aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2015-11-09tpm_tis: restore IRQ vector in IO memory after failed probingMartin Wilck1-1/+5
If the probing finishes without success, it will leave the value 15 in the TPM_IRQ_VECTOR register. If the driver is unloaded and reloaded, it will "think" that the hardware had been programmed with IRQ 15, and will not probe again. This patch restores the original value in the IO memory if no IRQ is probed. Signed-off-by: Martin Wilck <[email protected]> Acked-by: Jarkko Sakkinen <[email protected]> Signed-off-by: Jarkko Sakkinen <[email protected]> Acked-by: Peter Huewe <[email protected]>
2015-11-09tpm_tis: free irq after probingMartin Wilck1-0/+2
Release IRQs used for probing only. Otherwise the TPM will end up with all IRQs 3-15 assigned. Fixes: afb5abc262e9 ("tpm: two-phase chip management functions") Signed-off-by: Martin Wilck <[email protected]> Reviewed-by: Jarkko Sakkinen <[email protected]> Tested-by: Jarkko Sakkinen <[email protected]> Signed-off-by: Jarkko Sakkinen <[email protected]> Acked-by: Peter Huewe <[email protected]>
2015-11-09GFS2: Fix rgrp end rounding problem for bsize < page sizeBob Peterson1-2/+3
This patch fixes a bug introduced by commit 7005c3e. That patch tries to map a vm range for resource groups, but the calculation breaks down when the block size is less than the page size. Signed-off-by: Bob Peterson <[email protected]>
2015-11-09arm64: fix R/O permissions of FDT mappingArd Biesheuvel2-1/+2
The mapping permissions of the FDT are set to 'PAGE_KERNEL | PTE_RDONLY' in an attempt to map the FDT as read-only. However, not only does this break at build time under STRICT_MM_TYPECHECKS (since the two terms are of different types in that case), it also results in both the PTE_WRITE and PTE_RDONLY attributes to be set, which means the region is still writable under ARMv8.1 DBM (and an attempted write will simply clear the PT_RDONLY bit). So instead, define PAGE_KERNEL_RO (which already has an established meaning across architectures) and use that instead. Signed-off-by: Ard Biesheuvel <[email protected]> Signed-off-by: Catalin Marinas <[email protected]>
2015-11-09arm64: fix STRICT_MM_TYPECHECKS issue in PTE_CONT manipulationArd Biesheuvel1-1/+1
The new page table code that manipulates the PTE_CONT flags does so in a way that is inconsistent with STRICT_MM_TYPECHECKS. Fix it by using the correct combination of __pgprot() and pgprot_val(). Signed-off-by: Ard Biesheuvel <[email protected]> Signed-off-by: Catalin Marinas <[email protected]>
2015-11-09hwmon: (k10temp) Remove duplicate pci-id defineAdam Majer1-1/+0
PCI_DEVICE_ID_AMD_15H_M60H_NB_F3 is now defined in pci_ids.h Signed-off-by: Adam Majer <[email protected]> Acked-by: Clemens Ladisch <[email protected]> Signed-off-by: Jean Delvare <[email protected]>
2015-11-09nios2: Remove unnecessary #ifdef guardsTobias Klauser2-4/+0
__HAVE_ARCH_MEMMOVE and __HAVE_ARCH_MEMSET are unconditionally defined for nios2, so there is no need to protect the function definitions of memmove() and memset(). Signed-off-by: Tobias Klauser <[email protected]> Acked-by: Ley Foon Tan <[email protected]>
2015-11-09nios2: Switch to generic __xchg()Marek Vasut1-47/+0
The generic __xchg() implementation present in asm-generic/cmpxchg.h is correct on nios2 and even generates the same code. Switch to this generic implementation to trim down the amount of ad-hoc copies of the code. Signed-off-by: Marek Vasut <[email protected]> Acked-by: Ley Foon Tan <[email protected]>
2015-11-09nios2: Fix unused variable warningMarek Vasut1-1/+1
Fix the following compiler splat by adding __maybe_unused annotation to the variable. Using this particular annotation has the least ugly impact on the code compared to using ifdeffery. arch/nios2/kernel/setup.c: In function 'nios2_boot_init': arch/nios2/kernel/setup.c:107:7: warning: unused variable 'cmdline_passed' [-Wunused-variable] char cmdline_passed[COMMAND_LINE_SIZE] = { 0, }; ^ Signed-off-by: Marek Vasut <[email protected]> Acked-by: Ley Foon Tan <[email protected]>
