Fix type casting of channel descriptor to avoid memory access
violations during the descriptor usage.
Change-Id: I3d73bf327fd796f482e492265d18d3417f21dc9a
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Update card state change API to synchronize multiple invocation
of the same API from different drivers.
Change-Id: I03c18a9540bdb64fa462262faa9cabe158e9e199
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add ipc router sdio plugin config in Makefile and Kconfig
to enable it for compilation.
Change-Id: If3bc8e812240bd86f2ef3bc63452edcdf69deac6
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Disable logs to console after reporting error only once,to stop
excessive prints on console whcich causes watchdog bite due to
console lock up.
Change-Id: I26d9f3f258e0428c059440369c88691b2d523bba
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Free rx DMA buffere after the data is read from the buffer.
Change-Id: I2b8f97beb50b4752674fc8b4034feb84d2a36302
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add driver state information param to view from the user space.
Change-Id: I2d4cd86ae5d67ac948a93d28e24ea96394343513
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Disable retuning in function 1 driver by default. Created a sysfs
entry to enable it if needed.
Change-Id: I29b5ea4a8e0f1614b10226edb21c6cbeb7a327cd
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Ping with known good phase if last phase of tuning fails.
Change-Id: I3d8d8ce3d8a4dc00146668b815b69619d3a4cea8
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Update read write API to drop further messages during transport or
interface removal.
Change-Id: Ied43de8aa7cef1c36c9c520976357ba1e63017db
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add API to dynamically add and remove the SDIO card on demand.
Change-Id: I946115880da75804a6636616d9cf7cfb4a682b92
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add QCN SDIO core diver config into Makefile and
Kconfig to enable its compilation.
Change-Id: I19d184c628658175179d9b5c52a3a8bd07dd7999
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add QCN SDIO client driver config into Makefile and
Kconfig to enable its compilation.
Change-Id: I40ffa1844b2030aca53d0e36f1bee2900a3ea8d4
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
This snapshot is taken as of msm-4.4 commit dc86ce0faa ("soc: qcom:
ipc_rtr_xprt: Add support to set version in transport").
Change-Id: I729d3c910372d2c3c0bbaaae74a42150c9929732
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add support for SDIO transport in diag core framework.
Change-Id: Ic3ccfec3acd60b36a96aebb49a5681219a25e643
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Update code to support diag for peripheral devices over sdio bus.
Change-Id: I14678f301f94b239b61c4c94d3c0ff65204e739e
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
This commit is taken as of msm-4.4 commit f185067e36 ("diag: Fix
HSIC read complete work function").
Change-Id: I629fe0d9ed697fb20fd9263ecea6009157181e63
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Update the completion variable sequence to avoid multiple complete
call on a stale staack completion variable.
Change-Id: I582777431918c359cd10e4e072f33589dc600e62
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add QCN bridge client driver to provide interface to QCN SDIO
core function-1 driver. This driver currently provides interfacing
for diag, IPC router and sahara application.
Change-Id: Ie992aec91347bb68e46a01ee6f67a1d07ce40ecc
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add synchronization support between clients that operate on different
modes, i.e. synchronous or asynchronous. The synchronous client will
momentarily switch to asynchronous if any asynchronous transmission is
scheduled.
Change-Id: I41aec9bff950d2816a9ce55017e118e8504be77a
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
The current QCN core driver support four concurrent clients,
and there can be only two CRQ's that can be raised per client
from device.
Reserve 8 bufferes for data reception from device to host.
Change-Id: Iac72b5dd51988ffd1139bf55ecb9799bded77710
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Increase RW queue size to compensate higer data transfer from host
to device, when device is operating at lower frequencies.
Change-Id: I443aead06c757cde40ee99ea2c29a5f323249b8b
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Update read/write API in QCN sdio core driver to issue CMD53
instead of CMD52 for data equal or greater than 4 bytes.
Change-Id: I5641f32fa8896d3fafb087b440ff69ad936a83ee
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
Add QCN SDIO core driver as function-1 driver to provide streamlined
communication for peripheral device connected over sdio.
