4.15.8-stable review

This is the start of the stable review cycle for the 4.15.8 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 Fri Mar 9 19:16:43 UTC 2018. Anything received after that time might be too late.

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

	https://www.kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.15.8-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.15.y
and the diffstat can be found below.

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

    Linux 4.15.8-rc1

NeilBrown <[email protected]>
md: only allow remove_and_add_spares when no sync_thread running.

Nicholas Piggin <[email protected]>
powerpc/64s/radix: Boot-time NULL pointer protection using a guard-PID

Adam Ford <[email protected]>
ARM: dts: LogicPD Torpedo: Fix I2C1 pinmux

Adam Ford <[email protected]>
ARM: dts: LogicPD SOM-LV: Fix I2C1 pinmux

Kai Heng Feng <[email protected]>
ACPI / bus: Parse tables as term_list for Dell XPS 9570 and Precision M5530

Eric Biggers <[email protected]>
KVM/x86: remove WARN_ON() for when vm_munmap() fails

Radim Krčmář <[email protected]>
KVM: x86: fix vcpu initialization with userspace lapic

Paolo Bonzini <[email protected]>
KVM/VMX: Optimize vmx_vcpu_run() and svm_vcpu_run() by marking the RDMSR path as unlikely()

Paolo Bonzini <[email protected]>
KVM: x86: move LAPIC initialization after VMCS creation

Paolo Bonzini <[email protected]>
KVM/x86: Remove indirect MSR op calls from SPEC_CTRL

Wanpeng Li <[email protected]>
KVM: mmu: Fix overlap between public and private memslots

Wanpeng Li <[email protected]>
KVM: X86: Fix SMRAM accessing even if VM is shutdown

Arnd Bergmann <[email protected]>
ARM: kvm: fix building with gcc-8

Ulf Magnusson <[email protected]>

    ARM: mvebu: Fix broken PL310_ERRATA_753970 selects

Daniel Schultz <[email protected]>
ARM: dts: rockchip: Remove 1.8 GHz operation point from phycore som

Arnd Bergmann <[email protected]>
ARM: orion: fix orion_ge00_switch_board_info initialization

Jan Beulich <[email protected]>

    x86/mm: Fix {pmd,pud}_{set,clear}_flags()

Rasmus Villemoes <[email protected]>
nospec: Allow index argument to have const-qualified type

David Hildenbrand <[email protected]>
KVM: s390: consider epoch index on TOD clock syncs

David Hildenbrand <[email protected]>
KVM: s390: consider epoch index on hotplugged CPUs

David Hildenbrand <[email protected]>
KVM: s390: provide only a single function for setting the tod (fix SCK)

David Hildenbrand <[email protected]>
KVM: s390: take care of clock-comparator sign control

Anna Karbownik <[email protected]>
EDAC, sb_edac: Fix out of bound writes during DIMM configuration on KNL

Mauro Carvalho Chehab <[email protected]>
media: m88ds3103: don't call a non-initalized function

Ming Lei <[email protected]>

    blk-mq: don't call io sched's .requeue_request when requeueing rq to ->dispatch

Yuchung Cheng <[email protected]>
tcp: revert F-RTO extension to detect more spurious timeouts

Yuchung Cheng <[email protected]>
tcp: revert F-RTO middle-box workaround

Julian Wiedmann <[email protected]>
s390/qeth: fix IPA command submission race

Julian Wiedmann <[email protected]>
s390/qeth: fix IP address lookup for L3 devices

Julian Wiedmann <[email protected]>
Revert "s390/qeth: fix using of ref counter for rxip addresses"

Julian Wiedmann <[email protected]>
s390/qeth: fix double-free on IP add/remove race

Julian Wiedmann <[email protected]>
s390/qeth: fix IP removal on offline cards

Julian Wiedmann <[email protected]>
s390/qeth: fix overestimated count of buffer elements

Julian Wiedmann <[email protected]>
s390/qeth: fix SETIP command handling

Ursula Braun <[email protected]>
s390/qeth: fix underestimated count of buffer elements

James Chapman <[email protected]>
l2tp: fix tunnel lookup use-after-free race

James Chapman <[email protected]>
l2tp: fix race in pppol2tp_release with session object destroy

James Chapman <[email protected]>
l2tp: fix races with tunnel socket close

James Chapman <[email protected]>
l2tp: don't use inet_shutdown on ppp session destroy

James Chapman <[email protected]>
l2tp: don't use inet_shutdown on tunnel destroy

Song Liu <[email protected]>
tcp: tracepoint: only call trace_tcp_send_reset with full socket

Andrew Lunn <[email protected]>
net: phy: Restore phy_resume() locking assumption

Vlad Buslov <[email protected]>
net/mlx5: Fix error handling when adding flow rules

Rahul Lakkireddy <[email protected]>
cxgb4: fix trailing zero in CIM LA dump

Jason Wang <[email protected]>
virtio-net: disable NAPI only when enabled during XDP set

Jason Wang <[email protected]>
tuntap: disable preemption during XDP processing

Jason Wang <[email protected]>
tuntap: correctly add the missing XDP flush

Soheil Hassas Yeganeh <[email protected]>
tcp: purge write queue upon RST

Jason A. Donenfeld <[email protected]>
netlink: put module reference if dump start fails

Ido Schimmel <[email protected]>
mlxsw: spectrum_router: Do not unconditionally clear route offload indication

Paolo Abeni <[email protected]>
cls_u32: fix use after free in u32_destroy_key()

Tom Lendacky <[email protected]>
amd-xgbe: Restore PCI interrupt enablement setting on resume

Boris Pismenny <[email protected]>

    tls: Use correct sk->sk_prot for IPV6

Eran Ben Elisha <[email protected]>
net/mlx5e: Verify inline header size do not exceed SKB linear size

Ido Schimmel <[email protected]>
bridge: Fix VLAN reference count problem

Alexey Kodanev <[email protected]>
sctp: fix dst refcnt leak in sctp_v6_get_dst()

David Ahern <[email protected]>
net: ipv4: Set addr_type in hash_keys for forwarded case

Jiri Pirko <[email protected]>
mlxsw: spectrum_router: Fix error path in mlxsw_sp_vr_create

Xin Long <[email protected]>
sctp: do not pr_err for the duplicated node in transport rhlist

Ivan Vecera <[email protected]>
net/sched: cls_u32: fix cls_u32 on filter replace

Eric Dumazet <[email protected]>
net_sched: gen_estimator: fix broken estimators based on percpu stats

Inbar Karmy <[email protected]>
net/mlx5e: Fix loopback self test when GRO is off

Tonghao Zhang <[email protected]>
doc: Change the min default value of tcp_wmem/tcp_rmem.

Eric Dumazet <[email protected]>
tcp_bbr: better deal with suboptimal GSO

David Howells <[email protected]>
rxrpc: Fix send in rxrpc_send_data_packet()

Ilya Lesokhin <[email protected]>
tcp: Honor the eor bit in tcp_mtu_probe

Heiner Kallweit <[email protected]>
net: phy: fix phy_start to consider PHY_IGNORE_INTERRUPT

Gal Pressman <[email protected]>
net/mlx5e: Specify numa node when allocating drop rq

Shalom Toledo <[email protected]>
mlxsw: spectrum_switchdev: Check success of FDB add operation

Tommi Rantala <[email protected]>
sctp: fix dst refcnt leak in sctp_v4_get_dst

Gal Pressman <[email protected]>
net/mlx5e: Fix TCP checksum in LRO buffers

Alexey Kodanev <[email protected]>
udplite: fix partial checksum initialization

Alexey Kodanev <[email protected]>
sctp: verify size of a new chunk in _sctp_make_chunk()

Guillaume Nault <[email protected]>
ppp: prevent unregistered channels from connecting to PPP units

Roman Kapl <[email protected]>
net: sched: report if filter is too large to dump

Nicolas Dichtel <[email protected]>
netlink: ensure to loop over all netns in genlmsg_multicast_allns()

Sabrina Dubroca <[email protected]>
net: ipv4: don't allow setting net.ipv4.route.min_pmtu below 68

Jakub Kicinski <[email protected]>
net: fix race on decreasing number of TX queues

Grygorii Strashko <[email protected]>
net: ethernet: ti: cpsw: fix net watchdog timeout

Wolfram Sang <[email protected]>
net: amd-xgbe: fix comparison to bitshift when dealing with a mask

Arnd Bergmann <[email protected]>
ipv6 sit: work around bogus gcc-8 -Wrestrict warning

Denis Du <[email protected]>
hdlc_ppp: carrier detect ok, don't turn off negotiation

Stefano Brivio <[email protected]>
fib_semantics: Don't match route with mismatching tclassid

Xin Long <[email protected]>
bridge: check brport attr show in brport_show

Thomas Gleixner <[email protected]>
x86/cpu_entry_area: Sync cpu_entry_area to initial_page_table

Sebastian Panceac <[email protected]>
x86/platform/intel-mid: Handle Intel Edison reboot correctly

Juergen Gross <[email protected]>
x86/xen: Zero MSR_IA32_SPEC_CTRL before suspend

Jan Kara <[email protected]>
direct-io: Fix sleep in atomic due to sync AIO

Dan Williams <[email protected]>
dax: fix vma_is_fsdax() helper

Viresh Kumar <[email protected]>
cpufreq: s3c24xx: Fix broken s3c_cpufreq_init()

Dan Williams <[email protected]>
vfio: disable filesystem-dax page pinning

Ming Lei <[email protected]>
block: pass inclusive 'lend' parameter to truncate_inode_pages_range

Ming Lei <[email protected]>
block: kyber: fix domain token leak during requeue

Jiufei Xue <[email protected]>
block: fix the count of PGPGOUT for WRITE_SAME

Anand Jain <[email protected]>
btrfs: use proper endianness accessors for super_copy

Helge Deller <[email protected]>
parisc: Hide virtual kernel memory layout

John David Anglin <[email protected]>
parisc: Fix ordering of cache and TLB flushes

Helge Deller <[email protected]>
parisc: Reduce irq overhead when run in qemu

Helge Deller <[email protected]>
parisc: Use cr16 interval timers unconditionally on qemu

Lingutla Chandrasekhar <[email protected]>
timers: Forward timer base before migrating timers

Shawn Lin <[email protected]>
mmc: dw_mmc: Fix out-of-bounds access for slot's caps

Shawn Lin <[email protected]>
mmc: dw_mmc: Factor out dw_mci_init_slot_caps

Shawn Lin <[email protected]>
mmc: dw_mmc: Avoid accessing registers in runtime suspended state

Geert Uytterhoeven <[email protected]>
mmc: dw_mmc-k3: Fix out-of-bounds access through DT alias

Adrian Hunter <[email protected]>
mmc: sdhci-pci: Fix S0i3 for Intel BYT-based controllers

Takashi Iwai <[email protected]>
ALSA: hda - Fix pincfg at resume on Lenovo T470 dock

Hans de Goede <[email protected]>
ALSA: hda: Add a power_save blacklist

Takashi Iwai <[email protected]>
ALSA: x86: Fix missing spinlock and mutex initializations

Richard Fitzgerald <[email protected]>
ALSA: control: Fix memory corruption risk in snd_ctl_elem_read

Erik Veijola <[email protected]>
ALSA: usb-audio: Add a quirck for B&W PX headphones

Jeremy Boone <[email protected]>
tpm_tis: fix potential buffer overruns caused by bit glitches on the bus

Jeremy Boone <[email protected]>
tpm_i2c_nuvoton: fix potential buffer overruns caused by bit glitches on the bus

Jeremy Boone <[email protected]>
tpm_i2c_infineon: fix potential buffer overruns caused by bit glitches on the bus

Jeremy Boone <[email protected]>
tpm: fix potential buffer overruns caused by bit glitches on the bus

Jeremy Boone <[email protected]>
tpm: st33zp24: fix potential buffer overruns caused by bit glitches on the bus

Emil Tantilov <[email protected]>
ixgbe: fix crash in build_skb Rx code path

Hans de Goede <[email protected]>
Bluetooth: btusb: Use DMI matching for QCA reset_resume quirking

Sam Bobroff <[email protected]>
powerpc/pseries: Enable RAS hotplug events later

Mario Limonciello <[email protected]>
platform/x86: dell-laptop: Allocate buffer on heap rather than globally

Corey Minyard <[email protected]>
ipmi_si: Fix error handling of platform device

Anna-Maria Gleixner <[email protected]>
hrtimer: Ensure POSIX compliance (relative CLOCK_REALTIME hrtimers)

Adam Borowski <[email protected]>
vsprintf: avoid misleading "(null)" for %px

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

Diffstat:

 Documentation/networking/ip-sysctl.txt             |   4 +-
 Makefile                                           |   4 +-
 arch/arm/boot/dts/logicpd-som-lv.dtsi              |   9 +-
 arch/arm/boot/dts/logicpd-torpedo-som.dtsi         |   8 +
 arch/arm/boot/dts/rk3288-phycore-som.dtsi          |  20 ---
 arch/arm/kvm/hyp/Makefile                          |   5 +
 arch/arm/kvm/hyp/banked-sr.c                       |   4 +
 arch/arm/mach-mvebu/Kconfig                        |   4 +-
 arch/arm/plat-orion/common.c                       |  23 ++-
 arch/parisc/include/asm/cacheflush.h               |   1 +
 arch/parisc/include/asm/processor.h                |   2 +
 arch/parisc/kernel/cache.c                         |  57 ++++---
 arch/parisc/kernel/pacache.S                       |  22 +++
 arch/parisc/kernel/time.c                          |  11 +-
 arch/parisc/mm/init.c                              |   7 +-
 arch/powerpc/mm/pgtable-radix.c                    |  20 +++
 arch/powerpc/platforms/pseries/ras.c               |  31 +++-
 arch/s390/kvm/interrupt.c                          |  25 ++-
 arch/s390/kvm/kvm-s390.c                           |  79 +++++----
 arch/s390/kvm/kvm-s390.h                           |   5 +-
 arch/s390/kvm/priv.c                               |   9 +-
 arch/x86/include/asm/pgtable.h                     |   8 +-
 arch/x86/include/asm/pgtable_32.h                  |   1 +
 arch/x86/include/asm/pgtable_64.h                  |   1 +
 arch/x86/include/asm/pgtable_types.h               |  10 ++
 arch/x86/kernel/setup.c                            |  17 +-
 arch/x86/kernel/setup_percpu.c                     |  17 +-
 arch/x86/kvm/lapic.c                               |  11 +-
 arch/x86/kvm/mmu.c                                 |   2 +-
 arch/x86/kvm/svm.c                                 |   9 +-
 arch/x86/kvm/vmx.c                                 |   9 +-
 arch/x86/kvm/x86.c                                 |   8 +-
 arch/x86/mm/cpu_entry_area.c                       |   6 +
 arch/x86/mm/init_32.c                              |  15 ++
 arch/x86/platform/intel-mid/intel-mid.c            |   2 +-
 arch/x86/xen/suspend.c                             |  16 ++
 block/blk-core.c                                   |   2 +-
 block/blk-mq.c                                     |   4 +-
 block/ioctl.c                                      |   2 +-
 block/kyber-iosched.c                              |   1 +
 drivers/acpi/bus.c                                 |  38 ++++-
 drivers/bluetooth/btusb.c                          |  25 ++-
 drivers/char/ipmi/ipmi_si_intf.c                   |   9 +-
 drivers/char/tpm/st33zp24/st33zp24.c               |   4 +-
 drivers/char/tpm/tpm-interface.c                   |   4 +
 drivers/char/tpm/tpm2-cmd.c                        |   4 +
 drivers/char/tpm/tpm_i2c_infineon.c                |   5 +-
 drivers/char/tpm/tpm_i2c_nuvoton.c                 |   8 +-
 drivers/char/tpm/tpm_tis_core.c                    |   5 +-
 drivers/cpufreq/s3c24xx-cpufreq.c                  |   8 +-
 drivers/edac/sb_edac.c                             |   2 +-
 drivers/md/md.c                                    |   4 +
 drivers/media/dvb-frontends/m88ds3103.c            |   7 +-
 drivers/mmc/host/dw_mmc-exynos.c                   |   1 +
 drivers/mmc/host/dw_mmc-k3.c                       |   4 +
 drivers/mmc/host/dw_mmc-rockchip.c                 |   1 +
 drivers/mmc/host/dw_mmc-zx.c                       |   1 +
 drivers/mmc/host/dw_mmc.c                          |  84 +++++----
 drivers/mmc/host/dw_mmc.h                          |   2 +
 drivers/mmc/host/sdhci-pci-core.c                  |  35 +++-
 drivers/net/ethernet/amd/xgbe/xgbe-drv.c           |   2 +-
 drivers/net/ethernet/amd/xgbe/xgbe-pci.c           |   2 +
 drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c     |   2 +-
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c   |   2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c      |   8 +
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  10 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    |  49 ++++--
 .../net/ethernet/mellanox/mlx5/core/en_selftest.c  |   3 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_tx.c    |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  |  10 +-
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  |  35 ++--
 .../ethernet/mellanox/mlxsw/spectrum_switchdev.c   |  29 +++-
 drivers/net/ethernet/ti/cpsw.c                     |  16 +-
 drivers/net/phy/phy.c                              |   4 +-
 drivers/net/phy/phy_device.c                       |  18 +-
 drivers/net/ppp/ppp_generic.c                      |   9 +
 drivers/net/tun.c                                  |   7 +
 drivers/net/virtio_net.c                           |   8 +-
 drivers/net/wan/hdlc_ppp.c                         |   5 +-
 drivers/platform/x86/dell-laptop.c                 | 188 +++++++++++----------
 drivers/s390/net/qeth_core.h                       |   7 +-
 drivers/s390/net/qeth_core_main.c                  |  43 ++---
 drivers/s390/net/qeth_l3.h                         |  34 +++-
 drivers/s390/net/qeth_l3_main.c                    | 123 ++++++--------
 drivers/vfio/vfio_iommu_type1.c                    |  18 +-
 fs/btrfs/sysfs.c                                   |   8 +-
 fs/btrfs/transaction.c                             |  20 ++-
 fs/direct-io.c                                     |   3 +-
 include/linux/fs.h                                 |   2 +-
 include/linux/nospec.h                             |   3 +-
 include/linux/phy.h                                |   1 +
 include/net/udplite.h                              |   1 +
 kernel/time/hrtimer.c                              |   7 +-
 kernel/time/timer.c                                |   6 +
 lib/vsprintf.c                                     |   2 +-
 net/bridge/br_sysfs_if.c                           |   3 +
 net/bridge/br_vlan.c                               |   2 +
 net/core/dev.c                                     |  11 +-
 net/core/gen_estimator.c                           |   1 +
 net/ipv4/fib_semantics.c                           |   5 +
 net/ipv4/route.c                                   |  10 +-
 net/ipv4/tcp_input.c                               |  24 +--
 net/ipv4/tcp_ipv4.c                                |   3 +-
 net/ipv4/tcp_output.c                              |  34 +++-
 net/ipv4/udp.c                                     |   5 +
 net/ipv6/ip6_checksum.c                            |   5 +
 net/ipv6/sit.c                                     |   2 +-
 net/ipv6/tcp_ipv6.c                                |   3 +-
 net/l2tp/l2tp_core.c                               | 142 +++++-----------
 net/l2tp/l2tp_core.h                               |  23 +--
 net/l2tp/l2tp_ip.c                                 |  10 +-
 net/l2tp/l2tp_ip6.c                                |   8 +-
 net/l2tp/l2tp_ppp.c                                |  60 +++----
 net/netlink/af_netlink.c                           |   4 +-
 net/netlink/genetlink.c                            |  12 +-
 net/rxrpc/output.c                                 |   2 +-
 net/sched/cls_api.c                                |   7 +-
 net/sched/cls_u32.c                                |  24 +--
 net/sctp/input.c                                   |   5 +-
 net/sctp/ipv6.c                                    |  10 +-
 net/sctp/protocol.c                                |  10 +-
 net/sctp/sm_make_chunk.c                           |   7 +-
 net/tls/tls_main.c                                 |  52 ++++--
 sound/core/control.c                               |   2 +-
 sound/pci/hda/hda_intel.c                          |  38 ++++-
 sound/pci/hda/patch_realtek.c                      |   3 +-
 sound/usb/quirks-table.h                           |  47 ++++++
 sound/x86/intel_hdmi_audio.c                       |   2 +
 virt/kvm/kvm_main.c                                |   3 +-
 129 files changed, 1271 insertions(+), 737 deletions(-)

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

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

From: Adam Borowski <[email protected]>

commit 3a129cc2151425e5aeb69aeb25fbc994ec738137 upstream.

Like %pK already does, print "00000000" instead.

This confused people -- the convention is that "(null)" means you tried to dereference a null pointer as opposed to printing the address.

To: Sergey Senozhatsky <[email protected]>
To: Steven Rostedt <[email protected]>
To: [email protected]

---
 lib/vsprintf.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1849,7 +1849,7 @@ char *pointer(const char *fmt, char *buf
 {
 	const int default_width = 2 * sizeof(void *);
 
-	if (!ptr && *fmt != 'K') {
+	if (!ptr && *fmt != 'K' && *fmt != 'x') {
 		/*
 		 * Print (null) with the same width as a pointer so it makes
 		 * tabular output look nice.

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

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

From: Anna-Maria Gleixner <[email protected]>

commit 48d0c9becc7f3c66874c100c126459a9da0fdced upstream.

The POSIX specification defines that relative CLOCK_REALTIME timers are not affected by clock modifications. Those timers have to use CLOCK_MONOTONIC to ensure POSIX compliance.

The introduction of the additional HRTIMER_MODE_PINNED mode broke this requirement for pinned timers.

There is no user space visible impact because user space timers are not using pinned mode, but for consistency reasons this needs to be fixed.

Check whether the mode has the HRTIMER_MODE_REL bit set instead of comparing with HRTIMER_MODE_ABS.

---
 kernel/time/hrtimer.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

@@ -1104,7 +1104,12 @@ static void __hrtimer_init(struct hrtime
 
 	cpu_base = raw_cpu_ptr(&hrtimer_bases);
 
-	if (clock_id == CLOCK_REALTIME && mode != HRTIMER_MODE_ABS)
+	/*
+	 * POSIX magic: Relative CLOCK_REALTIME timers are not affected by
+	 * clock modifications, so they needs to become CLOCK_MONOTONIC to
+	 * ensure POSIX compliance.
+	 */
+	if (clock_id == CLOCK_REALTIME && mode & HRTIMER_MODE_REL)
 		clock_id = CLOCK_MONOTONIC;
 
 	base = hrtimer_clockid_to_base(clock_id);

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

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

From: Corey Minyard <[email protected]>

commit 174134ac760275457bb0d1560a0dbe6cf8a12ad6 upstream.

Cleanup of platform devices created by the IPMI driver was not being done correctly and could result in a memory leak. So create a local boolean to know how to clean up those platform devices.

---
 drivers/char/ipmi/ipmi_si_intf.c |    9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

@@ -2045,6 +2045,7 @@ static int try_smi_init(struct smi_info
 	int rv = 0;
 	int i;
 	char *init_name = NULL;
+	bool platform_device_registered = false;
 
 	pr_info(PFX "Trying %s-specified %s state machine at %s address 0x%lx, slave address 0x%x, irq %d\n",
 		ipmi_addr_src_to_str(new_smi->io.addr_source),

@@ -2173,6 +2174,7 @@ static int try_smi_init(struct smi_info
 				rv);
 			goto out_err;
 		}
+		platform_device_registered = true;
 	}
 
 	dev_set_drvdata(new_smi->io.dev, new_smi);

@@ -2279,10 +2281,11 @@ out_err:
 	}
 
 	if (new_smi->pdev) {
-		platform_device_unregister(new_smi->pdev);
+		if (platform_device_registered)
+			platform_device_unregister(new_smi->pdev);
+		else
+			platform_device_put(new_smi->pdev);
 		new_smi->pdev = NULL;
-	} else if (new_smi->pdev) {
-		platform_device_put(new_smi->pdev);
 	}
 
 	kfree(init_name);

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

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

From: Mario Limonciello <[email protected]>

commit 9862b43624a5450a097cc4122732857b869dbbca upstream.

There is no longer a need for the buffer to be defined in first 4GB physical address space.

Furthermore there may be race conditions with multiple different functions working on a module wide buffer causing incorrect results.