2015-11-08drivers: net: xgene: fix RGMII 10/100Mb modeIyappan Subramanian3-3/+48
This patch fixes the RGMII 10/100M mode by reprogramming the clock. Signed-off-by: Iyappan Subramanian <[email protected]> Tested-by: Fushen Chen <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2015-11-08Merge branch 'skb_to_full_sk'David S. Miller8-87/+149
Eric Dumazet says: ==================== net: add skb_to_full_sk() helper Many contexts need to reach listener socket from skb attached to a request socket. This patch series add skb_to_full_sk() to clearly express this need and use it where appropriate. ==================== Signed-off-by: David S. Miller <[email protected]>
2015-11-08netfilter: nft_meta: use skb_to_full_sk() helperEric Dumazet1-16/+20
SYNACK packets might be attached to request sockets. Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener") Signed-off-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2015-11-08net_sched: em_meta: use skb_to_full_sk() helperEric Dumazet1-46/+92
SYNACK packets might be attached to request sockets. Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener") Signed-off-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2015-11-08sched: cls_flow: use skb_to_full_sk() helperEric Dumazet1-4/+11
SYNACK packets might be attached to request sockets. Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener") Signed-off-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2015-11-08netfilter: xt_owner: use skb_to_full_sk() helperEric Dumazet1-2/+4
SYNACK packets might be attached to a request socket, xt_owner wants to gte the listener in this case. Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener") Signed-off-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2015-11-08smack: use skb_to_full_sk() helperEric Dumazet1-4/+7
This module wants to access sk->sk_security, which is not available for request sockets. Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener") Signed-off-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2015-11-08net: add skb_to_full_sk() helper and use it in selinux_netlbl_skbuff_setsid()Eric Dumazet3-15/+15
Generalize selinux_skb_sk() added in commit 212cd0895330 ("selinux: fix random read in selinux_ip_postroute_compat()") so that we can use it other contexts. Use it right away in selinux_netlbl_skbuff_setsid() Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener") Signed-off-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2015-11-08Merge tag 'nfc-fixes-4.4-1' of ↵David S. Miller4-33/+22
git://git.kernel.org/pub/scm/linux/kernel/git/sameo/nfc-fixes Samuel Ortiz says: ==================== NFC 4.4 fixes This is the 1st NFC fixes pull request for 4.4. It includes bug fixes and one fix for a build failure, all of them introduced with the first NFC pull request for 4.4. We have: - Fix nfcmrvl SPI driver potential build error due to a broken Kconfig dependency. - A few fixes for the firmware download implementation for the nfcmrvl UART driver. - A GPIO allocation leak for the nfcmrvl driver. - One code simplification for the nfcmrvl DT handling. ==================== Signed-off-by: David S. Miller <[email protected]>
2015-11-08bpf: doc: correct arch list for supported eBPF JITYang Shi1-3/+3
aarch64 and s390x support eBPF JIT too, correct document to reflect this and avoid any confusion. Signed-off-by: Yang Shi <[email protected]> Acked-by: Daniel Borkmann <[email protected]> Acked-by: Alexei Starovoitov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2015-11-08Merge tag 'xtensa-for-next-20151109' of git://github.com/jcmvbkbc/linux-xtensaChris Zankel29-248/+1210
Xtensa improvements for 4.4: - fix remaining issues with noMMU cores; - fix build for cores w/o cache or zero overhead loop options; - fix boot of secondary cores in SMP configuration; - add support for DMA to high memory pages; - add dma_to_phys and phys_to_dma functions.