Change-Id: Ia8aec77807b59d3e1476b0a12ee7016809335b39
Signed-off-by: Amandeep Singh <amansing@codeaurora.org>
The WALT time accounting breaks when CFS tasks are throttled by the CPU
bandwidth control mechanism of the CPU cgroups controller. This can
result in a negative cumulative_runnable_avg, which can then lead to a
kernel panic, and the device crashing.
Although the right fix would be add support for throttled CFS tasks to
WALT, the common kernel is now in stable maintenance mode and will not
get new features which could cause issues for partners downstream.
To work around the issue, make the CFS_BANDWIDTH Kconfig option depend
on SCHED_WALT=n, hence preventing these two things from being enabled
simultaneously. This should not be an issue for most partners (nobody
had noticed the breakage for years), and those who do need the better
fix can apply it in their device kernel.
Bug: 139071966
Bug: 120440300
Change-Id: Ieb3c367ae7893ac93fb5b38c1580dc59151aacce
Signed-off-by: Quentin Perret <quentin.perret@arm.com>
Check if the handle data type received from userspace is valid
for app loaded query request to avoid the offset boundary check
for qseecom_send_modfd_resp is bypassed.
Change-Id: I5f3611a8f830d6904213781c5ba70cfc0ba3e2e0
Signed-off-by: Zhen Kong <zkong@codeaurora.org>
Disable HSPHY autosuspend capability before going for core
initialization so as to avoid frivolous phy PLL lock warnings on
PHY initialization.
Change-Id: I1bdb7a21c3a2fabd3f44f7088d265fa29da3f7a5
Signed-off-by: Ajay Agarwal <ajaya@codeaurora.org>
This reverts commit e4250ce4ec.
AB/IB voting is not required to reset CPP AXI/HW during
camera close.
Change-Id: I230276b9d2c821dbc651fc5dd2a458e4e987baf9
Signed-off-by: Venu Raidu <vraidu@codeaurora.org>
This represents a rollup of a series of reverts, simplified are
modifications to remove fiq_glue and fiq_debugger references in:
arch/arm/common/Kconfig
arch/arm/common/Makefile
drivers/staging/android/Kconfig
drivers/staging/android/Makefile
And deletion of:
arch/arm/common/fiq_glue.S
arch/arm/common/fiq_glue_setup.c
drivers/staging/android/fiq_debugger/
Signed-off-by: Mark Salyzyn <salyzyn@google.com>
Bug: 32402555
Bug: 36101220
Change-Id: I3f74b1ff5e4971d619bcb37a911fed68fbb538d5
In stability reboot tests, deferred audio drivers
are not invoked after lpass loading sometimes and
results in sound card failure. Change APR to platform
device and after APR status changes to loaded state,
add dummy module child device under APR which invokes
deferred audio drivers and sound card registers successfully.
In analog codec driver defers until Q6 core ready.
Change-Id: I1c82d9da55e771299df356a2771de2b2e62348cf
Signed-off-by: Soumya Managoli <smanag@codeaurora.org>
This will be used by eBPF and the iorapd project for high speed
inode/dev numbers to file path lookup. Look at the inodemap CL for more
details about of eBPF and iorapd using the tracepoint.
This is planned to be used by the inodemap BPF program. Also, ART folks
have been using this tracepoint for debugging "unknown inode numer"
issues.
The tracepoint will be out of tree, and not sent upstream, since VFS
developers don't accept tracepoints strictly.
Test: Run "find /" command in emulator and measure completion time
with/without treacepoint. find does a flood of lookups which stresses
the tracepoint. No performance change observed.
Test: eBPF prototypes (wip) successfully read data from the tracepoint.