---
 drivers/platform/x86/dell-laptop.c |  188 ++++++++++++++++++++-----------------
 1 file changed, 103 insertions(+), 85 deletions(-)

@@ -78,7 +78,6 @@ static struct platform_driver platform_d
 	}
 };
 
-static struct calling_interface_buffer *buffer;
 static struct platform_device *platform_device;
 static struct backlight_device *dell_backlight_device;
 static struct rfkill *wifi_rfkill;

@@ -286,7 +285,8 @@ static const struct dmi_system_id dell_q
 	{ }
 };
 
-void dell_set_arguments(u32 arg0, u32 arg1, u32 arg2, u32 arg3)
+static void dell_fill_request(struct calling_interface_buffer *buffer,
+			      u32 arg0, u32 arg1, u32 arg2, u32 arg3)
 {
 	memset(buffer, 0, sizeof(struct calling_interface_buffer));
 	buffer->input[0] = arg0;

@@ -295,7 +295,8 @@ void dell_set_arguments(u32 arg0, u32 ar
 	buffer->input[3] = arg3;
 }
 
-int dell_send_request(u16 class, u16 select)
+static int dell_send_request(struct calling_interface_buffer *buffer,
+			     u16 class, u16 select)
 {
 	int ret;
 

@@ -432,21 +433,22 @@ static int dell_rfkill_set(void *data, b
 	int disable = blocked ? 1 : 0;
 	unsigned long radio = (unsigned long)data;
 	int hwswitch_bit = (unsigned long)data - 1;
+	struct calling_interface_buffer buffer;
 	int hwswitch;
 	int status;
 	int ret;
 
-	dell_set_arguments(0, 0, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
+	dell_fill_request(&buffer, 0, 0, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
 	if (ret)
 		return ret;
-	status = buffer->output[1];
+	status = buffer.output[1];
 
-	dell_set_arguments(0x2, 0, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
+	dell_fill_request(&buffer, 0x2, 0, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
 	if (ret)
 		return ret;
-	hwswitch = buffer->output[1];
+	hwswitch = buffer.output[1];
 
 	/* If the hardware switch controls this radio, and the hardware
 	   switch is disabled, always disable the radio */

@@ -454,8 +456,8 @@ static int dell_rfkill_set(void *data, b
 	    (status & BIT(0)) && !(status & BIT(16)))
 		disable = 1;
 
-	dell_set_arguments(1 | (radio<<8) | (disable << 16), 0, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
+	dell_fill_request(&buffer, 1 | (radio<<8) | (disable << 16), 0, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
 	return ret;
 }
 

@@ -464,9 +466,11 @@ static void dell_rfkill_update_sw_state(
 {
 	if (status & BIT(0)) {
 		/* Has hw-switch, sync sw_state to BIOS */
+		struct calling_interface_buffer buffer;
 		int block = rfkill_blocked(rfkill);
-		dell_set_arguments(1 | (radio << 8) | (block << 16), 0, 0, 0);
-		dell_send_request(CLASS_INFO, SELECT_RFKILL);
+		dell_fill_request(&buffer,
+				   1 | (radio << 8) | (block << 16), 0, 0, 0);
+		dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
 	} else {
 		/* No hw-switch, sync BIOS state to sw_state */
 		rfkill_set_sw_state(rfkill, !!(status & BIT(radio + 16)));

@@ -483,21 +487,22 @@ static void dell_rfkill_update_hw_state(
 static void dell_rfkill_query(struct rfkill *rfkill, void *data)
 {
 	int radio = ((unsigned long)data & 0xF);
+	struct calling_interface_buffer buffer;
 	int hwswitch;
 	int status;
 	int ret;
 
-	dell_set_arguments(0, 0, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
-	status = buffer->output[1];
+	dell_fill_request(&buffer, 0, 0, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
+	status = buffer.output[1];
 
 	if (ret != 0 || !(status & BIT(0))) {
 		return;
 	}
 
-	dell_set_arguments(0, 0x2, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
-	hwswitch = buffer->output[1];
+	dell_fill_request(&buffer, 0, 0x2, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
+	hwswitch = buffer.output[1];
 
 	if (ret != 0)
 		return;

@@ -514,22 +519,23 @@ static struct dentry *dell_laptop_dir;
 
 static int dell_debugfs_show(struct seq_file *s, void *data)
 {
+	struct calling_interface_buffer buffer;
 	int hwswitch_state;
 	int hwswitch_ret;
 	int status;
 	int ret;
 
-	dell_set_arguments(0, 0, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
+	dell_fill_request(&buffer, 0, 0, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
 	if (ret)
 		return ret;
-	status = buffer->output[1];
+	status = buffer.output[1];
 
-	dell_set_arguments(0, 0x2, 0, 0);
-	hwswitch_ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
+	dell_fill_request(&buffer, 0, 0x2, 0, 0);
+	hwswitch_ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
 	if (hwswitch_ret)
 		return hwswitch_ret;
-	hwswitch_state = buffer->output[1];
+	hwswitch_state = buffer.output[1];
 
 	seq_printf(s, "return:\t%d\n", ret);
 	seq_printf(s, "status:\t0x%X\n", status);

@@ -610,22 +616,23 @@ static const struct file_operations dell
 
 static void dell_update_rfkill(struct work_struct *ignored)
 {
+	struct calling_interface_buffer buffer;
 	int hwswitch = 0;
 	int status;
 	int ret;
 
-	dell_set_arguments(0, 0, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
-	status = buffer->output[1];
+	dell_fill_request(&buffer, 0, 0, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
+	status = buffer.output[1];
 
 	if (ret != 0)
 		return;
 
-	dell_set_arguments(0, 0x2, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
+	dell_fill_request(&buffer, 0, 0x2, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
 
 	if (ret == 0 && (status & BIT(0)))
-		hwswitch = buffer->output[1];
+		hwswitch = buffer.output[1];
 
 	if (wifi_rfkill) {
 		dell_rfkill_update_hw_state(wifi_rfkill, 1, status, hwswitch);

@@ -683,6 +690,7 @@ static struct notifier_block dell_laptop
 
 static int __init dell_setup_rfkill(void)
 {
+	struct calling_interface_buffer buffer;
 	int status, ret, whitelisted;
 	const char *product;
 

@@ -698,9 +706,9 @@ static int __init dell_setup_rfkill(void
 	if (!force_rfkill && !whitelisted)
 		return 0;
 
-	dell_set_arguments(0, 0, 0, 0);
-	ret = dell_send_request(CLASS_INFO, SELECT_RFKILL);
-	status = buffer->output[1];
+	dell_fill_request(&buffer, 0, 0, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_INFO, SELECT_RFKILL);
+	status = buffer.output[1];
 
 	/* dell wireless info smbios call is not supported */
 	if (ret != 0)

@@ -853,6 +861,7 @@ static void dell_cleanup_rfkill(void)
 
 static int dell_send_intensity(struct backlight_device *bd)
 {
+	struct calling_interface_buffer buffer;
 	struct calling_interface_token *token;
 	int ret;
 

@@ -860,17 +869,21 @@ static int dell_send_intensity(struct ba
 	if (!token)
 		return -ENODEV;
 
-	dell_set_arguments(token->location, bd->props.brightness, 0, 0);
+	dell_fill_request(&buffer,
+			   token->location, bd->props.brightness, 0, 0);
 	if (power_supply_is_system_supplied() > 0)
-		ret = dell_send_request(CLASS_TOKEN_WRITE, SELECT_TOKEN_AC);
+		ret = dell_send_request(&buffer,
+					CLASS_TOKEN_WRITE, SELECT_TOKEN_AC);
 	else
-		ret = dell_send_request(CLASS_TOKEN_WRITE, SELECT_TOKEN_BAT);
+		ret = dell_send_request(&buffer,
+					CLASS_TOKEN_WRITE, SELECT_TOKEN_BAT);
 
 	return ret;
 }
 
 static int dell_get_intensity(struct backlight_device *bd)
 {
+	struct calling_interface_buffer buffer;
 	struct calling_interface_token *token;
 	int ret;
 

@@ -878,14 +891,17 @@ static int dell_get_intensity(struct bac
 	if (!token)
 		return -ENODEV;
 
-	dell_set_arguments(token->location, 0, 0, 0);
+	dell_fill_request(&buffer, token->location, 0, 0, 0);
 	if (power_supply_is_system_supplied() > 0)
-		ret = dell_send_request(CLASS_TOKEN_READ, SELECT_TOKEN_AC);
+		ret = dell_send_request(&buffer,
+					CLASS_TOKEN_READ, SELECT_TOKEN_AC);
 	else
-		ret = dell_send_request(CLASS_TOKEN_READ, SELECT_TOKEN_BAT);
+		ret = dell_send_request(&buffer,
+					CLASS_TOKEN_READ, SELECT_TOKEN_BAT);
 
 	if (ret == 0)
-		ret = buffer->output[1];
+		ret = buffer.output[1];
+
 	return ret;
 }
 

@@ -1149,31 +1165,33 @@ static DEFINE_MUTEX(kbd_led_mutex);
 
 static int kbd_get_info(struct kbd_info *info)
 {
+	struct calling_interface_buffer buffer;
 	u8 units;
 	int ret;
 
-	dell_set_arguments(0, 0, 0, 0);
-	ret = dell_send_request(CLASS_KBD_BACKLIGHT, SELECT_KBD_BACKLIGHT);
+	dell_fill_request(&buffer, 0, 0, 0, 0);
+	ret = dell_send_request(&buffer,
+				CLASS_KBD_BACKLIGHT, SELECT_KBD_BACKLIGHT);
 	if (ret)
 		return ret;
 
-	info->modes = buffer->output[1] & 0xFFFF;
-	info->type = (buffer->output[1] >> 24) & 0xFF;
-	info->triggers = buffer->output[2] & 0xFF;
-	units = (buffer->output[2] >> 8) & 0xFF;
-	info->levels = (buffer->output[2] >> 16) & 0xFF;
+	info->modes = buffer.output[1] & 0xFFFF;
+	info->type = (buffer.output[1] >> 24) & 0xFF;
+	info->triggers = buffer.output[2] & 0xFF;
+	units = (buffer.output[2] >> 8) & 0xFF;
+	info->levels = (buffer.output[2] >> 16) & 0xFF;
 
 	if (quirks && quirks->kbd_led_levels_off_1 && info->levels)
 		info->levels--;
 
 	if (units & BIT(0))
-		info->seconds = (buffer->output[3] >> 0) & 0xFF;
+		info->seconds = (buffer.output[3] >> 0) & 0xFF;
 	if (units & BIT(1))
-		info->minutes = (buffer->output[3] >> 8) & 0xFF;
+		info->minutes = (buffer.output[3] >> 8) & 0xFF;
 	if (units & BIT(2))
-		info->hours = (buffer->output[3] >> 16) & 0xFF;
+		info->hours = (buffer.output[3] >> 16) & 0xFF;
 	if (units & BIT(3))
-		info->days = (buffer->output[3] >> 24) & 0xFF;
+		info->days = (buffer.output[3] >> 24) & 0xFF;
 
 	return ret;
 }

@@ -1233,31 +1251,34 @@ static int kbd_set_level(struct kbd_stat
 
 static int kbd_get_state(struct kbd_state *state)
 {
+	struct calling_interface_buffer buffer;
 	int ret;
 
-	dell_set_arguments(0x1, 0, 0, 0);
-	ret = dell_send_request(CLASS_KBD_BACKLIGHT, SELECT_KBD_BACKLIGHT);
+	dell_fill_request(&buffer, 0, 0, 0, 0);
+	ret = dell_send_request(&buffer,
+				CLASS_KBD_BACKLIGHT, SELECT_KBD_BACKLIGHT);
 	if (ret)
 		return ret;
 
-	state->mode_bit = ffs(buffer->output[1] & 0xFFFF);
+	state->mode_bit = ffs(buffer.output[1] & 0xFFFF);
 	if (state->mode_bit != 0)
 		state->mode_bit--;
 
-	state->triggers = (buffer->output[1] >> 16) & 0xFF;
-	state->timeout_value = (buffer->output[1] >> 24) & 0x3F;
-	state->timeout_unit = (buffer->output[1] >> 30) & 0x3;
-	state->als_setting = buffer->output[2] & 0xFF;
-	state->als_value = (buffer->output[2] >> 8) & 0xFF;
-	state->level = (buffer->output[2] >> 16) & 0xFF;
-	state->timeout_value_ac = (buffer->output[2] >> 24) & 0x3F;
-	state->timeout_unit_ac = (buffer->output[2] >> 30) & 0x3;
+	state->triggers = (buffer.output[1] >> 16) & 0xFF;
+	state->timeout_value = (buffer.output[1] >> 24) & 0x3F;
+	state->timeout_unit = (buffer.output[1] >> 30) & 0x3;
+	state->als_setting = buffer.output[2] & 0xFF;
+	state->als_value = (buffer.output[2] >> 8) & 0xFF;
+	state->level = (buffer.output[2] >> 16) & 0xFF;
+	state->timeout_value_ac = (buffer.output[2] >> 24) & 0x3F;
+	state->timeout_unit_ac = (buffer.output[2] >> 30) & 0x3;
 
 	return ret;
 }
 
 static int kbd_set_state(struct kbd_state *state)
 {
+	struct calling_interface_buffer buffer;
 	int ret;
 	u32 input1;
 	u32 input2;

@@ -1270,8 +1291,9 @@ static int kbd_set_state(struct kbd_stat
 	input2 |= (state->level & 0xFF) << 16;
 	input2 |= (state->timeout_value_ac & 0x3F) << 24;
 	input2 |= (state->timeout_unit_ac & 0x3) << 30;
-	dell_set_arguments(0x2, input1, input2, 0);
-	ret = dell_send_request(CLASS_KBD_BACKLIGHT, SELECT_KBD_BACKLIGHT);
+	dell_fill_request(&buffer, 0x2, input1, input2, 0);
+	ret = dell_send_request(&buffer,
+				CLASS_KBD_BACKLIGHT, SELECT_KBD_BACKLIGHT);
 
 	return ret;
 }

@@ -1298,6 +1320,7 @@ static int kbd_set_state_safe(struct kbd
 
 static int kbd_set_token_bit(u8 bit)
 {
+	struct calling_interface_buffer buffer;
 	struct calling_interface_token *token;
 	int ret;
 

@@ -1308,14 +1331,15 @@ static int kbd_set_token_bit(u8 bit)
 	if (!token)
 		return -EINVAL;
 
-	dell_set_arguments(token->location, token->value, 0, 0);
-	ret = dell_send_request(CLASS_TOKEN_WRITE, SELECT_TOKEN_STD);
+	dell_fill_request(&buffer, token->location, token->value, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_TOKEN_WRITE, SELECT_TOKEN_STD);
 
 	return ret;
 }
 
 static int kbd_get_token_bit(u8 bit)
 {
+	struct calling_interface_buffer buffer;
 	struct calling_interface_token *token;
 	int ret;
 	int val;

@@ -1327,9 +1351,9 @@ static int kbd_get_token_bit(u8 bit)
 	if (!token)
 		return -EINVAL;
 