2015-11-09xtensa: implement dma_to_phys and phys_to_dmaMax Filippov1-0/+10
This fixes the following build error seen in -next: drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c:143:2: error: implicit declaration of function 'dma_to_phys' Signed-off-by: Max Filippov <[email protected]>
2015-11-09xtensa: support DMA to high memoryMax Filippov2-17/+39
- don't bugcheck if high memory page is passed to xtensa_map_page; - turn empty dcache flush macros into functions so that they could be passed as function parameters; - use kmap_atomic to map high memory pages for cache invalidation/ flushing performed by xtensa_sync_single_for_{cpu,device}. Signed-off-by: Max Filippov <[email protected]>
2015-11-09raid5-cache: add crc32c Kconfig dependencyArnd Bergmann1-0/+1
The recent change of the raid5-cache code to use crc32c instead of crc32 causes link errors when CONFIG_LIBCRC32C is disabled: drivers/built-in.o: In function crc32c' core.c:(.text+0x1c6060): undefined reference to `crc32c' This adds an explicit 'select' statement like all other users of this function do. Signed-off-by: Arnd Bergmann <[email protected]> Fixes: 5cb2fbd6ea0d ("raid5-cache: use crc32c checksum") Signed-off-by: NeilBrown <[email protected]>
2015-11-08NTB: fix 32-bit compiler warningArnd Bergmann1-2/+2
resource_size_t may be 32-bit wide on some architectures, which causes this warning when building the NTB code: drivers/ntb/ntb_transport.c: In function 'ntb_transport_link_work': drivers/ntb/ntb_transport.c:828:46: warning: right shift count >= width of type [-Wshift-count-overflow] The warning is harmless but can be avoided by using the upper_32_bits() macro. Signed-off-by: Arnd Bergmann <[email protected]> Fixes: e26a5843f7f5 ("NTB: Split ntb_hw_intel and ntb_transport drivers") Signed-off-by: Jon Mason <[email protected]>
2015-11-08NTB: unify translation addressesDave Jiang2-18/+13
There is no need for the upstream and downstream addresses to be different for the NTB configs. Go to using a single set of address. It is still possible to configure them differently using module parameter override however. Signed-off-by: Dave Jiang <[email protected]> Acked and Tested-by: Allen Hubbe <[email protected]> Signed-off-by: Jon Mason <[email protected]>
2015-11-08NTB: invalid buf pointer in multi-MW setupsJon Mason1-2/+2
Order of operations issue with the QP Num and MW count, which would result in the receive buffer pointer being invalid if there are more than 1 MW. Corrected with parenthesis to enforce the proper order of operations. Reported-by: John I. Kading <[email protected]> Signed-off-by: Jon Mason <[email protected]>
2015-11-08NTB: remove unused variableSudip Mukherjee1-4/+0
These variables were not used anywhere. So remove them. Signed-off-by: Sudip Mukherjee <[email protected]> Signed-off-by: Jon Mason <[email protected]>
2015-11-08NTB: fix access of free-ed pointerSudip Mukherjee1-8/+7
We were accessing nt->mw_vec after freeing it. Fix the error path so that we free nt->mw_vec after we have finished using it. Signed-off-by: Sudip Mukherjee <[email protected]> Signed-off-by: Jon Mason <[email protected]>
2015-11-08NTB: Fix issue where we may be accessing NULL ptrDave Jiang1-8/+9
smatch detected an issue in the function ntb_transport_max_size() where we could be dereferencing a dma channel pointer when it is NULL. Reported-by: Dan Carpenter <[email protected]> Signed-off-by: Dave Jiang <[email protected]> Signed-off-by: Jon Mason <[email protected]>
2015-11-08parisc: Fixes and cleanups in kernel uapi header filesHelge Deller7-21/+44
This patch fixes some bugs and partly cleans up the parisc uapi header files to what glibc defined: - compat_semid64_ds was wrong and did not take the endianess into account - ipc64_perm exported userspace types which broke building userspace packages on debian (e.g. trinity) - ipc64_perm needs to use a 32bit mode_t on 64bit kernel - msqid64_ds and semid64_ds needs unsigned longs for various struct members - shmid64_ds exported size_t instead of __kernel_size_t And finally add some compile-time checks for the sizes of those structs to avoid future breakage. Runtime-tested with the Linux Test Project (LTP) testsuite. Cc: <[email protected]> # 3.18+ Reviewed-by: Arnd Bergmann <[email protected]> Signed-off-by: Helge Deller <[email protected]>