OOT Bug: 139663736
Bug: 135143784
Bug: 137393447
Change-Id: I657f374659673a9c8853530d73c0622dbdbab146
Signed-off-by: Joel Fernandes <joelaf@google.com>
(cherry picked from commit 987732fcbbe3ea78368c28e5a0d0d236be61420f)
(cherry picked from commit 2104283a8d7349011860d9bffb8a3d25456e6d20)
* refs/heads/tmp-ac7fbca
Linux 4.4.190
bonding: Add vlan tx offload to hw_enc_features
sctp: fix the transport error_count check
net/mlx5e: Only support tx/rx pause setting for port owner
xen/netback: Reset nr_frags before freeing skb
net/packet: fix race in tpacket_snd()
x86/boot: Disable the address-of-packed-member compiler warning
iommu/amd: Move iommu_init_pci() to .init section
x86/vdso: Remove direct HPET access through the vDSO
IB/mlx5: Make coding style more consistent
RDMA: Directly cast the sockaddr union to sockaddr
scsi: fcoe: Embed fc_rport_priv in fcoe_rport structure
asm-generic: default BUG_ON(x) to if(x)BUG()
Input: psmouse - fix build error of multiple definition
arm64: compat: Allow single-byte watchpoints on all addresses
include/linux/module.h: copy __init/__exit attrs to init/cleanup_module
Backport minimal compiler_attributes.h to support GCC 9
USB: serial: option: Add Motorola modem UARTs
USB: serial: option: add the BroadMobi BM818 card
USB: serial: option: Add support for ZTE MF871A
USB: serial: option: add D-Link DWM-222 device ID
usb: cdc-acm: make sure a refcount is taken early enough
USB: core: Fix races in character device registration and deregistraion
staging: comedi: dt3000: Fix rounding up of timer divisor
staging: comedi: dt3000: Fix signed integer overflow 'divider * base'
asm-generic: fix -Wtype-limits compiler warnings
ocfs2: remove set but not used variable 'last_hash'
IB/core: Add mitigation for Spectre V1
kbuild: modpost: handle KBUILD_EXTRA_SYMBOLS only for external modules
ata: libahci: do not complain in case of deferred probe
scsi: hpsa: correct scsi command status issue after reset
libata: zpodd: Fix small read overflow in zpodd_get_mech_type()
perf header: Fix use of unitialized value warning
perf header: Fix divide by zero error if f_header.attr_size==0
irqchip/irq-imx-gpcv2: Forward irq type to parent
xen/pciback: remove set but not used variable 'old_state'
net: usb: pegasus: fix improper read if get_registers() fail
Input: iforce - add sanity checks
Input: kbtab - sanity check for endpoint type
HID: hiddev: do cleanup in failure of opening a device
HID: hiddev: avoid opening a disconnected device
HID: holtek: test for sanity of intfdata
ALSA: hda - Fix a memory leak bug
mm/memcontrol.c: fix use after free in mem_cgroup_iter()
USB: gadget: f_midi: fixing a possible double-free in f_midi
usb: gadget: f_midi: fail if set_alt fails to allocate requests
sh: kernel: hw_breakpoint: Fix missing break in switch statement
scsi: mpt3sas: Use 63-bit DMA addressing on SAS35 HBA
mwifiex: fix 802.11n/WPA detection
smb3: send CAP_DFS capability during session setup
SMB3: Fix deadlock in validate negotiate hits reconnect
mac80211: don't WARN on short WMM parameters from AP
ALSA: firewire: fix a memory leak bug
hwmon: (nct7802) Fix wrong detection of in4 presence
can: peak_usb: pcan_usb_fd: Fix info-leaks to USB devices
can: peak_usb: pcan_usb_pro: Fix info-leaks to USB devices
perf/core: Fix creating kernel counters for PMUs that override event->cpu
tty/ldsem, locking/rwsem: Add missing ACQUIRE to read_failed sleep loop
scsi: ibmvfc: fix WARN_ON during event pool release
scsi: megaraid_sas: fix panic on loading firmware crashdump
ARM: davinci: fix sleep.S build error on ARMv4
perf probe: Avoid calling freeing routine multiple times for same pointer
ALSA: compress: Be more restrictive about when a drain is allowed
ALSA: compress: Prevent bypasses of set_params
ALSA: compress: Fix regression on compressed capture streams
s390/qdio: add sanity checks to the fast-requeue path
cpufreq/pasemi: fix use-after-free in pas_cpufreq_cpu_init()
hwmon: (nct6775) Fix register address and added missed tolerance for nct6106
mac80211: don't warn about CW params when not using them
iscsi_ibft: make ISCSI_IBFT dependson ACPI instead of ISCSI_IBFT_FIND
netfilter: nfnetlink: avoid deadlock due to synchronous request_module
can: peak_usb: fix potential double kfree_skb()
usb: yurex: Fix use-after-free in yurex_delete
perf db-export: Fix thread__exec_comm()
mm/vmalloc: Sync unmappings in __purge_vmap_area_lazy()
x86/mm: Sync also unmappings in vmalloc_sync_all()
x86/mm: Check for pfn instead of page in vmalloc_sync_one()
sound: fix a memory leak bug
usb: iowarrior: fix deadlock on disconnect
ANDROID: fix binder change in merge of 4.4.183
Conflicts:
sound/core/compress_offload.c
Change-Id: I30e498c36dc295fbfa0e1d455e31f192fd99479e
Signed-off-by: Srinivasarao P <spathi@codeaurora.org>
Currently ks_bridge driver queues requests to USB hardware in
open callback, but killing those urbs in disconnect() callback.