-	dell_set_arguments(token->location, 0, 0, 0);
-	ret = dell_send_request(CLASS_TOKEN_READ, SELECT_TOKEN_STD);
-	val = buffer->output[1];
+	dell_fill_request(&buffer, token->location, 0, 0, 0);
+	ret = dell_send_request(&buffer, CLASS_TOKEN_READ, SELECT_TOKEN_STD);
+	val = buffer.output[1];
 
 	if (ret)
 		return ret;

@@ -2046,6 +2070,7 @@ static struct notifier_block dell_laptop
 
 int dell_micmute_led_set(int state)
 {
+	struct calling_interface_buffer buffer;
 	struct calling_interface_token *token;
 
 	if (state == 0)

@@ -2058,8 +2083,8 @@ int dell_micmute_led_set(int state)
 	if (!token)
 		return -ENODEV;
 
-	dell_set_arguments(token->location, token->value, 0, 0);
-	dell_send_request(CLASS_TOKEN_WRITE, SELECT_TOKEN_STD);
+	dell_fill_request(&buffer, token->location, token->value, 0, 0);
+	dell_send_request(&buffer, CLASS_TOKEN_WRITE, SELECT_TOKEN_STD);
 
 	return state;
 }

@@ -2090,13 +2115,6 @@ static int __init dell_init(void)
 	if (ret)
 		goto fail_platform_device2;
 
-	buffer = kzalloc(sizeof(struct calling_interface_buffer), GFP_KERNEL);
-	if (!buffer) {
-		ret = -ENOMEM;
-		goto fail_buffer;
-	}
-
-
 	ret = dell_setup_rfkill();
 
 	if (ret) {

@@ -2121,10 +2139,13 @@ static int __init dell_init(void)
 
 	token = dell_smbios_find_token(BRIGHTNESS_TOKEN);
 	if (token) {
-		dell_set_arguments(token->location, 0, 0, 0);
-		ret = dell_send_request(CLASS_TOKEN_READ, SELECT_TOKEN_AC);
+		struct calling_interface_buffer buffer;
+
+		dell_fill_request(&buffer, token->location, 0, 0, 0);
+		ret = dell_send_request(&buffer,
+					CLASS_TOKEN_READ, SELECT_TOKEN_AC);
 		if (ret)
-			max_intensity = buffer->output[3];
+			max_intensity = buffer.output[3];
 	}
 
 	if (max_intensity) {

@@ -2158,8 +2179,6 @@ static int __init dell_init(void)
 fail_get_brightness:
 	backlight_device_unregister(dell_backlight_device);
 fail_backlight:
-	kfree(buffer);
-fail_buffer:
 	dell_cleanup_rfkill();
 fail_rfkill:
 	platform_device_del(platform_device);

@@ -2179,7 +2198,6 @@ static void __exit dell_exit(void)
 		touchpad_led_exit();
 	kbd_led_exit();
 	backlight_device_unregister(dell_backlight_device);
-	kfree(buffer);
 	dell_cleanup_rfkill();
 	if (platform_device) {
 		platform_device_unregister(platform_device);

Greg,

Can you please make sure that the one that came in right after this is also applied at the same time?
https://www.spinics.net/lists/stable/msg215268.html

As you recommended before they weren't squashed together, but they should both come in at the same time.

Thanks,

Oops, for some reason I thought that was going to have a stable@ tag on it. Now queued up.

greg k-h

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

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

From: Sam Bobroff <[email protected]>

commit c9dccf1d074a67d36c510845f663980d69e3409b upstream.

Currently if the kernel receives a memory hot-unplug event early enough, it may get stuck in an infinite loop in dissolve_free_huge_pages(). This appears as a stall just after:

pseries-hotplug-mem: Attempting to hot-remove XX LMB(s) at YYYYYYYY

It appears to be caused by "minimum_order" being uninitialized, due to init_ras_IRQ() executing before hugetlb_init().

To correct this, extract the part of init_ras_IRQ() that enables hotplug event processing and place it in the machine_late_initcall phase, which is guaranteed to be after hugetlb_init() is called.

---
 arch/powerpc/platforms/pseries/ras.c |   31 ++++++++++++++++++++++---------
 1 file changed, 22 insertions(+), 9 deletions(-)

@@ -49,6 +49,28 @@ static irqreturn_t ras_error_interrupt(i
 
 
 /*
+ * Enable the hotplug interrupt late because processing them may touch other
+ * devices or systems (e.g. hugepages) that have not been initialized at the
+ * subsys stage.
+ */
+int __init init_ras_hotplug_IRQ(void)
+{
+	struct device_node *np;
+
+	/* Hotplug Events */
+	np = of_find_node_by_path("/event-sources/hot-plug-events");
+	if (np != NULL) {
+		if (dlpar_workqueue_init() == 0)
+			request_event_sources_irqs(np, ras_hotplug_interrupt,
+						   "RAS_HOTPLUG");
+		of_node_put(np);
+	}
+
+	return 0;
+}
+machine_late_initcall(pseries, init_ras_hotplug_IRQ);
+
+/*
  * Initialize handlers for the set of interrupts caused by hardware errors
  * and power system events.
  */

@@ -66,15 +88,6 @@ static int __init init_ras_IRQ(void)
 		of_node_put(np);
 	}
 
-	/* Hotplug Events */
-	np = of_find_node_by_path("/event-sources/hot-plug-events");
-	if (np != NULL) {
-		if (dlpar_workqueue_init() == 0)
-			request_event_sources_irqs(np, ras_hotplug_interrupt,
-					   "RAS_HOTPLUG");
-		of_node_put(np);
-	}
-
 	/* EPOW Events */
 	np = of_find_node_by_path("/event-sources/epow-events");
 	if (np != NULL) {

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

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

From: Hans de Goede <[email protected]>

commit 1fdb926974695d3dbc05a429bafa266fdd16510e upstream.

Commit 61f5acea8737 ("Bluetooth: btusb: Restore QCA Rome suspend/resume fix with a "rewritten" version") applied the USB_QUIRK_RESET_RESUME to all QCA USB Bluetooth modules. But it turns out that the resume problems are not caused by the QCA Rome chipset, on most platforms it resumes fine. The resume problems are actually a platform problem (likely the platform cutting all power when suspended).

The USB_QUIRK_RESET_RESUME quirk also disables runtime suspend, so by matching on usb-ids, we're causing all boards with these chips to use extra power, to fix resume problems which only happen on some boards.

This commit fixes this by applying the quirk based on DMI matching instead of on usb-ids, so that we match the platform and not the chipset.

Here is the /sys/kernel/debug/usb/devices for the Bluetooth module:

T:  Bus=01 Lev=01 Prnt=01 Port=07 Cnt=04 Dev#=  5 Spd=12   MxCh= 0
D:  Ver= 2.01 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=0cf3 ProdID=e300 Rev= 0.01
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms

---
 drivers/bluetooth/btusb.c |   25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

@@ -21,6 +21,7 @@
  *
  */
 
+#include <linux/dmi.h>
 #include <linux/module.h>
 #include <linux/usb.h>
 #include <linux/usb/quirks.h>

@@ -376,6 +377,21 @@ static const struct usb_device_id blackl
 	{ }	/* Terminating entry */
 };
 
+/* The Bluetooth USB module build into some devices needs to be reset on resume,
+ * this is a problem with the platform (likely shutting off all power) not with
+ * the module itself. So we use a DMI list to match known broken platforms.
+ */
+static const struct dmi_system_id btusb_needs_reset_resume_table[] = {
+	{
+		/* Lenovo Yoga 920 (QCA Rome device 0cf3:e300) */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo YOGA 920"),
+		},
+	},
+	{}
+};
+
 #define BTUSB_MAX_ISOC_FRAMES	10
 
 #define BTUSB_INTR_RUNNING	0

@@ -3031,6 +3047,9 @@ static int btusb_probe(struct usb_interf
 	hdev->send   = btusb_send_frame;
 	hdev->notify = btusb_notify;
 
+	if (dmi_check_system(btusb_needs_reset_resume_table))
+		interface_to_usbdev(intf)->quirks |= USB_QUIRK_RESET_RESUME;
+
 #ifdef CONFIG_PM
 	err = btusb_config_oob_wake(hdev);
 	if (err)

@@ -3117,12 +3136,6 @@ static int btusb_probe(struct usb_interf
 	if (id->driver_info & BTUSB_QCA_ROME) {
 		data->setup_on_usb = btusb_setup_qca;
 		hdev->set_bdaddr = btusb_set_bdaddr_ath3012;
-
-		/* QCA Rome devices lose their updated firmware over suspend,
-		 * but the USB hub doesn't notice any status change.
-		 * explicitly request a device reset on resume.
-		 */
-		interface_to_usbdev(intf)->quirks |= USB_QUIRK_RESET_RESUME;
 	}
 
 #ifdef CONFIG_BT_HCIBTUSB_RTL

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

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

From: Emil Tantilov <[email protected]>

commit 0c5661ecc5dd7ce296870a3eb7b62b1b280a5e89 upstream.

Add check for build_skb enabled ring in ixgbe_dma_sync_frag(). In that case &skb_shinfo(skb)->frags[0] may not always be set which can lead to a crash. Instead we derive the page offset from skb->data.

---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |    8 ++++++++
 1 file changed, 8 insertions(+)

@@ -1878,6 +1878,14 @@ static void ixgbe_dma_sync_frag(struct i
 				     ixgbe_rx_pg_size(rx_ring),
 				     DMA_FROM_DEVICE,
 				     IXGBE_RX_DMA_ATTR);
+	} else if (ring_uses_build_skb(rx_ring)) {
+		unsigned long offset = (unsigned long)(skb->data) & ~PAGE_MASK;
+
+		dma_sync_single_range_for_cpu(rx_ring->dev,
+					      IXGBE_CB(skb)->dma,
+					      offset,
+					      skb_headlen(skb),
+					      DMA_FROM_DEVICE);
 	} else {
 		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0];
 

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

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

From: Jeremy Boone <[email protected]>

commit 6d24cd186d9fead3722108dec1b1c993354645ff upstream.

Discrete TPMs are often connected over slow serial buses which, on some platforms, can have glitches causing bit flips. In all the driver _recv() functions, we need to use a u32 to unmarshal the response size, otherwise a bit flip of the 31st bit would cause the expected variable to go negative, which would then try to read a huge amount of data. Also sanity check that the expected amount of data is large enough for the TPM header.

---
 drivers/char/tpm/st33zp24/st33zp24.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -457,7 +457,7 @@ static int st33zp24_recv(struct tpm_chip
 			    size_t count)
 {
 	int size = 0;
-	int expected;
+	u32 expected;
 
 	if (!chip)
 		return -EBUSY;

@@ -474,7 +474,7 @@ static int st33zp24_recv(struct tpm_chip
 	}
 
 	expected = be32_to_cpu(*(__be32 *)(buf + 2));
-	if (expected > count) {
+	if (expected > count || expected < TPM_HEADER_SIZE) {
 		size = -EIO;
 		goto out;
 	}

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

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

From: Jeremy Boone <[email protected]>

commit 3be23274755ee85771270a23af7691dc9b3a95db upstream.

Discrete TPMs are often connected over slow serial buses which, on some platforms, can have glitches causing bit flips. If a bit does flip it could cause an overrun if it's in one of the size parameters, so sanity check that we're not overrunning the provided buffer when doing a memcpy().

---
 drivers/char/tpm/tpm-interface.c |    4 ++++
 drivers/char/tpm/tpm2-cmd.c      |    4 ++++
 2 files changed, 8 insertions(+)

@@ -1228,6 +1228,10 @@ int tpm_get_random(u32 chip_num, u8 *out
 			break;
 
 		recd = be32_to_cpu(tpm_cmd.params.getrandom_out.rng_data_len);
+		if (recd > num_bytes) {
+			total = -EFAULT;
+			break;
+		}
 
 		rlength = be32_to_cpu(tpm_cmd.header.out.length);
 		if (rlength < offsetof(struct tpm_getrandom_out, rng_data) +

@@ -683,6 +683,10 @@ static int tpm2_unseal_cmd(struct tpm_ch
 	if (!rc) {
 		data_len = be16_to_cpup(
 			(__be16 *) &buf.data[TPM_HEADER_SIZE + 4]);
+		if (data_len < MIN_KEY_SIZE ||  data_len > MAX_KEY_SIZE + 1) {
+			rc = -EFAULT;
+			goto out;
+		}
 
 		rlength = be32_to_cpu(((struct tpm2_cmd *)&buf)
 					->header.out.length);

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

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

From: Jeremy Boone <[email protected]>

commit 9b8cb28d7c62568a5916bdd7ea1c9176d7f8f2ed upstream.

Discrete TPMs are often connected over slow serial buses which, on some platforms, can have glitches causing bit flips. In all the driver _recv() functions, we need to use a u32 to unmarshal the response size, otherwise a bit flip of the 31st bit would cause the expected variable to go negative, which would then try to read a huge amount of data. Also sanity check that the expected amount of data is large enough for the TPM header.