2015-11-08parisc: Drop hpux_stat64 struct from stat.h header fileHelge Deller1-31/+0
The struct hpux_stat64 is not needed any longer since we dropped HP-UX support in commit 04c1614 ("parisc: hpux - Drop support for HP-UX binaries"). Signed-off-by: Helge Deller <[email protected]>
2015-11-08h8300: Don't set CROSS_COMPILE unconditionallyJavier Martinez Canillas1-0/+2
The h8300 architecture Makefile sets the CROSS_COMPILE variable to "h8300-unknown-linux-" unconditionally but doing that prevents the variable to be used if a toolchain with a different binary name is installed in the machine. So only set CROSS_COMPILE to that default if the variable has not been set by the user before. Signed-off-by: Javier Martinez Canillas <[email protected]>
2015-11-08asm-generic: {get,put}_user ptr argument evaluate only 1 timeYoshinori Sato1-4/+6
Current implemantation ptr argument evaluate 2 times. It'll be an unexpected result. Changes v5: Remove unnecessary const. Changes v4: Temporary pointer type change to const void* Changes v3: Some build error fix. Changes v2: Argument x protect. Signed-off-by: Yoshinori Sato <[email protected]>
2015-11-08h8300: bit io fixYoshinori Sato1-6/+6
"WU" constraint required memory address. Signed-off-by: Yoshinori Sato <[email protected]>
2015-11-08h8300: zImage fixYoshinori Sato4-11/+7
- stack fix - configure base address Signed-off-by: Yoshinori Sato <[email protected]>
2015-11-08h8300: register address fixYoshinori Sato1-3/+3
Signed-off-by: Yoshinori Sato <[email protected]>
2015-11-08h8300: Fix alignment for .dataYoshinori Sato2-10/+8
Signed-off-by: Yoshinori Sato <[email protected]>
2015-11-08h8300: unaligned divcr register support.Yoshinori Sato1-1/+5
DIVCR is unaligned long word. So we need adjustment for long word align. Signed-off-by: Yoshinori Sato <[email protected]>
2015-11-08rtc: Add a driver for Micro Crystal RV8803Alexandre Belloni3-0/+531
This driver supports the following functions: - reading and settings time - alarms when connected to an IRQ - reading and clearing the voltage low flags - nvram Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: s3c: Set year, month, day value for setting alarmKrzysztof Kozlowski1-0/+16
This patch sets year, month, day value for set_alarm function. The current driver omits to set the values. This fixes setting wake alarm for dates different than current day. Without the patch the alarm scheduled for tomorrow would fire today on chosen time. Signed-off-by: Donggeun Kim <[email protected]> Signed-off-by: MyungJoo Ham <[email protected]> Signed-off-by: KyungMin Park <[email protected]> [k.kozlowski: Rebase and test the patch, update commit message] Signed-off-by: Krzysztof Kozlowski <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: ds1307: Fix alarm programming for mcp794xxTero Kristo1-2/+2
mcp794xx alarm registers must be written in BCD format. However, the alarm programming logic neglected this by adding one to the value after bin2bcd conversion has been already done, writing bad values to month register in case the alarm being set is in October. In this case, the alarm month value becomes 0x0a instead of the expected 0x10. Fix by moving the +1 addition within the bin2bcd call also. Fixes: 1d1945d261a2 ("drivers/rtc/rtc-ds1307.c: add alarm support for mcp7941x chips") Signed-off-by: Tero Kristo <[email protected]> Acked-by: Nishanth Menon <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: isl12057: enable support for the standard "wakeup-source" propertySudeep Holla1-5/+5