Due to this, USB host controller tries to ping on these endpoints
of ks_bridge driver, even though release callback is called. This
results in not using bandwidth for usecase data transfers in highspeed
mode. Fix this by killing those anchored urbs in release callback.
Change-Id: Iebe60125bfa3311337b0c9e5aaca44197a67c2be
Signed-off-by: Vijayavardhan Vennapusa <vvreddy@codeaurora.org>
-----BEGIN PGP SIGNATURE-----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=Kv4T
-----END PGP SIGNATURE-----
Merge 4.4.190 into android-4.4
Changes in 4.4.190
usb: iowarrior: fix deadlock on disconnect
sound: fix a memory leak bug
x86/mm: Check for pfn instead of page in vmalloc_sync_one()
x86/mm: Sync also unmappings in vmalloc_sync_all()
mm/vmalloc: Sync unmappings in __purge_vmap_area_lazy()
perf db-export: Fix thread__exec_comm()
usb: yurex: Fix use-after-free in yurex_delete
can: peak_usb: fix potential double kfree_skb()
netfilter: nfnetlink: avoid deadlock due to synchronous request_module
iscsi_ibft: make ISCSI_IBFT dependson ACPI instead of ISCSI_IBFT_FIND
mac80211: don't warn about CW params when not using them
hwmon: (nct6775) Fix register address and added missed tolerance for nct6106
cpufreq/pasemi: fix use-after-free in pas_cpufreq_cpu_init()
s390/qdio: add sanity checks to the fast-requeue path
ALSA: compress: Fix regression on compressed capture streams
ALSA: compress: Prevent bypasses of set_params
ALSA: compress: Be more restrictive about when a drain is allowed
perf probe: Avoid calling freeing routine multiple times for same pointer
ARM: davinci: fix sleep.S build error on ARMv4
scsi: megaraid_sas: fix panic on loading firmware crashdump
scsi: ibmvfc: fix WARN_ON during event pool release
tty/ldsem, locking/rwsem: Add missing ACQUIRE to read_failed sleep loop
perf/core: Fix creating kernel counters for PMUs that override event->cpu
can: peak_usb: pcan_usb_pro: Fix info-leaks to USB devices
can: peak_usb: pcan_usb_fd: Fix info-leaks to USB devices
hwmon: (nct7802) Fix wrong detection of in4 presence
ALSA: firewire: fix a memory leak bug
mac80211: don't WARN on short WMM parameters from AP
SMB3: Fix deadlock in validate negotiate hits reconnect
smb3: send CAP_DFS capability during session setup
mwifiex: fix 802.11n/WPA detection
scsi: mpt3sas: Use 63-bit DMA addressing on SAS35 HBA
sh: kernel: hw_breakpoint: Fix missing break in switch statement
usb: gadget: f_midi: fail if set_alt fails to allocate requests
USB: gadget: f_midi: fixing a possible double-free in f_midi
mm/memcontrol.c: fix use after free in mem_cgroup_iter()
ALSA: hda - Fix a memory leak bug
HID: holtek: test for sanity of intfdata
HID: hiddev: avoid opening a disconnected device
HID: hiddev: do cleanup in failure of opening a device
Input: kbtab - sanity check for endpoint type
Input: iforce - add sanity checks
net: usb: pegasus: fix improper read if get_registers() fail
xen/pciback: remove set but not used variable 'old_state'
irqchip/irq-imx-gpcv2: Forward irq type to parent
perf header: Fix divide by zero error if f_header.attr_size==0
perf header: Fix use of unitialized value warning
libata: zpodd: Fix small read overflow in zpodd_get_mech_type()
