (scmd->request); +unsigned int xfer_len =scsi_out (scmd)->length;unsigned int prot_op =scsi_get_prot_op (scmd);unsigned int sector_size = scmd->device->sector_size;
Greg Kroah-Hartman
maintainerPATCH 3/122 ibmvscsi: Abort init sequence during error recovery
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Brian King <[email protected]>
commit 9ee755974bea2f9880e517ec985dc9dede1b3a36 upstream.
If a CRQ reset is triggered for some reason while in the middle of performing VSCSI adapter initialization, we don't want to call the done function for the initialization MAD commands as this will only result in two threads attempting initialization at the same time, resulting in failures.
--- drivers/scsi/ibmvscsi/ibmvscsi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/ibmvscsi/ibmvscsi.c +2 -1
@@ -797,7 +797,8 @@ static void purge_requests(struct ibmvsc evt->hostdata->dev);if (evt->cmnd_done) evt->cmnd_done (evt->cmnd); - }else if (evt->done) + }else if (evt->done && evt->crq.format !=VIOSRP_MAD_FORMAT && + evt->iu.srp.login_req.opcode !=SRP_LOGIN_REQ ) evt->done (evt);free_event_struct (&evt->hostdata->pool, evt);spin_lock_irqsave (hostdata->host->host_lock, flags);
Greg Kroah-Hartman
maintainerPATCH 4/122 ibmvscsi: Add memory barriers for send / receive
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Brian King <[email protected]>
commit 7114aae02742d6b5c5a0d39a41deb61d415d3717 upstream.
Add a memory barrier prior to sending a new command to the VIOS to ensure the VIOS does not receive stale data in the command buffer. Also add a memory barrier when processing the CRQ for completed commands.
--- drivers/scsi/ibmvscsi/ibmvscsi.c | 10 ++++++++++ 1 file changed, 10 insertions(+)
drivers/scsi/ibmvscsi/ibmvscsi.c +10 -0
@@ -185,6 +185,11 @@ static struct viosrp_crq *crq_queue_nextif (crq->valid &0x80 ) {if (++queue->cur == queue->size) queue->cur =0 ; + +/* Ensure the read of the valid bit occurs before reading any +* other bits of the CRQ entry +*/ +rmb (); }else crq =NULL ;spin_unlock_irqrestore (&queue->lock, flags);
@@ -203,6 +208,11 @@ static int ibmvscsi_send_crq(struct ibmv {struct vio_dev *vdev =to_vio_dev (hostdata->dev); +/* +* Ensure the command buffer is flushed to memory before handing it +* over to the VIOS to prevent it from fetching any stale data. +*/ +mb ();return plpar_hcall_norets (H_SEND_CRQ , vdev->unit_address, word1, word2); }
Greg Kroah-Hartman
maintainerPATCH 5/122 virtio-scsi: avoid cancelling uninitialized work items
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Paolo Bonzini <[email protected]>
commit cdda0e5acbb78f7b777049f8c27899e5c5bb368f upstream.
Calling the workqueue interface on uninitialized work items isn't a good idea even if they're zeroed. It's not failing catastrophically only through happy accidents.
--- drivers/scsi/virtio_scsi.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/scsi/virtio_scsi.c +3 -1
@@ -291,6 +291,8 @@ static void virtscsi_ctrl_done(struct vivirtscsi_vq_done (vscsi, &vscsi->ctrl_vq, virtscsi_complete_free); }; +static void virtscsi_handle_event (struct work_struct *work); +static int virtscsi_kick_event (struct virtio_scsi *vscsi,struct virtio_scsi_event_node *event_node) {
@@ -298,6 +300,7 @@ static int virtscsi_kick_event(struct vistruct scatterlist sg;unsigned long flags; +INIT_WORK (&event_node->work, virtscsi_handle_event);sg_init_one (&sg, &event_node->event,sizeof (struct virtio_scsi_event ));spin_lock_irqsave (&vscsi->event_vq.vq_lock, flags);
@@ -415,7 +418,6 @@ static void virtscsi_complete_event(stru {struct virtio_scsi_event_node *event_node = buf; -INIT_WORK (&event_node->work, virtscsi_handle_event);schedule_work (&event_node->work); }
Greg Kroah-Hartman
maintainerPATCH 6/122 scsi_error: fix invalid setting of host byte
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ulrich Obergfell <[email protected]>
commit 8922a908908ff947f1f211e07e2e97f1169ad3cb upstream.
After scsi_try_to_abort_cmd returns, the eh_abort_handler may have already found that the command has completed in the device, causing the host_byte to be nonzero (e.g. it could be DID_ABORT). When this happens, ORing DID_TIME_OUT into the host byte will corrupt the result field and initiate an unwanted command retry.
Fix this by using set_host_byte instead, following the model of commit 2082ebc45af9c9c648383b8cde0dc1948eadbf31.
--- drivers/scsi/scsi_error.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/scsi/scsi_error.c +4 -4
@@ -131,7 +131,7 @@ scmd_eh_abort_handler(struct work_struct "aborting command %p\n", scmd)); rtn =scsi_try_to_abort_cmd (sdev->host->hostt, scmd);if (rtn ==SUCCESS ) { - scmd->result |=DID_TIME_OUT <<16 ; +set_host_byte (scmd,DID_TIME_OUT );if (scsi_host_eh_past_deadline (sdev->host)) { SCSI_LOG_ERROR_RECOVERY(3, scmd_printk(KERN_INFO, scmd,
@@ -167,7 +167,7 @@ scmd_eh_abort_handler(struct work_structscmd_printk (KERN_WARNING , scmd,"scmd %p terminate " "aborted command\n", scmd)); - scmd->result |=DID_TIME_OUT <<16 ; +set_host_byte (scmd,DID_TIME_OUT );scsi_finish_command (scmd); } }
@@ -291,7 +291,7 @@ enum blk_eh_timer_return scsi_times_out(if (scsi_abort_command (scmd) ==SUCCESS )return BLK_EH_NOT_HANDLED ; - scmd->result |=DID_TIME_OUT <<16 ; +set_host_byte (scmd,DID_TIME_OUT );if (unlikely (rtn ==BLK_EH_NOT_HANDLED && !scsi_eh_scmd_add (scmd,SCSI_EH_CANCEL_CMD )))
@@ -1776,7 +1776,7 @@ int scsi_decide_disposition(struct scsi_break ;case DID_ABORT :if (scmd->eh_eflags &SCSI_EH_ABORT_SCHEDULED ) { - scmd->result |=DID_TIME_OUT <<16 ; +set_host_byte (scmd,DID_TIME_OUT );return SUCCESS ; }case DID_NO_CONNECT :
Greg Kroah-Hartman
maintainerPATCH 7/122 virtio-scsi: fix various bad behavior on aborted requests
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Paolo Bonzini <[email protected]>
commit 8faeb529b2dabb9df691d614dda18910a43d05c9 upstream.
Even though the virtio-scsi spec guarantees that all requests related to the TMF will have been completed by the time the TMF itself completes, the request queue's callback might not have run yet. This causes requests to be completed more than once, and as a result triggers a variety of BUGs or oopses.
--- drivers/scsi/virtio_scsi.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)
drivers/scsi/virtio_scsi.c +22 -0
@@ -273,6 +273,16 @@ static void virtscsi_req_done(struct virvirtscsi_vq_done (vscsi, req_vq, virtscsi_complete_cmd); }; +static void virtscsi_poll_requests (struct virtio_scsi *vscsi) +{ +int i, num_vqs; + + num_vqs = vscsi->num_queues; +for (i =0 ; i < num_vqs; i++) +virtscsi_vq_done (vscsi, &vscsi->req_vqs[i], + virtscsi_complete_cmd); +} +static void virtscsi_complete_free (struct virtio_scsi *vscsi,void *buf) {struct virtio_scsi_cmd *cmd = buf;
@@ -607,6 +617,18 @@ static int virtscsi_tmf(struct virtio_sc cmd->resp.tmf.response == VIRTIO_SCSI_S_FUNCTION_SUCCEEDED) ret =SUCCESS ; +/* +* The spec guarantees that all requests related to the TMF have +* been completed, but the callback might not have run yet if +* we're using independent interrupts (e.g. MSI). Poll the +* virtqueues once. +* +* In the abort case, sc->scsi_done will do nothing, because +* the block layer must have detected a timeout and as a result +* REQ_ATOM_COMPLETE has been set. +*/ +virtscsi_poll_requests (vscsi); + out:mempool_free (cmd, virtscsi_cmd_pool);return ret;
Greg Kroah-Hartman
maintainerPATCH 8/122 xhci: Use correct SLOT ID when handling a reset device command
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mathias Nyman <[email protected]>
commit 6fcfb0d682a8212d321a6131adc94daf0905992a upstream.
Command completion events normally include command completion status, SLOT_ID, and a pointer to the original command. Reset device command completion SLOT_ID may be zero according to xhci specs 4.6.11.
VIA controllers set the SLOT_ID to zero, triggering a WARN_ON in the command completion handler.
Use the SLOT ID found from the original command instead.
This patch should be applied to stable kernels since 3.13 that contain the commit 20e7acb13ff48fbc884d5918c3697c27de63922a
"xhci: use completion event's slot id rather than dig it out of command"
--- drivers/usb/host/xhci-ring.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/usb/host/xhci-ring.c +5 -2
@@ -1611,8 +1611,11 @@ static void handle_cmd_completion(structxhci_handle_cmd_reset_ep (xhci, slot_id, cmd_trb, cmd_comp_code);break ;case TRB_RESET_DEV : -WARN_ON (slot_id !=TRB_TO_SLOT_ID ( -le32_to_cpu (cmd_trb->generic.field[3 ]))); +/* SLOT_ID field in reset device cmd completion event TRB is 0. +* Use the SLOT_ID from the command TRB instead (xhci 4.6.11) +*/ + slot_id =TRB_TO_SLOT_ID ( +le32_to_cpu (cmd_trb->generic.field[3 ]));xhci_handle_cmd_reset_dev (xhci, slot_id, event);break ;case TRB_NEC_GET_FW :
Greg Kroah-Hartman
maintainerPATCH 9/122 xhci: correct burst count field for isoc transfers on 1.0 xhci hosts
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mathias Nyman <[email protected]>
commit 3213b151387df0b95f4eada104f68eb1c1409cb3 upstream.
The transfer burst count (TBC) field in xhci 1.0 hosts should be set to the number of bursts needed to transfer all packets in a isoc TD. Supported values are 0-2 (1 to 3 bursts per service interval).
Formula for TBC calculation is given in xhci spec section 4.11.2.3:
TBC = roundup( Transfer Descriptor Packet Count / Max Burst Size +1 ) - 1
This patch should be applied to stable kernels since 3.0 that contain the commit 5cd43e33b9519143f06f507dd7cbee6b7a621885
"xhci 1.0: Set transfer burst count field."
--- drivers/usb/host/xhci-ring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/host/xhci-ring.c +1 -1
@@ -3699,7 +3699,7 @@ static unsigned int xhci_get_burst_countreturn 0 ; max_burst = urb->ep->ss_ep_comp.bMaxBurst; -return roundup (total_packet_count, max_burst +1 ) -1 ; +return DIV_ROUND_UP (total_packet_count, max_burst +1 ) -1 ; }/*
Greg Kroah-Hartman
maintainerPATCH 10/122 xhci: Fix runtime suspended xhci from blocking system suspend.
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Wang, Yu" <[email protected]>
commit d6236f6d1d885aa19d1cd7317346fe795227a3cc upstream.
The system suspend flow as following:
1, Freeze all user processes and kenrel threads.
2, Try to suspend all devices.
2.1, If pci device is in RPM suspended state, then pci driver will try to resume it to RPM active state in the prepare stage.
2.2, xhci_resume function calls usb_hcd_resume_root_hub to queue two workqueue items to resume usb2&usb3 roothub devices.
2.3, Call suspend callbacks of devices.
2.3.1, All suspend callbacks of all hcd's children, including roothub devices are called.
2.3.2, Finally, hcd_pci_suspend callback is called.
Due to workqueue threads were already frozen in step 1, the workqueue items can't be scheduled, and the roothub devices can't be resumed in this flow. The HCD_FLAG_WAKEUP_PENDING flag which is set in usb_hcd_resume_root_hub won't be cleared. Finally, hcd_pci_suspend will return -EBUSY, and system suspend fails.
The reason why this issue doesn't show up very often is due to that choose_wakeup will be called in step 2.3.1. In step 2.3.1, if udev->do_remote_wakeup is not equal to device_may_wakeup(&udev->dev), then udev will resume to RPM active for changing the wakeup settings. This has been a lucky hit which hides this issue.
For some special xHCI controllers which have no USB2 port, then roothub will not match hub driver due to probe failed. Then its do_remote_wakeup will be set to zero, and we won't be as lucky.
xhci driver doesn't need to resume roothub devices everytime like in the above case. It's only needed when there are pending event TRBs.
This patch should be back-ported to kernels as old as 3.2, that contains the commit f69e3120df82391a0ee8118e0a156239a06b2afb "USB: XHCI: resume root hubs when the controller resumes"
--- drivers/usb/host/xhci.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/usb/host/xhci.c +7 -3
@@ -932,7 +932,7 @@ int xhci_suspend(struct xhci_hcd *xhci)*/ int xhci_resume (struct xhci_hcd *xhci,bool hibernated) { -u32 command, temp =0 ; +u32 command, temp =0 , status;struct usb_hcd *hcd =xhci_to_hcd (xhci);struct usb_hcd *secondary_hcd;int retval =0 ;
@@ -1050,8 +1050,12 @@ int xhci_resume(struct xhci_hcd *xhci, b done:if (retval ==0 ) { -usb_hcd_resume_root_hub (hcd); -usb_hcd_resume_root_hub (xhci->shared_hcd); +/* Resume root hubs only when have pending events. */ + status =readl (&xhci->op_regs->status); +if (status &STS_EINT ) { +usb_hcd_resume_root_hub (hcd); +usb_hcd_resume_root_hub (xhci->shared_hcd); + } }/*
Greg Kroah-Hartman
maintainerPATCH 11/122 USB: option: add device ID for SpeedUp SU9800 usb 3g modem
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Oliver Neukum <[email protected]>
commit 1cab4c68e339086cdaff7535848e878e8f261fca upstream.
Reported by Alif Mubarak Ahmad:
This device vendor and product id is 1c9e:9800
It is working as serial interface with generic usbserial driver. I thought it is more suitable to use usbserial option driver, which has better capability distinguishing between modem serial interface and micro sd storage interface.
[ johan: style changes ]
--- drivers/usb/serial/option.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/usb/serial/option.c +4 -0
@@ -352,6 +352,9 @@ static void option_instat_callback(struc/* Zoom */ #define ZOOM_PRODUCT_4597 0x9607 +/* SpeedUp SU9800 usb 3g modem */ +#define SPEEDUP_PRODUCT_SU9800 0x9800 +/* Haier products */ #define HAIER_VENDOR_ID 0x201e#define HAIER_PRODUCT_CE100 0x2009
@@ -1577,6 +1580,7 @@ static const struct usb_device_id option { USB_DEVICE(LONGCHEER_VENDOR_ID, FOUR_G_SYSTEMS_PRODUCT_W14), .driver_info = (kernel_ulong_t)&four_g_w14_blacklist }, + { USB_DEVICE_INTERFACE_CLASS(LONGCHEER_VENDOR_ID, SPEEDUP_PRODUCT_SU9800, 0xff) }, { USB_DEVICE(LONGCHEER_VENDOR_ID, ZOOM_PRODUCT_4597) }, { USB_DEVICE(LONGCHEER_VENDOR_ID, IBALL_3_5G_CONNECT) }, { USB_DEVICE(HAIER_VENDOR_ID, HAIER_PRODUCT_CE100) },
Greg Kroah-Hartman
maintainerPATCH 13/122 usb: musb: ux500: dont propagate the OF node
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Linus Walleij <[email protected]>
commit 82363cf2eeafeea6ba88849f5e2febdc8a05943f upstream.
There is a regression in the upcoming v3.16-rc1, that is caused by a problem that has been around for a while but now finally hangs the system. The bootcrawl looks like this:
pinctrl-nomadik soc:pinctrl: pin GPIO256_AF28 already requested by a03e0000.usb_per5; cannot claim for musb-hdrc.0.auto pinctrl-nomadik soc:pinctrl: pin-256 (musb-hdrc.0.auto) status -22 pinctrl-nomadik soc:pinctrl: could not request pin 256 (GPIO256_AF28) from group usb_a_1 on device pinctrl-nomadik musb-hdrc musb-hdrc.0.auto: Error applying setting, reverse things back
HS USB OTG: no transceiver configured
musb-hdrc musb-hdrc.0.auto: musb_init_controller failed with status -517
platform musb-hdrc.0.auto: Driver musb-hdrc requests probe deferral
(...)
The ux500 MUSB driver propagates the OF node to the dynamically created musb-hdrc device, which is incorrect as it makes the OF core believe there are two devices spun from the very same DT node, which confuses other parts of the device core, notably the pin control subsystem, which will try to apply all the pin control settings also to the HDRC device as it gets instantiated. (The OMAP2430 for example, does not set the of_node member.)
--- drivers/usb/musb/ux500.c | 1 - 1 file changed, 1 deletion(-)
drivers/usb/musb/ux500.c +0 -1
@@ -274,7 +274,6 @@ static int ux500_probe(struct platform_d musb->dev.parent = &pdev->dev; musb->dev.dma_mask = &pdev->dev.coherent_dma_mask; musb->dev.coherent_dma_mask = pdev->dev.coherent_dma_mask; - musb->dev.of_node = pdev->dev.of_node; glue->dev = &pdev->dev; glue->musb = musb;
Greg Kroah-Hartman
maintainerPATCH 14/122 usb: musb: Ensure that cppi41 timer gets armed on premature DMA TX irq
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Gleixner <[email protected]>
commit c58d80f523ffc15ef4d062fc7aeb03793fe39701 upstream.
Some TI chips raise the DMA complete interrupt before the actual transfer has been completed. The code tries to busy wait for a few microseconds and if that fails it arms an hrtimer to recheck. So far so good, but that has the following issue:
CPU 0 CPU1
start_next_transfer(RQ1);
DMA interrupt
if (premature_irq(RQ1))
if (!hrtimer_active(timer))
hrtimer_start(timer);hrtimer expires
timer->state = CALLBACK_RUNNING;
timer->fn()
cppi41_recheck_tx_req()
complete_request(RQ1);
if (requests_pending())
start_next_transfer(RQ2);
DMA interrupt
if (premature_irq(RQ2))
if (!hrtimer_active(timer))
hrtimer_start(timer); timer->state = INACTIVE;
The premature interrupt of request2 on CPU1 does not arm the timer and therefor the request completion never happens because it checks for !hrtimer_active(). hrtimer_active() evaluates:
timer->state != HRTIMER_STATE_INACTIVE
which of course evaluates to true in the above case as timer->state is CALLBACK_RUNNING.
That's clearly documented:
* A timer is active, when it is enqueued into the rbtree or the
* callback function is running or it's in the state of being migrated
* to another cpu.
But that's not what the code wants to check. The code wants to check whether the timer is queued, i.e. whether its armed and waiting for expiry.
We have a helper function for this: hrtimer_is_queued(). This evaluates:
timer->state & HRTIMER_STATE_QUEUED
So in the above case this evaluates to false and therefor forces the DMA interrupt on CPU1 to call hrtimer_start().
Use hrtimer_is_queued() instead of hrtimer_active() and evrything is good.
--- drivers/usb/musb/musb_cppi41.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/musb/musb_cppi41.c +1 -1
@@ -318,7 +318,7 @@ static void cppi41_dma_callback(void *pr }list_add_tail (&cppi41_channel->tx_check, &controller->early_tx_list); -if (!hrtimer_active (&controller->early_tx)) { +if (!hrtimer_is_queued (&controller->early_tx)) { hrtimer_start_range_ns(&controller->early_tx, ktime_set(0, 140 * NSEC_PER_USEC), 40 * NSEC_PER_USEC,
Greg Kroah-Hartman
maintainerPATCH 15/122 usb: musb: Fix panic upon musb_am335x module removal
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ezequiel Garcia <[email protected]>
commit 7adb5c876e9c0677078a1e1094c6eafd29c30b74 upstream.
At probe time, the musb_am335x driver register its childs by calling of_platform_populate(), which registers all childs in the devicetree hierarchy recursively.
On the other side, the driver's remove() function uses of_device_unregister() to remove each child of musb_am335x's.
However, when musb_dsps is loaded, its devices are attached to the musb_am335x device as musb_am335x childs. Hence, musb_am335x remove() will attempt to unregister the devices registered by musb_dsps, which produces a kernel panic.
In other words, the childs in the "struct device" hierarchy are not the same as the childs in the "devicetree" hierarchy.
Ideally, we should enforce the removal of the devices registered by musb_am335x *only*, instead of all its child devices. However, because of the recursive nature of of_platform_populate, this doesn't seem possible.
Therefore, as the only solution at hand, this commit disables musb_am335x driver removal capability, preventing it from being ever removed. This was originally suggested by Sebastian Siewior:
https://www.mail-archive.com/[email protected]/msg104946.html
And for reference, here's the panic upon module removal:
musb-hdrc musb-hdrc.0.auto: remove, state 4 usb usb1: USB disconnect, device number 1 musb-hdrc musb-hdrc.0.auto: USB bus 1 deregistered
Unable to handle kernel NULL pointer dereference at virtual address 0000008c pgd = de11c000 [0000008c] *pgd=9e174831, *pte=00000000, *ppte=00000000 Internal error: Oops: 17 [#1] ARM Modules linked in: musb_am335x(-) musb_dsps musb_hdrc usbcore usb_common CPU: 0 PID: 623 Comm: modprobe Not tainted 3.15.0-rc4-00001-g24efd13 #69 task: de1b7500 ti: de122000 task.ti: de122000 PC is at am335x_shutdown+0x10/0x28 LR is at am335x_shutdown+0xc/0x28 pc : [<c0327798>] lr : [<c0327794>] psr: a0000013
sp : de123df8 ip : 00000004 fp : 00028f00
r10: 00000000 r9 : de122000 r8 : c000e6c4
r7 : de0e3c10 r6 : de0e3800 r5 : de624010 r4 : de1ec750 r3 : de0e3810 r2 : 00000000 r1 : 00000001 r0 : 00000000
Flags: NzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user
Control: 10c5387d Table: 9e11c019 DAC: 00000015 Process modprobe (pid: 623, stack limit = 0xde122240)
Stack: (0xde123df8 to 0xde124000)
3de0: de0e3810 bf054488
3e00: bf05444c de624010 60000013 bf043650 000012fc de624010 de0e3810 bf043a20
3e20: de0e3810 bf04b240 c0635b88 c02ca37c c02ca364 c02c8db0 de1b7500 de0e3844
3e40: de0e3810 c02c8e28 c0635b88 de02824c de0e3810 c02c884c de0e3800 de0e3810
3e60: de0e3818 c02c5b20 bf05417c de0e3800 de0e3800 c0635b88 de0f2410 c02ca838
3e80: bf05417c de0e3800 bf055438 c02ca8cc de0e3c10 bf054194 de0e3c10 c02ca37c
3ea0: c02ca364 c02c8db0 de1b7500 de0e3c44 de0e3c10 c02c8e28 c0635b88 de02824c
3ec0: de0e3c10 c02c884c de0e3c10 de0e3c10 de0e3c18 c02c5b20 de0e3c10 de0e3c10
3ee0: 00000000 bf059000 a0000013 c02c5bc0 00000000 bf05900c de0e3c10 c02c5c48
3f00: de0dd0c0 de1ec970 de0f2410 bf05929c de0f2444 bf05902c de0f2410 c02ca37c
3f20: c02ca364 c02c8db0 bf05929c de0f2410 bf05929c c02c94c8 bf05929c 00000000
3f40: 00000800 c02c8ab4 bf0592e0 c007fc40 c00dd820 6273756d 336d615f 00783533
3f60: c064a0ac de1b7500 de122000 de1b7500 c000e590 00000001 c000e6c4 c0060160
3f80: 00028e70 00028e70 00028ea4 00000081 60000010 00028e70 00028e70 00028ea4
3fa0: 00000081 c000e500 00028e70 00028e70 00028ea4 00000800 becb59f8 00027608
3fc0: 00028e70 00028e70 00028ea4 00000081 00000001 00000001 00000000 00028f00
3fe0: b6e6b6f0 becb59d4 000160e8 b6e6b6fc 60000010 00028ea4 00000000 00000000 [<c0327798>] (am335x_shutdown) from [<bf054488>] (dsps_musb_exit+0x3c/0x4c [musb_dsps]) [<bf054488>] (dsps_musb_exit [musb_dsps]) from [<bf043650>] (musb_shutdown+0x80/0x90 [musb_hdrc]) [<bf043650>] (musb_shutdown [musb_hdrc]) from [<bf043a20>] (musb_remove+0x24/0x68 [musb_hdrc]) [<bf043a20>] (musb_remove [musb_hdrc]) from [<c02ca37c>] (platform_drv_remove+0x18/0x1c) [<c02ca37c>] (platform_drv_remove) from [<c02c8db0>] (__device_release_driver+0x70/0xc8) [<c02c8db0>] (__device_release_driver) from [<c02c8e28>] (device_release_driver+0x20/0x2c) [<c02c8e28>] (device_release_driver) from [<c02c884c>] (bus_remove_device+0xdc/0x10c) [<c02c884c>] (bus_remove_device) from [<c02c5b20>] (device_del+0x104/0x198)
[<c02c5b20>] (device_del) from [<c02ca838>] (platform_device_del+0x14/0x9c)
[<c02ca838>] (platform_device_del) from [<c02ca8cc>] (platform_device_unregister+0xc/0x20) [<c02ca8cc>] (platform_device_unregister) from [<bf054194>] (dsps_remove+0x18/0x38 [musb_dsps]) [<bf054194>] (dsps_remove [musb_dsps]) from [<c02ca37c>] (platform_drv_remove+0x18/0x1c) [<c02ca37c>] (platform_drv_remove) from [<c02c8db0>] (__device_release_driver+0x70/0xc8) [<c02c8db0>] (__device_release_driver) from [<c02c8e28>] (device_release_driver+0x20/0x2c) [<c02c8e28>] (device_release_driver) from [<c02c884c>] (bus_remove_device+0xdc/0x10c) [<c02c884c>] (bus_remove_device) from [<c02c5b20>] (device_del+0x104/0x198)
[<c02c5b20>] (device_del) from [<c02c5bc0>] (device_unregister+0xc/0x20)
[<c02c5bc0>] (device_unregister) from [<bf05900c>] (of_remove_populated_child+0xc/0x14 [musb_am335x]) [<bf05900c>] (of_remove_populated_child [musb_am335x]) from [<c02c5c48>] (device_for_each_child+0x44/0x70) [<c02c5c48>] (device_for_each_child) from [<bf05902c>] (am335x_child_remove+0x18/0x30 [musb_am335x]) [<bf05902c>] (am335x_child_remove [musb_am335x]) from [<c02ca37c>] (platform_drv_remove+0x18/0x1c) [<c02ca37c>] (platform_drv_remove) from [<c02c8db0>] (__device_release_driver+0x70/0xc8) [<c02c8db0>] (__device_release_driver) from [<c02c94c8>] (driver_detach+0xb4/0xb8) [<c02c94c8>] (driver_detach) from [<c02c8ab4>] (bus_remove_driver+0x4c/0xa0)
[<c02c8ab4>] (bus_remove_driver) from [<c007fc40>] (SyS_delete_module+0x128/0x1cc) [<c007fc40>] (SyS_delete_module) from [<c000e500>] (ret_fast_syscall+0x0/0x48)
--- drivers/usb/musb/musb_am335x.c | 23 ++++++----------------- 1 file changed, 6 insertions(+), 17 deletions(-)
drivers/usb/musb/musb_am335x.c +6 -17
@@ -19,21 +19,6 @@ err:return ret; } -static int of_remove_populated_child (struct device *dev,void *d) -{ -struct platform_device *pdev =to_platform_device (dev); - -of_device_unregister (pdev); -return 0 ; -} - -static int am335x_child_remove (struct platform_device *pdev) -{ -device_for_each_child (&pdev->dev,NULL , of_remove_populated_child); -pm_runtime_disable (&pdev->dev); -return 0 ; -} -static const struct of_device_id am335x_child_of_match[] = { { .compatible ="ti,am33xx-usb" }, { },
@@ -42,13 +27,17 @@ MODULE_DEVICE_TABLE(of, am335x_child_of_static struct platform_driver am335x_child_driver = { .probe = am335x_child_probe, - .remove = am335x_child_remove, .driver = { .name ="am335x-usb-childs" , .of_match_table = am335x_child_of_match, }, }; -module_platform_driver (am335x_child_driver); +static int __init am335x_child_init (void ) +{ +return platform_driver_register (&am335x_child_driver); +} +module_init (am335x_child_init); +MODULE_DESCRIPTION ("AM33xx child devices" );MODULE_LICENSE ("GPL v2" );
Greg Kroah-Hartman
maintainerPATCH 16/122 usb: chipidea: udc: delete td from reqs td list at ep_dequeue
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Peter Chen <[email protected]>
commit e4adcff09ca39ecbcc4851d40d0f0a5458e7b77a upstream.
We need to delete un-finished td from current request's td list at ep_dequeue API, otherwise, this non-user td will be remained at td list before this request is freed. So if we do ep_queue-> ep_dequeue->ep_queue sequence, when the complete interrupt for the second ep_queue comes, we search td list for this request, the first td (added by the first ep_queue) will be handled, and its status is still active, so we will consider the this transfer still not be completed, but in fact, it has completed. It causes the peripheral side considers it never receives current data for this transfer.