---
 drivers/char/tpm/tpm_i2c_infineon.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -473,7 +473,8 @@ static int recv_data(struct tpm_chip *ch
 static int tpm_tis_i2c_recv(struct tpm_chip *chip, u8 *buf, size_t count)
 {
 	int size = 0;
-	int expected, status;
+	int status;
+	u32 expected;
 
 	if (count < TPM_HEADER_SIZE) {
 		size = -EIO;

@@ -488,7 +489,7 @@ static int tpm_tis_i2c_recv(struct tpm_c
 	}
 
 	expected = be32_to_cpu(*(__be32 *)(buf + 2));
-	if ((size_t) expected > count) {
+	if (((size_t) expected > count) || (expected < TPM_HEADER_SIZE)) {
 		size = -EIO;
 		goto out;
 	}

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

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

From: Jeremy Boone <[email protected]>

commit f9d4d9b5a5ef2f017bc344fb65a58a902517173b upstream.

Discrete TPMs are often connected over slow serial buses which, on some platforms, can have glitches causing bit flips. In all the driver _recv() functions, we need to use a u32 to unmarshal the response size, otherwise a bit flip of the 31st bit would cause the expected variable to go negative, which would then try to read a huge amount of data. Also sanity check that the expected amount of data is large enough for the TPM header.

---
 drivers/char/tpm/tpm_i2c_nuvoton.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

@@ -281,7 +281,11 @@ static int i2c_nuvoton_recv(struct tpm_c
 	struct device *dev = chip->dev.parent;
 	struct i2c_client *client = to_i2c_client(dev);
 	s32 rc;
-	int expected, status, burst_count, retries, size = 0;
+	int status;
+	int burst_count;
+	int retries;
+	int size = 0;
+	u32 expected;
 
 	if (count < TPM_HEADER_SIZE) {
 		i2c_nuvoton_ready(chip);    /* return to idle */

@@ -323,7 +327,7 @@ static int i2c_nuvoton_recv(struct tpm_c
 		 * to machine native
 		 */
 		expected = be32_to_cpu(*(__be32 *) (buf + 2));
-		if (expected > count) {
+		if (expected > count || expected < size) {
 			dev_err(dev, "%s() expected > count\n", __func__);
 			size = -EIO;
 			continue;

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

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

From: Jeremy Boone <[email protected]>

commit 6bb320ca4a4a7b5b3db8c8d7250cc40002046878 upstream.

Discrete TPMs are often connected over slow serial buses which, on some platforms, can have glitches causing bit flips. In all the driver _recv() functions, we need to use a u32 to unmarshal the response size, otherwise a bit flip of the 31st bit would cause the expected variable to go negative, which would then try to read a huge amount of data. Also sanity check that the expected amount of data is large enough for the TPM header.

---
 drivers/char/tpm/tpm_tis_core.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -202,7 +202,8 @@ static int tpm_tis_recv(struct tpm_chip
 {
 	struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
 	int size = 0;
-	int expected, status;
+	int status;
+	u32 expected;
 
 	if (count < TPM_HEADER_SIZE) {
 		size = -EIO;

@@ -217,7 +218,7 @@ static int tpm_tis_recv(struct tpm_chip
 	}
 
 	expected = be32_to_cpu(*(__be32 *) (buf + 2));
-	if (expected > count) {
+	if (expected > count || expected < TPM_HEADER_SIZE) {
 		size = -EIO;
 		goto out;
 	}

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

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

From: Erik Veijola <[email protected]>

commit 240a8af929c7c57dcde28682725b29cf8474e8e5 upstream.

The capture interface doesn't work and the playback interface only supports 48 kHz sampling rate even though it advertises more rates.

---
 sound/usb/quirks-table.h |   47 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)

@@ -3277,4 +3277,51 @@ AU0828_DEVICE(0x2040, 0x7270, "Hauppauge
 	}
 },
 
+{
+	/*
+	 * Bower's & Wilkins PX headphones only support the 48 kHz sample rate
+	 * even though it advertises more. The capture interface doesn't work
+	 * even on windows.
+	 */
+	USB_DEVICE(0x19b5, 0x0021),
+	.driver_info = (unsigned long) &(const struct snd_usb_audio_quirk) {
+		.ifnum = QUIRK_ANY_INTERFACE,
+		.type = QUIRK_COMPOSITE,
+		.data = (const struct snd_usb_audio_quirk[]) {
+			{
+				.ifnum = 0,
+				.type = QUIRK_AUDIO_STANDARD_MIXER,
+			},
+			/* Capture */
+			{
+				.ifnum = 1,
+				.type = QUIRK_IGNORE_INTERFACE,
+			},
+			/* Playback */
+			{
+				.ifnum = 2,
+				.type = QUIRK_AUDIO_FIXED_ENDPOINT,
+				.data = &(const struct audioformat) {
+					.formats = SNDRV_PCM_FMTBIT_S16_LE,
+					.channels = 2,
+					.iface = 2,
+					.altsetting = 1,
+					.altset_idx = 1,
+					.attributes = UAC_EP_CS_ATTR_FILL_MAX |
+						UAC_EP_CS_ATTR_SAMPLE_RATE,
+					.endpoint = 0x03,
+					.ep_attr = USB_ENDPOINT_XFER_ISOC,
+					.rates = SNDRV_PCM_RATE_48000,
+					.rate_min = 48000,
+					.rate_max = 48000,
+					.nr_rates = 1,
+					.rate_table = (unsigned int[]) {
+						48000
+					}
+				}
+			},
+		}
+	}
+},
+
 #undef USB_DEVICE_VENDOR_SPEC

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

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

From: Richard Fitzgerald <[email protected]>

commit 5a23699a39abc5328921a81b89383d088f6ba9cc upstream.

The patch "ALSA: control: code refactoring for ELEM_READ/ELEM_WRITE operations" introduced a potential for kernel memory corruption due to an incorrect if statement allowing non-readable controls to fall through and call the get function. For TLV controls a driver can omit SNDRV_CTL_ELEM_ACCESS_READ to ensure that only the TLV get function can be called. Instead the normal get() can be invoked unexpectedly and as the driver expects that this will only be called for controls <= 512 bytes, potentially try to copy >512 bytes into the 512 byte return array, so corrupting kernel memory.

The problem is an attempt to refactor the snd_ctl_elem_read function to invert the logic so that it conditionally aborted if the control is unreadable instead of conditionally executing. But the if statement wasn't inverted correctly.

The correct inversion of

    if (a && !b)

is

    if (!a || b)

---
 sound/core/control.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -888,7 +888,7 @@ static int snd_ctl_elem_read(struct snd_
 
 	index_offset = snd_ctl_get_ioff(kctl, &control->id);
 	vd = &kctl->vd[index_offset];
-	if (!(vd->access & SNDRV_CTL_ELEM_ACCESS_READ) && kctl->get == NULL)
+	if (!(vd->access & SNDRV_CTL_ELEM_ACCESS_READ) || kctl->get == NULL)
 		return -EPERM;
 
 	snd_ctl_build_ioff(&control->id, kctl, index_offset);

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

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

From: Takashi Iwai <[email protected]>

commit 350144069abf351c743d766b2fba9cb9b7cd32a1 upstream.

The commit change for supporting the multiple ports moved involved some code shuffling, and there the initializations of spinlock and mutex in snd_intelhad object were dropped mistakenly.

This patch adds the missing initializations again for each port.

---
 sound/x86/intel_hdmi_audio.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -1827,6 +1827,8 @@ static int hdmi_lpe_audio_probe(struct p
 		ctx->port = port;
 		ctx->pipe = -1;
 
+		spin_lock_init(&ctx->had_spinlock);
+		mutex_init(&ctx->mutex);
 		INIT_WORK(&ctx->hdmi_audio_wq, had_audio_wq);
 
 		ret = snd_pcm_new(card, INTEL_HAD, port, MAX_PB_STREAMS,

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

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

From: Hans de Goede <[email protected]>

commit 1ba8f9d308174e647b864c36209b4d7934d99888 upstream.

On some boards setting power_save to a non 0 value leads to clicking / popping sounds when ever we enter/leave powersaving mode. Ideally we would figure out how to avoid these sounds, but that is not always feasible.

This commit adds a blacklist for devices where powersaving is known to cause problems and disables it on these devices.

Note I tried to put this blacklist in userspace first:
https://github.com/systemd/systemd/pull/8128

But the systemd maintainers rightfully pointed out that it would be impossible to then later remove entries once we actually find a way to make power-saving work on listed boards without issues. Having this list in the kernel will allow removal of the blacklist entry in the same commit which fixes the clicks / plops.

The blacklist only applies to the default power_save module-option value, if a user explicitly sets the module-option then the blacklist is not used.

[ added an ifdef CONFIG_PM for the build error -- tiwai]

---
 sound/pci/hda/hda_intel.c |   38 ++++++++++++++++++++++++++++++++++++--
 1 file changed, 36 insertions(+), 2 deletions(-)

@@ -181,7 +181,7 @@ static const struct kernel_param_ops par
 };
 #define param_check_xint param_check_int
 
-static int power_save = CONFIG_SND_HDA_POWER_SAVE_DEFAULT;
+static int power_save = -1;
 module_param(power_save, xint, 0644);
 MODULE_PARM_DESC(power_save, "Automatic power-saving timeout "
 		 "(in second, 0 = disable).");

@@ -2186,6 +2186,24 @@ out_free:
 	return err;
 }
 
+#ifdef CONFIG_PM
+/* On some boards setting power_save to a non 0 value leads to clicking /
+ * popping sounds when ever we enter/leave powersaving mode. Ideally we would
+ * figure out how to avoid these sounds, but that is not always feasible.
+ * So we keep a list of devices where we disable powersaving as its known
+ * to causes problems on these devices.
+ */
+static struct snd_pci_quirk power_save_blacklist[] = {
+	/* https://bugzilla.redhat.com/show_bug.cgi?id=1525104 */
+	SND_PCI_QUIRK(0x1849, 0x0c0c, "Asrock B85M-ITX", 0),
+	/* https://bugzilla.redhat.com/show_bug.cgi?id=1525104 */
+	SND_PCI_QUIRK(0x1043, 0x8733, "Asus Prime X370-Pro", 0),
+	/* https://bugzilla.kernel.org/show_bug.cgi?id=198611 */
+	SND_PCI_QUIRK(0x17aa, 0x2227, "Lenovo X1 Carbon 3rd Gen", 0),
+	{}
+};
+#endif /* CONFIG_PM */
+
 /* number of codec slots for each chipset: 0 = default slots (i.e. 4) */
 static unsigned int azx_max_codecs[AZX_NUM_DRIVERS] = {
 	[AZX_DRIVER_NVIDIA] = 8,

@@ -2198,6 +2216,7 @@ static int azx_probe_continue(struct azx
 	struct hdac_bus *bus = azx_bus(chip);
 	struct pci_dev *pci = chip->pci;
 	int dev = chip->dev_index;
+	int val;
 	int err;
 
 	hda->probe_continued = 1;

@@ -2278,7 +2297,22 @@ static int azx_probe_continue(struct azx
 
 	chip->running = 1;
 	azx_add_card_list(chip);
-	snd_hda_set_power_save(&chip->bus, power_save * 1000);
+
+	val = power_save;
+#ifdef CONFIG_PM
+	if (val == -1) {
+		const struct snd_pci_quirk *q;
+
+		val = CONFIG_SND_HDA_POWER_SAVE_DEFAULT;
+		q = snd_pci_quirk_lookup(chip->pci, power_save_blacklist);
+		if (q && val) {
+			dev_info(chip->card->dev, "device %04x:%04x is on the power_save blacklist, forcing power_save to 0\n",
+				 q->subvendor, q->subdevice);
+			val = 0;
+		}
+	}
+#endif /* CONFIG_PM */
+	snd_hda_set_power_save(&chip->bus, val * 1000);
 	if (azx_has_pm_runtime(chip) || hda->use_vga_switcheroo)
 		pm_runtime_put_autosuspend(&pci->dev);
 

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

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

From: Takashi Iwai <[email protected]>

commit 71db96ddfa72671bd43cacdcc99ca178d90ba267 upstream.

We've added a quirk to enable the recent Lenovo dock support, where it overwrites the pin configs of NID 0x17 and 19, not only updating the pin config cache. It works right after the boot, but the problem is that the pin configs are occasionally cleared when the machine goes to PM. Meanwhile the quirk writes the pin configs only at the pre-probe, so this won't be applied any longer.

For addressing that issue, this patch moves the code to overwrite the pin configs into HDA_FIXUP_ACT_INIT section so that it's always applied at both probe and resume time.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=195161

---
 sound/pci/hda/patch_realtek.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -4875,13 +4875,14 @@ static void alc_fixup_tpt470_dock(struct
 
 	if (action == HDA_FIXUP_ACT_PRE_PROBE) {
 		spec->parse_flags = HDA_PINCFG_NO_HP_FIXUP;
+		snd_hda_apply_pincfgs(codec, pincfgs);
+	} else if (action == HDA_FIXUP_ACT_INIT) {
 		/* Enable DOCK device */
 		snd_hda_codec_write(codec, 0x17, 0,
 			    AC_VERB_SET_CONFIG_DEFAULT_BYTES_3, 0);
 		/* Enable DOCK device */
 		snd_hda_codec_write(codec, 0x19, 0,
 			    AC_VERB_SET_CONFIG_DEFAULT_BYTES_3, 0);
-		snd_hda_apply_pincfgs(codec, pincfgs);
 	}
 }
 

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

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

From: Adrian Hunter <[email protected]>

commit f8870ae6e2d6be75b1accc2db981169fdfbea7ab upstream.