scsi: hpsa: correct scsi command status issue after reset
ata: libahci: do not complain in case of deferred probe
kbuild: modpost: handle KBUILD_EXTRA_SYMBOLS only for external modules
IB/core: Add mitigation for Spectre V1
ocfs2: remove set but not used variable 'last_hash'
asm-generic: fix -Wtype-limits compiler warnings
staging: comedi: dt3000: Fix signed integer overflow 'divider * base'
staging: comedi: dt3000: Fix rounding up of timer divisor
USB: core: Fix races in character device registration and deregistraion
usb: cdc-acm: make sure a refcount is taken early enough
USB: serial: option: add D-Link DWM-222 device ID
USB: serial: option: Add support for ZTE MF871A
USB: serial: option: add the BroadMobi BM818 card
USB: serial: option: Add Motorola modem UARTs
Backport minimal compiler_attributes.h to support GCC 9
include/linux/module.h: copy __init/__exit attrs to init/cleanup_module
arm64: compat: Allow single-byte watchpoints on all addresses
Input: psmouse - fix build error of multiple definition
asm-generic: default BUG_ON(x) to if(x)BUG()
scsi: fcoe: Embed fc_rport_priv in fcoe_rport structure
RDMA: Directly cast the sockaddr union to sockaddr
IB/mlx5: Make coding style more consistent
x86/vdso: Remove direct HPET access through the vDSO
iommu/amd: Move iommu_init_pci() to .init section
x86/boot: Disable the address-of-packed-member compiler warning
net/packet: fix race in tpacket_snd()
xen/netback: Reset nr_frags before freeing skb
net/mlx5e: Only support tx/rx pause setting for port owner
sctp: fix the transport error_count check
bonding: Add vlan tx offload to hw_enc_features
Linux 4.4.190
Change-Id: I2af7fee66e6ce77c41266cec8cfa7b7c4a78a05c
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
[ Upstream commit d595b03de2cb0bdf9bcdf35ff27840cc3a37158f ]
As commit 30d8177e8ac7 ("bonding: Always enable vlan tx offload")
said, we should always enable bonding's vlan tx offload, pass the
vlan packets to the slave devices with vlan tci, let them to handle
vlan implementation.
Now if encapsulation protocols like VXLAN is used, skb->encapsulation
may be set, then the packet is passed to vlan device which based on
bonding device. However in netif_skb_features(), the check of
hw_enc_features:
if (skb->encapsulation)
features &= dev->hw_enc_features;
clears NETIF_F_HW_VLAN_CTAG_TX/NETIF_F_HW_VLAN_STAG_TX. This results
in same issue in commit 30d8177e8ac7 like this:
vlan_dev_hard_start_xmit
-->dev_queue_xmit
-->validate_xmit_skb
-->netif_skb_features //NETIF_F_HW_VLAN_CTAG_TX is cleared
-->validate_xmit_vlan
-->__vlan_hwaccel_push_inside //skb->tci is cleared
...
--> bond_start_xmit
--> bond_xmit_hash //BOND_XMIT_POLICY_ENCAP34
--> __skb_flow_dissect // nhoff point to IP header
--> case htons(ETH_P_8021Q)
// skb_vlan_tag_present is false, so
vlan = __skb_header_pointer(skb, nhoff, sizeof(_vlan),
//vlan point to ip header wrongly
Fixes: b2a103e6d0 ("bonding: convert to ndo_fix_features")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit a1794de8b92ea6bc2037f445b296814ac826693e ]
As the annotation says in sctp_do_8_2_transport_strike():
"If the transport error count is greater than the pf_retrans
threshold, and less than pathmaxrtx ..."