We met this problem when do "Error Recovery Test - Device Configured" test item for USBCV2 MSC test, the host has never received ACK for the IN token for CSW due to peripheral considers it does not get this CBW, the USBCV test log like belows:
--------------------------------------------------------------------------
INFO
Issuing BOT MSC Reset, reset should always succeed
INFO
Retrieving status on CBW endpoint
INFO
CBW endpoint status = 0x0
INFO
Retrieving status on CSW endpoint
INFO
CSW endpoint status = 0x0
INFO
Issuing required command (Test Unit Ready) to verify device has recovered INFO
Issuing CBW (attempt #1):
INFO
|----- CBW LUN = 0x0
INFO
|----- CBW Flags = 0x0
INFO
|----- CBW Data Transfer Length = 0x0
INFO
|----- CBW CDB Length = 0x6
INFO
|----- CBW CDB-00 = 0x0
INFO
|----- CBW CDB-01 = 0x0
INFO
|----- CBW CDB-02 = 0x0
INFO
|----- CBW CDB-03 = 0x0
INFO
|----- CBW CDB-04 = 0x0
INFO
|----- CBW CDB-05 = 0x0
INFO
Issuing CSW : try 1
INFO
CSW Bulk Request timed out!
ERROR
Failed CSW phase : should have been success or stall
FAIL
(5.3.4) The CSW status value must be 0x00, 0x01, or 0x02. ERROR
BOTCommonMSCRequest failed: error=80004000
--- drivers/usb/chipidea/udc.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/usb/chipidea/udc.c +7 -0
@@ -1328,6 +1328,7 @@ static int ep_dequeue(struct usb_ep *ep, struct ci_hw_ep *hwep = container_of(ep, struct ci_hw_ep, ep); struct ci_hw_req *hwreq = container_of(req, struct ci_hw_req, req);unsigned long flags; +struct td_node *node, *tmpnode; if (ep == NULL || req == NULL || hwreq->req.status != -EALREADY || hwep->ep.desc == NULL || list_empty(&hwreq->queue) ||
@@ -1338,6 +1339,12 @@ static int ep_dequeue(struct usb_ep *ep,hw_ep_flush (hwep->ci, hwep->num, hwep->dir); +list_for_each_entry_safe (node, tmpnode, &hwreq->tds, td) { +dma_pool_free (hwep->td_pool, node->ptr, node->dma); +list_del (&node->td); +kfree (node); + } +/* pop request */ list_del_init (&hwreq->queue);
Greg Kroah-Hartman
maintainerPATCH 17/122 USB: ftdi_sio: fix null deref at port probe
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hovold <[email protected]>
commit aea1ae8760314e072bf1b773521e9de5d5dda10d upstream.
Fix NULL-pointer dereference when probing an interface with no endpoints.
These devices have two bulk endpoints per interface, but this avoids crashing the kernel if a user forces a non-FTDI device to be probed.
Note that the iterator variable was made unsigned in order to avoid a maybe-uninitialized compiler warning for ep_desc after the loop.
calculation")
--- drivers/usb/serial/ftdi_sio.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/usb/serial/ftdi_sio.c +5 -2
@@ -1566,14 +1566,17 @@ static void ftdi_set_max_packet_size(strstruct usb_device *udev = serial->dev;struct usb_interface *interface = serial->interface; -struct usb_endpoint_descriptor *ep_desc = &interface->cur_altsetting->endpoint[1 ].desc; +struct usb_endpoint_descriptor *ep_desc;unsigned num_endpoints; -int i; +unsigned i; num_endpoints = interface->cur_altsetting->desc.bNumEndpoints;dev_info (&udev->dev,"Number of endpoints %d\n" , num_endpoints); +if (!num_endpoints) +return ; +/* NOTE: some customers have programmed FT232R/FT245R devices * with an endpoint size of 0 - not good. In this case, we * want to override the endpoint descriptor setting and use a
Greg Kroah-Hartman
maintainerPATCH 20/122 rt2x00: disable TKIP on USB
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stanislaw Gruszka <[email protected]>
commit 8edcb0ba0d56f5914eef11eda6db8bfe74eb9ca8 upstream.
On USB we can not get atomically TKIP key. We have to disable support for TKIP acceleration on USB hardware to avoid bug as showed bellow.
[ 860.827243] BUG: scheduling while atomic: hostapd/3397/0x00000002 <snip> [ 860.827280] Call Trace: [ 860.827282] [<ffffffff81682ea6>] dump_stack+0x4d/0x66 [ 860.827284] [<ffffffff8167eb9b>] __schedule_bug+0x47/0x55 [ 860.827285] [<ffffffff81685bb3>] __schedule+0x733/0x7b0 [ 860.827287] [<ffffffff81685c59>] schedule+0x29/0x70 [ 860.827289] [<ffffffff81684f8a>] schedule_timeout+0x15a/0x2b0 [ 860.827291] [<ffffffff8105ac50>] ? ftrace_raw_event_tick_stop+0xc0/0xc0 [ 860.827294] [<ffffffff810c13c2>] ? __module_text_address+0x12/0x70 [ 860.827296] [<ffffffff81686823>] wait_for_completion_timeout+0xb3/0x140 [ 860.827298] [<ffffffff81080fc0>] ? wake_up_state+0x20/0x20 [ 860.827301] [<ffffffff814d5b3d>] usb_start_wait_urb+0x7d/0x150 [ 860.827303] [<ffffffff814d5cd5>] usb_control_msg+0xc5/0x110 [ 860.827305] [<ffffffffa02fb0c6>] rt2x00usb_vendor_request+0xc6/0x160 [rt2x00usb] [ 860.827307] [<ffffffffa02fb215>] rt2x00usb_vendor_req_buff_lock+0x75/0x150 [rt2x00usb] [ 860.827309] [<ffffffffa02fb393>] rt2x00usb_vendor_request_buff+0xa3/0xe0 [rt2x00usb] [ 860.827311] [<ffffffffa023d1a3>] rt2x00usb_register_multiread+0x33/0x40 [rt2800usb] [ 860.827314] [<ffffffffa05805f9>] rt2800_get_tkip_seq+0x39/0x50 [rt2800lib] [ 860.827321] [<ffffffffa0480f88>] ieee80211_get_key+0x218/0x2a0 [mac80211] [ 860.827322] [<ffffffff815cc68c>] ? __nlmsg_put+0x6c/0x80 [ 860.827329] [<ffffffffa051b02e>] nl80211_get_key+0x22e/0x360 [cfg80211]
--- drivers/net/wireless/rt2x00/rt2x00mac.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/wireless/rt2x00/rt2x00mac.c +2 -0
@@ -487,6 +487,8 @@ int rt2x00mac_set_key(struct ieee80211_h crypto.cipher =rt2x00crypto_key_to_cipher (key);if (crypto.cipher ==CIPHER_NONE )return -EOPNOTSUPP ; +if (crypto.cipher ==CIPHER_TKIP &&rt2x00_is_usb (rt2x00dev)) +return -EOPNOTSUPP ; crypto.cmd = cmd;
Greg Kroah-Hartman
maintainerPATCH 21/122 rt2x00: fix rfkill regression on rt2500pci
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stanislaw Gruszka <[email protected]>
commit 616a8394b5df8c88f4dd416f4527439a4e365034 upstream.
As reported by Niels, starting rfkill polling during device probe (commit e2bc7c5, generally sane change) broke rfkill on rt2500pci device. I considered that bug as some initalization issue, which should be fixed on rt2500pci specific code. But after several attempts (see bug report for details) we fail to find working solution. Hence I decided to revert to old behaviour on rt2500pci to fix regression.
Additionally patch also unregister rfkill on device remove instead of ifconfig down, what was another issue introduced by bad commit.
Bug report:
https://bugzilla.kernel.org/show_bug.cgi?id=73821
--- drivers/net/wireless/rt2x00/rt2500pci.c | 7 ++++++- drivers/net/wireless/rt2x00/rt2x00.h | 1 + drivers/net/wireless/rt2x00/rt2x00dev.c | 24 +++++++++++++++++++++--- 3 files changed, 28 insertions(+), 4 deletions(-)
drivers/net/wireless/rt2x00/rt2500pci.c +6 -1
@@ -1681,8 +1681,13 @@ static int rt2500pci_init_eeprom(struct/* * Detect if this device has an hardware controlled radio. */ -if (rt2x00_get_field16 (eeprom,EEPROM_ANTENNA_HARDWARE_RADIO )) +if (rt2x00_get_field16 (eeprom,EEPROM_ANTENNA_HARDWARE_RADIO )) {__set_bit (CAPABILITY_HW_BUTTON , &rt2x00dev->cap_flags); +/* +* On this device RFKILL initialized during probe does not work. +*/ +__set_bit (REQUIRE_DELAYED_RFKILL , &rt2x00dev->cap_flags); + }/* * Check if the BBP tuning should be enabled.
drivers/net/wireless/rt2x00/rt2x00.h +1 -0
@@ -693,6 +693,7 @@ enum rt2x00_capability_flags {REQUIRE_SW_SEQNO ,REQUIRE_HT_TX_DESC ,REQUIRE_PS_AUTOWAKE , + REQUIRE_DELAYED_RFKILL,/* * Capabilities
drivers/net/wireless/rt2x00/rt2x00dev.c +21 -3
@@ -1126,9 +1126,10 @@ static void rt2x00lib_uninitialize(strucreturn ;/* -* Unregister extra components. +* Stop rfkill polling. */ -rt2x00rfkill_unregister (rt2x00dev); +if (test_bit (REQUIRE_DELAYED_RFKILL , &rt2x00dev->cap_flags)) +rt2x00rfkill_unregister (rt2x00dev);/* * Allow the HW to uninitialize.
@@ -1166,6 +1167,12 @@ static int rt2x00lib_initialize(struct rset_bit (DEVICE_STATE_INITIALIZED , &rt2x00dev->flags); +/* +* Start rfkill polling. +*/ +if (test_bit (REQUIRE_DELAYED_RFKILL , &rt2x00dev->cap_flags)) +rt2x00rfkill_register (rt2x00dev); +return 0 ; }
@@ -1375,7 +1382,12 @@ int rt2x00lib_probe_dev(struct rt2x00_dert2x00link_register (rt2x00dev);rt2x00leds_register (rt2x00dev);rt2x00debug_register (rt2x00dev); -rt2x00rfkill_register (rt2x00dev); + +/* +* Start rfkill polling. +*/ +if (!test_bit (REQUIRE_DELAYED_RFKILL , &rt2x00dev->cap_flags)) +rt2x00rfkill_register (rt2x00dev);return 0 ;
@@ -1391,6 +1403,12 @@ void rt2x00lib_remove_dev(struct rt2x00_clear_bit (DEVICE_STATE_PRESENT , &rt2x00dev->flags);/* +* Stop rfkill polling. +*/ +if (!test_bit (REQUIRE_DELAYED_RFKILL , &rt2x00dev->cap_flags)) +rt2x00rfkill_unregister (rt2x00dev); + +/* * Disable radio. */ rt2x00lib_disable_radio (rt2x00dev);
Greg Kroah-Hartman
maintainerPATCH 22/122 mtd: eLBC NAND: fix subpage write support
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Pekon Gupta <[email protected]>
commit f034d87def51f026b735d1e2877e9387011b2ba3 upstream.
As subpage write is enabled by default for all drivers, nand_write_subpage_hwecc causes a crash if the driver did not register ecc->hwctl or ecc->calculate. This behavior was introduced in
commit 837a6ba4f3b6d23026674e6af6b6849a4634fff9
"mtd: nand: subpage write support for hardware based ECC schemes".
This fixes a crash by emulating subpage write support by padding sub-page data with 0xff on either sides to make it full page compatible.
--- drivers/mtd/nand/fsl_elbc_nand.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+)
drivers/mtd/nand/fsl_elbc_nand.c +14 -0
@@ -723,6 +723,19 @@ static int fsl_elbc_write_page(struct mtreturn 0 ; } +/* ECC will be calculated automatically, and errors will be detected in +* waitfunc. +*/ +static int fsl_elbc_write_subpage (struct mtd_info *mtd,struct nand_chip *chip, +uint32_t offset,uint32_t data_len, +const uint8_t *buf,int oob_required) +{ +fsl_elbc_write_buf (mtd, buf, mtd->writesize); +fsl_elbc_write_buf (mtd, chip->oob_poi, mtd->oobsize); + +return 0 ; +} +static int fsl_elbc_chip_init (struct fsl_elbc_mtd *priv) {struct fsl_lbc_ctrl *ctrl = priv->ctrl;
@@ -761,6 +774,7 @@ static int fsl_elbc_chip_init(struct fsl chip->ecc.read_page = fsl_elbc_read_page; chip->ecc.write_page = fsl_elbc_write_page; + chip->ecc.write_subpage = fsl_elbc_write_subpage;/* If CS Base Register selects full hardware ECC then use it */ if ((in_be32(&lbc->bank[priv->bank].br) & BR_DECC) ==
Greg Kroah-Hartman
maintainerPATCH 23/122 mtd: nand: omap: fix BCHx ecc.correct to return detected bit-flips in erased-page
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: pekon gupta <[email protected]>
commit f306e8c3b667632952f1a4a74ffb910bbc06255f upstream.
mtd: nand: omap2: Support for hardware BCH error correction.
In omap_elm_correct_data(), if bitflip_count in an erased-page is within the correctable limit (< ecc.strength), then it is not indicated back to the caller ecc->read_page().
This mis-guides upper layers like MTD and UBIFS layer to assume erased-page as perfectly clean and use it for writing even if actual bitflip_count was dangerously high (bitflip_count > mtd->bitflip_threshold).
This patch fixes this above issue, by returning 'stats' to caller ecc->read_page() under all scenarios.
--- drivers/mtd/nand/omap2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mtd/nand/omap2.c +1 -1
@@ -1382,7 +1382,7 @@ static int omap_elm_correct_data(struct/* Check if any error reported */ if (!is_error_reported) -return 0 ; +return stat;/* Decode BCH error using ELM module */ elm_decode_bch_error_page (info->elm_dev, ecc_vec, err_vec);
Greg Kroah-Hartman
maintainerPATCH 24/122 mtd: pxa3xx_nand: make the driver work on big-endian systems
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Petazzoni <[email protected]>
commit b7e460624f0f3c31150f3b09e75b0d009e22ba5f upstream.
The pxa3xx_nand driver currently uses __raw_writel() and __raw_readl() to access I/O registers. However, those functions do not do any endianness swapping, which means that they won't work when the CPU runs in big-endian but the I/O registers are little endian, which is the common situation for ARM systems running big endian.
Since __raw_writel() and __raw_readl() do not include any memory barriers and the pxa3xx_nand driver can only be compiled for ARM platforms, the closest I/o accessors functions that do endianess swapping are writel_relaxed() and readl_relaxed().
This patch has been verified to work on Armada XP GP: without the patch, the NAND is not detected when the kernel runs big endian while it is properly detected when the kernel runs little endian. With the patch applied, the NAND is properly detected in both situations (little and big endian).
--- drivers/mtd/nand/pxa3xx_nand.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/mtd/nand/pxa3xx_nand.c +2 -2
@@ -127,10 +127,10 @@/* macros for registers read/write */ #define nand_writel (info, off, val) \ - __raw_writel((val), (info)->mmio_base + (off)) + writel_relaxed((val), (info)->mmio_base + (off))#define nand_readl (info, off) \ - __raw_readl((info)->mmio_base + (off)) + readl_relaxed((info)->mmio_base + (off))/* error code and state */ enum {
Greg Kroah-Hartman
maintainerPATCH 25/122 vgaswitcheroo: switch the mux to the igp on power down when runpm is enabled
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit f2bc561610962693be61425cf913778586d8f9c1 upstream.
Avoids blank screens on muxed systems when runpm is active.
bug:
https://bugs.freedesktop.org/show_bug.cgi?id=75917
--- drivers/gpu/vga/vga_switcheroo.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/vga/vga_switcheroo.c +2 -1
@@ -623,7 +623,8 @@ static int vga_switcheroo_runtime_suspen ret = dev->bus->pm->runtime_suspend (dev);if (ret)return ret; - +if (vgasr_priv.handler->switchto) + vgasr_priv.handler->switchto (VGA_SWITCHEROO_IGD );vga_switcheroo_power_switch (pdev,VGA_SWITCHEROO_OFF );return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 26/122 drm/nouveau/kms/nv04-nv40: fix pageflip events via special case.
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mario Kleiner <[email protected]>
commit af4870e406126b7ac0ae7c7ce5751f25ebe60f28 upstream.
Cards with nv04 display engine can't reliably use vblank counts and timestamps computed via drm_handle_vblank(), as the function gets invoked after sending the pageflip events.
Fix this by defaulting to the old crtcid = -1 fallback path on <= NV-50 cards, and only using the precise path on NV-50 and later.
--- drivers/gpu/drm/nouveau/nouveau_display.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/gpu/drm/nouveau/nouveau_display.c +8 -2
@@ -798,6 +798,7 @@ nouveau_finish_page_flip(struct nouveau_struct drm_device *dev = drm->dev;struct nouveau_page_flip_state *s;unsigned long flags; +int crtcid =-1 ;spin_lock_irqsave (&dev->event_lock, flags);
@@ -808,8 +809,13 @@ nouveau_finish_page_flip(struct nouveau_ } s = list_first_entry(&fctx->flip, struct nouveau_page_flip_state, head); -if (s->event) -drm_send_vblank_event (dev, s->crtc, s->event); +if (s->event) { +/* Vblank timestamps/counts are only correct on >= NV-50 */ +if (nv_device (drm->device)->card_type >=NV_50 ) + crtcid = s->crtc; + +drm_send_vblank_event (dev, crtcid, s->event); + }list_del (&s->head);if (ps)
Greg Kroah-Hartman
maintainerPATCH 27/122 drm/nouveau/disp/nv04-nv40: abort scanoutpos query on vga analog.
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mario Kleiner <[email protected]>
commit e291af3f2259e9e105dfd72416fd5796513791a4 upstream.
nv04_disp_scanoutpos() must abort to trigger simple timestamping fallback if vtotal/htotal regs return zero. This happens if the output isn't a digital output, but a vga analog output, as the regs don't get initialized in that case.
Fixes timestamping failure on nv-40 and earlier with vga output.
--- drivers/gpu/drm/nouveau/core/engine/disp/nv04.c | 8 ++++++++ 1 file changed, 8 insertions(+)
drivers/gpu/drm/nouveau/core/engine/disp/nv04.c +8 -0
@@ -51,6 +51,14 @@ nv04_disp_scanoutpos(struct nouveau_obje args->htotal =nv_rd32 (priv,0x680824 + (head *0x2000 )) &0xffff ; args->hblanke = args->htotal -1 ; +/* +* If output is vga instead of digital then vtotal/htotal is invalid +* so we have to give up and trigger the timestamping fallback in the +* drm core. +*/ +if (!args->vtotal || !args->htotal) +return -ENOTSUPP ; + args->time[0 ] =ktime_to_ns (ktime_get ()); line =nv_rd32 (priv,0x600868 + (head *0x2000 )); args->time[1 ] =ktime_to_ns (ktime_get ());
Greg Kroah-Hartman
maintainerPATCH 28/122 drm/nouveau/kms: reference vblank for crtc during pageflip.
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mario Kleiner <[email protected]>
commit ba124a41058b300a5464206d2d33803cc3dc82ec upstream.
Need to drm_vblank_get/put() the crtc involved in a pending pageflip, or we might not get vblank irqs and updates of vblank counts and timestamps for pageflip events and flip completion.
--- drivers/gpu/drm/nouveau/nouveau_display.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/gpu/drm/nouveau/nouveau_display.c +7 -0
@@ -736,6 +736,9 @@ nouveau_crtc_page_flip(struct drm_crtc * fb->bits_per_pixel, fb->pitches[0 ], crtc->x, crtc->y, new_bo->bo.offset }; +/* Keep vblanks on during flip, for the target crtc of this flip */ +drm_vblank_get (dev,nouveau_crtc (crtc)->index); +/* Emit a page flip */ if (nv_device (drm->device)->card_type >=NV_50 ) { ret =nv50_display_flip_next (crtc, fb, chan, swap_interval);
@@ -779,6 +782,7 @@ nouveau_crtc_page_flip(struct drm_crtc *return 0 ; fail_unreserve: +drm_vblank_put (dev,nouveau_crtc (crtc)->index);ttm_bo_unreserve (&old_bo->bo); fail_unpin:mutex_unlock (&chan->cli->mutex);
@@ -817,6 +821,9 @@ nouveau_finish_page_flip(struct nouveau_drm_send_vblank_event (dev, crtcid, s->event); } +/* Give up ownership of vblank for page-flipped crtc */ +drm_vblank_put (dev, s->crtc); +list_del (&s->head);if (ps) *ps = *s;
Greg Kroah-Hartman
maintainerPATCH 29/122 drm/radeon: only apply hdmi bpc pll flags when encoder mode is hdmi
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit 7d5ab3009a8ca777174f6f469277b3922d56fd4b upstream.
May fix display issues with non-HDMI displays.
--- drivers/gpu/drm/radeon/atombios_crtc.c | 48 +++++++++++++++++---------------- 1 file changed, 26 insertions(+), 22 deletions(-)
drivers/gpu/drm/radeon/atombios_crtc.c +26 -22
@@ -862,14 +862,16 @@ static void atombios_crtc_program_pll(st args.v5.ucMiscInfo =0 ;/* HDMI depth, etc. */ if (ss_enabled && (ss->type &ATOM_EXTERNAL_SS_MASK )) args.v5.ucMiscInfo |=PIXEL_CLOCK_V5_MISC_REF_DIV_SRC ; -switch (bpc) { -case 8 : -default : - args.v5.ucMiscInfo |=PIXEL_CLOCK_V5_MISC_HDMI_24BPP ; -break ; -case 10 : - args.v5.ucMiscInfo |=PIXEL_CLOCK_V5_MISC_HDMI_30BPP ; -break ; +if (encoder_mode ==ATOM_ENCODER_MODE_HDMI ) { +switch (bpc) { +case 8 : +default : + args.v5.ucMiscInfo |=PIXEL_CLOCK_V5_MISC_HDMI_24BPP ; +break ; +case 10 : + args.v5.ucMiscInfo |=PIXEL_CLOCK_V5_MISC_HDMI_30BPP ; +break ; + } } args.v5.ucTransmitterID = encoder_id; args.v5.ucEncoderMode = encoder_mode;
@@ -884,20 +886,22 @@ static void atombios_crtc_program_pll(st args.v6.ucMiscInfo =0 ;/* HDMI depth, etc. */ if (ss_enabled && (ss->type &ATOM_EXTERNAL_SS_MASK )) args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_REF_DIV_SRC ; -switch (bpc) { -case 8 : -default : - args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_HDMI_24BPP ; -break ; -case 10 : - args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_HDMI_30BPP ; -break ; -case 12 : - args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_HDMI_36BPP ; -break ; -case 16 : - args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_HDMI_48BPP ; -break ; +if (encoder_mode ==ATOM_ENCODER_MODE_HDMI ) { +switch (bpc) { +case 8 : +default : + args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_HDMI_24BPP ; +break ; +case 10 : + args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_HDMI_30BPP ; +break ; +case 12 : + args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_HDMI_36BPP ; +break ; +case 16 : + args.v6.ucMiscInfo |=PIXEL_CLOCK_V6_MISC_HDMI_48BPP ; +break ; + } } args.v6.ucTransmitterID = encoder_id; args.v6.ucEncoderMode = encoder_mode;
Greg Kroah-Hartman
maintainerPATCH 30/122 drm/radeon: fix typo in radeon_connector_is_dp12_capable()
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit af5d36539dfe043f1cf0f8b7334d6bb12cd14e75 upstream.
We were checking the ext clock rather than the display clock.
Noticed by ArtForz on IRC.
--- drivers/gpu/drm/radeon/radeon_connectors.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/radeon/radeon_connectors.c +1 -1
@@ -1387,7 +1387,7 @@ bool radeon_connector_is_dp12_capable(ststruct radeon_device *rdev = dev->dev_private;if (ASIC_IS_DCE5 (rdev) && - (rdev->clock.dp_extclk >=53900 ) && + (rdev->clock.default_dispclk >=53900 ) &&radeon_connector_encoder_is_hbr2 (connector)) {return true; }
Greg Kroah-Hartman
maintainerPATCH 31/122 drm/radeon/dp: fix lane/clock setup for dp 1.2 capable devices
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit 3b6d9fd23e015b5397c438fd3cd74147d2c805b6 upstream.
Only DCE5+ asics support DP 1.2.
Noticed by ArtForz on IRC.
--- drivers/gpu/drm/radeon/atombios_dp.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-)
drivers/gpu/drm/radeon/atombios_dp.c +15 -2
@@ -281,6 +281,19 @@ static int dp_get_max_dp_pix_clock(int l/***** radeon specific DP functions *****/ +static int radeon_dp_get_max_link_rate (struct drm_connector *connector, +u8 dpcd[DP_DPCD_SIZE ]) +{ +int max_link_rate; + +if (radeon_connector_is_dp12_capable (connector)) + max_link_rate =min (drm_dp_max_link_rate (dpcd),540000 ); +else + max_link_rate =min (drm_dp_max_link_rate (dpcd),270000 ); + +return max_link_rate; +} +/* First get the min lane# when low rate is used according to pixel clock * (prefer low rate), second check max lane# supported by DP panel, * if the max lane# < low rate lane# then use max lane# instead.
@@ -290,7 +303,7 @@ static int radeon_dp_get_dp_lane_number( int pix_clock) {int bpp =convert_bpc_to_bpp (radeon_get_monitor_bpc (connector)); -int max_link_rate =drm_dp_max_link_rate (dpcd); +int max_link_rate =radeon_dp_get_max_link_rate (connector, dpcd);int max_lane_num =drm_dp_max_lane_count (dpcd);int lane_num;int max_dp_pix_clock;
@@ -328,7 +341,7 @@ static int radeon_dp_get_dp_link_clock(sreturn 540000 ; } -return drm_dp_max_link_rate (dpcd); +return radeon_dp_get_max_link_rate (connector, dpcd); } static u8 radeon_dp_encoder_service(struct radeon_device *rdev,
Greg Kroah-Hartman
maintainerPATCH 32/122 drm/radeon/atom: fix dithering on certain panels
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit 642528355c694f5ed68f6bff9ff520326a249f99 upstream.
We need to specify the encoder mode as LVDS for eDP when using the Crtc_Source atom table in order to properly set up the FMT hardware.
bug:
https://bugs.freedesktop.org/show_bug.cgi?id=73911
--- drivers/gpu/drm/radeon/atombios_encoders.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/gpu/drm/radeon/atombios_encoders.c +4 -1
@@ -1884,8 +1884,11 @@ atombios_set_encoder_crtc_source(struct args.v2.ucEncodeMode =ATOM_ENCODER_MODE_CRT ; else args.v2.ucEncodeMode = atombios_get_encoder_mode(encoder); - }else + }else if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT )) { + args.v2.ucEncodeMode =ATOM_ENCODER_MODE_LVDS ; + }else { args.v2.ucEncodeMode =atombios_get_encoder_mode (encoder); + }switch (radeon_encoder->encoder_id) {case ENCODER_OBJECT_ID_INTERNAL_UNIPHY :case ENCODER_OBJECT_ID_INTERNAL_UNIPHY1 :
Greg Kroah-Hartman
maintainerPATCH 33/122 drm/radeon: add missing vce init case for hawaii
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit d71c48f69cc03912578472bced4cc43069fe07e1 upstream.
Hawaii has the same version of VCE as other CIK parts.
--- drivers/gpu/drm/radeon/radeon_vce.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/radeon/radeon_vce.c +1 -0
@@ -66,6 +66,7 @@ int radeon_vce_init(struct radeon_devicecase CHIP_BONAIRE :case CHIP_KAVERI :case CHIP_KABINI : +case CHIP_HAWAII :case CHIP_MULLINS : fw_name =FIRMWARE_BONAIRE ;break ;
Greg Kroah-Hartman
maintainerPATCH 34/122 drm/radeon/dpm: fix typo in vddci setup for eg/btc
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit e07929810f0a19ddd756558290c7d72827cbfcd9 upstream.
We were using the vddc mask rather than the vddci mask.
Bug:
https://bugzilla.kernel.org/show_bug.cgi?id=79071
Possibly also fixes:
https://bugzilla.kernel.org/show_bug.cgi?id=68571
--- drivers/gpu/drm/radeon/cypress_dpm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/radeon/cypress_dpm.c +1 -1
@@ -1551,7 +1551,7 @@ int cypress_populate_smc_voltage_tables( table->voltageMaskTable.highMask[RV770_SMC_VOLTAGEMASK_VDDCI ] =0 ; table->voltageMaskTable.lowMask[RV770_SMC_VOLTAGEMASK_VDDCI ] = -cpu_to_be32 (eg_pi->vddc_voltage_table.mask_low); +cpu_to_be32 (eg_pi->vddci_voltage_table.mask_low); }return 0 ;
Greg Kroah-Hartman
maintainerPATCH 36/122 drm/radeon/cik: fix typo in EOP packet
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Deucher <[email protected]>
commit b397207b7475afa9df2f94541f978100ff1ea47e upstream.
Volatile bit was in the wrong location. This bit is not used at the moment.
--- drivers/gpu/drm/radeon/cikd.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/radeon/cikd.h +1 -1
@@ -1751,12 +1751,12 @@#define EOP_TC_WB_ACTION_EN (1 << 15)/* L2 */ #define EOP_TCL1_ACTION_EN (1 << 16)#define EOP_TC_ACTION_EN (1 << 17)/* L2 */ +#define EOP_TCL2_VOLATILE (1 << 24)#define EOP_CACHE_POLICY (x) ((x) << 25)/* 0 - LRU * 1 - Stream * 2 - Bypass */ -#define EOP_TCL2_VOLATILE (1 << 27)#define DATA_SEL (x) ((x) << 29)/* 0 - discard * 1 - send low 32bit data
Greg Kroah-Hartman
maintainerPATCH 37/122 drm/nv50-/mc: fix kms pageflip events by reordering irq handling order.