Tuning can leave the IP in an active state (Buffer Read Enable bit set) which prevents the entry to low power states (i.e. S0i3). Data reset will clear it.

Generally tuning is followed by a data transfer which will anyway sort out the state, so it is rare that S0i3 is actually prevented.

---
 drivers/mmc/host/sdhci-pci-core.c |   35 +++++++++++++++++++++++++++++++----
 1 file changed, 31 insertions(+), 4 deletions(-)

@@ -593,9 +593,36 @@ static void byt_read_dsm(struct sdhci_pc
 	slot->chip->rpm_retune = intel_host->d3_retune;
 }
 
-static int byt_emmc_probe_slot(struct sdhci_pci_slot *slot)
+static int intel_execute_tuning(struct mmc_host *mmc, u32 opcode)
+{
+	int err = sdhci_execute_tuning(mmc, opcode);
+	struct sdhci_host *host = mmc_priv(mmc);
+
+	if (err)
+		return err;
+
+	/*
+	 * Tuning can leave the IP in an active state (Buffer Read Enable bit
+	 * set) which prevents the entry to low power states (i.e. S0i3). Data
+	 * reset will clear it.
+	 */
+	sdhci_reset(host, SDHCI_RESET_DATA);
+
+	return 0;
+}
+
+static void byt_probe_slot(struct sdhci_pci_slot *slot)
 {
+	struct mmc_host_ops *ops = &slot->host->mmc_host_ops;
+
 	byt_read_dsm(slot);
+
+	ops->execute_tuning = intel_execute_tuning;
+}
+
+static int byt_emmc_probe_slot(struct sdhci_pci_slot *slot)
+{
+	byt_probe_slot(slot);
 	slot->host->mmc->caps |= MMC_CAP_8_BIT_DATA | MMC_CAP_NONREMOVABLE |
 				 MMC_CAP_HW_RESET | MMC_CAP_1_8V_DDR |
 				 MMC_CAP_CMD_DURING_TFR |

@@ -650,7 +677,7 @@ static int ni_byt_sdio_probe_slot(struct
 {
 	int err;
 
-	byt_read_dsm(slot);
+	byt_probe_slot(slot);
 
 	err = ni_set_max_freq(slot);
 	if (err)

@@ -663,7 +690,7 @@ static int ni_byt_sdio_probe_slot(struct
 
 static int byt_sdio_probe_slot(struct sdhci_pci_slot *slot)
 {
-	byt_read_dsm(slot);
+	byt_probe_slot(slot);
 	slot->host->mmc->caps |= MMC_CAP_POWER_OFF_CARD | MMC_CAP_NONREMOVABLE |
 				 MMC_CAP_WAIT_WHILE_BUSY;
 	return 0;

@@ -671,7 +698,7 @@ static int byt_sdio_probe_slot(struct sd
 
 static int byt_sd_probe_slot(struct sdhci_pci_slot *slot)
 {
-	byt_read_dsm(slot);
+	byt_probe_slot(slot);
 	slot->host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY |
 				 MMC_CAP_AGGRESSIVE_PM | MMC_CAP_CD_WAKE;
 	slot->cd_idx = 0;

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

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

From: Geert Uytterhoeven <[email protected]>

commit 325501d9360eb42c7c51e6daa0d733844c1e790b upstream.

The hs_timing_cfg[] array is indexed using a value derived from the "mshcN" alias in DT, which may lead to an out-of-bounds access.

Fix this by adding a range check.

---
 drivers/mmc/host/dw_mmc-k3.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -135,6 +135,9 @@ static int dw_mci_hi6220_parse_dt(struct
 	if (priv->ctrl_id < 0)
 		priv->ctrl_id = 0;
 
+	if (priv->ctrl_id >= TIMING_MODE)
+		return -EINVAL;
+
 	host->priv = priv;
 	return 0;
 }

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

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

From: Shawn Lin <[email protected]>

commit 5b43df8b4c1a7f0c3fbf793c9566068e6b1e570c upstream.

cat /sys/kernel/debug/mmc0/regs will hang up the system since it's in runtime suspended state, so the genpd and biu_clk is off. This patch fixes this problem by calling pm_runtime_get_sync to wake it up before reading the registers.

---
 drivers/mmc/host/dw_mmc.c |    4 ++++
 1 file changed, 4 insertions(+)

@@ -165,6 +165,8 @@ static int dw_mci_regs_show(struct seq_f
 {
 	struct dw_mci *host = s->private;
 
+	pm_runtime_get_sync(host->dev);
+
 	seq_printf(s, "STATUS:\t0x%08x\n", mci_readl(host, STATUS));
 	seq_printf(s, "RINTSTS:\t0x%08x\n", mci_readl(host, RINTSTS));
 	seq_printf(s, "CMD:\t0x%08x\n", mci_readl(host, CMD));

@@ -172,6 +174,8 @@ static int dw_mci_regs_show(struct seq_f
 	seq_printf(s, "INTMASK:\t0x%08x\n", mci_readl(host, INTMASK));
 	seq_printf(s, "CLKENA:\t0x%08x\n", mci_readl(host, CLKENA));
 
+	pm_runtime_put_autosuspend(host->dev);
+
 	return 0;
 }
 

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

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

From: Shawn Lin <[email protected]>

commit a4faa4929ed3be15e2d500d2405f992f6dedc8eb upstream.

Factor out dw_mci_init_slot_caps to consolidate parsing all differents types of capabilities from host contrllers. No functional change intended.

---
 drivers/mmc/host/dw_mmc.c |   73 +++++++++++++++++++++++++++-------------------
 1 file changed, 43 insertions(+), 30 deletions(-)

@@ -2782,12 +2782,50 @@ static irqreturn_t dw_mci_interrupt(int
 	return IRQ_HANDLED;
 }
 
+static int dw_mci_init_slot_caps(struct dw_mci_slot *slot)
+{
+	struct dw_mci *host = slot->host;
+	const struct dw_mci_drv_data *drv_data = host->drv_data;
+	struct mmc_host *mmc = slot->mmc;
+	int ctrl_id;
+
+	if (host->pdata->caps)
+		mmc->caps = host->pdata->caps;
+
+	/*
+	 * Support MMC_CAP_ERASE by default.
+	 * It needs to use trim/discard/erase commands.
+	 */
+	mmc->caps |= MMC_CAP_ERASE;
+
+	if (host->pdata->pm_caps)
+		mmc->pm_caps = host->pdata->pm_caps;
+
+	if (host->dev->of_node) {
+		ctrl_id = of_alias_get_id(host->dev->of_node, "mshc");
+		if (ctrl_id < 0)
+			ctrl_id = 0;
+	} else {
+		ctrl_id = to_platform_device(host->dev)->id;
+	}
+	if (drv_data && drv_data->caps)
+		mmc->caps |= drv_data->caps[ctrl_id];
+
+	if (host->pdata->caps2)
+		mmc->caps2 = host->pdata->caps2;
+
+	/* Process SDIO IRQs through the sdio_irq_work. */
+	if (mmc->caps & MMC_CAP_SDIO_IRQ)
+		mmc->caps2 |= MMC_CAP2_SDIO_IRQ_NOTHREAD;
+
+	return 0;
+}
+
 static int dw_mci_init_slot(struct dw_mci *host)
 {
 	struct mmc_host *mmc;
 	struct dw_mci_slot *slot;
-	const struct dw_mci_drv_data *drv_data = host->drv_data;
-	int ctrl_id, ret;
+	int ret;
 	u32 freq[2];
 
 	mmc = mmc_alloc_host(sizeof(struct dw_mci_slot), host->dev);

@@ -2821,38 +2859,13 @@ static int dw_mci_init_slot(struct dw_mc
 	if (!mmc->ocr_avail)
 		mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
 
-	if (host->pdata->caps)
-		mmc->caps = host->pdata->caps;
-
-	/*
-	 * Support MMC_CAP_ERASE by default.
-	 * It needs to use trim/discard/erase commands.
-	 */
-	mmc->caps |= MMC_CAP_ERASE;
-
-	if (host->pdata->pm_caps)
-		mmc->pm_caps = host->pdata->pm_caps;
-
-	if (host->dev->of_node) {
-		ctrl_id = of_alias_get_id(host->dev->of_node, "mshc");
-		if (ctrl_id < 0)
-			ctrl_id = 0;
-	} else {
-		ctrl_id = to_platform_device(host->dev)->id;
-	}
-	if (drv_data && drv_data->caps)
-		mmc->caps |= drv_data->caps[ctrl_id];
-
-	if (host->pdata->caps2)
-		mmc->caps2 = host->pdata->caps2;
-
 	ret = mmc_of_parse(mmc);
 	if (ret)
 		goto err_host_allocated;
 
-	/* Process SDIO IRQs through the sdio_irq_work. */
-	if (mmc->caps & MMC_CAP_SDIO_IRQ)
-		mmc->caps2 |= MMC_CAP2_SDIO_IRQ_NOTHREAD;
+	ret = dw_mci_init_slot_caps(slot);
+	if (ret)
+		goto err_host_allocated;
 
 	/* Useful defaults if platform data is unset. */
 	if (host->use_dma == TRANS_MODE_IDMAC) {

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

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

From: Shawn Lin <[email protected]>

commit 0d84b9e5631d923744767dc6608672df906dd092 upstream.

Add num_caps field for dw_mci_drv_data to validate the controller id from DT alias and non-DT ways.

---
 drivers/mmc/host/dw_mmc-exynos.c   |    1 +
 drivers/mmc/host/dw_mmc-k3.c       |    1 +
 drivers/mmc/host/dw_mmc-rockchip.c |    1 +
 drivers/mmc/host/dw_mmc-zx.c       |    1 +
 drivers/mmc/host/dw_mmc.c          |    9 ++++++++-
 drivers/mmc/host/dw_mmc.h          |    2 ++
 6 files changed, 14 insertions(+), 1 deletion(-)

@@ -487,6 +487,7 @@ static unsigned long exynos_dwmmc_caps[4
 
 static const struct dw_mci_drv_data exynos_drv_data = {
 	.caps			= exynos_dwmmc_caps,
+	.num_caps		= ARRAY_SIZE(exynos_dwmmc_caps),
 	.init			= dw_mci_exynos_priv_init,
 	.set_ios		= dw_mci_exynos_set_ios,
 	.parse_dt		= dw_mci_exynos_parse_dt,

@@ -210,6 +210,7 @@ static int dw_mci_hi6220_execute_tuning(
 
 static const struct dw_mci_drv_data hi6220_data = {
 	.caps			= dw_mci_hi6220_caps,
+	.num_caps		= ARRAY_SIZE(dw_mci_hi6220_caps),
 	.switch_voltage		= dw_mci_hi6220_switch_voltage,
 	.set_ios		= dw_mci_hi6220_set_ios,
 	.parse_dt		= dw_mci_hi6220_parse_dt,

@@ -319,6 +319,7 @@ static const struct dw_mci_drv_data rk29
 
 static const struct dw_mci_drv_data rk3288_drv_data = {
 	.caps			= dw_mci_rk3288_dwmmc_caps,
+	.num_caps		= ARRAY_SIZE(dw_mci_rk3288_dwmmc_caps),
 	.set_ios		= dw_mci_rk3288_set_ios,
 	.execute_tuning		= dw_mci_rk3288_execute_tuning,
 	.parse_dt		= dw_mci_rk3288_parse_dt,

@@ -195,6 +195,7 @@ static unsigned long zx_dwmmc_caps[3] =
 
 static const struct dw_mci_drv_data zx_drv_data = {
 	.caps			= zx_dwmmc_caps,
+	.num_caps		= ARRAY_SIZE(zx_dwmmc_caps),
 	.execute_tuning		= dw_mci_zx_execute_tuning,
 	.prepare_hs400_tuning	= dw_mci_zx_prepare_hs400_tuning,
 	.parse_dt               = dw_mci_zx_parse_dt,

@@ -2808,8 +2808,15 @@ static int dw_mci_init_slot_caps(struct
 	} else {
 		ctrl_id = to_platform_device(host->dev)->id;
 	}
-	if (drv_data && drv_data->caps)
+
+	if (drv_data && drv_data->caps) {
+		if (ctrl_id >= drv_data->num_caps) {
+			dev_err(host->dev, "invalid controller id %d\n",
+				ctrl_id);
+			return -EINVAL;
+		}
 		mmc->caps |= drv_data->caps[ctrl_id];
+	}
 
 	if (host->pdata->caps2)
 		mmc->caps2 = host->pdata->caps2;

@@ -543,6 +543,7 @@ struct dw_mci_slot {
 /**
  * dw_mci driver data - dw-mshc implementation specific driver data.
  * @caps: mmc subsystem specified capabilities of the controller(s).
+ * @num_caps: number of capabilities specified by @caps.
  * @init: early implementation specific initialization.
  * @set_ios: handle bus specific extensions.
  * @parse_dt: parse implementation specific device tree properties.