It should be transport->error_count checked with pathmaxrxt,
instead of asoc->pf_retrans.
Fixes: 5aa93bcf66 ("sctp: Implement quick failover draft from tsvwg")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 466df6eb4a9e813b3cfc674363316450c57a89c5 ]
Only support changing tx/rx pause frame setting if the net device
is the vport group manager.
Fixes: 3c2d18ef22 ("net/mlx5e: Support ethtool get/set_pauseparam")
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 3a0233ddec554b886298de2428edb5c50a20e694 ]
At this point nr_frags has been incremented but the frag does not yet
have a page assigned so freeing the skb results in a crash. Reset
nr_frags before freeing the skb to prevent this.
Signed-off-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 20c6c189045539d29f4854d92b7ea9c329e1edfc upstream.
The clang warning 'address-of-packed-member' is disabled for the general
kernel code, also disable it for the x86 boot code.
This suppresses a bunch of warnings like this when building with clang:
./arch/x86/include/asm/processor.h:535:30: warning: taking address of
packed member 'sp0' of class or structure 'x86_hw_tss' may result in an
unaligned pointer value [-Waddress-of-packed-member]
return this_cpu_read_stable(cpu_tss.x86_tss.sp0);
^~~~~~~~~~~~~~~~~~~
./arch/x86/include/asm/percpu.h:391:59: note: expanded from macro
'this_cpu_read_stable'
#define this_cpu_read_stable(var) percpu_stable_op("mov", var)
^~~
./arch/x86/include/asm/percpu.h:228:16: note: expanded from macro
'percpu_stable_op'
: "p" (&(var)));
^~~
Signed-off-by: Matthias Kaehlcke <mka@chromium.org>
Cc: Doug Anderson <dianders@chromium.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20170725215053.135586-1-mka@chromium.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 24d2c521749d8547765b555b7a85cca179bb2275 upstream.
The function is only called from another __init function, so
it should be moved to .init too.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1ed95e52d902035e39a715ff3a314a893a96e5b7 upstream.
Allowing user code to map the HPET is problematic. HPET
implementations are notoriously buggy, and there are probably many
machines on which even MMIO reads from bogus HPET addresses are
problematic.
We have a report that the Dell Precision M2800 with:
ACPI: HPET 0x00000000C8FE6238 000038 (v01 DELL CBX3 01072009 AMI. 00000005)
is either so slow when accessing the HPET or actually hangs in some
regard, causing soft lockups to be reported if users do unexpected
things to the HPET.
The vclock HPET code has also always been a questionable speedup.
Accessing an HPET is exceedingly slow (on the order of several
microseconds), so the added overhead in requiring a syscall to read
the HPET is a small fraction of the total code of accessing it.
To avoid future problems, let's just delete the code entirely.
In the long run, this could actually be a speedup. Waiman Long as a
patch to optimize the case where multiple CPUs contend for the HPET,
but that won't help unless all the accesses are mediated by the
kernel.
Reported-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Waiman Long <Waiman.Long@hpe.com>
Cc: Waiman Long <waiman.long@hpe.com>
Link: http://lkml.kernel.org/r/d2f90bba98db9905041cff294646d290d378f67a.1460074438.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 0025b0bdeae7c13b8ab1dce64b0108ed9c071e2e upstream.
These three related functions can't agree whether to put the
umrwr on the stack dirty and then memset it, or to initialize
it on the stack. Make them all agree.
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 641114d2af312d39ca9bbc2369d18a5823da51c6 upstream.
gcc 9 now does allocation size tracking and thinks that passing the member
of a union and then accessing beyond that member's bounds is an overflow.
Instead of using the union member, use the entire union with a cast to
get to the sockaddr. gcc will now know that the memory extends the full
size of the union.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 023358b136d490ca91735ac6490db3741af5a8bd upstream.
Gcc-9 complains for a memset across pointer boundaries, which happens as
the code tries to allocate a flexible array on the stack. Turns out we
cannot do this without relying on gcc-isms, so with this patch we'll embed
the fc_rport_priv structure into fcoe_rport, can use the normal
'container_of' outcast, and will only have to do a memset over one
structure.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