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mario Kleiner <[email protected]>
commit dcfb1009df3b4ad8d2e0779dd45b438629d6858a upstream.
Whenever a single nouveau_mc_intr() main gpu irq-handler invocation was responsible for calling both, the vblank-irq handler (display engine irq) and kms-pageflip completion handler (from fifo irq), the order of invocation was wrong. nouveau_finish_flip() was called before drm_handle_vblank() for the vblank of pageflip completion, so the emitted pageflip event contained stale vblank count and timestamp from previous vblank. This caused failure in userspace to timestamp properly.
Reorder order of invocation of engine irq handlers: Put NVDEV_ENGINE_DISP always on top, and thereby before NVDEV_ENGINE_FIFO, so that drm_handle_vblank() gets called to update vblank timestamps and count before potential pageflip events make use of that information.
This works on nv-50 and later, where kms-pageflip completion triggers an irq either after a separate vblank irq, or both pageflip and vblank trigger one common irq invocation, but never before vblank irqs.
v2 (Ben):
- removed mods for nv04-nv40, it doesn't help there anyway
- this is considered a hack, and a better solution should be found
--- drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c | 2 +- drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c | 2 +- drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c +1 -1
@@ -26,6 +26,7 @@const struct nouveau_mc_intr nv50_mc_intr[] = { + {0x04000000 ,NVDEV_ENGINE_DISP },/* DISP before FIFO, so pageflip-timestamping works! */ {0x00000001 ,NVDEV_ENGINE_MPEG }, {0x00000100 ,NVDEV_ENGINE_FIFO }, {0x00001000 ,NVDEV_ENGINE_GR },
@@ -34,7 +35,6 @@ nv50_mc_intr[] = { { 0x00020000, NVDEV_ENGINE_VP }, /* NV84- */ { 0x00100000, NVDEV_SUBDEV_TIMER }, { 0x00200000, NVDEV_SUBDEV_GPIO }, - { 0x04000000, NVDEV_ENGINE_DISP }, { 0x10000000, NVDEV_SUBDEV_BUS }, { 0x80000000, NVDEV_ENGINE_SW }, { 0x0002d101, NVDEV_SUBDEV_FB },
drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c +1 -1
@@ -26,6 +26,7 @@static const struct nouveau_mc_intr nv98_mc_intr[] = { + {0x04000000 ,NVDEV_ENGINE_DISP },/* DISP first, so pageflip timestamps work */ {0x00000001 ,NVDEV_ENGINE_PPP }, {0x00000100 ,NVDEV_ENGINE_FIFO }, {0x00001000 ,NVDEV_ENGINE_GR },
@@ -37,7 +38,6 @@ nv98_mc_intr[] = { { 0x00100000, NVDEV_SUBDEV_TIMER }, { 0x00200000, NVDEV_SUBDEV_GPIO }, { 0x00400000, NVDEV_ENGINE_COPY0 }, /* NVA3- */ - { 0x04000000, NVDEV_ENGINE_DISP }, { 0x10000000, NVDEV_SUBDEV_BUS }, { 0x80000000, NVDEV_ENGINE_SW }, { 0x0042d101, NVDEV_SUBDEV_FB },
drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c +1 -1
@@ -26,6 +26,7 @@const struct nouveau_mc_intr nvc0_mc_intr[] = { + {0x04000000 ,NVDEV_ENGINE_DISP },/* DISP first, so pageflip timestamps work. */ {0x00000001 ,NVDEV_ENGINE_PPP }, {0x00000020 ,NVDEV_ENGINE_COPY0 }, {0x00000040 ,NVDEV_ENGINE_COPY1 },
@@ -40,7 +41,6 @@ nvc0_mc_intr[] = { { 0x00200000, NVDEV_SUBDEV_GPIO }, { 0x01000000, NVDEV_SUBDEV_PWR }, { 0x02000000, NVDEV_SUBDEV_LTCG }, - { 0x04000000, NVDEV_ENGINE_DISP }, { 0x08000000, NVDEV_SUBDEV_FB }, { 0x10000000, NVDEV_SUBDEV_BUS }, { 0x40000000, NVDEV_SUBDEV_IBUS },
Greg Kroah-Hartman
maintainerPATCH 38/122 drm/gk208/gr: add missing registers to grctx init
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ilia Mirkin <[email protected]>
commit 255b329ca7f0e9b5fa6da3a68bb713684fe10305 upstream.
This fixes hangs on GK208 which happen instantaneously on trying to use a geometry shader.
--- drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c +2 -0
@@ -545,10 +545,12 @@ nv108_grctx_generate_mods(struct nvc0_grmmio_list (0x408010 ,0x80000000 ,0 ,0 );mmio_list (0x419004 ,0x00000000 ,8 ,1 );mmio_list (0x419008 ,0x00000000 ,0 ,0 ); +mmio_list (0x4064cc ,0x80000000 ,0 ,0 );mmio_list (0x408004 ,0x00000000 ,8 ,0 );mmio_list (0x408008 ,0x80000030 ,0 ,0 );mmio_list (0x418808 ,0x00000000 ,8 ,0 );mmio_list (0x41880c ,0x80000030 ,0 ,0 ); +mmio_list (0x4064c8 ,0x00c20200 ,0 ,0 );mmio_list (0x418810 ,0x80000000 ,12 ,2 );mmio_list (0x419848 ,0x10000000 ,12 ,2 );
Greg Kroah-Hartman
maintainerPATCH 39/122 drm/i915/bdw: Only use 2g GGTT for 32b platforms
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ben Widawsky <[email protected]>
commit 562d55d991b39ce376c492df2f7890fd6a541ffc upstream.
Daniel requested in the bug that I use a 3GB fallback size. Since this is not in the spec as a valid size, I decided against it. We could potentially add a patch to bump it to 3GB on top of this one.
This probably should be CC: stable - but I'll let the powers that be decide that one.
Regression from a revert of the revert:
commit 7907f45bf9f67a1c5e5d4ae05bab428d7c2f43b2
Author: Ben Widawsky <[email protected]> Date: Wed Feb 19 22:05:46 2014 -0800
Revert "drm/i915/bdw: Limit GTT to 2GB"
v2: Change ifdef to 32b, instead of ifndef
update comment
v3. Update comment to not wrap (Daniel).
Update commit message
v4: s/CONFIG_32/CONFIG_X86_32 (Jani).
v5: s/CONFIG_x86_32BIT/CONFIG_x86_32, as meant in v4 s/32B/32b (chris)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=76619
--- drivers/gpu/drm/i915/i915_gem_gtt.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/gpu/drm/i915/i915_gem_gtt.c +7 -0
@@ -1811,6 +1811,13 @@ static inline unsigned int gen8_get_tota bdw_gmch_ctl &=BDW_GMCH_GGMS_MASK ;if (bdw_gmch_ctl) bdw_gmch_ctl =1 << bdw_gmch_ctl; + +#ifdef CONFIG_X86_32 +/* Limit 32b platforms to a 2GB GGTT: 4 << 20 / pte size * PAGE_SIZE */ +if (bdw_gmch_ctl >4 ) + bdw_gmch_ctl =4 ; +#endif +return bdw_gmch_ctl <<20 ; }
Greg Kroah-Hartman
maintainerPATCH 40/122 drm/i915: Reorder semaphore deadlock check
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chris Wilson <[email protected]>
commit 4be173813e57c7298103a83155c2391b5b167b4c upstream.
If a semaphore is waiting on another ring, which in turn happens to be waiting on the first ring, but that second semaphore has been signalled, we will be able to kick the second ring and so can treat the first ring as a valid WAIT and not as HUNG.
v2: Be paranoid and cap the potential recursion depth whilst visiting the semaphore signallers. (Mika)
References: https://bugs.freedesktop.org/show_bug.cgi?id=54226
References: https://bugs.freedesktop.org/show_bug.cgi?id=75502
--- drivers/gpu/drm/i915/i915_irq.c | 18 ++++++++++++++---- drivers/gpu/drm/i915/intel_ringbuffer.h | 2 +- 2 files changed, 15 insertions(+), 5 deletions(-)
drivers/gpu/drm/i915/i915_irq.c +14 -4
@@ -2561,10 +2561,14 @@ static int semaphore_passed(struct intelstruct intel_ring_buffer *signaller;u32 seqno, ctl; - ring->hangcheck.deadlock = true; + ring->hangcheck.deadlock++; signaller =semaphore_waits_for (ring, &seqno); -if (signaller ==NULL || signaller->hangcheck.deadlock) +if (signaller ==NULL ) +return -1 ; + +/* Prevent pathological recursion due to driver bugs */ +if (signaller->hangcheck.deadlock >=I915_NUM_RINGS )return -1 ;/* cursory check for an unkickable deadlock */
@@ -2572,7 +2576,13 @@ static int semaphore_passed(struct intelif (ctl &RING_WAIT_SEMAPHORE &&semaphore_passed (signaller) <0 )return -1 ; -return i915_seqno_passed (signaller->get_seqno (signaller, false), seqno); +if (i915_seqno_passed (signaller->get_seqno (signaller, false), seqno)) +return 1 ; + +if (signaller->hangcheck.deadlock) +return -1 ; + +return 0 ; }static void semaphore_clear_deadlocks (struct drm_i915_private *dev_priv)
@@ -2581,7 +2591,7 @@ static void semaphore_clear_deadlocks(stint i;for_each_ring (ring, dev_priv, i) - ring->hangcheck.deadlock = false; + ring->hangcheck.deadlock =0 ; } static enum intel_ring_hangcheck_action
drivers/gpu/drm/i915/intel_ringbuffer.h +1 -1
@@ -51,7 +51,7 @@ struct intel_ring_hangcheck {u32 seqno;int score;enum intel_ring_hangcheck_action action; -bool deadlock; +int deadlock; };struct intel_ring_buffer {
Greg Kroah-Hartman
maintainerPATCH 41/122 drm/i915: Disable FBC by default also on Haswell and later
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chris Wilson <[email protected]>
commit 0368920e51ae0cded0eb518c340a4dd17764d461 upstream.
It causes black screen on bootup and is approximately 100x slower than running with FBC disabled, so the GPU runs at a high frequency for much longer - completely contrary to the power saving claims. It also still has mutex deadlocks in multi-head scenarios, which can lead to a system/X lockup. These bugs were known before FBC was enabled by default on Haswell and still have not been fixed.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=79716
--- drivers/gpu/drm/i915/intel_pm.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/gpu/drm/i915/intel_pm.c +1 -2
@@ -511,8 +511,7 @@ void intel_update_fbc(struct drm_device obj = intel_fb->obj; adjusted_mode = &intel_crtc->config.adjusted_mode; -if (i915.enable_fbc <0 && -INTEL_INFO (dev)->gen <=7 && !IS_HASWELL (dev)) { +if (i915.enable_fbc <0 ) {if (set_no_fbc_reason (dev_priv,FBC_CHIP_DEFAULT ))DRM_DEBUG_KMS ("disabled per chip default\n" ); goto out_disable;
Greg Kroah-Hartman
maintainerPATCH 43/122 drm/i915: set backlight duty cycle after backlight enable for gen4
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jani Nikula <[email protected]>
commit 2e7eeeb59a92d09144fdb7d2dc1af77a10a7945b upstream.
For reasons I can't claim to fully understand gen4 seems to require backlight duty cycle setting after the backlight has been enabled, or else black screen follows. I don't have documentation for the correct sequence on gen4 either. Confirmed on Dell Latitude D630 and MacBook4,1.
This fixes a regression introduced by
commit b35684b8fa94e04f55fd38bf672b737741d2f9e2
Author: Jani Nikula <[email protected]> Date: Thu Nov 14 12:13:41 2013 +0200
drm/i915: do full backlight setup at enable time
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=75791
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=79423
--- drivers/gpu/drm/i915/intel_panel.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_panel.c +2 -3
@@ -723,9 +723,6 @@ static void i965_enable_backlight(struct ctl = freq <<16 ;I915_WRITE (BLC_PWM_CTL , ctl); -/* XXX: combine this into above write? */ -intel_panel_actually_set_backlight (connector, panel->backlight.level); - ctl2 =BLM_PIPE (pipe);if (panel->backlight.combination_mode) ctl2 |=BLM_COMBINATION_MODE ;
@@ -734,6 +731,8 @@ static void i965_enable_backlight(structI915_WRITE (BLC_PWM_CTL2 , ctl2);POSTING_READ (BLC_PWM_CTL2 );I915_WRITE (BLC_PWM_CTL2 , ctl2 |BLM_PWM_ENABLE ); + +intel_panel_actually_set_backlight (connector, panel->backlight.level); }static void vlv_enable_backlight (struct intel_connector *connector)
Greg Kroah-Hartman
maintainerPATCH 44/122 drm/i915, HD-audio: Dont continue probing when nomodeset is given
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Takashi Iwai <[email protected]>
commit 74b0c2d75fb4cc89173944e6d8f9eb47aca0c343 upstream.
When a machine is booted with nomodeset option, i915 driver skips the whole initialization. Meanwhile, HD-audio tries to bind wth i915 just by request_symbol() without knowing that the initialization was skipped, and eventually it hits WARN_ON() in i915_request_power_well() and i915_release_power_well() wrongly but still continues probing, even though it doesn't work at all.
In this patch, both functions are changed to return an error in case of uninitialized state instead of WARN_ON(), so that HD-audio driver can give up HDMI controller initialization at the right time.
--- drivers/gpu/drm/i915/intel_pm.c | 14 ++++++++------ include/drm/i915_powerwell.h | 4 ++-- sound/pci/hda/hda_i915.c | 12 ++++++------ sound/pci/hda/hda_i915.h | 4 ++-- sound/pci/hda/hda_intel.c | 7 ++++++- 5 files changed, 24 insertions(+), 17 deletions(-)
drivers/gpu/drm/i915/intel_pm.c +8 -6
@@ -5705,30 +5705,32 @@ void intel_display_power_put(struct drm_static struct i915_power_domains *hsw_pwr;/* Display audio driver power well request */ -void i915_request_power_well (void ) +int i915_request_power_well (void ) {struct drm_i915_private *dev_priv; -if (WARN_ON (!hsw_pwr)) -return ; +if (!hsw_pwr) +return -ENODEV ; dev_priv = container_of(hsw_pwr, struct drm_i915_private, power_domains);intel_display_power_get (dev_priv,POWER_DOMAIN_AUDIO ); +return 0 ; }EXPORT_SYMBOL_GPL (i915_request_power_well);/* Display audio driver power well release */ -void i915_release_power_well (void ) +int i915_release_power_well (void ) {struct drm_i915_private *dev_priv; -if (WARN_ON (!hsw_pwr)) -return ; +if (!hsw_pwr) +return -ENODEV ; dev_priv = container_of(hsw_pwr, struct drm_i915_private, power_domains);intel_display_power_put (dev_priv,POWER_DOMAIN_AUDIO ); +return 0 ; }EXPORT_SYMBOL_GPL (i915_release_power_well);
include/drm/i915_powerwell.h +2 -2
@@ -30,7 +30,7 @@#define _I915_POWERWELL_H_/* For use by hda_i915 driver */ -extern void i915_request_power_well (void ); -extern void i915_release_power_well (void ); +extern int i915_request_power_well (void ); +extern int i915_release_power_well (void );#endif /* _I915_POWERWELL_H_ */
sound/pci/hda/hda_i915.c +6 -6
@@ -22,20 +22,20 @@#include <drm/i915_powerwell.h> #include "hda_i915.h" -static void (*get_power)(void ); -static void (*put_power)(void ); +static int (*get_power)(void ); +static int (*put_power)(void ); -void hda_display_power (bool enable) +int hda_display_power (bool enable) {if (!get_power || !put_power) -return ; +return -ENODEV ;pr_debug ("HDA display power %s \n" , enable ?"Enable" :"Disable" );if (enable) -get_power (); +return get_power ();else -put_power (); +return put_power (); }int hda_i915_init (void )
sound/pci/hda/hda_i915.h +2 -2
@@ -17,11 +17,11 @@#define __SOUND_HDA_I915_H#ifdef CONFIG_SND_HDA_I915 -void hda_display_power (bool enable); +int hda_display_power (bool enable);int hda_i915_init (void );int hda_i915_exit (void );#else -static inline void hda_display_power (bool enable) {} +static inline int hda_display_power (bool enable) {return 0 ; }static inline int hda_i915_init (void ) {return -ENODEV ;
sound/pci/hda/hda_intel.c +6 -1
@@ -1707,8 +1707,13 @@ static int azx_probe_continue(struct azx "Error request power-well from i915\n"); goto out_free; } + err =hda_display_power (true); +if (err <0 ) { +dev_err (chip->card->dev, +"Cannot turn on display power on i915\n" ); + goto out_free; + }#endif -hda_display_power (true); } err =azx_first_init (chip);
Greg Kroah-Hartman
maintainerPATCH 45/122 drm/i915: Hold the table lock whilst walking the files idr and counting the objects in debugfs
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chris Wilson <[email protected]>
commit 5b5ffff0d25060ab0e21fa0f6cd16428e87bf1ea upstream.
Fixes an issue whereby we may race with the table updates (before the core takes the struct_mutex) and so risk dereferencing a stale pointer in the iterator for /debugfs/.../i915_gem_objects. For example,
[ 1524.757545] BUG: unable to handle kernel paging request at f53af748 [ 1524.757572] IP: [<c1406982>] per_file_stats+0x12/0x100 [ 1524.757599] *pdpt = 0000000001b13001 *pde = 00000000379fb067 *pte = 80000000353af060 [ 1524.757621] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC [ 1524.757637] Modules linked in: ctr ccm arc4 ath9k ath9k_common ath9k_hw ath snd_hda_codec_conexant mac80211 snd_hda_codec_generic snd_hda_intel snd_hda_controller snd_hda_codec bnep snd_hwdep rfcomm snd_pcm gpio_ich dell_wmi sparse_keymap snd_seq_midi hid_multitouch uvcvideo snd_seq_midi_event dell_laptop snd_rawmidi dcdbas snd_seq videobuf2_vmalloc videobuf2_memops videobuf2_core usbhid videodev snd_seq_device coretemp snd_timer hid joydev kvm_intel cfg80211 ath3k kvm btusb bluetooth serio_raw snd microcode soundcore lpc_ich wmi mac_hid parport_pc ppdev lp parport psmouse ahci libahci [ 1524.757825] CPU: 3 PID: 1911 Comm: intel-gpu-overl Tainted: G W OE 3.15.0-rc3+ #96 [ 1524.757840] Hardware name: Dell Inc. Inspiron 1090/Inspiron 1090, BIOS A06 08/23/2011 [ 1524.757855] task: f52f36c0 ti: f4cbc000 task.ti: f4cbc000 [ 1524.757869] EIP: 0060:[<c1406982>] EFLAGS: 00210202 CPU: 3 [ 1524.757884] EIP is at per_file_stats+0x12/0x100 [ 1524.757896] EAX: 0000002d EBX: 00000000 ECX: f4cbdefc EDX: f53af700 [ 1524.757909] ESI: c1406970 EDI: f53af700 EBP: f4cbde6c ESP: f4cbde5c [ 1524.757922] DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 [ 1524.757934] CR0: 80050033 CR2: f53af748 CR3: 356af000 CR4: 000007f0 [ 1524.757945] Stack: [ 1524.757957] f4cbdefc 00000000 c1406970 f53af700 f4cbdea8 c12e5f15 f4cbdefc c1406970 [ 1524.757993] 0000ffff f4cbde90 0000002d f5dc5cd0 e4e80438 c1181d59 f4cbded8 f4d89900 [ 1524.758027] f5631b40 e5131074 c1903f37 f4cbdf28 c14068e6 f52648a0 c1927748 c1903f37 [ 1524.758062] Call Trace: [ 1524.758084] [<c1406970>] ? i915_gem_object_info+0x510/0x510 [ 1524.758106] [<c12e5f15>] idr_for_each+0xa5/0x100 [ 1524.758126] [<c1406970>] ? i915_gem_object_info+0x510/0x510 [ 1524.758148] [<c1181d59>] ? seq_vprintf+0x29/0x50 [ 1524.758168] [<c14068e6>] i915_gem_object_info+0x486/0x510 [ 1524.758189] [<c11823a6>] seq_read+0xd6/0x380 [ 1524.758208] [<c116d11d>] ? final_putname+0x1d/0x40 [ 1524.758227] [<c11822d0>] ? seq_hlist_next_percpu+0x90/0x90 [ 1524.758246] [<c1163e52>] vfs_read+0x82/0x150 [ 1524.758265] [<c11645d6>] SyS_read+0x46/0x90 [ 1524.758285] [<c16b8d8c>] sysenter_do_call+0x12/0x22 [ 1524.758298] Code: f5 8f 2a 00 83 c4 6c 31 c0 5b 5e 5f 5d c3 8d 74 26 00 8d bc 27 00 00 00 00 55 89 e5 57 56 53 83 ec 04 3e 8d 74 26 00 83 41 04 01 <8b> 42 48 01 41 08 8b 42 4c 89 d7 85 c0 75 07 8b 42 60 85 c0 74 [ 1524.758461] EIP: [<c1406982>] per_file_stats+0x12/0x100 SS:ESP 0068:f4cbde5c [ 1524.758485] CR2: 00000000f53af748
--- drivers/gpu/drm/i915/i915_debugfs.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/gpu/drm/i915/i915_debugfs.c +2 -0
@@ -446,7 +446,9 @@ static int i915_gem_object_info(struct smemset (&stats,0 ,sizeof (stats)); stats.file_priv = file->driver_priv; +spin_lock (&file->table_lock);idr_for_each (&file->object_idr, per_file_stats, &stats); +spin_unlock (&file->table_lock);/* * Although we have a valid reference on file->pid, that does * not guarantee that the task_struct who called get_pid() is
Greg Kroah-Hartman
maintainerPATCH 46/122 drm/i915: default to having backlight if VBT not available
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jani Nikula <[email protected]>
commit 56c4b63aaf4c2cd91966b2a5e69e5367bea60bbe upstream.
Apparently there are Apple laptops with magic smoke for a VBIOS, which we fail to find and use. Default to having and setting up backlight in this case.
This fixes a regression introduced by
commit c675949ec58ca50d5a3ae3c757892f1560f6e896
Author: Jani Nikula <[email protected]> Date: Wed Apr 9 11:31:37 2014 +0300
drm/i915: do not setup backlight if not available according to VBT
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=77831
--- drivers/gpu/drm/i915/intel_bios.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_bios.c +3 -3
@@ -287,9 +287,6 @@ parse_lfp_backlight(struct drm_i915_privconst struct bdb_lfp_backlight_data *backlight_data;const struct bdb_lfp_backlight_data_entry *entry; -/* Err to enabling backlight if no backlight block. */ - dev_priv->vbt.backlight.present = true; - backlight_data =find_section (bdb,BDB_LVDS_BACKLIGHT );if (!backlight_data)return ;
@@ -839,6 +836,9 @@ init_vbt_defaults(struct drm_i915_privat dev_priv->vbt.crt_ddc_pin =GMBUS_PORT_VGADDC ; +/* Default to having backlight */ + dev_priv->vbt.backlight.present = true; +/* LFP panel data */ dev_priv->vbt.lvds_dither =1 ; dev_priv->vbt.lvds_vbt =0 ;
Greg Kroah-Hartman
maintainerPATCH 47/122 drm/i95: Initialize active ring->pid to -1
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chris Wilson <[email protected]>
commit eee73b46261325eb140d899b5371f49b02d88f63 upstream.
Otherwise we print out spurious processes on unused rings in the error state.
--- drivers/gpu/drm/i915/i915_gpu_error.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_gpu_error.c +2 -1
@@ -894,6 +894,8 @@ static void i915_gem_record_rings(structfor (i =0 ; i <I915_NUM_RINGS ; i++) {struct intel_ring_buffer *ring = &dev_priv->ring[i]; + error->ring[i].pid =-1 ; +if (ring->dev ==NULL )continue ;
@@ -901,7 +903,6 @@ static void i915_gem_record_rings(structi915_record_ring_state (dev, ring, &error->ring[i]); - error->ring[i].pid =-1 ; request =i915_gem_find_active_request (ring);if (request) {/* We need to copy these to an anonymous buffer
Greg Kroah-Hartman
maintainerPATCH 48/122 drm: fix NULL pointer access by wrong ioctl
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zhaowei Yuan <[email protected]>
commit 1539fb9bd405ee32282ea0a38404f9e008ac5b7a upstream.
If user uses wrong ioctl command with _IOC_NONE and argument size greater than 0, it can cause NULL pointer access from memset of line
463. If _IOC_NONE, don't memset to 0 for kdata.
--- drivers/gpu/drm/drm_drv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/drm_drv.c +2 -1
@@ -419,8 +419,9 @@ long drm_ioctl(struct file *filp, retcode = -EFAULT ; goto err_i1; } - }else + }else if (cmd &IOC_OUT ) {memset (kdata,0 , usize); + }if (ioctl->flags &DRM_UNLOCKED ) retcode =func (dev, kdata, file_priv);
Greg Kroah-Hartman
maintainerPATCH 49/122 drm/i915: provide interface for audio driver to query cdclk
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jani Nikula <[email protected]>
commit c149dcb5c60bfea8871f16dfcc0690255eeb825f upstream.
For Haswell and Broadwell, if the display power well has been disabled, the display audio controller divider values EM4 M VALUE and EM5 N VALUE will have been lost. The CDCLK frequency is required for reprogramming them to generate 24MHz HD-A link BCLK. So provide a private interface for the audio driver to query CDCLK.
This is a stopgap solution until a more generic interface between audio and display drivers has been implemented.
--- drivers/gpu/drm/i915/intel_pm.c | 21 +++++++++++++++++++++ include/drm/i915_powerwell.h | 1 + 2 files changed, 22 insertions(+)
drivers/gpu/drm/i915/intel_pm.c +21 -0
@@ -5734,6 +5734,27 @@ int i915_release_power_well(void) }EXPORT_SYMBOL_GPL (i915_release_power_well); +/* +* Private interface for the audio driver to get CDCLK in kHz. +* +* Caller must request power well using i915_request_power_well() prior to +* making the call. +*/ +int i915_get_cdclk_freq (void ) +{ +struct drm_i915_private *dev_priv; + +if (!hsw_pwr) +return -ENODEV ; + + dev_priv = container_of(hsw_pwr, struct drm_i915_private, + power_domains); + +return intel_ddi_get_cdclk_freq (dev_priv); +} +EXPORT_SYMBOL_GPL (i915_get_cdclk_freq); + +#define POWER_DOMAIN_MASK (BIT(POWER_DOMAIN_NUM) - 1)#define HSW_ALWAYS_ON_POWER_DOMAINS ( \
include/drm/i915_powerwell.h +1 -0
@@ -32,5 +32,6 @@/* For use by hda_i915 driver */ extern int i915_request_power_well (void );extern int i915_release_power_well (void ); +extern int i915_get_cdclk_freq (void );#endif /* _I915_POWERWELL_H_ */
Greg Kroah-Hartman
maintainerPATCH 50/122 ALSA: hda - restore BCLK M/N value as per CDCLK for HSW/BDW display HDA controller
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mengdong Lin <[email protected]>
commit e4d9e513dedb5ac4e166c1053314fa935ddecc8c upstream.
For HSW/BDW display HD-A controller, hda_set_bclk() is defined to set BCLK by programming the M/N values as per the core display clock (CDCLK) queried from i915 display driver.
And the audio driver will also set BCLK in azx_first_init() since the display driver can turn off the shared power in boot phase if only eDP is connected and M/N values will be lost and must be reprogrammed.
--- sound/pci/hda/hda_i915.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++ sound/pci/hda/hda_i915.h | 2 + sound/pci/hda/hda_intel.c | 50 +++++++---------------------------------- 3 files changed, 66 insertions(+), 41 deletions(-)
sound/pci/hda/hda_i915.c +55 -0
@@ -20,10 +20,20 @@#include <linux/module.h> #include <sound/core.h> #include <drm/i915_powerwell.h> +#include "hda_priv.h" #include "hda_i915.h" +/* Intel HSW/BDW display HDA controller Extended Mode registers. +* EM4 (M value) and EM5 (N Value) are used to convert CDClk (Core Display +* Clock) to 24MHz BCLK: BCLK = CDCLK * M / N +* The values will be lost when the display power well is disabled. +*/ +#define ICH6_REG_EM4 0x100c +#define ICH6_REG_EM5 0x1010 +static int (*get_power)(void );static int (*put_power)(void ); +static int (*get_cdclk)(void );int hda_display_power (bool enable) {
@@ -38,6 +48,43 @@ int hda_display_power(bool enable)return put_power (); } +void haswell_set_bclk (struct azx *chip) +{ +int cdclk_freq; +unsigned int bclk_m, bclk_n; + +if (!get_cdclk) +return ; + + cdclk_freq =get_cdclk (); +switch (cdclk_freq) { +case 337500 : + bclk_m =16 ; + bclk_n =225 ; +break ; + +case 450000 : +default :/* default CDCLK 450MHz */ + bclk_m =4 ; + bclk_n =75 ; +break ; + +case 540000 : + bclk_m =4 ; + bclk_n =90 ; +break ; + +case 675000 : + bclk_m =8 ; + bclk_n =225 ; +break ; + } + +azx_writew (chip,EM4 , bclk_m); +azx_writew (chip,EM5 , bclk_n); +} + +int hda_i915_init (void ) {int err =0 ;
@@ -55,6 +102,10 @@ int hda_i915_init(void)return -ENODEV ; } + get_cdclk =symbol_request (i915_get_cdclk_freq); +if (!get_cdclk)/* may have abnormal BCLK and audio playback rate */ +pr_warn ("hda-i915: get_cdclk symbol get fail\n" ); +pr_debug ("HDA driver get symbol successfully from i915 module\n" );return err;
@@ -70,6 +121,10 @@ int hda_i915_exit(void)symbol_put (i915_release_power_well); put_power =NULL ; } +if (get_cdclk) { +symbol_put (i915_get_cdclk_freq); + get_cdclk =NULL ; + }return 0 ; }
sound/pci/hda/hda_i915.h +2 -0
@@ -18,10 +18,12 @@#ifdef CONFIG_SND_HDA_I915 int hda_display_power (bool enable); +void haswell_set_bclk (struct azx *chip);int hda_i915_init (void );int hda_i915_exit (void );#else static inline int hda_display_power (bool enable) {return 0 ; } +static inline void haswell_set_bclk (struct azx *chip) {return ; }static inline int hda_i915_init (void ) {return -ENODEV ;
sound/pci/hda/hda_intel.c +9 -41
103 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 51/122 drm/vmwgfx: Fix incorrect write to read-only register v2:
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Hellstrom <[email protected]>
commit 4e578080ed3262ed2c3985868539bc66218d25c0 upstream.