@@ -554,6 +555,7 @@ struct dw_mci_slot {
  */
 struct dw_mci_drv_data {
 	unsigned long	*caps;
+	u32		num_caps;
 	int		(*init)(struct dw_mci *host);
 	void		(*set_ios)(struct dw_mci *host, struct mmc_ios *ios);
 	int		(*parse_dt)(struct dw_mci *host);

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

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

From: Lingutla Chandrasekhar <[email protected]>

commit c52232a49e203a65a6e1a670cd5262f59e9364a0 upstream.

On CPU hotunplug the enqueued timers of the unplugged CPU are migrated to a live CPU. This happens from the control thread which initiated the unplug.

If the CPU on which the control thread runs came out from a longer idle period then the base clock of that CPU might be stale because the control thread runs prior to any event which forwards the clock.

In such a case the timers from the unplugged CPU are queued on the live CPU based on the stale clock which can cause large delays due to increased granularity of the outer timer wheels which are far away from base:;clock.

But there is a worse problem than that. The following sequence of events illustrates it:

- CPU0 timer1 is queued expires = 59969 and base->clk = 59131.

The timer is queued at wheel level 2, with resulting expiry time = 60032 (due to level granularity).

- CPU1 enters idle @60007, with next timer expiry @60020.

- CPU0 is hotplugged at @60009

- CPU1 exits idle and runs the control thread which migrates the timers from CPU0

timer1 is now queued in level 0 for immediate handling in the next softirq because the requested expiry time 59969 is before CPU1 base->clk 60007

- CPU1 runs code which forwards the base clock which succeeds because the next expiring timer. which was collected at idle entry time is still set to 60020.

So it forwards beyond 60007 and therefore misses to expire the migrated timer1. That timer gets expired when the wheel wraps around again, which takes between 63 and 630ms depending on the HZ setting.

Address both problems by invoking forward_timer_base() for the control CPUs timer base. All other places, which might run into a similar problem (mod_timer()/add_timer_on()) already invoke forward_timer_base() to avoid that.

[ tglx: Massaged comment and changelog ]

---
 kernel/time/timer.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -1886,6 +1886,12 @@ int timers_dead_cpu(unsigned int cpu)
 		raw_spin_lock_irq(&new_base->lock);
 		raw_spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING);
 
+		/*
+		 * The current CPUs base clock might be stale. Update it
+		 * before moving the timers over.
+		 */
+		forward_timer_base(new_base);
+
 		BUG_ON(old_base->running_timer);
 
 		for (i = 0; i < WHEEL_SIZE; i++)

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

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

From: Helge Deller <[email protected]>

commit 5ffa8518851f1401817c15d2a7eecc0373c26ff9 upstream.

When running on qemu we know that the (emulated) cr16 cpu-internal clocks are syncronized. So let's use them unconditionally on qemu.

---
 arch/parisc/include/asm/processor.h |    2 ++
 arch/parisc/kernel/time.c           |    2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

@@ -316,6 +316,8 @@ extern int _parisc_requires_coherency;
 #define parisc_requires_coherency()	(0)
 #endif
 
+extern int running_on_qemu;
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* __ASM_PARISC_PROCESSOR_H */

@@ -248,7 +248,7 @@ static int __init init_cr16_clocksource(
 	 * different sockets, so mark them unstable and lower rating on
 	 * multi-socket SMP systems.
 	 */
-	if (num_online_cpus() > 1) {
+	if (num_online_cpus() > 1 && !running_on_qemu) {
 		int cpu;
 		unsigned long cpu0_loc;
 		cpu0_loc = per_cpu(cpu_data, 0).cpu_loc;

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

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

From: Helge Deller <[email protected]>

commit 636a415bcc7f4fd020ece8fd5fc648c4cef19c34 upstream.

When run under QEMU, calling mfctl(16) creates some overhead because the qemu timer has to be scaled and moved into the register. This patch reduces the number of calls to mfctl(16) by moving the calls out of the loops.

Additionally, increase the minimal time interval to 8000 cycles instead of 500 to compensate possible QEMU delays when delivering interrupts.

---
 arch/parisc/kernel/time.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

@@ -76,10 +76,10 @@ irqreturn_t __irq_entry timer_interrupt(
 	next_tick = cpuinfo->it_value;
 
 	/* Calculate how many ticks have elapsed. */
+	now = mfctl(16);
 	do {
 		++ticks_elapsed;
 		next_tick += cpt;
-		now = mfctl(16);
 	} while (next_tick - now > cpt);
 
 	/* Store (in CR16 cycles) up to when we are accounting right now. */

@@ -103,16 +103,17 @@ irqreturn_t __irq_entry timer_interrupt(
 	 * if one or the other wrapped. If "now" is "bigger" we'll end up
 	 * with a very large unsigned number.
 	 */
-	while (next_tick - mfctl(16) > cpt)
+	now = mfctl(16);
+	while (next_tick - now > cpt)
 		next_tick += cpt;
 
 	/* Program the IT when to deliver the next interrupt.
 	 * Only bottom 32-bits of next_tick are writable in CR16!
 	 * Timer interrupt will be delivered at least a few hundred cycles
-	 * after the IT fires, so if we are too close (<= 500 cycles) to the
+	 * after the IT fires, so if we are too close (<= 8000 cycles) to the
 	 * next cycle, simply skip it.
 	 */
-	if (next_tick - mfctl(16) <= 500)
+	if (next_tick - now <= 8000)
 		next_tick += cpt;
 	mtctl(next_tick, 16);
 

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

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

From: John David Anglin <[email protected]>

commit 0adb24e03a124b79130c9499731936b11ce2677d upstream.

The change to flush_kernel_vmap_range() wasn't sufficient to avoid the SMP stalls.  The problem is some drivers call these routines with interrupts disabled.  Interrupts need to be enabled for flush_tlb_all() and flush_cache_all() to work.  This version adds checks to ensure interrupts are not disabled before calling routines that need IPI interrupts.  When interrupts are disabled, we now drop into slower code.

The attached change fixes the ordering of cache and TLB flushes in several cases.  When we flush the cache using the existing PTE/TLB entries, we need to flush the TLB after doing the cache flush.  We don't need to do this when we flush the entire instruction and data caches as these flushes don't use the existing TLB entries.  The same is true for tmpalias region flushes.

The flush_kernel_vmap_range() and invalidate_kernel_vmap_range() routines have been updated.

Secondly, we added a new purge_kernel_dcache_range_asm() routine to pacache.S and use it in invalidate_kernel_vmap_range().  Nominally, purges are faster than flushes as the cache lines don't have to be written back to memory.

Hopefully, this is sufficient to resolve the remaining problems due to cache speculation.  So far, testing indicates that this is the case.  I did work up a patch using tmpalias flushes, but there is a performance hit because we need the physical address for each page, and we also need to sequence access to the tmpalias flush code.  This increases the probability of stalls.

---
 arch/parisc/include/asm/cacheflush.h |    1 
 arch/parisc/kernel/cache.c           |   57 +++++++++++++++++++----------------
 arch/parisc/kernel/pacache.S         |   22 +++++++++++++
 3 files changed, 54 insertions(+), 26 deletions(-)

@@ -26,6 +26,7 @@ void flush_user_icache_range_asm(unsigne
 void flush_kernel_icache_range_asm(unsigned long, unsigned long);
 void flush_user_dcache_range_asm(unsigned long, unsigned long);
 void flush_kernel_dcache_range_asm(unsigned long, unsigned long);
+void purge_kernel_dcache_range_asm(unsigned long, unsigned long);
 void flush_kernel_dcache_page_asm(void *);
 void flush_kernel_icache_page(void *);
 

@@ -1110,6 +1110,28 @@ ENTRY_CFI(flush_kernel_dcache_range_asm)
 	.procend
 ENDPROC_CFI(flush_kernel_dcache_range_asm)
 
+ENTRY_CFI(purge_kernel_dcache_range_asm)
+	.proc
+	.callinfo NO_CALLS
+	.entry
+
+	ldil		L%dcache_stride, %r1
+	ldw		R%dcache_stride(%r1), %r23
+	ldo		-1(%r23), %r21
+	ANDCM		%r26, %r21, %r26
+
+1:      cmpb,COND(<<),n	%r26, %r25,1b
+	pdc,m		%r23(%r26)
+
+	sync
+	syncdma
+	bv		%r0(%r2)
+	nop
+	.exit
+
+	.procend
+ENDPROC_CFI(purge_kernel_dcache_range_asm)
+
 ENTRY_CFI(flush_user_icache_range_asm)
 	.proc
 	.callinfo NO_CALLS

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

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

From: Helge Deller <[email protected]>

commit fd8d0ca2563151204f3fe555dc8ca4bcfe8677a3 upstream.

For security reasons do not expose the virtual kernel memory layout to userspace.

---
 arch/parisc/mm/init.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

@@ -629,7 +629,12 @@ void __init mem_init(void)
 #endif
 
 	mem_init_print_info(NULL);
-#ifdef CONFIG_DEBUG_KERNEL /* double-sanity-check paranoia */
+
+#if 0
+	/*
+	 * Do not expose the virtual kernel memory layout to userspace.
+	 * But keep code for debugging purposes.
+	 */
 	printk("virtual kernel memory layout:\n"
 	       "    vmalloc : 0x%px - 0x%px   (%4ld MB)\n"
 	       "    memory  : 0x%px - 0x%px   (%4ld MB)\n"

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

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

From: Anand Jain <[email protected]>

commit 3c181c12c431fe33b669410d663beb9cceefcd1b upstream.

The fs_info::super_copy is a byte copy of the on-disk structure and all members must use the accessor macros/functions to obtain the right value. This was missing in update_super_roots and in sysfs readers.

Moving between opposite endianness hosts will report bogus numbers in sysfs, and mount may fail as the root will not be restored correctly. If the filesystem is always used on a same endian host, this will not be a problem.

Fix this by using the btrfs_set_super...() functions to set fs_info::super_copy values, and for the sysfs, use the cached fs_info::nodesize/sectorsize values.

---
 fs/btrfs/sysfs.c       |    8 +++-----
 fs/btrfs/transaction.c |   20 ++++++++++++--------
 2 files changed, 15 insertions(+), 13 deletions(-)

@@ -423,7 +423,7 @@ static ssize_t btrfs_nodesize_show(struc
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->nodesize);
+	return snprintf(buf, PAGE_SIZE, "%u\n", fs_info->nodesize);
 }
 
 BTRFS_ATTR(, nodesize, btrfs_nodesize_show);

@@ -433,8 +433,7 @@ static ssize_t btrfs_sectorsize_show(str
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n",
-			fs_info->super_copy->sectorsize);
+	return snprintf(buf, PAGE_SIZE, "%u\n", fs_info->sectorsize);
 }
 
 BTRFS_ATTR(, sectorsize, btrfs_sectorsize_show);

@@ -444,8 +443,7 @@ static ssize_t btrfs_clone_alignment_sho
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n",
-			fs_info->super_copy->sectorsize);
+	return snprintf(buf, PAGE_SIZE, "%u\n", fs_info->sectorsize);
 }
 
 BTRFS_ATTR(, clone_alignment, btrfs_clone_alignment_show);

@@ -1723,19 +1723,23 @@ static void update_super_roots(struct bt
 
 	super = fs_info->super_copy;
 
+	/* update latest btrfs_super_block::chunk_root refs */
 	root_item = &fs_info->chunk_root->root_item;
-	super->chunk_root = root_item->bytenr;
-	super->chunk_root_generation = root_item->generation;
-	super->chunk_root_level = root_item->level;
+	btrfs_set_super_chunk_root(super, root_item->bytenr);
+	btrfs_set_super_chunk_root_generation(super, root_item->generation);
+	btrfs_set_super_chunk_root_level(super, root_item->level);
 
+	/* update latest btrfs_super_block::root refs */
 	root_item = &fs_info->tree_root->root_item;
-	super->root = root_item->bytenr;
-	super->generation = root_item->generation;
-	super->root_level = root_item->level;
+	btrfs_set_super_root(super, root_item->bytenr);
+	btrfs_set_super_generation(super, root_item->generation);
+	btrfs_set_super_root_level(super, root_item->level);
+
 	if (btrfs_test_opt(fs_info, SPACE_CACHE))
-		super->cache_generation = root_item->generation;
+		btrfs_set_super_cache_generation(super, root_item->generation);
 	if (test_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags))
-		super->uuid_tree_generation = root_item->generation;
+		btrfs_set_super_uuid_tree_generation(super,
+						     root_item->generation);
 }
 
 int btrfs_transaction_in_commit(struct btrfs_fs_info *info)

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

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

From: Jiufei Xue <[email protected]>

commit 7c5a0dcf557c6511a61e092ba887de28882fe857 upstream.

The vm counters is counted in sectors, so we should do the conversation in submit_bio.

---
 block/blk-core.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -2401,7 +2401,7 @@ blk_qc_t submit_bio(struct bio *bio)
 		unsigned int count;
 
 		if (unlikely(bio_op(bio) == REQ_OP_WRITE_SAME))
-			count = queue_logical_block_size(bio->bi_disk->queue);
+			count = queue_logical_block_size(bio->bi_disk->queue) >> 9;
 		else
 			count = bio_sectors(bio);
 

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

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

From: Ming Lei <[email protected]>

commit ba989a01469d027861e55c8f1121edadef757797 upstream.