Commit "drm/vmwgfx: correct fb_fix_screeninfo.line_length", while fixing a vmwgfx fbdev bug, also writes the pitch to a supposedly read-only register:SVGA_REG_BYTES_PER_LINE, while it should be (and also in fact is) written to SVGA_REG_PITCHLOCK.
This patch is Cc'd stable because of the unknown effects writing to this register might have, particularly on older device versions.
v2: Updated log message.
--- drivers/gpu/drm/vmwgfx/vmwgfx_fb.c | 1 - 1 file changed, 1 deletion(-)
drivers/gpu/drm/vmwgfx/vmwgfx_fb.c +0 -1
@@ -179,7 +179,6 @@ static int vmw_fb_set_par(struct fb_infovmw_write (vmw_priv,SVGA_REG_DISPLAY_POSITION_Y , info->var.yoffset);vmw_write (vmw_priv,SVGA_REG_DISPLAY_WIDTH , info->var.xres);vmw_write (vmw_priv,SVGA_REG_DISPLAY_HEIGHT , info->var.yres); -vmw_write (vmw_priv,SVGA_REG_BYTES_PER_LINE , info->fix.line_length);vmw_write (vmw_priv,SVGA_REG_DISPLAY_ID ,SVGA_ID_INVALID ); }
Greg Kroah-Hartman
maintainerPATCH 53/122 Bluetooth: Fix incorrectly overriding conn->src_type
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hedberg <[email protected]>
commit b62b65055bcc5372d5c3f4103629176cb8db3678 upstream.
The src_type member of struct hci_conn should always reflect the address type of the src_member. It should never be overridden. There is already code in place in the command status handler of HCI_LE_Create_Connection to copy the right initiator address into conn->init_addr_type.
Without this patch, if privacy is enabled, we will send the wrong address type in the SMP identity address information PDU (it'll e.g. contain our public address but a random address type).
--- net/bluetooth/hci_conn.c | 5 ----- 1 file changed, 5 deletions(-)
net/bluetooth/hci_conn.c +0 -5
@@ -584,11 +584,6 @@ static void hci_req_add_le_create_conn(sif (hci_update_random_address (req, false, &own_addr_type))return ; -/* Save the address type used for this connnection attempt so we able -* to retrieve this information if we need it. -*/ - conn->src_type = own_addr_type; - cp.scan_interval =cpu_to_le16 (hdev->le_scan_interval); cp.scan_window =cpu_to_le16 (hdev->le_scan_window);bacpy (&cp.peer_addr, &conn->dst);
Greg Kroah-Hartman
maintainerPATCH 54/122 Bluetooth: Fix SSP acceptor just-works confirmation without MITM
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hedberg <[email protected]>
commit ba15a58b179ed76a7e887177f2b06de12c58ec8f upstream.
Quoted 1 line
From the Bluetooth Core Specification 4.1 page 1958:
"if both devices have set the Authentication_Requirements parameter to one of the MITM Protection Not Required options, authentication stage 1 shall function as if both devices set their IO capabilities to DisplayOnly (e.g., Numeric comparison with automatic confirmation on both devices)"
So far our implementation has done user confirmation for all just-works cases regardless of the MITM requirements, however following the specification to the word means that we should not be doing confirmation when neither side has the MITM flag set.
--- net/bluetooth/hci_event.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
net/bluetooth/hci_event.c +5 -2
@@ -3503,8 +3503,11 @@ static void hci_user_confirm_request_evt/* If we're not the initiators request authorization to * proceed from user space (mgmt_user_confirm with -* confirm_hint set to 1). */ -if (!test_bit (HCI_CONN_AUTH_PEND , &conn->flags)) { +* confirm_hint set to 1). The exception is if neither +* side had MITM in which case we do auto-accept. +*/ +if (!test_bit (HCI_CONN_AUTH_PEND , &conn->flags) && + (loc_mitm || rem_mitm)) {BT_DBG ("Confirming auto-accept as acceptor" ); confirm_hint =1 ; goto confirm;
Greg Kroah-Hartman
maintainerPATCH 55/122 Bluetooth: Fix check for connection encryption
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hedberg <[email protected]>
commit e694788d73efe139b24f78b036deb97fe57fa8cb upstream.
The conn->link_key variable tracks the type of link key in use. It is set whenever we respond to a link key request as well as when we get a link key notification event.
These two events do not however always guarantee that encryption is
enabled: getting a link key request and responding to it may only mean that the remote side has requested authentication but not encryption. On the other hand, the encrypt change event is a certain guarantee that encryption is enabled. The real encryption state is already tracked in the conn->link_mode variable through the HCI_LM_ENCRYPT bit.
This patch fixes a check for encryption in the hci_conn_auth function to use the proper conn->link_mode value and thereby eliminates the chance of a false positive result.
--- net/bluetooth/hci_conn.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/bluetooth/hci_conn.c +1 -1
@@ -821,7 +821,7 @@ static int hci_conn_auth(struct hci_conn/* If we're already encrypted set the REAUTH_PEND flag, * otherwise set the ENCRYPT_PEND. */ -if (conn->key_type !=0xff ) +if (conn->link_mode &HCI_LM_ENCRYPT )set_bit (HCI_CONN_REAUTH_PEND , &conn->flags);else set_bit (HCI_CONN_ENCRYPT_PEND , &conn->flags);
Greg Kroah-Hartman
maintainerPATCH 56/122 Bluetooth: Fix indicating discovery state when canceling inquiry
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hedberg <[email protected]>
commit 50143a433b70e3145bcf8a4a4e54f0c11bdee32b upstream.
When inquiry is canceled through the HCI_Cancel_Inquiry command there is no Inquiry Complete event generated. Instead, all we get is the command complete for the HCI_Inquiry_Cancel command. This means that we must call the hci_discovery_set_state() function from the respective command complete handler in order to ensure that user space knows the correct discovery state.
--- net/bluetooth/hci_event.c | 4 ++++ 1 file changed, 4 insertions(+)
net/bluetooth/hci_event.c +4 -0
@@ -48,6 +48,10 @@ static void hci_cc_inquiry_cancel(structsmp_mb__after_clear_bit ();/* wake_up_bit advises about this barrier */ wake_up_bit (&hdev->flags,HCI_INQUIRY ); +hci_dev_lock (hdev); +hci_discovery_set_state (hdev,DISCOVERY_STOPPED ); +hci_dev_unlock (hdev); +hci_conn_check_pending (hdev); }
Greg Kroah-Hartman
maintainerPATCH 57/122 Bluetooth: Refactor discovery stopping into its own function
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hedberg <[email protected]>
commit 21a60d307ddc2180cfa542a995d943d1034cf5c5 upstream.
We'll need to reuse the same logic for stopping discovery also when cleaning up HCI state when powering off. This patch refactors the code out to its own function that can later (in a subsequent patch) be used also for the power off case.
--- net/bluetooth/mgmt.c | 93 ++++++++++++++++++++++++++------------------------- 1 file changed, 49 insertions(+), 44 deletions(-)
net/bluetooth/mgmt.c +49 -44
@@ -1045,6 +1045,43 @@ static void clean_up_hci_complete(struct } } +static void hci_stop_discovery (struct hci_request *req) +{ +struct hci_dev *hdev = req->hdev; +struct hci_cp_remote_name_req_cancel cp; +struct inquiry_entry *e; + +switch (hdev->discovery.state) { +case DISCOVERY_FINDING : +if (test_bit (HCI_INQUIRY , &hdev->flags)) { +hci_req_add (req,HCI_OP_INQUIRY_CANCEL ,0 ,NULL ); + }else { +cancel_delayed_work (&hdev->le_scan_disable); +hci_req_add_le_scan_disable (req); + } + +break ; + +case DISCOVERY_RESOLVING : + e =hci_inquiry_cache_lookup_resolve (hdev,BDADDR_ANY , +NAME_PENDING ); +if (!e) +return ; + +bacpy (&cp.bdaddr, &e->data.bdaddr); +hci_req_add (req,HCI_OP_REMOTE_NAME_REQ_CANCEL ,sizeof (cp), + &cp); + +break ; + +default : +/* Passive scanning */ +if (test_bit (HCI_LE_SCAN , &hdev->dev_flags)) +hci_req_add_le_scan_disable (req); +break ; + } +} +static int clean_up_hci_state (struct hci_dev *hdev) {struct hci_request req;
@@ -3570,8 +3607,6 @@ static int stop_discovery(struct sock *s {struct mgmt_cp_stop_discovery *mgmt_cp = data;struct pending_cmd *cmd; -struct hci_cp_remote_name_req_cancel cp; -struct inquiry_entry *e;struct hci_request req;int err;
@@ -3601,52 +3636,22 @@ static int stop_discovery(struct sock *shci_req_init (&req, hdev); -switch (hdev->discovery.state) { -case DISCOVERY_FINDING : -if (test_bit (HCI_INQUIRY , &hdev->flags)) { -hci_req_add (&req,HCI_OP_INQUIRY_CANCEL ,0 ,NULL ); - }else { -cancel_delayed_work (&hdev->le_scan_disable); - -hci_req_add_le_scan_disable (&req); - } - -break ; - -case DISCOVERY_RESOLVING : - e =hci_inquiry_cache_lookup_resolve (hdev,BDADDR_ANY , -NAME_PENDING ); -if (!e) { -mgmt_pending_remove (cmd); - err =cmd_complete (sk, hdev->id, -MGMT_OP_STOP_DISCOVERY ,0 , - &mgmt_cp->type, -sizeof (mgmt_cp->type)); -hci_discovery_set_state (hdev,DISCOVERY_STOPPED ); - goto unlock; - } - -bacpy (&cp.bdaddr, &e->data.bdaddr); -hci_req_add (&req,HCI_OP_REMOTE_NAME_REQ_CANCEL ,sizeof (cp), - &cp); - -break ; +hci_stop_discovery (&req); -default : -BT_DBG ("unknown discovery state %u" , hdev->discovery.state); - -mgmt_pending_remove (cmd); - err =cmd_complete (sk, hdev->id,MGMT_OP_STOP_DISCOVERY , -MGMT_STATUS_FAILED , &mgmt_cp->type, -sizeof (mgmt_cp->type)); + err =hci_req_run (&req, stop_discovery_complete); +if (!err) { +hci_discovery_set_state (hdev,DISCOVERY_STOPPING ); goto unlock; } - err =hci_req_run (&req, stop_discovery_complete); -if (err <0 ) -mgmt_pending_remove (cmd); -else -hci_discovery_set_state (hdev,DISCOVERY_STOPPING ); +mgmt_pending_remove (cmd); + +/* If no HCI commands were sent we're done */ +if (err == -ENODATA ) { + err =cmd_complete (sk, hdev->id,MGMT_OP_STOP_DISCOVERY ,0 , + &mgmt_cp->type,sizeof (mgmt_cp->type)); +hci_discovery_set_state (hdev,DISCOVERY_STOPPED ); + } unlock:hci_dev_unlock (hdev);
Greg Kroah-Hartman
maintainerPATCH 58/122 Bluetooth: Reuse hci_stop_discovery function when cleaning up HCI state
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hedberg <[email protected]>
commit f8680f128b01212895a9afb31032f6ffe91bd771 upstream.
When cleaning up the HCI state as part of the power-off procedure we can reuse the hci_stop_discovery() function instead of explicitly sending HCI command related to discovery. The added benefit of this is that it takes care of canceling name resolving and inquiry which were not previously covered by the code.
--- net/bluetooth/mgmt.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
net/bluetooth/mgmt.c +1 -3
@@ -1098,9 +1098,7 @@ static int clean_up_hci_state(struct hciif (test_bit (HCI_ADVERTISING , &hdev->dev_flags))disable_advertising (&req); -if (test_bit (HCI_LE_SCAN , &hdev->dev_flags)) { -hci_req_add_le_scan_disable (&req); - } +hci_stop_discovery (&req);list_for_each_entry (conn, &hdev->conn_hash.list, list) {struct hci_cp_disconnect dc;
Greg Kroah-Hartman
maintainerPATCH 59/122 Bluetooth: Fix setting correct authentication information for SMP STK
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hedberg <[email protected]>
commit fff3490f47810e2d34b91fb9e31103e923b11c2f upstream.
When we store the STK in slave role we should set the correct authentication information for it. If the pairing is producing a HIGH security level the STK is considered authenticated, and otherwise it's considered unauthenticated. This patch fixes the value passed to the hci_add_ltk() function when adding the STK on the slave side.
--- net/bluetooth/smp.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
net/bluetooth/smp.c +7 -2
@@ -533,7 +533,7 @@ static void random_work(struct work_struhci_le_start_enc (hcon, ediv, rand, stk); hcon->enc_key_size = smp->enc_key_size; }else { -u8 stk[16 ]; +u8 stk[16 ], auth;__le64 rand =0 ;__le16 ediv =0 ;
@@ -545,8 +545,13 @@ static void random_work(struct work_strumemset (stk + smp->enc_key_size,0 ,SMP_MAX_ENC_KEY_SIZE - smp->enc_key_size); +if (hcon->pending_sec_level ==BT_SECURITY_HIGH ) + auth =1 ; +else + auth =0 ; +hci_add_ltk (hcon->hdev, &hcon->dst, hcon->dst_type, -HCI_SMP_STK_SLAVE ,0 , stk, smp->enc_key_size, +HCI_SMP_STK_SLAVE , auth, stk, smp->enc_key_size, ediv, rand); }
Greg Kroah-Hartman
maintainerPATCH 60/122 Bluetooth: Fix deadlock in l2cap_conn_del()
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jukka Taimisto <[email protected]>
commit 7ab56c3a6eccb215034b0cb096e0313441cbf2a4 upstream.
A deadlock occurs when PDU containing invalid SMP opcode is received on Security Manager Channel over LE link and conn->pending_rx_work worker has not run yet.
When LE link is created l2cap_conn_ready() is called and before returning it schedules conn->pending_rx_work worker to hdev->workqueue. Incoming data to SMP fixed channel is handled by l2cap_recv_frame() which calls smp_sig_channel() to handle the SMP PDU. If smp_sig_channel() indicates failure l2cap_conn_del() is called to delete the connection. When deleting the connection, l2cap_conn_del() purges the pending_rx queue and calls flush_work() to wait for the pending_rx_work worker to complete.
Since incoming data is handled by a worker running from the same workqueue as the pending_rx_work is being scheduled on, we will deadlock on waiting for pending_rx_work to complete.
This patch fixes the deadlock by calling cancel_work_sync() instead of flush_work().
--- net/bluetooth/l2cap_core.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
net/bluetooth/l2cap_core.c +7 -1
@@ -1657,7 +1657,13 @@ static void l2cap_conn_del(struct hci_cokfree_skb (conn->rx_skb);skb_queue_purge (&conn->pending_rx); -flush_work (&conn->pending_rx_work); + +/* We can not call flush_work(&conn->pending_rx_work) here since we +* might block if we are running on a worker from the same workqueue +* pending_rx_work is waiting on. +*/ +if (work_pending (&conn->pending_rx_work)) +cancel_work_sync (&conn->pending_rx_work);l2cap_unregister_all_users (conn);
Greg Kroah-Hartman
maintainerPATCH 61/122 Bluetooth: Fix locking of hdev when calling into SMP code
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Johan Hedberg <[email protected]>
commit c73f94b8c093a615ce80eabbde0ac6eb9abfe31a upstream.
The SMP code expects hdev to be unlocked since e.g. crypto functions will try to (re)lock it. Therefore, we need to release the lock before calling into smp.c from mgmt.c. Without this we risk a deadlock whenever the smp_user_confirm_reply() function is called.
--- net/bluetooth/mgmt.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
net/bluetooth/mgmt.c +6 -1
@@ -3032,8 +3032,13 @@ static int user_pairing_resp(struct sock }if (addr->type ==BDADDR_LE_PUBLIC || addr->type ==BDADDR_LE_RANDOM ) { -/* Continue with pairing via SMP */ +/* Continue with pairing via SMP. The hdev lock must be +* released as SMP may try to recquire it for crypto +* purposes. +*/ +hci_dev_unlock (hdev); err =smp_user_confirm_reply (conn, mgmt_op, passkey); +hci_dev_lock (hdev);if (!err) err = cmd_complete(sk, hdev->id, mgmt_op,
Greg Kroah-Hartman
maintainerPATCH 62/122 Bluetooth: Allow change security level on ATT_CID in slave role
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marcin Kraglak <[email protected]>
commit 92d1372e1a9fec00e146b74e8b9ad7a385b9b37f upstream.
Kernel supports SMP Security Request so don't block increasing security when we are slave.
--- net/bluetooth/l2cap_sock.c | 5 ----- 1 file changed, 5 deletions(-)
net/bluetooth/l2cap_sock.c +0 -5
@@ -787,11 +787,6 @@ static int l2cap_sock_setsockopt(struct/*change security for LE channels */ if (chan->scid ==L2CAP_CID_ATT ) { -if (!conn->hcon->out) { - err = -EINVAL ; -break ; - } -if (smp_conn_security (conn->hcon, sec.level))break ; sk->sk_state =BT_CONFIG ;
Greg Kroah-Hartman
maintainerPATCH 63/122 dm era: check for a non-NULL metadata object before closing it
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Joe Thornber <[email protected]>
commit 989f26f5ad308f40a95f280bf9cd75e558d4f18d upstream.
era_ctr() may call era_destroy() before era->md is initialized so era_destory() must only close the metadata object if it is not NULL.
--- drivers/md/dm-era-target.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/md/dm-era-target.c +2 -1
@@ -1391,7 +1391,8 @@ static int era_is_congested(struct dm_tastatic void era_destroy (struct era *era) { -metadata_close (era->md); +if (era->md) +metadata_close (era->md);if (era->wq)destroy_workqueue (era->wq);
Greg Kroah-Hartman
maintainerPATCH 64/122 dm thin: update discard_granularity to reflect the thin-pool blocksize
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lukas Czerner <[email protected]>
commit 09869de57ed2728ae3c619803932a86cb0e2c4f8 upstream.
DM thinp already checks whether the discard_granularity of the data device is a factor of the thin-pool block size. But when using the dm-thin-pool's discard passdown support, DM thinp was not selecting the max of the underlying data device's discard_granularity and the thin-pool's block size.
Update set_discard_limits() to set discard_granularity to the max of these values. This enables blkdev_issue_discard() to properly align the discards that are sent to the DM thin device on a full block boundary. As such each discard will now cover an entire DM thin-pool block and the block will be reclaimed.
--- drivers/md/dm-thin.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/md/dm-thin.c +2 -1
@@ -3068,7 +3068,8 @@ static void set_discard_limits(struct po*/ if (pt->adjusted_pf.discard_passdown) { data_limits = &bdev_get_queue (pt->data_dev->bdev)->limits; - limits->discard_granularity = data_limits->discard_granularity; + limits->discard_granularity =max (data_limits->discard_granularity, + pool->sectors_per_block <<SECTOR_SHIFT ); }else limits->discard_granularity = pool->sectors_per_block <<SECTOR_SHIFT ; }
Greg Kroah-Hartman
maintainerPATCH 65/122 rbd: use reference counts for image requests
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alex Elder <[email protected]>
commit 0f2d5be792b0466b06797f637cfbb0f64dbb408c upstream.
Each image request contains a reference count, but to date it has not actually been used. (I think this was just an oversight.) A recent report involving rbd failing an assertion shed light on why and where we need to use these reference counts.
Every OSD request associated with an object request uses rbd_osd_req_callback() as its callback function. That function will call a helper function (dependent on the type of OSD request) that will set the object request's "done" flag if the object request if appropriate. If that "done" flag is set, the object request is passed to rbd_obj_request_complete().
In rbd_obj_request_complete(), requests are processed in sequential order. So if an object request completes before one of its predecessors in the image request, the completion is deferred. Otherwise, if it's a completing object's "turn" to be completed, it is passed to rbd_img_obj_end_request(), which records the result of the operation, accumulates transferred bytes, and so on. Next, the successor to this request is checked and if it is marked "done", (deferred) completion processing is performed on that request, and so on. If the last object request in an image request is completed, rbd_img_request_complete() is called, which (typically) destroys the image request.
There is a race here, however. The instant an object request is marked "done" it can be provided (by a thread handling completion of one of its predecessor operations) to rbd_img_obj_end_request(), which (for the last request) can then lead to the image request getting torn down. And this can happen *before* that object has itself entered rbd_img_obj_end_request(). As a result, once it *does* enter that function, the image request (and even the object request itself) may have been freed and become invalid.
All that's necessary to avoid this is to properly count references to the image requests. We tear down an image request's object requests all at once--only when the entire image request has completed. So there's no need for an image request to count references for its object requests. However, we don't want an image request to go away until the last of its object requests has passed through rbd_img_obj_callback(). In other words, we don't want rbd_img_request_complete() to necessarily result in the image request being destroyed, because it may get called before we've finished processing on all of its object requests.
So the fix is to add a reference to an image request for each of its object requests. The reference can be viewed as representing an object request that has not yet finished its call to rbd_img_obj_callback(). That is emphasized by getting the reference right after assigning that as the image object's callback function. The corresponding release of that reference is done at the end of rbd_img_obj_callback(), which every image object request passes through exactly once.
--- drivers/block/rbd.c | 9 +++++++++ 1 file changed, 9 insertions(+)
drivers/block/rbd.c +9 -0
@@ -1382,6 +1382,13 @@ static void rbd_obj_request_put(struct rkref_put (&obj_request->kref, rbd_obj_request_destroy); } +static void rbd_img_request_get (struct rbd_img_request *img_request) +{ +dout ("%s: img %p (was %d)\n" , __func__, img_request, +atomic_read (&img_request->kref.refcount)); +kref_get (&img_request->kref); +} +static bool img_request_child_test (struct rbd_img_request *img_request);static void rbd_parent_request_destroy (struct kref *kref);static void rbd_img_request_destroy (struct kref *kref);
@@ -2142,6 +2149,7 @@ static void rbd_img_obj_callback(struct img_request->next_completion = which; out:spin_unlock_irq (&img_request->completion_lock); +rbd_img_request_put (img_request);if (!more)rbd_img_request_complete (img_request);
@@ -2242,6 +2250,7 @@ static int rbd_img_request_fill(struct r goto out_unwind; obj_request->osd_req = osd_req; obj_request->callback = rbd_img_obj_callback; +rbd_img_request_get (img_request);if (write_request) { osd_req_op_alloc_hint_init(osd_req, which,
Greg Kroah-Hartman
maintainerPATCH 66/122 rbd: handle parent_overlap on writes correctly
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ilya Dryomov <[email protected]>
commit 9638556a276125553549fdfe349c464481ec2f39 upstream.
The following check in rbd_img_obj_request_submit()
rbd_dev->parent_overlap <= obj_request->img_offset
allows the fall through to the non-layered write case even if both parent_overlap and obj_request->img_offset belong to the same RADOS object. This leads to data corruption, because the area to the left of parent_overlap ends up unconditionally zero-filled instead of being populated with parent data. Suppose we want to write 1M to offset 6M of image bar, which is a clone of foo@snap; object_size is 4M, parent_overlap is 5M:
rbd_data.<id>.0000000000000001
---------------------|----------------------|------------
| should be copyup'ed | should be zeroed out | write ...
---------------------|----------------------|------------
4M 5M 6M
parent_overlap obj_request->img_offset4..5M should be copyup'ed from foo, yet it is zero-filled, just like 5..6M is.
Given that the only striping mode kernel client currently supports is chunking (i.e. stripe_unit == object_size, stripe_count == 1), round parent_overlap up to the next object boundary for the purposes of the overlap check.
--- drivers/block/rbd.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/block/rbd.c +9 -1
@@ -1366,6 +1366,14 @@ static bool obj_request_exists_test(strureturn test_bit (OBJ_REQ_EXISTS , &obj_request->flags) !=0 ; } +static bool obj_request_overlaps_parent (struct rbd_obj_request *obj_request) +{ +struct rbd_device *rbd_dev = obj_request->img_request->rbd_dev; + +return obj_request->img_offset < +round_up (rbd_dev->parent_overlap,rbd_obj_bytes (&rbd_dev->header)); +} +static void rbd_obj_request_get (struct rbd_obj_request *obj_request) { dout("%s: obj %p (was %d)\n", __func__, obj_request,
@@ -2683,7 +2691,7 @@ static int rbd_img_obj_request_submit(st*/ if (!img_request_write_test (img_request) || !img_request_layered_test (img_request) || - rbd_dev->parent_overlap <= obj_request->img_offset || + !obj_request_overlaps_parent (obj_request) || ((known =obj_request_known_test (obj_request)) &&obj_request_exists_test (obj_request))) {
Greg Kroah-Hartman
maintainerPATCH 67/122 hwmon: (ina2xx) Cast to s16 on shunt and current regs
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Fabio Baltieri <[email protected]>
commit c0214f98943b1fe43f7be61b7782b0c8f0836f28 upstream.
All devices supported by ina2xx are bidirectional and report the measured shunt voltage and power values as a signed 16 bit, but the current driver implementation caches all registers as u16, leading to an incorrect sign extension when reporting to userspace in ina2xx_get_value().
This patch fixes the problem by casting the signed registers to s16. Tested on an INA219.
--- drivers/hwmon/ina2xx.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/hwmon/ina2xx.c +4 -3
@@ -148,7 +148,8 @@ static int ina2xx_get_value(struct ina2xswitch (reg) {case INA2XX_SHUNT_VOLTAGE : - val =DIV_ROUND_CLOSEST (data->regs[reg], +/* signed register */ + val =DIV_ROUND_CLOSEST ((s16 )data->regs[reg], data->config->shunt_div);break ;case INA2XX_BUS_VOLTAGE :
@@ -160,8 +161,8 @@ static int ina2xx_get_value(struct ina2x val = data->regs[reg] * data->config->power_lsb;break ;case INA2XX_CURRENT : -/* LSB=1mA (selected). Is in mA */ - val = data->regs[reg]; +/* signed register, LSB=1mA (selected), in mA */ + val = (s16 )data->regs[reg];break ;default :/* programmer goofed */
Greg Kroah-Hartman
maintainerPATCH 68/122 intel_pstate: Correct rounding in busy calculation
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Doug Smythies <[email protected]>
commit 51d211e9c334b9eca3505f4052afa660c3e0606b upstream.
There was a mistake in the actual rounding portion this previous patch:f0fe3cd7e12d (intel_pstate: Correct rounding in busy calculation) such that the rounding was asymetric and incorrect.
Severity: Not very serious, but can increase target pstate by one extra value. For real world work flows the issue should self correct (but I have no proof). It is the equivalent of different PID gains for positive and negative numbers.
Examples:
-3.000000 used to round to -4, rounds to -3 with this patch. -3.503906 used to round to -5, rounds to -4 with this patch.
--- drivers/cpufreq/intel_pstate.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-)
drivers/cpufreq/intel_pstate.c +1 -4
@@ -200,10 +200,7 @@ static signed int pid_calc(struct _pid * pid->last_err = fp_error; result = pterm +mul_fp (pid->integral, pid->i_gain) + dterm; -if (result >=0 ) - result = result + (1 << (FRAC_BITS -1 )); -else - result = result - (1 << (FRAC_BITS -1 )); + result = result + (1 << (FRAC_BITS -1 ));return (signed int )fp_toint (result); }
Greg Kroah-Hartman
maintainerPATCH 69/122 twl4030-madc: Request processed values in twl4030_get_madc_conversion
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Paul Kocialkowski <[email protected]>
commit e0326be0cded13dfc3a24cbeece1f1ae64348a0e upstream.
Not setting the raw parameter in the request causes it to be randomly initialized to a value that might be different from zero or zero. This leads to values that are randomly either raw or processed, making it very difficult to make reliable use of the values.
--- drivers/iio/adc/twl4030-madc.c | 1 + 1 file changed, 1 insertion(+)
drivers/iio/adc/twl4030-madc.c +1 -0
@@ -645,6 +645,7 @@ int twl4030_get_madc_conversion(int chan req.channels = (1 << channel_no); req.method =TWL4030_MADC_SW2 ; req.active =0 ; + req.raw =0 ; req.func_cb =NULL ; ret =twl4030_madc_conversion (&req);if (ret <0 )
Greg Kroah-Hartman
maintainerPATCH 71/122 mac80211: dont check netdev state for debugfs read/write
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arik Nemtsov <[email protected]>
commit 923eaf367206e01f22c97aee22300e332d071916 upstream.
Doing so will lead to an oops for a p2p-dev interface, since it has no netdev.
--- net/mac80211/debugfs_netdev.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
net/mac80211/debugfs_netdev.c +2 -4
@@ -34,8 +34,7 @@ static ssize_t ieee80211_if_read(ssize_t ret = -EINVAL ;read_lock (&dev_base_lock); -if (sdata->dev->reg_state ==NETREG_REGISTERED ) - ret = (*format)(sdata, buf,sizeof (buf)); + ret = (*format)(sdata, buf,sizeof (buf));read_unlock (&dev_base_lock);if (ret >=0 )
@@ -62,8 +61,7 @@ static ssize_t ieee80211_if_write( ret = -ENODEV ;rtnl_lock (); -if (sdata->dev->reg_state ==NETREG_REGISTERED ) - ret = (*write)(sdata, buf, count); + ret = (*write)(sdata, buf, count);rtnl_unlock ();return ret;
Greg Kroah-Hartman
maintainerPATCH 73/122 iwlwifi: pcie: try to get ownership several times
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Emmanuel Grumbach <[email protected]>
commit 501fd9895c1d7d8161ed56698ae2fccb10ef14f5 upstream.
Some races with the hardware can happen when we take ownership of the device. Don't give up after the first try.
--- drivers/net/wireless/iwlwifi/pcie/trans.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-)
drivers/net/wireless/iwlwifi/pcie/trans.c +18 -12
@@ -454,6 +454,7 @@ static int iwl_pcie_prepare_card_hw(stru {int ret;int t =0 ; +int iter;IWL_DEBUG_INFO (trans,"iwl_trans_prepare_card_hw enter\n" );
@@ -462,18 +463,23 @@ static int iwl_pcie_prepare_card_hw(struif (ret >=0 )return 0 ; -/* If HW is not ready, prepare the conditions to check again */ -iwl_set_bit (trans,CSR_HW_IF_CONFIG_REG , -CSR_HW_IF_CONFIG_REG_PREPARE ); - -do { - ret =iwl_pcie_set_hw_ready (trans); -if (ret >=0 ) -return 0 ; - -usleep_range (200 ,1000 ); - t +=200 ; - }while (t <150000 ); +for (iter =0 ; iter <10 ; iter++) { +/* If HW is not ready, prepare the conditions to check again */ +iwl_set_bit (trans,CSR_HW_IF_CONFIG_REG , +CSR_HW_IF_CONFIG_REG_PREPARE ); + +do { + ret =iwl_pcie_set_hw_ready (trans); +if (ret >=0 ) +return 0 ; + +usleep_range (200 ,1000 ); + t +=200 ; + }while (t <150000 ); +msleep (25 ); + } + +IWL_DEBUG_INFO (trans,"got NIC after %d iterations\n" , iter);return ret; }
Greg Kroah-Hartman
maintainerPATCH 74/122 mm: nommu: per-thread vma cache fix
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Miao <[email protected]>
commit e020d5bd8a730757b565b18d620240f71c3e21fe upstream.
mm could be removed from current task struct, using previous vma->vm_mm
It will crash on blackfin after updated to Linux 3.15. The commit "mm:
per-thread vma caching" caused the crash. mm could be removed from current task struct before
mmput()->
exit_mmap()->
delete_vma_from_mm()the detailed fault information:
NULL pointer access
Kernel OOPS in progress
Deferred Exception context
CURRENT PROCESS:
COMM=modprobe PID=278 CPU=0
invalid mm
return address: [0x000531de]; contents of:
0x000531b0: c727 acea 0c42 181d 0000 0000 0000 a0a8
0x000531c0: b090 acaa 0c42 1806 0000 0000 0000 a0e8
0x000531d0: b0d0 e801 0000 05b3 0010 e522 0046 [a090]
0x000531e0: 6408 b090 0c00 17cc 3042 e3ff f37b 2fc8
CPU: 0 PID: 278 Comm: modprobe Not tainted 3.15.0-ADI-2014R1-pre-00345-gea9f446 #25
task: 0572b720 ti: 0569e000 task.ti: 0569e000
Compiled for cpu family 0x27fe (Rev 0), but running on:0x0000 (Rev 0)
ADSP-BF609-0.0 500(MHz CCLK) 125(MHz SCLK) (mpu off)
Linux version 3.15.0-ADI-2014R1-pre-00345-gea9f446 (steven@steven-OptiPlex-390) (gcc version 4.3.5 (ADI-trunk/svn-5962) ) #25 Tue Jun 10 17:47:46 CST 2014
SEQUENCER STATUS: Not tainted
SEQSTAT: 00000027 IPEND: 8008 IMASK: ffff SYSCFG: 2806
EXCAUSE : 0x27
physical IVG3 asserted : <0xffa00744> { _trap + 0x0 }
physical IVG15 asserted : <0xffa00d68> { _evt_system_call + 0x0 }
logical irq 6 mapped : <0xffa003bc> { _bfin_coretmr_interrupt + 0x0 }
logical irq 7 mapped : <0x00008828> { _bfin_fault_routine + 0x0 }
logical irq 11 mapped : <0x00007724> { _l2_ecc_err + 0x0 }
logical irq 13 mapped : <0x00008828> { _bfin_fault_routine + 0x0 }
logical irq 39 mapped : <0x00150788> { _bfin_twi_interrupt_entry + 0x0 }
logical irq 40 mapped : <0x00150788> { _bfin_twi_interrupt_entry + 0x0 }
RETE: <0x00000000> /* Maybe null pointer? */
RETN: <0x0569fe50> /* kernel dynamic memory (maybe user-space) */
RETX: <0x00000480> /* Maybe fixed code section */
RETS: <0x00053384> { _exit_mmap + 0x28 }
PC : <0x000531de> { _delete_vma_from_mm + 0x92 }
DCPLB_FAULT_ADDR: <0x00000008> /* Maybe null pointer? */
ICPLB_FAULT_ADDR: <0x000531de> { _delete_vma_from_mm + 0x92 }
PROCESSOR STATE:
R0 : 00000004 R1 : 0569e000 R2 : 00bf3db4 R3 : 00000000
R4 : 057f9800 R5 : 00000001 R6 : 0569ddd0 R7 : 0572b720
P0 : 0572b854 P1 : 00000004 P2 : 00000000 P3 : 0569dda0
P4 : 0572b720 P5 : 0566c368 FP : 0569fe5c SP : 0569fd74
LB0: 057f523f LT0: 057f523e LC0: 00000000
LB1: 0005317c LT1: 00053172 LC1: 00000002
B0 : 00000000 L0 : 00000000 M0 : 0566f5bc I0 : 00000000
B1 : 00000000 L1 : 00000000 M1 : 00000000 I1 : ffffffff
B2 : 00000001 L2 : 00000000 M2 : 00000000 I2 : 00000000
B3 : 00000000 L3 : 00000000 M3 : 00000000 I3 : 057f8000
A0.w: 00000000 A0.x: 00000000 A1.w: 00000000 A1.x: 00000000
USP : 056ffcf8 ASTAT: 02003024
Hardware Trace:
0 Target : <0x00003fb8> { _trap_c + 0x0 }
Source : <0xffa006d8> { _exception_to_level5 + 0xa0 } JUMP.L
1 Target : <0xffa00638> { _exception_to_level5 + 0x0 }
Source : <0xffa004f2> { _bfin_return_from_exception + 0x6 } RTX
2 Target : <0xffa004ec> { _bfin_return_from_exception + 0x0 }
Source : <0xffa00590> { _ex_trap_c + 0x70 } JUMP.S
3 Target : <0xffa00520> { _ex_trap_c + 0x0 }
Source : <0xffa0076e> { _trap + 0x2a } JUMP (P4)
4 Target : <0xffa00744> { _trap + 0x0 }
FAULT : <0x000531de> { _delete_vma_from_mm + 0x92 } P0 = W[P2 + 2]
Source : <0x000531da> { _delete_vma_from_mm + 0x8e } P2 = [P4 + 0x18]
5 Target : <0x000531da> { _delete_vma_from_mm + 0x8e }
Source : <0x00053176> { _delete_vma_from_mm + 0x2a } IF CC JUMP pcrel
6 Target : <0x0005314c> { _delete_vma_from_mm + 0x0 }
Source : <0x00053380> { _exit_mmap + 0x24 } JUMP.L
7 Target : <0x00053378> { _exit_mmap + 0x1c }
Source : <0x00053394> { _exit_mmap + 0x38 } IF !CC JUMP pcrel (BP)
8 Target : <0x00053390> { _exit_mmap + 0x34 }
Source : <0xffa020e0> { __cond_resched + 0x20 } RTS
9 Target : <0xffa020c0> { __cond_resched + 0x0 }
Source : <0x0005338c> { _exit_mmap + 0x30 } JUMP.L
10 Target : <0x0005338c> { _exit_mmap + 0x30 }
Source : <0x0005333a> { _delete_vma + 0xb2 } RTS
11 Target : <0x00053334> { _delete_vma + 0xac }
Source : <0x0005507a> { _kmem_cache_free + 0xba } RTS
12 Target : <0x00055068> { _kmem_cache_free + 0xa8 }
Source : <0x0005505e> { _kmem_cache_free + 0x9e } IF !CC JUMP pcrel (BP)
13 Target : <0x00055052> { _kmem_cache_free + 0x92 }
Source : <0x0005501a> { _kmem_cache_free + 0x5a } IF CC JUMP pcrel
14 Target : <0x00054ff4> { _kmem_cache_free + 0x34 }
Source : <0x00054fce> { _kmem_cache_free + 0xe } IF CC JUMP pcrel (BP)
15 Target : <0x00054fc0> { _kmem_cache_free + 0x0 }
Source : <0x00053330> { _delete_vma + 0xa8 } JUMP.L
Kernel Stack
Stack info:
SP: [0x0569ff24] <0x0569ff24> /* kernel dynamic memory (maybe user-space) */
Memory from 0x0569ff20 to 056a0000
0569ff20: 00000001 [04e8da5a] 00008000 00000000 00000000 056a0000 04e8da5a 04e8da5a
0569ff40: 04eb9eea ffa00dce 02003025 04ea09c5 057f523f 04ea09c4 057f523e 00000000
0569ff60: 00000000 00000000 00000000 00000000 00000000 00000000 00000001 00000000
0569ff80: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
0569ffa0: 0566f5bc 057f8000 057f8000 00000001 04ec0170 056ffcf8 056ffd04 057f9800
0569ffc0: 04d1d498 057f9800 057f8fe4 057f8ef0 00000001 057f928c 00000001 00000001
0569ffe0: 057f9800 00000000 00000008 00000007 00000001 00000001 00000001 <00002806>
Return addresses in stack:
address : <0x00002806> { _show_cpuinfo + 0x2d2 }
Modules linked in:
Kernel panic - not syncing: Kernel exception
[ end Kernel panic - not syncing: Kernel exception--- mm/nommu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
mm/nommu.c +1 -1
@@ -783,7 +783,7 @@ static void delete_vma_from_mm(struct vmfor (i =0 ; i <VMACACHE_SIZE ; i++) {/* if the vma is cached, invalidate the entire cache */ if (curr->vmacache[i] == vma) { -vmacache_invalidate (curr->mm); +vmacache_invalidate (mm);break ; } }
Greg Kroah-Hartman
maintainerPATCH 75/122 hugetlb: fix copy_hugetlb_page_range() to handle migration/hwpoisoned entry
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Naoya Horiguchi <[email protected]>
commit 4a705fef986231a3e7a6b1a6d3c37025f021f49f upstream.
There's a race between fork() and hugepage migration, as a result we try to "dereference" a swap entry as a normal pte, causing kernel panic. The cause of the problem is that copy_hugetlb_page_range() can't handle "swap entry" family (migration entry and hwpoisoned entry) so let's fix it.
--- mm/hugetlb.c | 71 +++++++++++++++++++++++++++++++++++------------------------ 1 file changed, 43 insertions(+), 28 deletions(-)
mm/hugetlb.c +43 -28
@@ -2377,6 +2377,31 @@ static void set_huge_ptep_writable(strucupdate_mmu_cache (vma, address, ptep); } +static int is_hugetlb_entry_migration (pte_t pte) +{ +swp_entry_t swp; + +if (huge_pte_none (pte) ||pte_present (pte)) +return 0 ; + swp =pte_to_swp_entry (pte); +if (non_swap_entry (swp) &&is_migration_entry (swp)) +return 1 ; +else +return 0 ; +} + +static int is_hugetlb_entry_hwpoisoned (pte_t pte) +{ +swp_entry_t swp; + +if (huge_pte_none (pte) ||pte_present (pte)) +return 0 ; + swp =pte_to_swp_entry (pte); +if (non_swap_entry (swp) &&is_hwpoison_entry (swp)) +return 1 ; +else +return 0 ; +}int copy_hugetlb_page_range (struct mm_struct *dst,struct mm_struct *src,struct vm_area_struct *vma)
@@ -2416,10 +2441,26 @@ int copy_hugetlb_page_range(struct mm_st dst_ptl =huge_pte_lock (h, dst, dst_pte); src_ptl =huge_pte_lockptr (h, src, src_pte);spin_lock_nested (src_ptl,SINGLE_DEPTH_NESTING ); -if (!huge_pte_none (huge_ptep_get (src_pte))) { + entry =huge_ptep_get (src_pte); +if (huge_pte_none (entry)) {/* skip none entry */ + ; + }else if (unlikely (is_hugetlb_entry_migration (entry) || +is_hugetlb_entry_hwpoisoned (entry))) { +swp_entry_t swp_entry =pte_to_swp_entry (entry); + +if (is_write_migration_entry (swp_entry) && cow) { +/* +* COW mappings require pages in both +* parent and child to be set to read. +*/ +make_migration_entry_read (&swp_entry); + entry =swp_entry_to_pte (swp_entry); +set_huge_pte_at (src, addr, src_pte, entry); + } +set_huge_pte_at (dst, addr, dst_pte, entry); + }else {if (cow)huge_ptep_set_wrprotect (src, addr, src_pte); - entry =huge_ptep_get (src_pte); ptepage =pte_page (entry);get_page (ptepage);page_dup_rmap (ptepage);
@@ -2435,32 +2476,6 @@ int copy_hugetlb_page_range(struct mm_streturn ret; } -static int is_hugetlb_entry_migration (pte_t pte) -{ -swp_entry_t swp; - -if (huge_pte_none (pte) ||pte_present (pte)) -return 0 ; - swp =pte_to_swp_entry (pte); -if (non_swap_entry (swp) &&is_migration_entry (swp)) -return 1 ; -else -return 0 ; -} - -static int is_hugetlb_entry_hwpoisoned (pte_t pte) -{ -swp_entry_t swp; - -if (huge_pte_none (pte) ||pte_present (pte)) -return 0 ; - swp =pte_to_swp_entry (pte); -if (non_swap_entry (swp) &&is_hwpoison_entry (swp)) -return 1 ; -else -return 0 ; -} -void __unmap_hugepage_range (struct mmu_gather *tlb,struct vm_area_struct *vma,unsigned long start,unsigned long end,struct page *ref_page)
Greg Kroah-Hartman
maintainerPATCH 76/122 mm, pcp: allow restoring percpu_pagelist_fraction default
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Rientjes <[email protected]>
commit 7cd2b0a34ab8e4db971920eef8982f985441adfb upstream.
Oleg reports a division by zero error on zero-length write() to the percpu_pagelist_fraction sysctl:
divide error: 0000 [#1] SMP DEBUG_PAGEALLOC
CPU: 1 PID: 9142 Comm: badarea_io Not tainted 3.15.0-rc2-vm-nfs+ #19
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
task: ffff8800d5aeb6e0 ti: ffff8800d87a2000 task.ti: ffff8800d87a2000
RIP: 0010: percpu_pagelist_fraction_sysctl_handler+0x84/0x120
RSP: 0018:ffff8800d87a3e78 EFLAGS: 00010246
RAX: 0000000000000f89 RBX: ffff88011f7fd000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000001 RDI: 0000000000000010
RBP: ffff8800d87a3e98 R08: ffffffff81d002c8 R09: ffff8800d87a3f50
R10: 000000000000000b R11: 0000000000000246 R12: 0000000000000060
R13: ffffffff81c3c3e0 R14: ffffffff81cfddf8 R15: ffff8801193b0800
FS: 00007f614f1e9740(0000) GS:ffff88011f440000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 00007f614f1fa000 CR3: 00000000d9291000 CR4: 00000000000006e0
Call Trace:
proc_sys_call_handler+0xb3/0xc0
proc_sys_write+0x14/0x20
vfs_write+0xba/0x1e0
SyS_write+0x46/0xb0
tracesys+0xe1/0xe6However, if the percpu_pagelist_fraction sysctl is set by the user, it is also impossible to restore it to the kernel default since the user cannot write 0 to the sysctl.
This patch allows the user to write 0 to restore the default behavior. It still requires a fraction equal to or larger than 8, however, as stated by the documentation for sanity. If a value in the range [1, 7] is written, the sysctl will return EINVAL.
This successfully solves the divide by zero issue at the same time.
--- Documentation/sysctl/vm.txt | 3 ++- kernel/sysctl.c | 3 +-- mm/page_alloc.c | 40 ++++++++++++++++++++++++++++------------ 3 files changed, 31 insertions(+), 15 deletions(-)
Documentation/sysctl/vm.txt +2 -1
@@ -702,7 +702,8 @@ The batch value of each per cpu pagelist set to pcp->high/4. The upper limit of batch is (PAGE_SHIFT * 8) The initial value is zero. Kernel does not use this value at boot time to set -the high water marks for each per cpu page list. +the high water marks for each per cpu page list. If the user writes '0' to this +sysctl, it will revert to this default behavior. ==============================================================
kernel/sysctl.c +1 -2
@@ -136,7 +136,6 @@ static unsigned long dirty_bytes_min = 2/* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */ static int maxolduid =65535 ;static int minolduid; -static int min_percpu_pagelist_fract =8 ;static int ngroups_max =NGROUPS_MAX ;static const int cap_last_cap =CAP_LAST_CAP ;
@@ -1305,7 +1304,7 @@ static struct ctl_table vm_table[] = { .maxlen = sizeof(percpu_pagelist_fraction), .mode = 0644, .proc_handler = percpu_pagelist_fraction_sysctl_handler, - .extra1 = &min_percpu_pagelist_fract, + .extra1 = &zero, },#ifdef CONFIG_MMU {
mm/page_alloc.c +28 -12
@@ -69,6 +69,7 @@/* prevent >1 _updater_ of zone percpu pageset ->high and ->batch fields */ static DEFINE_MUTEX (pcp_batch_high_lock); +#define MIN_PERCPU_PAGELIST_FRACTION (8)#ifdef CONFIG_USE_PERCPU_NUMA_NODE_ID DEFINE_PER_CPU (int, numa_node);
@@ -4107,7 +4108,7 @@ static void __meminit zone_init_free_lis memmap_init_zone((size), (nid), (zone), (start_pfn), MEMMAP_EARLY)#endif -static int __meminit zone_batchsize (struct zone *zone) +static int zone_batchsize (struct zone *zone) {#ifdef CONFIG_MMU int batch;
@@ -4223,8 +4224,8 @@ static void pageset_set_high(struct per_pageset_update (&p->pcp, high, batch); } -static void __meminit pageset_set_high_and_batch (struct zone *zone, -struct per_cpu_pageset *pcp) +static void pageset_set_high_and_batch (struct zone *zone, +struct per_cpu_pageset *pcp) {if (percpu_pagelist_fraction) pageset_set_high(pcp,
@@ -5850,23 +5851,38 @@ int percpu_pagelist_fraction_sysctl_hand void __user *buffer, size_t *length, loff_t *ppos) { struct zone *zone; -unsigned int cpu; +int old_percpu_pagelist_fraction;int ret; +mutex_lock (&pcp_batch_high_lock); + old_percpu_pagelist_fraction = percpu_pagelist_fraction; + ret =proc_dointvec_minmax (table, write, buffer, length, ppos); -if (!write || (ret <0 )) -return ret; +if (!write || ret <0 ) + goto out; + +/* Sanity checking to avoid pcp imbalance */ +if (percpu_pagelist_fraction && + percpu_pagelist_fraction <MIN_PERCPU_PAGELIST_FRACTION ) { + percpu_pagelist_fraction = old_percpu_pagelist_fraction; + ret = -EINVAL ; + goto out; + } + +/* No change? */ +if (percpu_pagelist_fraction == old_percpu_pagelist_fraction) + goto out; -mutex_lock (&pcp_batch_high_lock);for_each_populated_zone (zone) { -unsigned long high; - high = zone->managed_pages / percpu_pagelist_fraction; +unsigned int cpu; +for_each_possible_cpu (cpu) -pageset_set_high (per_cpu_ptr (zone->pageset, cpu), - high); +pageset_set_high_and_batch (zone, +per_cpu_ptr (zone->pageset, cpu)); } +out:mutex_unlock (&pcp_batch_high_lock); -return 0 ; +return ret; }int hashdist =HASHDIST_DEFAULT ;
Greg Kroah-Hartman
maintainerPATCH 77/122 ia64: arch/ia64/include/uapi/asm/fcntl.h needs personality.h
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Andrew Morton <[email protected]>
commit f9af420fc8208d3add2fe3198dc5d8215f5a81ba upstream.
fs/notify/fanotify/fanotify_user.c: In function 'SYSC_fanotify_init':fs/notify/fanotify/fanotify_user.c:726: error: implicit declaration of function 'personality' fs/notify/fanotify/fanotify_user.c:726: error: 'PER_LINUX32' undeclared (first use in this function) fs/notify/fanotify/fanotify_user.c:726: error: (Each undeclared identifier is reported only once fs/notify/fanotify/fanotify_user.c:726: error: for each function it appears in.)
--- arch/ia64/include/uapi/asm/fcntl.h | 1 + 1 file changed, 1 insertion(+)
arch/ia64/include/uapi/asm/fcntl.h +1 -0
@@ -8,6 +8,7 @@#define force_o_largefile () \ (personality(current->personality) != PER_LINUX32) +#include <linux/personality.h> #include <asm-generic/fcntl.h> #endif /* _ASM_IA64_FCNTL_H */
Greg Kroah-Hartman
maintainerPATCH 78/122 ARM: mvebu: Fix the improper use of the compatible string armada38x using a wildcard
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gregory CLEMENT <[email protected]>
commit 8dbdb8e704db34085f5978c335c10256b0fb9629 upstream.
Wildcards in compatible strings should be avoid. "marvell,armada38x" was recently introduced but was not yet used.
The armada 385 SoC is a superset of the armada 380 SoC (with more CPUs and more PCIe slots). So this patch replaces the use of "marvell,armada38x" by the "marvell,armada380" string.
--- Documentation/devicetree/bindings/arm/armada-38x.txt | 14 ++++++++++++-- arch/arm/boot/dts/armada-380.dtsi | 2 +- arch/arm/boot/dts/armada-385-db.dts | 2 +- arch/arm/boot/dts/armada-385-rd.dts | 2 +- arch/arm/boot/dts/armada-385.dtsi | 2 +- arch/arm/boot/dts/armada-38x.dtsi | 2 +- 6 files changed, 17 insertions(+), 7 deletions(-)
Documentation/devicetree/bindings/arm/armada-38x.txt +12 -2
@@ -6,5 +6,15 @@ following property: Required root node property: - - compatible: must contain either "marvell,armada380" or - "marvell,armada385" depending on the variant of the SoC being used. + - compatible: must contain "marvell,armada380" + +In addition, boards using the Marvell Armada 385 SoC shall have the +following property before the previous one: + +Required root node property: + +compatible: must contain "marvell,armada385" + +Example: + +compatible = "marvell,a385-rd", "marvell,armada385", "marvell,armada380";
arch/arm/boot/dts/armada-380.dtsi +1 -1
@@ -16,7 +16,7 @@ / { model = "Marvell Armada 380 family SoC"; - compatible = "marvell,armada380", "marvell,armada38x"; + compatible = "marvell,armada380"; cpus { #address-cells = <1>;
arch/arm/boot/dts/armada-385-db.dts +1 -1
@@ -16,7 +16,7 @@ / { model = "Marvell Armada 385 Development Board"; - compatible = "marvell,a385-db", "marvell,armada385", "marvell,armada38x"; + compatible = "marvell,a385-db", "marvell,armada385", "marvell,armada380"; chosen { bootargs = "console=ttyS0,115200 earlyprintk";
arch/arm/boot/dts/armada-385-rd.dts +1 -1
@@ -17,7 +17,7 @@ / { model = "Marvell Armada 385 Reference Design"; - compatible = "marvell,a385-rd", "marvell,armada385", "marvell,armada38x"; + compatible = "marvell,a385-rd", "marvell,armada385", "marvell,armada380"; chosen { bootargs = "console=ttyS0,115200 earlyprintk";
arch/arm/boot/dts/armada-385.dtsi +1 -1
@@ -16,7 +16,7 @@ / { model = "Marvell Armada 385 family SoC"; - compatible = "marvell,armada385", "marvell,armada38x"; + compatible = "marvell,armada385", "marvell,armada380"; cpus { #address-cells = <1>;
arch/arm/boot/dts/armada-38x.dtsi +1 -1
@@ -20,7 +20,7 @@ / { model = "Marvell Armada 38x family SoC"; - compatible = "marvell,armada38x"; + compatible = "marvell,armada380"; aliases { gpio0 = &gpio0;
Greg Kroah-Hartman
maintainerPATCH 79/122 arm64: mm: Make icache synchronisation logic huge page aware
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steve Capper <[email protected]>
commit 923b8f5044da753e4985ab15c1374ced2cdf616c upstream.
The __sync_icache_dcache routine will only flush the dcache for the first page of a compound page, potentially leading to stale icache data residing further on in a hugetlb page.
This patch addresses this issue by taking into consideration the order of the page when flushing the dcache.
--- arch/arm64/mm/flush.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
arch/arm64/mm/flush.c +2 -1
@@ -79,7 +79,8 @@ void __sync_icache_dcache(pte_t pte, unsreturn ;if (!test_and_set_bit (PG_dcache_clean, &page->flags)) { -__flush_dcache_area (page_address (page),PAGE_SIZE ); +__flush_dcache_area (page_address (page), +PAGE_SIZE <<compound_order (page));__flush_icache_all (); }else if (icache_is_aivivt ()) {__flush_icache_all ();
Greg Kroah-Hartman
maintainerPATCH 80/122 ARM: OMAP2+: Fix parser-bug in platform muxing code
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: "David R. Piegdon" <[email protected]>
commit c021f241f4fab2bb4fc4120a38a828a03dd3f970 upstream.
Fix a parser-bug in the omap2 muxing code where muxtable-entries will be wrongly selected if the requested muxname is a *prefix* of their m0-entry and they have a matching mN-entry. Fix by additionally checking that the length of the m0_entry is equal.
For example muxing of "dss_data2.dss_data2" on omap32xx will fail because the prefix "dss_data2" will match the mux-entries "dss_data2" as well as "dss_data20", with the suffix "dss_data2" matching m0 (for dss_data2) and m4 (for dss_data20). Thus both are recognized as signal path candidates:
Relevant muxentries from mux34xx.c:
_OMAP3_MUXENTRY(DSS_DATA20, 90,
"dss_data20", NULL, "mcspi3_somi", "dss_data2",
"gpio_90", NULL, NULL, "safe_mode"),
_OMAP3_MUXENTRY(DSS_DATA2, 72,
"dss_data2", NULL, NULL, NULL,
"gpio_72", NULL, NULL, "safe_mode"),This will result in a failure to mux the pin at all:
_omap_mux_get_by_name: Multiple signal paths (2) for dss_data2.dss_data2
Patch should apply to linus' latest master down to rather old linux-2.6 trees.
--- arch/arm/mach-omap2/mux.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
arch/arm/mach-omap2/mux.c +4 -2
@@ -183,8 +183,10 @@ static int __init _omap_mux_get_by_name( m0_entry = mux->muxnames[0 ];/* First check for full name in mode0.muxmode format */ -if (mode0_len &&strncmp (muxname, m0_entry, mode0_len)) -continue ; +if (mode0_len) +if (strncmp (muxname, m0_entry, mode0_len) || + (strlen (m0_entry) != mode0_len)) +continue ;/* Then check for muxmode only */ for (i =0 ; i <OMAP_MUX_NR_MODES ; i++) {
Greg Kroah-Hartman
maintainerPATCH 82/122 net: allwinner: emac: Add missing free_irq
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Maxime Ripard <[email protected]>
commit b91113282bf44df46aba374a0b8f88a75bfd4b3f upstream.
If the mdio probe function fails in emac_open, the interrupt we just requested isn't freed. If emac_open is called again, for example because we try to set up the interface again, the kernel will oops because the interrupt wasn't properly released.
--- drivers/net/ethernet/allwinner/sun4i-emac.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/ethernet/allwinner/sun4i-emac.c +1 -0
@@ -736,6 +736,7 @@ static int emac_open(struct net_device * ret =emac_mdio_probe (dev);if (ret <0 ) { +free_irq (dev->irq, dev);netdev_err (dev,"cannot probe MDIO bus\n" );return ret; }
Greg Kroah-Hartman
maintainerPATCH 85/122 CIFS: fix mount failure with broken pathnames when smb3 mount with mapchars option
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steve French <[email protected]>
commit ce36d9ab3bab06b7b5522f5c8b68fac231b76ffb upstream.