When requeuing request, the domain token should have been freed before re-inserting the request to io scheduler. Otherwise, the assigned domain token will be leaked, and IO hang can be caused.

---
 block/kyber-iosched.c |    1 +
 1 file changed, 1 insertion(+)

@@ -833,6 +833,7 @@ static struct elevator_type kyber_sched
 		.limit_depth = kyber_limit_depth,
 		.prepare_request = kyber_prepare_request,
 		.finish_request = kyber_finish_request,
+		.requeue_request = kyber_finish_request,
 		.completed_request = kyber_completed_request,
 		.dispatch_request = kyber_dispatch_request,
 		.has_work = kyber_has_work,

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

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

From: Ming Lei <[email protected]>

commit 0bd1ed4860d0f5f836aa8371797689a3779d1bf5 upstream.

The 'lend' parameter of truncate_inode_pages_range is required to be inclusive, so follow the rule.

This patch fixes one memory corruption triggered by discard.

---
 block/ioctl.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -225,7 +225,7 @@ static int blk_ioctl_discard(struct bloc
 
 	if (start + len > i_size_read(bdev->bd_inode))
 		return -EINVAL;
-	truncate_inode_pages_range(mapping, start, start + len);
+	truncate_inode_pages_range(mapping, start, start + len - 1);
 	return blkdev_issue_discard(bdev, start >> 9, len >> 9,
 				    GFP_KERNEL, flags);
 }

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

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

From: Dan Williams <[email protected]>

commit 94db151dc89262bfa82922c44e8320cea2334667 upstream.

Filesystem-DAX is incompatible with 'longterm' page pinning. Without page cache indirection a DAX mapping maps filesystem blocks directly. This means that the filesystem must not modify a file's block map while any page in a mapping is pinned. In order to prevent the situation of userspace holding of filesystem operations indefinitely, disallow 'longterm' Filesystem-DAX mappings.

RDMA has the same conflict and the plan there is to add a 'with lease' mechanism to allow the kernel to notify userspace that the mapping is being torn down for block-map maintenance. Perhaps something similar can be put in place for vfio.

Note that xfs and ext4 still report:

"DAX enabled. Warning: EXPERIMENTAL, use at your own risk"

...at mount time, and resolving the dax-dma-vs-truncate problem is one of the last hurdles to remove that designation.

---
 drivers/vfio/vfio_iommu_type1.c |   18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

@@ -338,11 +338,12 @@ static int vaddr_get_pfn(struct mm_struc
 {
 	struct page *page[1];
 	struct vm_area_struct *vma;
+	struct vm_area_struct *vmas[1];
 	int ret;
 
 	if (mm == current->mm) {
-		ret = get_user_pages_fast(vaddr, 1, !!(prot & IOMMU_WRITE),
-					  page);
+		ret = get_user_pages_longterm(vaddr, 1, !!(prot & IOMMU_WRITE),
+					      page, vmas);
 	} else {
 		unsigned int flags = 0;
 

@@ -351,7 +352,18 @@ static int vaddr_get_pfn(struct mm_struc
 
 		down_read(&mm->mmap_sem);
 		ret = get_user_pages_remote(NULL, mm, vaddr, 1, flags, page,
-					    NULL, NULL);
+					    vmas, NULL);
+		/*
+		 * The lifetime of a vaddr_get_pfn() page pin is
+		 * userspace-controlled. In the fs-dax case this could
+		 * lead to indefinite stalls in filesystem operations.
+		 * Disallow attempts to pin fs-dax pages via this
+		 * interface.
+		 */
+		if (ret > 0 && vma_is_fsdax(vmas[0])) {
+			ret = -EOPNOTSUPP;
+			put_page(page[0]);
+		}
 		up_read(&mm->mmap_sem);
 	}
 

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

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

From: Viresh Kumar <[email protected]>

commit 0373ca74831b0f93cd4cdbf7ad3aec3c33a479a5 upstream.

commit a307a1e6bc0d "cpufreq: s3c: use cpufreq_generic_init()" accidentally broke cpufreq on s3c2410 and s3c2412.

These two platforms don't have a CPU frequency table and used to skip calling cpufreq_table_validate_and_show() for them. But with the above commit, we started calling it unconditionally and that will eventually fail as the frequency table pointer is NULL.

Fix this by calling cpufreq_table_validate_and_show() conditionally again.

---
 drivers/cpufreq/s3c24xx-cpufreq.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

@@ -351,7 +351,13 @@ struct clk *s3c_cpufreq_clk_get(struct d
 static int s3c_cpufreq_init(struct cpufreq_policy *policy)
 {
 	policy->clk = clk_arm;
-	return cpufreq_generic_init(policy, ftab, cpu_cur.info->latency);
+
+	policy->cpuinfo.transition_latency = cpu_cur.info->latency;
+
+	if (ftab)
+		return cpufreq_table_validate_and_show(policy, ftab);
+
+	return 0;
 }
 
 static int __init s3c_cpufreq_initclks(void)

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

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

From: Dan Williams <[email protected]>

commit 230f5a8969d8345fc9bbe3683f068246cf1be4b8 upstream.

Gerd reports that ->i_mode may contain other bits besides S_IFCHR. Use S_ISCHR() instead. Otherwise, get_user_pages_longterm() may fail on device-dax instances when those are meant to be explicitly allowed.

---
 include/linux/fs.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -3204,7 +3204,7 @@ static inline bool vma_is_fsdax(struct v
 	if (!vma_is_dax(vma))
 		return false;
 	inode = file_inode(vma->vm_file);
-	if (inode->i_mode == S_IFCHR)
+	if (S_ISCHR(inode->i_mode))
 		return false; /* device-dax */
 	return true;
 }

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

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

From: Jan Kara <[email protected]>

commit d9c10e5b8863cfb6886d1640386455075c6e979d upstream.

Commit e864f39569f4 "fs: add RWF_DSYNC aand RWF_SYNC" added additional way for direct IO to become synchronous and thus trigger fsync from the IO completion handler. Then commit 9830f4be159b "fs: Use RWF_* flags for AIO operations" allowed these flags to be set for AIO as well. However that commit forgot to update the condition checking whether the IO completion handling should be defered to a workqueue and thus AIO DIO with RWF_[D]SYNC set will call fsync() from IRQ context resulting in sleep in atomic.

Fix the problem by checking directly iocb flags (the same way as it is done in dio_complete()) instead of checking all conditions that could lead to IO being synchronous.

---
 fs/direct-io.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

@@ -1252,8 +1252,7 @@ do_blockdev_direct_IO(struct kiocb *iocb
 	 */
 	if (dio->is_async && iov_iter_rw(iter) == WRITE) {
 		retval = 0;
-		if ((iocb->ki_filp->f_flags & O_DSYNC) ||
-		    IS_SYNC(iocb->ki_filp->f_mapping->host))
+		if (iocb->ki_flags & IOCB_DSYNC)
 			retval = dio_set_defer_completion(dio);
 		else if (!dio->inode->i_sb->s_dio_done_wq) {
 			/*

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

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

From: Juergen Gross <[email protected]>

commit 71c208dd54ab971036d83ff6d9837bae4976e623 upstream.

Older Xen versions (4.5 and before) might have problems migrating pv guests with MSR_IA32_SPEC_CTRL having a non-zero value. So before suspending zero that MSR and restore it after being resumed.

---
 arch/x86/xen/suspend.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)

@@ -1,12 +1,15 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/types.h>
 #include <linux/tick.h>
+#include <linux/percpu-defs.h>
 
 #include <xen/xen.h>
 #include <xen/interface/xen.h>
 #include <xen/grant_table.h>
 #include <xen/events.h>
 
+#include <asm/cpufeatures.h>
+#include <asm/msr-index.h>
 #include <asm/xen/hypercall.h>
 #include <asm/xen/page.h>
 #include <asm/fixmap.h>

@@ -15,6 +18,8 @@
 #include "mmu.h"
 #include "pmu.h"
 
+static DEFINE_PER_CPU(u64, spec_ctrl);
+
 void xen_arch_pre_suspend(void)
 {
 	xen_save_time_memory_area();

@@ -35,6 +40,9 @@ void xen_arch_post_suspend(int cancelled
 
 static void xen_vcpu_notify_restore(void *data)
 {
+	if (xen_pv_domain() && boot_cpu_has(X86_FEATURE_SPEC_CTRL))
+		wrmsrl(MSR_IA32_SPEC_CTRL, this_cpu_read(spec_ctrl));
+
 	/* Boot processor notified via generic timekeeping_resume() */
 	if (smp_processor_id() == 0)
 		return;

@@ -44,7 +52,15 @@ static void xen_vcpu_notify_restore(void
 
 static void xen_vcpu_notify_suspend(void *data)
 {
+	u64 tmp;
+
 	tick_suspend_local();
+
+	if (xen_pv_domain() && boot_cpu_has(X86_FEATURE_SPEC_CTRL)) {
+		rdmsrl(MSR_IA32_SPEC_CTRL, tmp);
+		this_cpu_write(spec_ctrl, tmp);
+		wrmsrl(MSR_IA32_SPEC_CTRL, 0);
+	}
 }
 
 void xen_arch_resume(void)

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

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

From: Sebastian Panceac <[email protected]>

commit 028091f82eefd5e84f81cef81a7673016ecbe78b upstream.

When the Intel Edison module is powered with 3.3V, the reboot command makes the module stuck. If the module is powered at a greater voltage, like 4.4V (as the Edison Mini Breakout board does), reboot works OK.

The official Intel Edison BSP sends the IPCMSG_COLD_RESET message to the SCU by default. The IPCMSG_COLD_BOOT which is used by the upstream kernel is only sent when explicitely selected on the kernel command line.

Use IPCMSG_COLD_RESET unconditionally which makes reboot work independent of the power supply voltage.

[ tglx: Massaged changelog ]

---
 arch/x86/platform/intel-mid/intel-mid.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -79,7 +79,7 @@ static void intel_mid_power_off(void)
 
 static void intel_mid_reboot(void)
 {
-	intel_scu_ipc_simple_command(IPCMSG_COLD_BOOT, 0);
+	intel_scu_ipc_simple_command(IPCMSG_COLD_RESET, 0);
 }
 
 static unsigned long __init intel_mid_calibrate_tsc(void)

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

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

From: Thomas Gleixner <[email protected]>

commit 945fd17ab6bab8a4d05da6c3170519fbcfe62ddb upstream.

The separation of the cpu_entry_area from the fixmap missed the fact that on 32bit non-PAE kernels the cpu_entry_area mapping might not be covered in initial_page_table by the previous synchronizations.

This results in suspend/resume failures because 32bit utilizes initial page table for resume. The absence of the cpu_entry_area mapping results in a triple fault, aka. insta reboot.

With PAE enabled this works by chance because the PGD entry which covers the fixmap and other parts incindentally provides the cpu_entry_area mapping as well.

Synchronize the initial page table after setting up the cpu entry area. Instead of adding yet another copy of the same code, move it to a function and invoke it from the various places.

It needs to be investigated if the existing calls in setup_arch() and setup_per_cpu_areas() can be replaced by the later invocation from setup_cpu_entry_areas(), but that's beyond the scope of this fix.

---
 arch/x86/include/asm/pgtable_32.h |    1 +
 arch/x86/include/asm/pgtable_64.h |    1 +
 arch/x86/kernel/setup.c           |   17 +++++------------
 arch/x86/kernel/setup_percpu.c    |   17 ++++-------------
 arch/x86/mm/cpu_entry_area.c      |    6 ++++++
 arch/x86/mm/init_32.c             |   15 +++++++++++++++
 6 files changed, 32 insertions(+), 25 deletions(-)

@@ -32,6 +32,7 @@ extern pmd_t initial_pg_pmd[];
 static inline void pgtable_cache_init(void) { }
 static inline void check_pgt_cache(void) { }
 void paging_init(void);
+void sync_initial_page_table(void);
 
 /*
  * Define this if things work differently on an i386 and an i486:

@@ -28,6 +28,7 @@ extern pgd_t init_top_pgt[];
 #define swapper_pg_dir init_top_pgt
 
 extern void paging_init(void);
+static inline void sync_initial_page_table(void) { }
 
 #define pte_ERROR(e)					\
 	pr_err("%s:%d: bad pte %p(%016lx)\n",		\

@@ -1205,20 +1205,13 @@ void __init setup_arch(char **cmdline_p)
 
 	kasan_init();
 
-#ifdef CONFIG_X86_32
-	/* sync back kernel address range */
-	clone_pgd_range(initial_page_table + KERNEL_PGD_BOUNDARY,
-			swapper_pg_dir     + KERNEL_PGD_BOUNDARY,
-			KERNEL_PGD_PTRS);
-
 	/*
-	 * sync back low identity map too.  It is used for example
-	 * in the 32-bit EFI stub.
+	 * Sync back kernel address range.
+	 *
+	 * FIXME: Can the later sync in setup_cpu_entry_areas() replace
+	 * this call?
 	 */
-	clone_pgd_range(initial_page_table,
-			swapper_pg_dir     + KERNEL_PGD_BOUNDARY,
-			min(KERNEL_PGD_PTRS, KERNEL_PGD_BOUNDARY));
-#endif
+	sync_initial_page_table();
 
 	tboot_probe();