When we SMB3 mounted with mapchars (to allow reserved characters : \ / > < * ? via the Unicode Windows to POSIX remap range) empty paths
(eg when we open "" to query the root of the SMB3 directory on mount) were not null terminated so we sent garbarge as a path name on empty paths which caused SMB2/SMB2.1/SMB3 mounts to fail when mapchars was specified. mapchars is particularly important since Unix Extensions for SMB3 are not supported (yet)
--- fs/cifs/cifs_unicode.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
fs/cifs/cifs_unicode.c +4 -3
@@ -290,7 +290,8 @@ int cifsConvertToUTF16(__le16 *target, const char *source, int srclen, const struct nls_table *cp, int mapChars) { -int i, j, charlen; +int i, charlen; +int j =0 ;char src_char;__le16 dst_char;wchar_t tmp;
@@ -298,12 +299,11 @@ cifsConvertToUTF16(__le16 *target, constif (!mapChars)return cifs_strtoUTF16 (target, source,PATH_MAX , cp); -for (i =0 , j =0 ; i < srclen; j++) { +for (i =0 ; i < srclen; j++) { src_char = source[i]; charlen =1 ;switch (src_char) {case 0 : -put_unaligned (0 , &target[j]); goto ctoUTF16_out;case ':' : dst_char =cpu_to_le16 (UNI_COLON );
@@ -350,6 +350,7 @@ cifsConvertToUTF16(__le16 *target, const } ctoUTF16_out: +put_unaligned (0 , &target[j]);/* Null terminate target unicode string */ return j; }
Greg Kroah-Hartman
maintainerPATCH 86/122 regulator: tps65218: Add the missing of_node assignment in probe
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Keerthy <[email protected]>
commit d2fa87c3af0df7ed10463afc588affdab954fa92 upstream.
Add the missing of_node assignment in probe.
--- drivers/regulator/tps65218-regulator.c | 1 + 1 file changed, 1 insertion(+)
drivers/regulator/tps65218-regulator.c +1 -0
@@ -257,6 +257,7 @@ static int tps65218_regulator_probe(stru config.init_data = init_data; config.driver_data = tps; config.regmap = tps->regmap; + config.of_node = pdev->dev.of_node; rdev =devm_regulator_register (&pdev->dev, ®ulators[id], &config);if (IS_ERR (rdev)) {
Greg Kroah-Hartman
maintainerPATCH 87/122 regulator: tps65218: Correct the the config register for LDO1
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Keerthy <[email protected]>
commit 0eada6a1fc85a98ce69a199e46925abd6a7001c2 upstream.
Correct the the config register for LDO1.
TPS65218 PMIC)
--- drivers/regulator/tps65218-regulator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/regulator/tps65218-regulator.c +1 -1
@@ -226,7 +226,7 @@ static const struct regulator_desc regul 1, -1, -1, TPS65218_REG_ENABLE1, TPS65218_ENABLE1_DC6_EN, NULL, NULL, 0),TPS65218_REGULATOR ("LDO1" ,TPS65218_LDO_1 , tps65218_ldo1_dcdc34_ops,64 , -TPS65218_REG_CONTROL_DCDC4 , +TPS65218_REG_CONTROL_LDO1 ,TPS65218_CONTROL_LDO1_MASK ,TPS65218_REG_ENABLE2 ,TPS65218_ENABLE2_LDO1_EN ,NULL , ldo1_dcdc3_ranges, 2),
Greg Kroah-Hartman
maintainerPATCH 88/122 iommu/vt-d: fix bug in handling multiple RMRRs for the same PCI device
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jiang Liu <[email protected]>
commit 27e249501ca06a3010519c306206cc402b61b5ab upstream.
Function dmar_iommu_notify_scope_dev() makes a wrong assumption that there's one RMRR for each PCI device at most, which causes DMA failure on some HP platforms. So enhance dmar_iommu_notify_scope_dev() to handle multiple RMRRs for the same PCI device.
Fixbug: https://bugzilla.novell.com/show_bug.cgi?id=879482
--- drivers/iommu/intel-iommu.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-)
drivers/iommu/intel-iommu.c +3 -6
@@ -3799,14 +3799,11 @@ int dmar_iommu_notify_scope_dev(struct d ((void *)rmrr) + rmrr->header.length, rmrr->segment, rmrru->devices, rmrru->devices_cnt); -if (ret >0 ) -break ; -else if (ret <0 ) +if (ret <0 )return ret; }else if (info->event ==BUS_NOTIFY_DEL_DEVICE ) { -if (dmar_remove_dev_scope (info, rmrr->segment, - rmrru->devices, rmrru->devices_cnt)) -break ; +dmar_remove_dev_scope (info, rmrr->segment, + rmrru->devices, rmrru->devices_cnt); } }
Greg Kroah-Hartman
maintainerPATCH 89/122 blkcg: fix use-after-free in __blkg_release_rcu() by making blkcg_gq refcnt an atomic_t
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tejun Heo <[email protected]>
commit a5049a8ae34950249a7ae94c385d7c5c98914412 upstream.
Hello,
So, this patch should do. Joe, Vivek, can one of you guys please verify that the oops goes away with this patch?
Jens, the original thread can be read at
http://thread.gmane.org/gmane.linux.kernel/1720729
The fix converts blkg->refcnt from int to atomic_t. It does some overhead but it should be minute compared to everything else which is going on and the involved cacheline bouncing, so I think it's highly unlikely to cause any noticeable difference. Also, the refcnt in question should be converted to a perpcu_ref for blk-mq anyway, so the atomic_t is likely to go away pretty soon anyway.
Thanks.
------- 8< -------__blkg_release_rcu() may be invoked after the associated request_queue is released with a RCU grace period inbetween. As such, the function and callbacks invoked from it must not dereference the associated request_queue. This is clearly indicated in the comment above the function.
Unfortunately, while trying to fix a different issue, 2a4fd070ee85 ("blkcg: move bulk of blkcg_gq release operations to the RCU callback") ignored this and added [un]locking of @blkg->q->queue_lock to __blkg_release_rcu(). This of course can cause oops as the request_queue may be long gone by the time this code gets executed.
general protection fault: 0000 [#1] SMP CPU: 21 PID: 30 Comm: rcuos/21 Not tainted 3.15.0 #1 Hardware name: Stratus ftServer 6400/G7LAZ, BIOS BIOS Version 6.3:57 12/25/2013 task: ffff880854021de0 ti: ffff88085403c000 task.ti: ffff88085403c000 RIP: 0010:[<ffffffff8162e9e5>] [<ffffffff8162e9e5>] _raw_spin_lock_irq+0x15/0x60 RSP: 0018:ffff88085403fdf0 EFLAGS: 00010086 RAX: 0000000000020000 RBX: 0000000000000010 RCX: 0000000000000000 RDX: 000060ef80008248 RSI: 0000000000000286 RDI: 6b6b6b6b6b6b6b6b RBP: ffff88085403fdf0 R08: 0000000000000286 R09: 0000000000009f39 R10: 0000000000020001 R11: 0000000000020001 R12: ffff88103c17a130 R13: ffff88103c17a080 R14: 0000000000000000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff88107fca0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000006e5ab8 CR3: 000000000193d000 CR4: 00000000000407e0 Stack: ffff88085403fe18 ffffffff812cbfc2 ffff88103c17a130 0000000000000000 ffff88103c17a130 ffff88085403fec0 ffffffff810d1d28 ffff880854021de0 ffff880854021de0 ffff88107fcaec58 ffff88085403fe80 ffff88107fcaec30 Call Trace: [<ffffffff812cbfc2>] __blkg_release_rcu+0x72/0x150 [<ffffffff810d1d28>] rcu_nocb_kthread+0x1e8/0x300 [<ffffffff81091d81>] kthread+0xe1/0x100 [<ffffffff8163813c>] ret_from_fork+0x7c/0xb0 Code: ff 47 04 48 8b 7d 08 be 00 02 00 00 e8 55 48 a4 ff 5d c3 0f 1f 00 66 66 66 66 90 55 48 89 e5 +fa 66 66 90 66 66 90 b8 00 00 02 00 <f0> 0f c1 07 89 c2 c1 ea 10 66 39 c2 75 02 5d c3 83 e2 fe 0f +b7 RIP [<ffffffff8162e9e5>] _raw_spin_lock_irq+0x15/0x60 RSP <ffff88085403fdf0>
The request_queue locking was added because blkcg_gq->refcnt is an int protected with the queue lock and __blkg_release_rcu() needs to put the parent. Let's fix it by making blkcg_gq->refcnt an atomic_t and dropping queue locking in the function.
Given the general heavy weight of the current request_queue and blkcg operations, this is unlikely to cause any noticeable overhead. Moreover, blkcg_gq->refcnt is likely to be converted to percpu_ref in the near future, so whatever (most likely negligible) overhead it may add is temporary.
--- block/blk-cgroup.c | 7 ++----- block/blk-cgroup.h | 17 +++++++---------- 2 files changed, 9 insertions(+), 15 deletions(-)
block/blk-cgroup.c +2 -5
@@ -80,7 +80,7 @@ static struct blkcg_gq *blkg_alloc(struc blkg->q = q;INIT_LIST_HEAD (&blkg->q_node); blkg->blkcg = blkcg; - blkg->refcnt =1 ; +atomic_set (&blkg->refcnt,1 );/* root blkg uses @q->root_rl, init rl only for !root blkgs */ if (blkcg != &blkcg_root) {
@@ -399,11 +399,8 @@ void __blkg_release_rcu(struct rcu_head/* release the blkcg and parent blkg refs this blkg has been holding */ css_put (&blkg->blkcg->css); -if (blkg->parent) { -spin_lock_irq (blkg->q->queue_lock); +if (blkg->parent)blkg_put (blkg->parent); -spin_unlock_irq (blkg->q->queue_lock); - }blkg_free (blkg); }
block/blk-cgroup.h +7 -10
@@ -18,6 +18,7 @@#include <linux/seq_file.h> #include <linux/radix-tree.h> #include <linux/blkdev.h> +#include <linux/atomic.h> /* Max limits for throttle policy */ #define THROTL_IOPS_MAX UINT_MAX
@@ -104,7 +105,7 @@ struct blkcg_gq {struct request_list rl;/* reference count */ -int refcnt; +atomic_t refcnt;/* is this blkg online? protected by both blkcg and q locks */ bool online;
@@ -257,13 +258,12 @@ static inline int blkg_path(struct blkcg* blkg_get - get a blkg reference * @blkg: blkg to get * -* The caller should be holding queue_lock and an existing reference. +* The caller should be holding an existing reference. */ static inline void blkg_get (struct blkcg_gq *blkg) { -lockdep_assert_held (blkg->q->queue_lock); -WARN_ON_ONCE (!blkg->refcnt); - blkg->refcnt++; +WARN_ON_ONCE (atomic_read (&blkg->refcnt) <=0 ); +atomic_inc (&blkg->refcnt); }void __blkg_release_rcu (struct rcu_head *rcu);
@@ -271,14 +271,11 @@ void __blkg_release_rcu(struct rcu_head/** * blkg_put - put a blkg reference * @blkg: blkg to put -* -* The caller should be holding queue_lock. */ static inline void blkg_put (struct blkcg_gq *blkg) { -lockdep_assert_held (blkg->q->queue_lock); -WARN_ON_ONCE (blkg->refcnt <=0 ); -if (!--blkg->refcnt) +WARN_ON_ONCE (atomic_read (&blkg->refcnt) <=0 ); +if (atomic_dec_and_test (&blkg->refcnt))call_rcu (&blkg->rcu_head, __blkg_release_rcu); }
Greg Kroah-Hartman
maintainerPATCH 90/122 ext4: Fix buffer double free in ext4_alloc_branch()
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jan Kara <[email protected]>
commit c5c7b8ddfbf8cb3b2291e515a34ab1b8982f5a2d upstream.
Error recovery in ext4_alloc_branch() calls ext4_forget() even for buffer corresponding to indirect block it did not allocate. This leads to brelse() being called twice for that buffer (once from ext4_forget() and once from cleanup in ext4_ind_map_blocks()) leading to buffer use count misaccounting. Eventually (but often much later because there are other users of the buffer) we will see messages like:
VFS: brelse: Trying to free free buffer
Another manifestation of this problem is an error:
JBD2 unexpected failure: jbd2_journal_revoke: !buffer_revoked(bh); inconsistent data on disk
The fix is easy - don't forget buffer we did not allocate. Also add an explanatory comment because the indexing at ext4_alloc_branch() is somewhat subtle.
--- fs/ext4/indirect.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
fs/ext4/indirect.c +7 -1
@@ -389,7 +389,13 @@ static int ext4_alloc_branch(handle_t *hreturn 0 ; failed:for (; i >=0 ; i--) { -if (i != indirect_blks && branch[i].bh) +/* +* We want to ext4_forget() only freshly allocated indirect +* blocks. Buffer for new_blocks[i-1] is at branch[i].bh and +* buffer at branch[0].bh is indirect block / inode already +* existing before ext4_alloc_branch() was called. +*/ +if (i >0 && i != indirect_blks && branch[i].bh)ext4_forget (handle,1 , inode, branch[i].bh, branch[i].bh->b_blocknr); ext4_free_blocks(handle, inode, NULL, new_blocks[i],
Greg Kroah-Hartman
maintainerPATCH 91/122 ext4: Fix hole punching for files with indirect blocks
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jan Kara <[email protected]>
commit a93cd4cf86466caa49cfe64607bea7f0bde3f916 upstream.
Hole punching code for files with indirect blocks wrongly computed number of blocks which need to be cleared when traversing the indirect block tree. That could result in punching more blocks than actually requested and thus effectively cause a data loss. For example:
fallocate -n -p 10240000 4096
will punch the range 10240000 - 12632064 instead of the range 1024000 - 10244096. Fix the calculation.
--- fs/ext4/indirect.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-)
fs/ext4/indirect.c +10 -2
@@ -1318,16 +1318,24 @@ static int free_hole_blocks(handle_t *ha blk = *i_data;if (level >0 ) {ext4_lblk_t first2; +ext4_lblk_t count2; + bh =sb_bread (inode->i_sb,le32_to_cpu (blk));if (!bh) {EXT4_ERROR_INODE_BLOCK (inode,le32_to_cpu (blk),"Read failure" );return -EIO ; } - first2 = (first > offset) ? first - offset :0 ; +if (first > offset) { + first2 = first - offset; + count2 = count; + }else { + first2 =0 ; + count2 = count - (offset - first); + } ret =free_hole_blocks (handle, inode, bh, (__le32 *)bh->b_data, level -1 , - first2, count - offset, + first2, count2, inode->i_sb->s_blocksize >>2 );if (ret) {brelse (bh);
Greg Kroah-Hartman
maintainerPATCH 93/122 zram: revalidate disk after capacity change
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Minchan Kim <[email protected]>
commit 2e32baea46ce542c561a519414c840295b229c8f upstream.
Alexander reported mkswap on /dev/zram0 is failed if other process is opening the block device file.
Step is as follows,
0. Reset the unused zram device.
1. Use a program that opens /dev/zram0 with O_RDWR and sleeps until killed.
2. While that program sleeps, echo the correct value to /sys/block/zram0/disksize.
3. Verify (e.g. in /proc/partitions) that the disk size is applied correctly. It is.
4. While that program still sleeps, attempt to mkswap /dev/zram0. This fails: mkswap: error: swap area needs to be at least 40 KiB
When I investigated, the size get by ioctl(fd, BLKGETSIZE64, xxx) on mkswap to get a size of blockdev was zero although zram0 has right size by 2.
The reason is zram didn't revalidate disk after changing capacity so that size of blockdev's inode is not uptodate until all of file is close.
This patch should fix the BUG.
--- drivers/block/zram/zram_drv.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/block/zram/zram_drv.c +4 -1
@@ -622,8 +622,10 @@ static void zram_reset_device(struct zramemset (&zram->stats,0 ,sizeof (zram->stats)); zram->disksize =0 ; -if (reset_capacity) +if (reset_capacity) {set_capacity (zram->disk,0 ); +revalidate_disk (zram->disk); + }up_write (&zram->init_lock); }
@@ -664,6 +666,7 @@ static ssize_t disksize_store(struct dev zram->comp = comp; zram->disksize = disksize;set_capacity (zram->disk, zram->disksize >>SECTOR_SHIFT ); +revalidate_disk (zram->disk);up_write (&zram->init_lock);return len;
Re: [PATCH 3.15 093/122] zram: revalidate disk after capacity change
Quoted 45 lines · Greg Kroah-Hartman
Date: Mon, 7 Jul 2014 16:57:35 -0700
From: Greg Kroah-Hartman <[email protected]>
To: [email protected]
Cc: Greg Kroah-Hartman <[email protected]>,
[email protected], Minchan Kim <[email protected]>, "Alexander E. Patrakov" <[email protected]>, Sergey Senozhatsky <[email protected]>, Nitin Gupta <[email protected]>, Jerome Marchand <[email protected]>, Andrew Morton <[email protected]>, Linus Torvalds <[email protected]>
Subject: [PATCH 3.15 093/122] zram: revalidate disk after capacity change
X-Mailer: git-send-email 2.0.0.254.g50f84e3
User-Agent: quilt/0.63-13.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Minchan Kim <[email protected]>
commit
2e32baea46ce542c561a519414c840295b229c8fupstream.Alexander reported mkswap on
/dev/zram0is failed if other process is opening the block device file.Step is as follows,
0. Reset the unused zram device.
1. Use a program that opens/dev/zram0withO_RDWRand sleeps until killed.
2. While that program sleeps, echo the correct value to/sys/block/zram0/disksize.
3. Verify (e.g. in/proc/partitions) that the disk size is applied correctly. It is.
4. While that program still sleeps, attempt to mkswap/dev/zram0. This fails: mkswap: error: swap area needs to be at least 40 KiBWhen I investigated, the size get by
ioctl(fd, BLKGETSIZE64, xxx)on mkswap to get a size of blockdev was zero although zram0 has right size by 2.The reason is zram didn't revalidate disk after changing capacity so that size of blockdev's inode is not uptodate until all of file is close.
This patch should fix the BUG.
Quoted 9 lines
Signed-off-by Minchan Kim <[email protected]>Reported-by Alexander E. Patrakov <[email protected]>Reviewed-by Sergey Senozhatsky <[email protected]>Cc Nitin Gupta <[email protected]>Acked-by Jerome Marchand <[email protected]>Signed-off-by Andrew Morton <[email protected]>Signed-off-by Linus Torvalds <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>
Hello,
this patch is known to cause problems. please see
https://lkml.org/lkml/2014/7/3/885
-ss
Quoted 28 lines
--- drivers/block/zram/zram_drv.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)drivers/block/zram/zram_drv.c +4
-1@@ -622,8 +622,10 @@ static void zram_reset_device(struct zramemset (&zram->stats,0 ,sizeof (zram->stats)); zram->disksize =0 ; -if (reset_capacity) +if (reset_capacity) {set_capacity (zram->disk,0 ); +revalidate_disk (zram->disk); + }up_write (&zram->init_lock); }@@ -664,6 +666,7 @@ static ssize_t disksize_store(struct dev zram->comp = comp; zram->disksize = disksize;set_capacity (zram->disk, zram->disksize >>SECTOR_SHIFT ); +revalidate_disk (zram->disk);up_write (&zram->init_lock);return len;
Greg Kroah-Hartman
maintainerRe: [PATCH 3.15 093/122] zram: revalidate disk after capacity change
Quoted 56 lines · Sergey Senozhatsky
Quoted 45 lines · Greg Kroah-Hartman
Date: Mon, 7 Jul 2014 16:57:35 -0700
From: Greg Kroah-Hartman <[email protected]>
To: [email protected]
Cc: Greg Kroah-Hartman <[email protected]>,
[email protected], Minchan Kim <[email protected]>, "Alexander E. Patrakov" <[email protected]>, Sergey Senozhatsky <[email protected]>, Nitin Gupta <[email protected]>, Jerome Marchand <[email protected]>, Andrew Morton <[email protected]>, Linus Torvalds <[email protected]>
Subject: [PATCH 3.15 093/122] zram: revalidate disk after capacity change
X-Mailer: git-send-email 2.0.0.254.g50f84e3
User-Agent: quilt/0.63-13.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Minchan Kim <[email protected]>
commit
2e32baea46ce542c561a519414c840295b229c8fupstream.Alexander reported mkswap on
/dev/zram0is failed if other process is opening the block device file.Step is as follows,
0. Reset the unused zram device.
1. Use a program that opens/dev/zram0withO_RDWRand sleeps until killed.
2. While that program sleeps, echo the correct value to/sys/block/zram0/disksize.
3. Verify (e.g. in/proc/partitions) that the disk size is applied correctly. It is.
4. While that program still sleeps, attempt to mkswap/dev/zram0. This fails: mkswap: error: swap area needs to be at least 40 KiBWhen I investigated, the size get by
ioctl(fd, BLKGETSIZE64, xxx)on mkswap to get a size of blockdev was zero although zram0 has right size by 2.The reason is zram didn't revalidate disk after changing capacity so that size of blockdev's inode is not uptodate until all of file is close.
This patch should fix the BUG.
Quoted 9 lines
Signed-off-by Minchan Kim <[email protected]>Reported-by Alexander E. Patrakov <[email protected]>Reviewed-by Sergey Senozhatsky <[email protected]>Cc Nitin Gupta <[email protected]>Acked-by Jerome Marchand <[email protected]>Signed-off-by Andrew Morton <[email protected]>Signed-off-by Linus Torvalds <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>
Quoted 4 lines
Hello,
this patch is known to cause problems. please see
https://lkml.org/lkml/2014/7/3/885
Did the fix ever get merged to Linus's tree? If not, why not? I'd rather just take that fix as well.
thanks,
greg k-h
Re: [PATCH 3.15 093/122] zram: revalidate disk after capacity change
Hello Greg,
Quoted 57 lines · Greg Kroah-Hartman
Quoted 56 lines · Sergey Senozhatsky
Quoted 54 lines · Greg Kroah-Hartman
Date: Mon, 7 Jul 2014 16:57:35 -0700
From: Greg Kroah-Hartman <[email protected]>
To: [email protected]
Cc: Greg Kroah-Hartman <[email protected]>,
[email protected], Minchan Kim <[email protected]>, "Alexander E. Patrakov" <[email protected]>, Sergey Senozhatsky <[email protected]>, Nitin Gupta <[email protected]>, Jerome Marchand <[email protected]>, Andrew Morton <[email protected]>, Linus Torvalds <[email protected]>
Subject: [PATCH 3.15 093/122] zram: revalidate disk after capacity change
X-Mailer: git-send-email 2.0.0.254.g50f84e3
User-Agent: quilt/0.63-13.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Minchan Kim <[email protected]>
commit
2e32baea46ce542c561a519414c840295b229c8fupstream.Alexander reported mkswap on
/dev/zram0is failed if other process is opening the block device file.Step is as follows,
0. Reset the unused zram device.
1. Use a program that opens/dev/zram0withO_RDWRand sleeps until killed.
2. While that program sleeps, echo the correct value to/sys/block/zram0/disksize.
3. Verify (e.g. in/proc/partitions) that the disk size is applied correctly. It is.
4. While that program still sleeps, attempt to mkswap/dev/zram0. This fails: mkswap: error: swap area needs to be at least 40 KiBWhen I investigated, the size get by
ioctl(fd, BLKGETSIZE64, xxx)on mkswap to get a size of blockdev was zero although zram0 has right size by 2.The reason is zram didn't revalidate disk after changing capacity so that size of blockdev's inode is not uptodate until all of file is close.
This patch should fix the BUG.
Signed-off-by Minchan Kim <[email protected]>Reported-by Alexander E. Patrakov <[email protected]>Reviewed-by Sergey Senozhatsky <[email protected]>Cc Nitin Gupta <[email protected]>Acked-by Jerome Marchand <[email protected]>Signed-off-by Andrew Morton <[email protected]>Signed-off-by Linus Torvalds <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>
Quoted 7 lines
Quoted 4 lines
Hello,
this patch is known to cause problems. please see
https://lkml.org/lkml/2014/7/3/885Did the fix ever get merged to Linus's tree? If not, why not? I'd rather just take that fix as well.
Not yet.
Because Andrew picked his tree(ie, mmotm) today.
Quoted 9 lines
thanks,
greg k-h
--
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/
Signature · 2 lines
Kind regards, Minchan Kim
Greg Kroah-Hartman
maintainerPATCH 94/122 KVM: x86: Increase the number of fixed MTRR regs to 10
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Nadav Amit <[email protected]>
commit 682367c494869008eb89ef733f196e99415ae862 upstream.
Recent Intel CPUs have 10 variable range MTRRs. Since operating systems sometime make assumptions on CPUs while they ignore capability MSRs, it is better for KVM to be consistent with recent CPUs. Reporting more MTRRs than actually supported has no functional implications.
--- arch/x86/include/asm/kvm_host.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/kvm_host.h +1 -1
@@ -99,7 +99,7 @@ static inline gfn_t gfn_to_index(gfn_t g#define KVM_REFILL_PAGES 25#define KVM_MAX_CPUID_ENTRIES 80#define KVM_NR_FIXED_MTRR_REGION 88 -#define KVM_NR_VAR_MTRR 8 +#define KVM_NR_VAR_MTRR 10#define ASYNC_PF_PER_VCPU 64
Greg Kroah-Hartman
maintainerPATCH 95/122 KVM: x86: preserve the high 32-bits of the PAT register
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Paolo Bonzini <[email protected]>
commit 7cb060a91c0efc5ff94f83c6df3ed705e143cdb9 upstream.
KVM does not really do much with the PAT, so this went unnoticed for a long time. It is exposed however if you try to do rdmsr on the PAT register.
--- arch/x86/include/asm/kvm_host.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/kvm_host.h +1 -1
@@ -466,7 +466,7 @@ struct kvm_vcpu_arch {bool nmi_injected;/* Trying to inject an NMI this entry */ struct mtrr_state_type mtrr_state; -u32 pat; +u64 pat;unsigned switch_db_regs;unsigned long db[KVM_NR_DB_REGS ];
Greg Kroah-Hartman
maintainerPATCH 96/122 kvm: fix wrong address when writing Hyper-V tsc page
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xiaoming Gao <[email protected]>
commit e1fa108d24697b78348fd4e5a531029a50d0d36d upstream.
When kvm_write_guest writes the tsc_ref structure to the guest, or it will lead the low HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT bits of the TSC page address must be cleared, or the guest can see a non-zero sequence number.
Otherwise Windows guests would not be able to get a correct clocksource (QueryPerformanceCounter will always return 0) which causes serious chaos.
--- arch/x86/kvm/x86.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kvm/x86.c +1 -1
@@ -1912,7 +1912,7 @@ static int set_msr_hyperv_pw(struct kvm_if (!(data &HV_X64_MSR_TSC_REFERENCE_ENABLE ))break ; gfn = data >>HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT ; -if (kvm_write_guest (kvm, data, +if (kvm_write_guest (kvm, gfn <<HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT , &tsc_ref,sizeof (tsc_ref)))return 1 ;mark_page_dirty (kvm, gfn);
Greg Kroah-Hartman
maintainerPATCH 97/122 iio: of_iio_channel_get_by_name() returns non-null pointers for error legs
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Adam Thomson <[email protected]>
commit a2c12493ed7e63a18cef33a71686d12ffcd6600e upstream.
Currently in the inkern.c code for IIO framework, the function of_iio_channel_get_by_name() will return a non-NULL pointer when it cannot find a channel using of_iio_channel_get() and when it tries to search for 'io-channel-ranges' property and fails. This is incorrect behaviour as the function which calls this expects a NULL pointer for failure. This patch rectifies the issue.
--- drivers/iio/inkern.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/iio/inkern.c +4 -2
@@ -183,7 +183,7 @@ static struct iio_channel *of_iio_channeelse if (name && index >=0 ) {pr_err ("ERROR: could not get IIO channel %s:%s(%i)\n" , np->full_name, name ? name :"" , index); -return chan; +return NULL ; }/*
@@ -193,8 +193,9 @@ static struct iio_channel *of_iio_channe*/ np = np->parent;if (np && !of_get_property (np,"io-channel-ranges" ,NULL )) -break ; +return NULL ; } +return chan; }
@@ -317,6 +318,7 @@ struct iio_channel *iio_channel_get(struif (channel !=NULL )return channel; } +return iio_channel_get_sys (name, channel_name); }EXPORT_SYMBOL_GPL (iio_channel_get);
Greg Kroah-Hartman
maintainerPATCH 98/122 staging: iio/ad7291: fix error code in ad7291_probe()
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dan Carpenter <[email protected]>
commit b70e19c222a64018d308ebc80333575aff9f4e51 upstream.
We should be returning a negative error code instead of success here.
This would have been detected by GCC, except that the "ret" variable was initialized with a bogus value to disable GCC's uninitialized variable warnings. I've cleaned that up, as well.
--- drivers/staging/iio/adc/ad7291.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/staging/iio/adc/ad7291.c +2 -2
@@ -465,7 +465,7 @@ static int ad7291_probe(struct i2c_clienstruct ad7291_platform_data *pdata = client->dev.platform_data;struct ad7291_chip_info *chip;struct iio_dev *indio_dev; -int ret =0 ; +int ret; indio_dev =devm_iio_device_alloc (&client->dev,sizeof (*chip));if (!indio_dev)
@@ -475,7 +475,7 @@ static int ad7291_probe(struct i2c_clienif (pdata && pdata->use_external_ref) { chip->reg =devm_regulator_get (&client->dev,"vref" );if (IS_ERR (chip->reg)) -return ret; +return PTR_ERR (chip->reg); ret =regulator_enable (chip->reg);if (ret)
Greg Kroah-Hartman
maintainerPATCH 99/122 nfsd: fix rare symlink decoding bug
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: "J. Bruce Fields" <[email protected]>
commit 76f47128f9b33af1e96819746550d789054c9664 upstream.
An NFS operation that creates a new symlink includes the symlink data, which is xdr-encoded as a length followed by the data plus 0 to 3 bytes of zero-padding as required to reach a 4-byte boundary.
The vfs, on the other hand, wants null-terminated data.
The simple way to handle this would be by copying the data into a newly allocated buffer with space for the final null.
The current nfsd_symlink code tries to be more clever by skipping that step in the (likely) case where the byte following the string is already 0.
But that assumes that the byte following the string is ours to look at. In fact, it might be the first byte of a page that we can't read, or of some object that another task might modify.
Worse, the NFSv4 code tries to fix the problem by actually writing to that byte.
In the NFSv2/v3 cases this actually appears to be safe:
- nfs3svc_decode_symlinkargs explicitly null-terminates the data (after first checking its length and copying it to a new page).
- NFSv2 limits symlinks to 1k. The buffer holding the rpc request is always at least a page, and the link data (and previous fields) have maximum lengths that prevent the request from reaching the end of a page.
In the NFSv4 case the CREATE op is potentially just one part of a long compound so can end up on the end of a page if you're unlucky.
The minimal fix here is to copy and null-terminate in the NFSv4 case. The nfsd_symlink() interface here seems too fragile, though. It should really either do the copy itself every time or just require a null-terminated string.
--- fs/nfsd/nfs4proc.c | 9 --------- fs/nfsd/nfs4xdr.c | 13 ++++++++++++- 2 files changed, 12 insertions(+), 10 deletions(-)
fs/nfsd/nfs4proc.c +0 -9
@@ -618,15 +618,6 @@ nfsd4_create(struct svc_rqst *rqstp, strswitch (create->cr_type) {case NF4LNK : -/* ugh! we have to null-terminate the linktext, or -* vfs_symlink() will choke. it is always safe to -* null-terminate by brute force, since at worst we -* will overwrite the first byte of the create namelen -* in the XDR buffer, which has already been extracted -* during XDR decode. -*/ - create->cr_linkname[create->cr_linklen] =0 ; - status = nfsd_symlink(rqstp, &cstate->current_fh, create->cr_name, create->cr_namelen, create->cr_linkname, create->cr_linklen,
fs/nfsd/nfs4xdr.c +12 -1
@@ -602,7 +602,18 @@ nfsd4_decode_create(struct nfsd4_compounREAD_BUF (4 );READ32 (create->cr_linklen);READ_BUF (create->cr_linklen); -SAVEMEM (create->cr_linkname, create->cr_linklen); +/* +* The VFS will want a null-terminated string, and +* null-terminating in place isn't safe since this might +* end on a page boundary: +*/ + create->cr_linkname = +kmalloc (create->cr_linklen +1 ,GFP_KERNEL ); +if (!create->cr_linkname) +return nfserr_jukebox; +memcpy (create->cr_linkname, p, create->cr_linklen); + create->cr_linkname[create->cr_linklen] ='\0' ; +defer_free (argp, kfree, create->cr_linkname);break ;case NF4BLK :case NF4CHR :
Greg Kroah-Hartman
maintainerPATCH 100/122 tools: ffs-test: fix header values endianess
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Michal Nazarewicz <[email protected]>
commit f35f71244da6e51db4e1f2c7e318581f498ececf upstream.
It appears that no one ever run ffs-test on a big-endian machine, since it used cpu-endianess for fs_count and hs_count fields which should be in little-endian format. Fix by wrapping the numbers in cpu_to_le32.
--- tools/usb/ffs-test.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/usb/ffs-test.c +2 -2
@@ -116,8 +116,8 @@ static const struct { .header = { .magic = cpu_to_le32(FUNCTIONFS_DESCRIPTORS_MAGIC), .length = cpu_to_le32(sizeof descriptors), - .fs_count = 3, - .hs_count = 3, + .fs_count = cpu_to_le32(3), + .hs_count = cpu_to_le32(3), }, .fs_descs = { .intf = {
Greg Kroah-Hartman
maintainerPATCH 101/122 staging: tidspbridge: fix an erroneous removal of parentheses
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Suman Anna <[email protected]>
commit ff4f58f0ca5dee33a80a72393dd195de9284702b upstream.
Commit 4a9fdbb (staging: core: tiomap3430.c Fix line over 80 characters.) erroneously removed the parentheses around the function pointer leading to the following build error (when enabling the build of TI DSP/Bridge driver):
drivers/staging/tidspbridge/core/tiomap3430.c: In function 'bridge_brd_monitor':drivers/staging/tidspbridge/core/tiomap3430.c:283:10: error: invalid type argument of unary '*' (have 'u32') make[3]: *** [drivers/staging/tidspbridge/core/tiomap3430.o] Error 1
Fix this build error properly.
--- drivers/staging/tidspbridge/core/tiomap3430.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/staging/tidspbridge/core/tiomap3430.c +4 -2
@@ -280,8 +280,10 @@ static int bridge_brd_monitor(struct bri OMAP3430_IVA2_MOD, OMAP2_CM_CLKSTCTRL);/* Wait until the state has moved to ON */ -while (*pdata->dsp_prm_read (OMAP3430_IVA2_MOD ,OMAP2_PM_PWSTST )& -OMAP_INTRANSITION_MASK ); +while ((*pdata->dsp_prm_read)(OMAP3430_IVA2_MOD , +OMAP2_PM_PWSTST ) & +OMAP_INTRANSITION_MASK ) + ;/* Disable Automatic transition */ (*pdata->dsp_cm_write)(OMAP34XX_CLKSTCTRL_DISABLE_AUTO ,OMAP3430_IVA2_MOD ,OMAP2_CM_CLKSTCTRL );
Greg Kroah-Hartman
maintainerPATCH 102/122 tracing: Remove ftrace_stop/start() from reading the trace file
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Steven Rostedt (Red Hat)" <[email protected]>
commit 099ed151675cd1d2dbeae1dac697975f6a68716d upstream.
Disabling reading and writing to the trace file should not be able to disable all function tracing callbacks. There's other users today (like kprobes and perf). Reading a trace file should not stop those from happening.
--- kernel/trace/trace.c | 2 -- 1 file changed, 2 deletions(-)
kernel/trace/trace.c +0 -2
@@ -1367,7 +1367,6 @@ void tracing_start(void)arch_spin_unlock (&ftrace_max_lock); -ftrace_start (); out:raw_spin_unlock_irqrestore (&global_trace.start_lock, flags); }
@@ -1414,7 +1413,6 @@ void tracing_stop(void)struct ring_buffer *buffer;unsigned long flags; -ftrace_stop ();raw_spin_lock_irqsave (&global_trace.start_lock, flags);if (global_trace.stop_count++) goto out;
Greg Kroah-Hartman
maintainerPATCH 103/122 tracing/uprobes: Revert "Support mix of ftrace and perf"
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Oleg Nesterov <[email protected]>
commit 48212542067a7ff6cbe829dbae279c2ff7557b44 upstream.
This reverts commit 43fe98913c9f67e3b523615ee3316f9520a623e0.
This patch is very wrong. Firstly, this change leads to unbalanced uprobe_unregister(). Just for example,
# perf probe -x /lib/libc.so.6 syscall # echo 1 >> /sys/kernel/debug/tracing/events/probe_libc/enable # perf record -e probe_libc:syscall whatever
after that uprobe is dead (unregistered) but the user of ftrace/perf can't know this, and it looks as if nobody hits this probe.
This would be easy to fix, but there are other reasons why it is not simple to mix ftrace and perf. If nothing else, they can't share the same ->consumer.filter. This is fixable too, but probably we need to fix the poorly designed uprobe_register() interface first. At least "register" and "apply" should be clearly separated.
--- kernel/trace/trace_uprobe.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)
kernel/trace/trace_uprobe.c +8 -1
@@ -893,6 +893,9 @@ probe_event_enable(struct trace_uprobe *int ret;if (file) { +if (tu->tp.flags &TP_FLAG_PROFILE ) +return -EINTR ; + link =kmalloc (sizeof (*link),GFP_KERNEL );if (!link)return -ENOMEM ;
@@ -901,8 +904,12 @@ probe_event_enable(struct trace_uprobe *list_add_tail_rcu (&link->list, &tu->tp.files); tu->tp.flags |=TP_FLAG_TRACE ; - }else + }else { +if (tu->tp.flags &TP_FLAG_TRACE ) +return -EINTR ; + tu->tp.flags |=TP_FLAG_PROFILE ; + } ret =uprobe_buffer_enable ();if (ret <0 )
Greg Kroah-Hartman
maintainerPATCH 104/122 tracing/uprobes: Fix the usage of uprobe_buffer_enable() in probe_event_enable()
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Oleg Nesterov <[email protected]>
commit fb6bab6a5ad46d00b5ffa22268f21df1cd7c59df upstream.
The usage of uprobe_buffer_enable() added by dcad1a20 is very wrong,
1. uprobe_buffer_enable() and uprobe_buffer_disable() are not balanced, _enable() should be called only if !enabled.
2. If uprobe_buffer_enable() fails probe_event_enable() should clear tp.flags and free event_file_link.
3. If uprobe_register() fails it should do uprobe_buffer_disable().
--- kernel/trace/trace_uprobe.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-)
kernel/trace/trace_uprobe.c +19 -12
@@ -911,26 +911,33 @@ probe_event_enable(struct trace_uprobe * tu->tp.flags |=TP_FLAG_PROFILE ; } - ret =uprobe_buffer_enable (); -if (ret <0 ) -return ret; -WARN_ON (!uprobe_filter_is_empty (&tu->filter));if (enabled)return 0 ; + ret =uprobe_buffer_enable (); +if (ret) + goto err_flags; + tu->consumer.filter = filter; ret =uprobe_register (tu->inode, tu->offset, &tu->consumer); -if (ret) { -if (file) { -list_del (&link->list); -kfree (link); - tu->tp.flags &= ~TP_FLAG_TRACE ; - }else - tu->tp.flags &= ~TP_FLAG_PROFILE ; - } +if (ret) + goto err_buffer; + +return 0 ; + err_buffer: +uprobe_buffer_disable (); + + err_flags: +if (file) { +list_del (&link->list); +kfree (link); + tu->tp.flags &= ~TP_FLAG_TRACE ; + }else { + tu->tp.flags &= ~TP_FLAG_PROFILE ; + }return ret; }
Greg Kroah-Hartman
maintainerPATCH 105/122 btrfs: only unlock block in verify_parent_transid if we locked it
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Josef Bacik <[email protected]>
commit 472b909ff6f4884d235ef7b9d3847fad5efafbff upstream.
This is a regression from my patch a26e8c9f75b0bfd8cccc9e8f110737b136eb5994, we need to only unlock the block if we were the one who locked it. Otherwise this will trip BUG_ON()'s in locking.c Thanks,
--- fs/btrfs/disk-io.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
fs/btrfs/disk-io.c +2 -1
@@ -368,7 +368,8 @@ static int verify_parent_transid(struct out:unlock_extent_cached (io_tree, eb->start, eb->start + eb->len -1 , &cached_state,GFP_NOFS ); -btrfs_tree_read_unlock_blocking (eb); +if (need_lock) +btrfs_tree_read_unlock_blocking (eb);return ret; }
Greg Kroah-Hartman
maintainerPATCH 106/122 md: flush writes before starting a recovery.
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: NeilBrown <[email protected]>
commit 133d4527eab8d199a62eee6bd433f0776842df2e upstream.
When we write to a degraded array which has a bitmap, we make sure the relevant bit in the bitmap remains set when the write completes (so a 're-add' can quickly rebuilt a temporarily-missing device).
If, immediately after such a write starts, we incorporate a spare, commence recovery, and skip over the region where the write is happening (because the 'needs recovery' flag isn't set yet), then that write will not get to the new device.
Once the recovery finishes the new device will be trusted, but will have incorrect data, leading to possible corruption.
We cannot set the 'needs recovery' flag when we start the write as we do not know easily if the write will be "degraded" or not. That depends on details of the particular raid level and particular write request.
This patch fixes a corruption issue of long standing and so it suitable for any -stable kernel. It applied correctly to 3.0 at least and will minor editing to earlier kernels.
--- drivers/md/md.c | 13 +++++++++++++ 1 file changed, 13 insertions(+)
drivers/md/md.c +13 -0
@@ -7491,6 +7491,19 @@ void md_do_sync(struct md_thread *thread rdev->recovery_offset < j) j = rdev->recovery_offset;rcu_read_unlock (); + +/* If there is a bitmap, we need to make sure all +* writes that started before we added a spare +* complete before we start doing a recovery. +* Otherwise the write might complete and (via +* bitmap_endwrite) set a bit in the bitmap after the +* recovery has checked that bit and skipped that +* region. +*/ +if (mddev->bitmap) { + mddev->pers->quiesce (mddev,1 ); + mddev->pers->quiesce (mddev,0 ); + } }printk (KERN_INFO "md: %s of RAID array %s\n" , desc,mdname (mddev));
Greg Kroah-Hartman
maintainerPATCH 107/122 irqchip: spear_shirq: Fix interrupt offset
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Gleixner <[email protected]>
commit 4f4366033945419b0c52118c29d3057d7c558765 upstream.
The ras3 block on spear320 claims to have 3 interrupts. In fact it has one and 6 reserved interrupts. Account the 6 reserved to this block so it has 7 interrupts total. That matches the datasheet and the device tree entries.
Broken since commit 80515a5a(ARM: SPEAr3xx: shirq: simplify and move the shared irq multiplexor to DT). Testing is overrated....
--- drivers/irqchip/spear-shirq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/irqchip/spear-shirq.c +1 -1
@@ -125,7 +125,7 @@ static struct spear_shirq spear320_shirq };static struct spear_shirq spear320_shirq_ras3 = { - .irq_nr =3 , + .irq_nr =7 , .irq_bit_off =0 , .invalid_irq =1 , .regs = {
Greg Kroah-Hartman
maintainerPATCH 108/122 ARC: Fix build breakage for !CONFIG_ARC_DW2_UNWIND
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vineet Gupta <[email protected]>
commit ba25915fb2cd18152cb14b144dbe8bf2f2bd8e45 upstream.
--- arch/arc/kernel/ctx_sw_asm.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arc/kernel/ctx_sw_asm.S +1 -1
@@ -10,9 +10,9 @@ * -This is the more "natural" hand written assembler */ +#include <linux/linkage.h> #include <asm/entry.h> /* For the SAVE_* macros */ #include <asm/asm-offsets.h> -#include <asm/linkage.h> #define KSP_WORD_OFF ((TASK_THREAD + THREAD_KSP) / 4)
Greg Kroah-Hartman
maintainerPATCH 109/122 mlx4_core: Fix incorrect FLAGS1 bitmap test in mlx4_QUERY_FUNC_CAP
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jack Morgenstein <[email protected]>
commit bc82878baa10c2a6a4a6affaf52c152935112142 upstream.
Commit eb17711bc1d6 ("net/mlx4_core: Introduce nic_info new flag in QUERY_FUNC_CAP") did:
if (func_cap->flags1 & QUERY_FUNC_CAP_FLAGS1_OFFSET) {which should be:
if (func_cap->flags1 & QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN) {Fix that.
--- drivers/net/ethernet/mellanox/mlx4/fw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx4/fw.c +1 -1
@@ -414,7 +414,7 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_devMLX4_GET (func_cap->flags1, outbox,QUERY_FUNC_CAP_FLAGS1_OFFSET );if (dev->caps.port_type[gen_or_port] ==MLX4_PORT_TYPE_ETH ) { -if (func_cap->flags1 &QUERY_FUNC_CAP_FLAGS1_OFFSET ) { +if (func_cap->flags1 &QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN ) {mlx4_err (dev,"VLAN is enforced on this port\n" ); err = -EPROTONOSUPPORT ; goto out;
Greg Kroah-Hartman
maintainerPATCH 110/122 net/mlx4_core: Keep only one driver entry release mlx4_priv
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Wei Yang <[email protected]>
commit da1de8dfff09d33d4a5345762c21b487028e25f5 upstream.
Following commit befdf89 "net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one()", there are two mlx4 pci callbacks which will attempt to release the mlx4_priv object -- .shutdown and .remove.
This leads to a use-after-free access to the already freed mlx4_priv instance and trigger a "Kernel access of bad area" crash when both .shutdown and .remove are called.
During reboot or kexec, .shutdown is called, with the VFs probed to the host going through shutdown first and then the PF. Later, the PF will trigger VFs' .remove since VFs still have driver attached.
Fix that by keeping only one driver entry which releases mlx4_priv.
--- drivers/net/ethernet/mellanox/mlx4/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx4/main.c +1 -1
@@ -2800,7 +2800,7 @@ static struct pci_driver mlx4_driver = { .name = DRV_NAME, .id_table = mlx4_pci_table, .probe = mlx4_init_one, - .shutdown = mlx4_remove_one, + .shutdown = __mlx4_remove_one, .remove = mlx4_remove_one, .err_handler = &mlx4_err_handler, };
Greg Kroah-Hartman
maintainerPATCH 111/122 clk: qcom: Fix clk_rcg2_is_enabled() check
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stephen Boyd <[email protected]>
commit aa014149ba002155a084ec1e9328e95b70167cbb upstream.
If the bit is set the clock is off so we should be checking for a clear bit, not a set bit. Invert the logic.
--- drivers/clk/qcom/clk-rcg2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clk/qcom/clk-rcg2.c +1 -1
@@ -55,7 +55,7 @@ static int clk_rcg2_is_enabled(struct clif (ret)return ret; -return (cmd &CMD_ROOT_OFF ) !=0 ; +return (cmd &CMD_ROOT_OFF ) ==0 ; }static u8 clk_rcg2_get_parent (struct clk_hw *hw)
Greg Kroah-Hartman
maintainerPATCH 112/122 clk: qcom: Fix mmcc-8974s PLL configurations
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stephen Boyd <[email protected]>
commit 437ae6a1b8f2eedebfbf0f6572e19ca5c58a3f71 upstream.
We forgot to add the status bit for the PLLs and we were using the wrong register and masks for configuration, leading to unexpected PLL configurations. Fix this.
--- drivers/clk/qcom/mmcc-msm8974.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-)
drivers/clk/qcom/mmcc-msm8974.c +8 -5
@@ -169,6 +169,7 @@ static struct clk_pll mmpll0 = { .config_reg = 0x0014, .mode_reg = 0x0000, .status_reg = 0x001c, + .status_bit = 17, .clkr.hw.init = &(struct clk_init_data ){ .name ="mmpll0" , .parent_names = (const char *[]){"xo" },
@@ -192,9 +193,10 @@ static struct clk_pll mmpll1 = { .l_reg = 0x0044, .m_reg = 0x0048, .n_reg = 0x004c, - .config_reg = 0x0054, + .config_reg = 0x0050, .mode_reg = 0x0040, .status_reg = 0x005c, + .status_bit = 17, .clkr.hw.init = &(struct clk_init_data ){ .name ="mmpll1" , .parent_names = (const char *[]){"xo" },
@@ -218,7 +220,7 @@ static struct clk_pll mmpll2 = { .l_reg = 0x4104, .m_reg = 0x4108, .n_reg = 0x410c, - .config_reg = 0x4114, + .config_reg = 0x4110, .mode_reg = 0x4100, .status_reg = 0x411c, .clkr.hw.init = &(struct clk_init_data ){
@@ -233,9 +235,10 @@ static struct clk_pll mmpll3 = { .l_reg = 0x0084, .m_reg = 0x0088, .n_reg = 0x008c, - .config_reg = 0x0094, + .config_reg = 0x0090, .mode_reg = 0x0080, .status_reg = 0x009c, + .status_bit = 17, .clkr.hw.init = &(struct clk_init_data ){ .name ="mmpll3" , .parent_names = (const char *[]){"xo" },
@@ -2318,7 +2321,7 @@ static const struct pll_config mmpll1_co .vco_val = 0x0, .vco_mask = 0x3 << 20, .pre_div_val = 0x0, - .pre_div_mask = 0x3 << 12, + .pre_div_mask = 0x7 << 12, .post_div_val = 0x0, .post_div_mask = 0x3 << 8, .mn_ena_mask = BIT(24),
@@ -2332,7 +2335,7 @@ static struct pll_config mmpll3_config = .vco_val = 0x0, .vco_mask = 0x3 << 20, .pre_div_val = 0x0, - .pre_div_mask = 0x3 << 12, + .pre_div_mask = 0x7 << 12, .post_div_val = 0x0, .post_div_mask = 0x3 << 8, .mn_ena_mask = BIT(24),
Greg Kroah-Hartman
maintainerPATCH 113/122 serial: Fix IGNBRK handling
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Peter Hurley <[email protected]>
commit ef8b9ddcb45fa3b1e11acd72be2398001e807d14 upstream.
If IGNBRK is set without either BRKINT or PARMRK set, some uart drivers send a 0x00 byte for BREAK without the TTYBREAK flag to the line discipline, when it should send either nothing or the TTYBREAK flag set. This happens because the read_status_mask masks out the BI condition, which uart_insert_char() then interprets as a normal 0x00 byte.
SUS v3 is clear regarding the meaning of IGNBRK; Section 11.2.2, General Terminal Interface - Input Modes, states:
"If IGNBRK is set, a break condition detected on input shall be ignored;
that is, not put on the input queue and therefore not read by any process."
Fix read_status_mask to include the BI bit if IGNBRK is set; the lsr status retains the BI bit if a BREAK is recv'd, which is subsequently ignored in uart_insert_char() when masked with the ignore_status_mask.
Affected drivers:
8250 - allserial_txx9
mfd
amba-pl010
amba-pl011atmel_serialbfin_uart
dz
ip22zilog
max310x
mxs-auart
netx-serialpnx8xxx_uart
pxa
sb1250-duart
sccnxpserial_ks8695sirfsoc_uart
st-ascvr41xx_siu
zs
sunzilogfsl_lpuart
sunsabucc_uartbcm63xx_uart
sunsu
efm32-uartpmac_zilog
mpscmsm_serialm32r_sio
Unaffected drivers:
omap-serial
rp2
sa1100
imx
icom
Annotated for fixes:altera_uart
mcf
Drivers without break detection:
21285
xilinx-uartpsaltera_jtaguart
apbuart
arc-uart
clps711x
max3100
uartlitemsm_serial_hs
nwpserial
lantiqvt8500_serial
Unknown:
samsungmpc52xx_uartbfin_sport_uart
cpm_uart/core
--- drivers/tty/serial/8250/8250_core.c | 2 +- drivers/tty/serial/altera_uart.c | 6 ++++++ drivers/tty/serial/amba-pl010.c | 2 +- drivers/tty/serial/amba-pl011.c | 2 +- drivers/tty/serial/atmel_serial.c | 2 +- drivers/tty/serial/bcm63xx_uart.c | 2 +- drivers/tty/serial/bfin_uart.c | 2 +- drivers/tty/serial/dz.c | 2 +- drivers/tty/serial/efm32-uart.c | 2 +- drivers/tty/serial/fsl_lpuart.c | 2 +- drivers/tty/serial/ip22zilog.c | 2 +- drivers/tty/serial/m32r_sio.c | 2 +- drivers/tty/serial/max310x.c | 2 +- drivers/tty/serial/mcf.c | 6 ++++++ drivers/tty/serial/mfd.c | 2 +- drivers/tty/serial/mpsc.c | 2 +- drivers/tty/serial/msm_serial.c | 2 +- drivers/tty/serial/mxs-auart.c | 2 +- drivers/tty/serial/netx-serial.c | 2 +- drivers/tty/serial/pmac_zilog.c | 2 +- drivers/tty/serial/pnx8xxx_uart.c | 2 +- drivers/tty/serial/pxa.c | 2 +- drivers/tty/serial/sb1250-duart.c | 2 +- drivers/tty/serial/sccnxp.c | 2 +- drivers/tty/serial/serial_ks8695.c | 2 +- drivers/tty/serial/serial_txx9.c | 2 +- drivers/tty/serial/sirfsoc_uart.c | 2 +- drivers/tty/serial/st-asc.c | 2 +- drivers/tty/serial/sunsab.c | 2 +- drivers/tty/serial/sunsu.c | 2 +- drivers/tty/serial/sunzilog.c | 2 +- drivers/tty/serial/ucc_uart.c | 2 +- drivers/tty/serial/vr41xx_siu.c | 2 +- drivers/tty/serial/zs.c | 2 +- 34 files changed, 44 insertions(+), 32 deletions(-)
drivers/tty/serial/8250/8250_core.c +1 -1
@@ -2360,7 +2360,7 @@ serial8250_do_set_termios(struct uart_po port->read_status_mask =UART_LSR_OE |UART_LSR_THRE |UART_LSR_DR ;if (termios->c_iflag &INPCK ) port->read_status_mask |=UART_LSR_FE |UART_LSR_PE ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=UART_LSR_BI ;/*
drivers/tty/serial/altera_uart.c +6 -0
@@ -185,6 +185,12 @@ static void altera_uart_set_termios(struuart_update_timeout (port, termios->c_cflag, baud);altera_uart_writel (port, baudclk,ALTERA_UART_DIVISOR_REG );spin_unlock_irqrestore (&port->lock, flags); + +/* +* FIXME: port->read_status_mask and port->ignore_status_mask +* need to be initialized based on termios settings for +* INPCK, IGNBRK, IGNPAR, PARMRK, BRKINT +*/ }static void altera_uart_rx_chars (struct altera_uart *pp)
drivers/tty/serial/amba-pl010.c +1 -1
@@ -420,7 +420,7 @@ pl010_set_termios(struct uart_port *port uap->port.read_status_mask = UART01x_RSR_OE;if (termios->c_iflag &INPCK ) uap->port.read_status_mask |= UART01x_RSR_FE | UART01x_RSR_PE; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) uap->port.read_status_mask |= UART01x_RSR_BE;/*
drivers/tty/serial/amba-pl011.c +1 -1
@@ -1744,7 +1744,7 @@ pl011_set_termios(struct uart_port *port port->read_status_mask =UART011_DR_OE |255 ;if (termios->c_iflag &INPCK ) port->read_status_mask |=UART011_DR_FE |UART011_DR_PE ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=UART011_DR_BE ;/*
drivers/tty/serial/atmel_serial.c +1 -1
@@ -1784,7 +1784,7 @@ static void atmel_set_termios(struct uar port->read_status_mask =ATMEL_US_OVRE ;if (termios->c_iflag &INPCK ) port->read_status_mask |= (ATMEL_US_FRAME |ATMEL_US_PARE ); -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=ATMEL_US_RXBRK ;if (atmel_use_pdc_rx (port))
drivers/tty/serial/bcm63xx_uart.c +1 -1
@@ -567,7 +567,7 @@ static void bcm_uart_set_termios(struct port->read_status_mask |=UART_FIFO_FRAMEERR_MASK ; port->read_status_mask |=UART_FIFO_PARERR_MASK ; } -if (new->c_iflag & (BRKINT )) +if (new->c_iflag & (IGNBRK |BRKINT )) port->read_status_mask |=UART_FIFO_BRKDET_MASK ; port->ignore_status_mask =0 ;
drivers/tty/serial/bfin_uart.c +1 -1
@@ -833,7 +833,7 @@ bfin_serial_set_termios(struct uart_port port->read_status_mask =OE ;if (termios->c_iflag &INPCK ) port->read_status_mask |= (FE |PE ); -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=BI ;/*
drivers/tty/serial/dz.c +1 -1
@@ -625,7 +625,7 @@ static void dz_set_termios(struct uart_p dport->port.read_status_mask =DZ_OERR ;if (termios->c_iflag &INPCK ) dport->port.read_status_mask |=DZ_FERR |DZ_PERR ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) dport->port.read_status_mask |=DZ_BREAK ;/* characters to ignore */
drivers/tty/serial/efm32-uart.c +1 -1
@@ -407,7 +407,7 @@ static void efm32_uart_set_termios(strucif (new->c_iflag &INPCK ) port->read_status_mask |= UARTn_RXDATAX_FERR | UARTn_RXDATAX_PERR; -if (new->c_iflag & (BRKINT |PARMRK )) +if (new->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |= SW_UARTn_RXDATAX_BERR; port->ignore_status_mask =0 ;
drivers/tty/serial/fsl_lpuart.c +1 -1
@@ -902,7 +902,7 @@ lpuart_set_termios(struct uart_port *por sport->port.read_status_mask =0 ;if (termios->c_iflag &INPCK ) sport->port.read_status_mask |= (UARTSR1_FE |UARTSR1_PE ); -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) sport->port.read_status_mask |=UARTSR1_FE ;/* characters to ignore */
drivers/tty/serial/ip22zilog.c +1 -1
@@ -850,7 +850,7 @@ ip22zilog_convert_to_zs(struct uart_ip22 up->port.read_status_mask = Rx_OVR;if (iflag &INPCK ) up->port.read_status_mask |=CRC_ERR |PAR_ERR ; -if (iflag & (BRKINT |PARMRK )) +if (iflag & (IGNBRK |BRKINT |PARMRK )) up->port.read_status_mask |=BRK_ABRT ; up->port.ignore_status_mask =0 ;
drivers/tty/serial/m32r_sio.c +1 -1
@@ -737,7 +737,7 @@ static void m32r_sio_set_termios(struct up->port.read_status_mask =UART_LSR_OE |UART_LSR_THRE |UART_LSR_DR ;if (termios->c_iflag &INPCK ) up->port.read_status_mask |=UART_LSR_FE |UART_LSR_PE ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) up->port.read_status_mask |=UART_LSR_BI ;/*
drivers/tty/serial/max310x.c +1 -1
@@ -835,7 +835,7 @@ static void max310x_set_termios(struct uif (termios->c_iflag &INPCK ) port->read_status_mask |=MAX310X_LSR_RXPAR_BIT |MAX310X_LSR_FRERR_BIT ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=MAX310X_LSR_RXBRK_BIT ;/* Set status ignore mask */
drivers/tty/serial/mcf.c +6 -0
@@ -248,6 +248,12 @@ static void mcf_set_termios(struct uart_ mr1 |=MCFUART_MR1_PARITYNONE ; } +/* +* FIXME: port->read_status_mask and port->ignore_status_mask +* need to be initialized based on termios settings for +* INPCK, IGNBRK, IGNPAR, PARMRK, BRKINT +*/ +if (termios->c_cflag &CSTOPB ) mr2 |=MCFUART_MR2_STOP2 ;else
drivers/tty/serial/mfd.c +1 -1
@@ -977,7 +977,7 @@ serial_hsu_set_termios(struct uart_port up->port.read_status_mask =UART_LSR_OE |UART_LSR_THRE |UART_LSR_DR ;if (termios->c_iflag &INPCK ) up->port.read_status_mask |=UART_LSR_FE |UART_LSR_PE ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) up->port.read_status_mask |=UART_LSR_BI ;/* Characters to ignore */
drivers/tty/serial/mpsc.c +1 -1
@@ -1458,7 +1458,7 @@ static void mpsc_set_termios(struct uart pi->port.read_status_mask |=SDMA_DESC_CMDSTAT_PE |SDMA_DESC_CMDSTAT_FR ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) pi->port.read_status_mask |=SDMA_DESC_CMDSTAT_BR ;/* Characters/events to ignore */
drivers/tty/serial/msm_serial.c +1 -1
@@ -583,7 +583,7 @@ static void msm_set_termios(struct uart_ port->read_status_mask =0 ;if (termios->c_iflag &INPCK ) port->read_status_mask |=UART_SR_PAR_FRAME_ERR ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=UART_SR_RX_BREAK ;uart_update_timeout (port, termios->c_cflag, baud);
drivers/tty/serial/mxs-auart.c +1 -1
@@ -604,7 +604,7 @@ static void mxs_auart_settermios(structif (termios->c_iflag &INPCK ) u->read_status_mask |=AUART_STAT_PERR ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) u->read_status_mask |=AUART_STAT_BERR ;/*
drivers/tty/serial/netx-serial.c +1 -1
@@ -419,7 +419,7 @@ netx_set_termios(struct uart_port *port, } port->read_status_mask =0 ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=SR_BE ;if (termios->c_iflag &INPCK ) port->read_status_mask |=SR_PE |SR_FE ;
drivers/tty/serial/pmac_zilog.c +1 -1
@@ -1092,7 +1092,7 @@ static void pmz_convert_to_zs(struct uar uap->port.read_status_mask = Rx_OVR;if (iflag &INPCK ) uap->port.read_status_mask |=CRC_ERR |PAR_ERR ; -if (iflag & (BRKINT |PARMRK )) +if (iflag & (IGNBRK |BRKINT |PARMRK )) uap->port.read_status_mask |=BRK_ABRT ; uap->port.ignore_status_mask =0 ;
drivers/tty/serial/pnx8xxx_uart.c +1 -1
@@ -477,7 +477,7 @@ pnx8xxx_set_termios(struct uart_port *po sport->port.read_status_mask |=FIFO_TO_SM (PNX8XXX_UART_FIFO_RXFE ) |FIFO_TO_SM (PNX8XXX_UART_FIFO_RXPAR ); -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) sport->port.read_status_mask |=ISTAT_TO_SM (PNX8XXX_UART_INT_BREAK );
drivers/tty/serial/pxa.c +1 -1
@@ -492,7 +492,7 @@ serial_pxa_set_termios(struct uart_port up->port.read_status_mask =UART_LSR_OE |UART_LSR_THRE |UART_LSR_DR ;if (termios->c_iflag &INPCK ) up->port.read_status_mask |=UART_LSR_FE |UART_LSR_PE ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) up->port.read_status_mask |=UART_LSR_BI ;/*
drivers/tty/serial/sb1250-duart.c +1 -1
@@ -596,7 +596,7 @@ static void sbd_set_termios(struct uart_if (termios->c_iflag &INPCK ) uport->read_status_mask |=M_DUART_FRM_ERR |M_DUART_PARITY_ERR ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) uport->read_status_mask |=M_DUART_RCVD_BRK ; uport->ignore_status_mask =0 ;
drivers/tty/serial/sccnxp.c +1 -1
@@ -667,7 +667,7 @@ static void sccnxp_set_termios(struct ua port->read_status_mask =SR_OVR ;if (termios->c_iflag &INPCK ) port->read_status_mask |=SR_PE |SR_FE ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=SR_BRK ;/* Set status ignore mask */
drivers/tty/serial/serial_ks8695.c +1 -1
@@ -437,7 +437,7 @@ static void ks8695uart_set_termios(struc port->read_status_mask =URLS_URROE ;if (termios->c_iflag &INPCK ) port->read_status_mask |= (URLS_URFE |URLS_URPE ); -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=URLS_URBI ;/*
drivers/tty/serial/serial_txx9.c +1 -1
@@ -702,7 +702,7 @@ serial_txx9_set_termios(struct uart_portTXX9_SIDISR_TDIS |TXX9_SIDISR_RDIS ;if (termios->c_iflag &INPCK ) up->port.read_status_mask |=TXX9_SIDISR_UFER |TXX9_SIDISR_UPER ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) up->port.read_status_mask |=TXX9_SIDISR_UBRK ;/*
drivers/tty/serial/sirfsoc_uart.c +1 -1
@@ -907,7 +907,7 @@ static void sirfsoc_uart_set_termios(strif (termios->c_iflag &INPCK ) port->read_status_mask |= uint_en->sirfsoc_frm_err_en; } -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |= uint_en->sirfsoc_rxd_brk_en;if (sirfport->uart_reg->uart_type ==SIRF_REAL_UART ) {if (termios->c_iflag &IGNPAR )
drivers/tty/serial/st-asc.c +1 -1
@@ -547,7 +547,7 @@ static void asc_set_termios(struct uart_ ascport->port.read_status_mask =ASC_RXBUF_DUMMY_OE ;if (termios->c_iflag &INPCK ) ascport->port.read_status_mask |=ASC_RXBUF_FE |ASC_RXBUF_PE ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) ascport->port.read_status_mask |=ASC_RXBUF_DUMMY_BE ;/*
drivers/tty/serial/sunsab.c +1 -1
@@ -719,7 +719,7 @@ static void sunsab_convert_to_sab(structif (iflag &INPCK ) up->port.read_status_mask |= (SAB82532_ISR0_PERR |SAB82532_ISR0_FERR ); -if (iflag & (BRKINT |PARMRK )) +if (iflag & (IGNBRK |BRKINT |PARMRK )) up->port.read_status_mask |= (SAB82532_ISR1_BRK <<8 );/*
drivers/tty/serial/sunsu.c +1 -1
@@ -834,7 +834,7 @@ sunsu_change_speed(struct uart_port *por up->port.read_status_mask =UART_LSR_OE |UART_LSR_THRE |UART_LSR_DR ;if (iflag &INPCK ) up->port.read_status_mask |=UART_LSR_FE |UART_LSR_PE ; -if (iflag & (BRKINT |PARMRK )) +if (iflag & (IGNBRK |BRKINT |PARMRK )) up->port.read_status_mask |=UART_LSR_BI ;/*
drivers/tty/serial/sunzilog.c +1 -1
@@ -915,7 +915,7 @@ sunzilog_convert_to_zs(struct uart_sunzi up->port.read_status_mask = Rx_OVR;if (iflag &INPCK ) up->port.read_status_mask |=CRC_ERR |PAR_ERR ; -if (iflag & (BRKINT |PARMRK )) +if (iflag & (IGNBRK |BRKINT |PARMRK )) up->port.read_status_mask |=BRK_ABRT ; up->port.ignore_status_mask =0 ;
drivers/tty/serial/ucc_uart.c +1 -1
@@ -936,7 +936,7 @@ static void qe_uart_set_termios(struct u port->read_status_mask =BD_SC_EMPTY |BD_SC_OV ;if (termios->c_iflag &INPCK ) port->read_status_mask |=BD_SC_FR |BD_SC_PR ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=BD_SC_BR ;/*
drivers/tty/serial/vr41xx_siu.c +1 -1
@@ -559,7 +559,7 @@ static void siu_set_termios(struct uart_ port->read_status_mask =UART_LSR_THRE |UART_LSR_OE |UART_LSR_DR ;if (c_iflag &INPCK ) port->read_status_mask |=UART_LSR_FE |UART_LSR_PE ; -if (c_iflag & (BRKINT |PARMRK )) +if (c_iflag & (IGNBRK |BRKINT |PARMRK )) port->read_status_mask |=UART_LSR_BI ; port->ignore_status_mask =0 ;
drivers/tty/serial/zs.c +1 -1
@@ -923,7 +923,7 @@ static void zs_set_termios(struct uart_p uport->read_status_mask = Rx_OVR;if (termios->c_iflag &INPCK ) uport->read_status_mask |=FRM_ERR |PAR_ERR ; -if (termios->c_iflag & (BRKINT |PARMRK )) +if (termios->c_iflag & (IGNBRK |BRKINT |PARMRK )) uport->read_status_mask |= Rx_BRK; uport->ignore_status_mask =0 ;
Greg Kroah-Hartman
maintainerPATCH 114/122 tty: Correct INPCK handling
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Peter Hurley <[email protected]>
commit 66528f90669691c85c73bea4f0c9f4a5857c4cab upstream.
If INPCK is not set, input parity detection should be disabled. This means parity errors should not be received from the tty driver, and the data received should be treated normally.
SUS v3, 11.2.2, General Terminal Interface - Input Modes, states:
"If INPCK is set, input parity checking shall be enabled. If INPCK is
not set, input parity checking shall be disabled, allowing output parity generation without input parity errors. Note that whether input parity checking is enabled or disabled is independent of whether parity detection is enabled or disabled (see Control Modes). If parity detection is enabled but input parity checking is disabled, the hardware to which the terminal is connected shall recognize the parity bit, but the terminal special file shall not check whether or not this bit is correctly set."
Ignore parity errors reported by the tty driver when INPCK is not set, and handle the received data normally.
--- drivers/tty/n_tty.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-)
drivers/tty/n_tty.c +10 -9
@@ -1214,15 +1214,16 @@ static void n_tty_receive_parity_error(s {struct n_tty_data *ldata = tty->disc_data; -if (I_IGNPAR (tty)) -return ; -if (I_PARMRK (tty)) { -put_tty_queue ('\377' , ldata); -put_tty_queue ('\0' , ldata); -put_tty_queue (c, ldata); - }else if (I_INPCK (tty)) -put_tty_queue ('\0' , ldata); -else +if (I_INPCK (tty)) { +if (I_IGNPAR (tty)) +return ; +if (I_PARMRK (tty)) { +put_tty_queue ('\377' , ldata); +put_tty_queue ('\0' , ldata); +put_tty_queue (c, ldata); + }else +put_tty_queue ('\0' , ldata); + }else put_tty_queue (c, ldata);if (waitqueue_active (&tty->read_wait))wake_up_interruptible (&tty->read_wait);
Greg Kroah-Hartman
maintainerPATCH 115/122 netfilter: ctnetlink: fix dumping of dying/unconfirmed conntracks
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Pablo Neira Ayuso <[email protected]>
commit 266155b2de8fb721ae353688529b2f8bcdde2f90 upstream.
The dumping prematurely stops, it seems the callback argument that indicates that all entries have been dumped is set after iterating on the first cpu list. The dumping also may stop before the entire per-cpu list content is also dumped.
With this patch, conntrack -L dying now shows the dying list content again.
--- net/netfilter/nf_conntrack_netlink.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-)
net/netfilter/nf_conntrack_netlink.c +3 -5
@@ -1163,9 +1163,6 @@ ctnetlink_dump_list(struct sk_buff *skb,if (cb->args[2 ])return 0 ; -if (cb->args[0 ] == nr_cpu_ids) -return 0 ; -for (cpu = cb->args[0 ]; cpu < nr_cpu_ids; cpu++) {struct ct_pcpu *pcpu;
@@ -1194,6 +1191,7 @@ restart:rcu_read_unlock ();if (res <0 ) {nf_conntrack_get (&ct->ct_general); + cb->args[0 ] = cpu; cb->args[1 ] = (unsigned long )ct;spin_unlock_bh (&pcpu->lock); goto out;
@@ -1202,10 +1200,10 @@ restart:if (cb->args[1 ]) { cb->args[1 ] =0 ; goto restart; - }else - cb->args[2 ] =1 ; + }spin_unlock_bh (&pcpu->lock); } + cb->args[2 ] =1 ; out:if (last)nf_ct_put (last);
Greg Kroah-Hartman
maintainerPATCH 116/122 netfilter: nf_nat: fix oops on netns removal
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Florian Westphal <[email protected]>
commit 945b2b2d259d1a4364a2799e80e8ff32f8c6ee6f upstream.
Quoting Samu Kallio:
Basically what's happening is, during netns cleanup, nf_nat_net_exit gets called before ipv4_net_exit. As I understand it, nf_nat_net_exit is supposed to kill any conntrack entries which have NAT context (through nf_ct_iterate_cleanup), but for some reason this doesn't happen (perhaps something else is still holding refs to those entries?).
When ipv4_net_exit is called, conntrack entries (including those with NAT context) are cleaned up, but the
nat_bysource hashtable is long gone - freed in nf_nat_net_exit. The bug happens when attempting to free a conntrack entry whose NAT hash 'prev' field points to a slot in the freed hash table (head for that bin).
We ignore conntracks with null nat bindings. But this is wrong, as these are in bysource hash table as well.
Restore nat-cleaning for the netns-is-being-removed case.
bug:
https://bugzilla.kernel.org/show_bug.cgi?id=65191
--- net/netfilter/nf_nat_core.c | 35 ++++++++++++++++++++++++++++++++++- 1 file changed, 34 insertions(+), 1 deletion(-)
net/netfilter/nf_nat_core.c +34 -1
@@ -517,6 +517,39 @@ static int nf_nat_proto_remove(struct nfreturn i->status &IPS_NAT_MASK ?1 :0 ; } +static int nf_nat_proto_clean (struct nf_conn *ct,void *data) +{ +struct nf_conn_nat *nat =nfct_nat (ct); + +if (nf_nat_proto_remove (ct, data)) +return 1 ; + +if (!nat || !nat->ct) +return 0 ; + +/* This netns is being destroyed, and conntrack has nat null binding. +* Remove it from bysource hash, as the table will be freed soon. +* +* Else, when the conntrack is destoyed, nf_nat_cleanup_conntrack() +* will delete entry from already-freed table. +*/ +if (!del_timer (&ct->timeout)) +return 1 ; + +spin_lock_bh (&nf_nat_lock); +hlist_del_rcu (&nat->bysource); + ct->status &= ~IPS_NAT_DONE_MASK ; + nat->ct =NULL ; +spin_unlock_bh (&nf_nat_lock); + +add_timer (&ct->timeout); + +/* don't delete conntrack. Although that would make things a lot +* simpler, we'd end up flushing all conntracks on nat rmmod. +*/ +return 0 ; +} +static void nf_nat_l4proto_clean (u8 l3proto,u8 l4proto) {struct nf_nat_proto_clean clean = {
@@ -787,7 +820,7 @@ static void __net_exit nf_nat_net_exit(s {struct nf_nat_proto_clean clean = {}; -nf_ct_iterate_cleanup (net, &nf_nat_proto_remove, &clean,0 ,0 ); +nf_ct_iterate_cleanup (net, nf_nat_proto_clean, &clean,0 ,0 );synchronize_rcu ();nf_ct_free_hashtable (net->ct.nat_bysource, net->ct.nat_htable_size); }
Greg Kroah-Hartman
maintainerPATCH 117/122 netfilter: ctnetlink: fix refcnt leak in dying/unconfirmed list dumper
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Florian Westphal <[email protected]>
commit cd5f336f1780cb20e83146cde64d3d5779e175e6 upstream.
'last' keeps track of the ct that had its refcnt bumped during previous dump cycle. Thus it must not be overwritten until end-of-function.
Another (unrelated, theoretical) issue: Don't attempt to bump refcnt of a conntrack whose reference count is already 0. Such conntrack is being destroyed right now, its memory is freed once we release the percpu dying spinlock.
--- net/netfilter/nf_conntrack_netlink.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
net/netfilter/nf_conntrack_netlink.c +5 -3
@@ -1150,7 +1150,7 @@ static int ctnetlink_done_list(struct nestatic int ctnetlink_dump_list (struct sk_buff *skb,struct netlink_callback *cb,bool dying) { -struct nf_conn *ct, *last =NULL ; +struct nf_conn *ct, *last;struct nf_conntrack_tuple_hash *h;struct hlist_nulls_node *n;struct nfgenmsg *nfmsg =nlmsg_data (cb->nlh);
@@ -1163,6 +1163,8 @@ ctnetlink_dump_list(struct sk_buff *skb,if (cb->args[2 ])return 0 ; + last = (struct nf_conn *)cb->args[1 ]; +for (cpu = cb->args[0 ]; cpu < nr_cpu_ids; cpu++) {struct ct_pcpu *pcpu;
@@ -1171,7 +1173,6 @@ ctnetlink_dump_list(struct sk_buff *skb, pcpu =per_cpu_ptr (net->ct.pcpu_lists, cpu);spin_lock_bh (&pcpu->lock); - last = (struct nf_conn *)cb->args[1 ]; list = dying ? &pcpu->dying : &pcpu->unconfirmed; restart:hlist_nulls_for_each_entry (h, n, list, hnnode) {
@@ -1190,7 +1191,8 @@ restart: ct);rcu_read_unlock ();if (res <0 ) { -nf_conntrack_get (&ct->ct_general); +if (!atomic_inc_not_zero (&ct->ct_general.use)) +continue ; cb->args[0 ] = cpu; cb->args[1 ] = (unsigned long )ct;spin_unlock_bh (&pcpu->lock);
Greg Kroah-Hartman
maintainerPATCH 118/122 mmc: rtsx: add R1-no-CRC mmc command type handle
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Micky Ching <[email protected]>
commit 5027251eced6e34315a52bd841279df957f627bb upstream.
a27fbf2f067b0cd ("mmc: add ignorance case for CMD13 CRC error") produced a cmd.flags unhandled in realtek pci host driver. This will make MMC card fail to initialize, this patch is used to handle the new cmd.flags condition and MMC card can be used.
--- drivers/mmc/host/rtsx_pci_sdmmc.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/mmc/host/rtsx_pci_sdmmc.c +3 -0
@@ -236,6 +236,9 @@ static void sd_send_cmd_get_rsp(struct rcase MMC_RSP_R1 : rsp_type =SD_RSP_TYPE_R1 ;break ; +case MMC_RSP_R1 & ~MMC_RSP_CRC : + rsp_type =SD_RSP_TYPE_R1 |SD_NO_CHECK_CRC7 ; +break ;case MMC_RSP_R1B : rsp_type = SD_RSP_TYPE_R1b;break ;
Greg Kroah-Hartman
maintainerPATCH 119/122 drm/i915: fix display power sw state reporting
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Imre Deak <[email protected]>
commit b8c000d9bf23e7c1155ef421f595d1cbc25262da upstream.
Atm, we refcount both power domains and power wells and intel_display_power_enabled_sw() returns the power domain refcount. What the callers are really interested in though is the sw state of the underlying power wells. Due to this we will report incorrectly that a given power domain is off if its power wells were enabled via another power domain, for example POWER_DOMAIN_INIT which enables all power wells.
As a fix return instead the state based on the refcount of all power wells included in the passed in power domain.
References: https://bugs.freedesktop.org/show_bug.cgi?id=79505
References: https://bugs.freedesktop.org/show_bug.cgi?id=79038
--- drivers/gpu/drm/i915/intel_pm.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/intel_pm.c +16 -1
@@ -5313,10 +5313,25 @@ bool intel_display_power_enabled_sw(stru enum intel_display_power_domain domain) {struct i915_power_domains *power_domains; +struct i915_power_well *power_well; +bool is_enabled; +int i; + +if (dev_priv->pm.suspended) +return false; power_domains = &dev_priv->power_domains; + is_enabled = true; +for_each_power_well_rev (i, power_well,BIT (domain), power_domains) { +if (power_well->always_on) +continue ; -return power_domains->domain_use_count[domain]; +if (!power_well->count) { + is_enabled = false; +break ; + } + } +return is_enabled; } bool intel_display_power_enabled(struct drm_i915_private *dev_priv,
Greg Kroah-Hartman
maintainerPATCH 120/122 arch/unicore32/mm/alignment.c: include "asm/pgtable.h" to avoid compiling error
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chen Gang <[email protected]>
commit 1ff38c56cbd095c4c0dfa581a859ba3557830f78 upstream.
Need include "asm/pgtable.h" to include "asm-generic/pgtable-nopmd.h", so can let 'pmd_t' defined. The related error with allmodconfig:
CC arch/unicore32/mm/alignment.o
In file included from arch/unicore32/mm/alignment.c:24:
arch/unicore32/include/asm/tlbflush.h:135: error: expected .). before .*. token arch/unicore32/include/asm/tlbflush.h:154: error: expected .). before .*. token In file included from arch/unicore32/mm/alignment.c:27:
arch/unicore32/mm/mm.h:15: error: expected .=., .,., .;., .sm. or ._attribute__. before .*. token arch/unicore32/mm/mm.h:20: error: expected .=., .,., .;., .sm. or ._attribute__. before .*. token arch/unicore32/mm/mm.h:25: error: expected .=., .,., .;., .sm. or ._attribute__. before .*. token make[1]: *** [arch/unicore32/mm/alignment.o] Error 1
make: *** [arch/unicore32/mm] Error 2
--- arch/unicore32/mm/alignment.c | 1 + 1 file changed, 1 insertion(+)
arch/unicore32/mm/alignment.c +1 -0
@@ -21,6 +21,7 @@#include <linux/sched.h> #include <linux/uaccess.h> +#include <asm/pgtable.h> #include <asm/tlbflush.h> #include <asm/unaligned.h>
Greg Kroah-Hartman
maintainerPATCH 121/122 drivers/video/fbdev/fb-puv3.c: Add header files for function unifb_mmap
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zhichuang SUN <[email protected]>
commit fbc6c4a13bbfb420eedfdb26a0a859f9c07e8a7b upstream.
Function unifb_mmap calls functions which are defined in linux/mm.h and asm/pgtable.h
The related error (for unicore32 with unicore32_defconfig):
CC drivers/video/fbdev/fb-puv3.o
drivers/video/fbdev/fb-puv3.c: In function 'unifb_mmap':
drivers/video/fbdev/fb-puv3.c:646: error: implicit declaration of
function 'vm_iomap_memory'
drivers/video/fbdev/fb-puv3.c:646: error: implicit declaration of
function 'pgprot_noncached'
--- drivers/video/fbdev/fb-puv3.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/video/fbdev/fb-puv3.c +2 -0
@@ -18,8 +18,10 @@#include <linux/fb.h> #include <linux/init.h> #include <linux/console.h> +#include <linux/mm.h> #include <asm/sizes.h> +#include <asm/pgtable.h> #include <mach/hardware.h> /* Platform_data reserved for unifb registers. */
Greg Kroah-Hartman
maintainerPATCH 122/122 mm/numa: Remove BUG_ON() in __handle_mm_fault()
In reply to Greg Kroah-Hartman3.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Rik van Riel <[email protected]>
commit 107437febd495a50e2cd09c81bbaa84d30e57b07 upstream.
Changing PTEs and PMDs to pte_numa & pmd_numa is done with the mmap_sem held for reading, which means a pmd can be instantiated and turned into a numa one while __handle_mm_fault() is examining the value of old_pmd.
If that happens, __handle_mm_fault() should just return and let the page fault retry, instead of throwing an oops. This is handled by the test for pmd_trans_huge(*pmd) below.
Patrick McLean <[email protected]>
--- mm/memory.c | 3 --- 1 file changed, 3 deletions(-)
mm/memory.c +0 -3
@@ -3920,9 +3920,6 @@ static int __handle_mm_fault(struct mm_s } } -/* THP should already have been handled */ -BUG_ON (pmd_numa (*pmd)); -/* * Use __pte_alloc instead of pte_alloc_map, because we can't * run pte_offset_map on the pmd, if an huge pmd could
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 8 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 3.15.5 release. There are 122 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 Wed Jul 9 23:57:10 UTC 2014. Anything received after that time might be too late.
Build results:
total: 139 pass: 136 fail: 3
Failed builds:
powerpc:allmodconfig (binutils 2.23)
powerpc:allmodconfig (binutils 2.24)
unicore32:defconfig
Qemu tests all passed.
Results are as expected.
Details are available at http://server.roeck-us.net:8010/builders.
For the unicore32 build to pass in this release, the following additional patches would be needed.
73fa540618 drivers/rtc/rtc-puv3.c: remove "&dev->" for typo issue c863810cef drivers/rtc/rtc-puv3.c: use dev_dbg() instead of dev_debug() for typo issue
A workaround patch to address the powerpc build failures is pending. I'll send you a note after it is available if Ben doesn't Cc: stable.
Guenter
At Tue, 08 Jul 2014 06:26:21 -0700,
Quoted 25 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 3.15.5 release. There are 122 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 Wed Jul 9 23:57:10 UTC 2014. Anything received after that time might be too late.
Build results:
total: 139 pass: 136 fail: 3Failed builds:
powerpc:allmodconfig (binutils 2.23)
powerpc:allmodconfig (binutils 2.24)
unicore32:defconfigQemu tests all passed.
Results are as expected.
Details are available at http://server.roeck-us.net:8010/builders.
Quoted 8 lines
For the unicore32 build to pass in this release, the following additional patches would be needed.
73fa540618drivers/rtc/rtc-puv3.c: remove "&dev->" for typo issuec863810cefdrivers/rtc/rtc-puv3.c: usedev_dbg()instead ofdev_debug()for typo issueA workaround patch to address the powerpc build failures is pending. I'll send you a note after it is available if Ben doesn't Cc: stable.
3.15.5-rc1, 3.14.12-rc1, 3.10.48-rc1, and 3.4.98-rc1 passed my test. Build errors in 3.14.11 and 3.10.11-rc1 disappeared.
- Test Cases:
- Build this kernel.
- Boot this kernel.
- Build the latest mainline kernel with this kernel.
- Test Tool:
https://github.com/satoru-takeuchi/test-linux-stable
- Test Result (kernel .config, ktest config and test log):
http://satoru-takeuchi.org/test-linux-stable/results/<version>-<test datetime>.tar.xz
- Build Environment:
- OS: Debian Jessy x86_64
- CPU: Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz x 4
- memory: 8GB
- Test Target Environment:
- Debian Jessy x86_64 (KVM guest on the Build Environment)
- # of vCPU: 2
- memory: 2GB
Thanks,
Satoru
Quoted 3 lines
Guenter
Greg Kroah-Hartman
maintainerQuoted 55 lines · Satoru Takeuchi
At Tue, 08 Jul 2014 06:26:21 -0700,
Quoted 25 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 3.15.5 release. There are 122 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 Wed Jul 9 23:57:10 UTC 2014. Anything received after that time might be too late.
Build results:
total: 139 pass: 136 fail: 3Failed builds:
powerpc:allmodconfig (binutils 2.23)
powerpc:allmodconfig (binutils 2.24)
unicore32:defconfigQemu tests all passed.
Results are as expected.
Details are available at http://server.roeck-us.net:8010/builders.
Quoted 8 lines
For the unicore32 build to pass in this release, the following additional patches would be needed.
73fa540618drivers/rtc/rtc-puv3.c: remove "&dev->" for typo issuec863810cefdrivers/rtc/rtc-puv3.c: usedev_dbg()instead ofdev_debug()for typo issueA workaround patch to address the powerpc build failures is pending. I'll send you a note after it is available if Ben doesn't Cc: stable.
3.15.5-rc1, 3.14.12-rc1, 3.10.48-rc1, and 3.4.98-rc1 passed my test. Build errors in 3.14.11 and 3.10.11-rc1 disappeared.
- Test Cases:
- Build this kernel.
- Boot this kernel.
- Build the latest mainline kernel with this kernel.- Test Tool:
https://github.com/satoru-takeuchi/test-linux-stable- Test Result (kernel .config, ktest config and test log):
http://satoru-takeuchi.org/test-linux-stable/results/<version>-<test datetime>.tar.xz- Build Environment:
- OS: Debian Jessyx86_64
- CPU: Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz x 4
- memory: 8GB
Quoted 4 lines
- Test Target Environment:
- Debian Jessyx86_64(KVM guest on the Build Environment)
- # of vCPU: 2
- memory: 2GB
Thanks for testing these and letting me know.
greg k-h
Quoted 24 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 3.15.5 release. There are 122 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 Wed Jul 9 23:57:10 UTC 2014. Anything received after that time might be too late.
Build results:
total: 139 pass: 136 fail: 3Failed builds:
powerpc:allmodconfig (binutils 2.23)
powerpc:allmodconfig (binutils 2.24)
unicore32:defconfigQemu tests all passed.
Results are as expected.
Details are available at http://server.roeck-us.net:8010/builders.
Quoted 9 lines
For the unicore32 build to pass in this release, the following additional patches would be needed.
73fa540618drivers/rtc/rtc-puv3.c: remove "&dev->" for typo issuec863810cefdrivers/rtc/rtc-puv3.c: usedev_dbg()instead ofdev_debug()for typo issueA workaround patch to address the powerpc build failures is pending. I'll send you a note after it is available if Ben doesn't Cc: stable.
Greg,
The patch to fix the powerpc:allmodconfig build is now available upstream.
fb43e84 powerpc: Disable RELOCATABLE for COMPILE_TEST with PPC64
Please add this patch to 3.14.y and 3.15.y.
Thanks,
Guenter
Quoted 8 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 3.15.5 release. There are 122 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 Wed Jul 9 23:57:10 UTC 2014. Anything received after that time might be too late.
Quoted 8 lines
The whole patch series can be found in one patch at:
kernel.org/pub/linux/kernel/v3.0/stable-review/patch-3.15.5-rc1.gzand the diffstat can be found below.
thanks,
greg k-h
Compiled and booted on my test system. No demsg regressions.
-- Shuah
Signature · 4 lines
Shuah Khan Senior Linux Kernel Developer - Open Source Group Samsung Research America(Silicon Valley) [email protected] | (970) 672-0658