Though the isl12057 rtc driver should and will continue to support the legacy "isil,irq2-can-wakeup-machine" property to enable RTC as the wakeup source, we need to add support for the new standard property "wakeup-source". This patch adds support for "wakeup-source" property in addition to the existing "isil,irq2-can-wakeup-machine" property. Cc: Alessandro Zummo <[email protected]> Cc: Alexandre Belloni <[email protected]> Cc: [email protected] Signed-off-by: Sudeep Holla <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: opal: enable support for the stardard "wakeup-source" propertySudeep Holla1-2/+3
Though the opal rtc driver should and will continue to support the legacy "has-tpo" property to enable RTC as the wakeup source, we need to add support for the new standard property "wakeup-source" This patch adds support for "wakeup-source" property in addition to the existing "has-tpo" property. Cc: Alessandro Zummo <[email protected]> Cc: Alexandre Belloni <[email protected]> Cc: [email protected] Signed-off-by: Sudeep Holla <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: isl1208: Pass the IRQF_ONESHOT flagFabio Estevam1-1/+1
Since commit 1c6c69525b40eb76de8adf039409722015927dc3 ("genirq: Reject bogus threaded irq requests") threaded IRQs without a primary handler need to be requested with IRQF_ONESHOT, otherwise the request will fail. So pass the IRQF_ONESHOT flag in this case. The semantic patch that makes this change is available in scripts/coccinelle/misc/irqf_oneshot.cocci Signed-off-by: Fabio Estevam <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: pcf8563: add CLKOUT to common clock frameworkHeiko Schocher2-1/+194
Add the clkout output clk to the common clock framework. Disable the CLKOUT of the RTC after power-up. After power-up/reset of the RTC, CLKOUT is enabled by default, with CLKOUT enabled the RTC chip has 2-3 times higher power consumption. Signed-off-by: Heiko Schocher <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: davinci: remove incorrect reference to probe functionArnd Bergmann1-1/+0
The davinci rtc driver uses the module_platform_driver_probe() helper to call the probe function and mark it as __init, but it also puts a reference into its davinci_rtc_driver function. This will crash if we ever get a deferred probe and the probe function is called again after the init section has been removed. kbuild warns about this: WARNING: vmlinux.o(.data+0x1aa2b4): Section mismatch in reference from the variable davinci_rtc_driver to the function .init.text:davinci_rtc_probe() The variable davinci_rtc_driver references the function __init davinci_rtc_probe() This patch removes the .probe callback from the platform driver, which avoids those problems. Signed-off-by: Arnd Bergmann <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: at91rm9200: clear RTC alarm status flag prior to suspendingWenyou Yang1-0/+2
As said in the SAMA5D2 datasheet, "Prior to instructing the device to enter ULP mode 1, ... and the internal sources of wake-up must be cleared." This patch is to clear the RTC alarm status flag prior to suspending to avoid the erroneous wake-up activity, as it is often used as the wake-up source for the ULP mode 1. Signed-off-by: Wenyou Yang <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: pcf2127: remove useless driver versionUwe Kleine-König1-5/+0
A driver version is only really sensible for oot drivers. Also the dev_info about having found a chip only signals that allocating the driver data succeeded and so isn't worth much. Signed-off-by: Uwe Kleine-König <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: pcf2127: fix reading uninitialized value on RTC_READ_VL ioctlUwe Kleine-König1-18/+22
The flag reported on the RTC_READ_VL ioctl is only initialized when the date is read out. So the voltage low value doesn't represent reality but the status at the time the date was read (or 0 if the date was not read yet). Moreover when userspace requests a value via an ioctl there is no added benefit to also make a prosa representation of this (and other) values appear in the kernel log so remove the calls to dev_info and the driver data members to track their state. Signed-off-by: Uwe Kleine-König <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
2015-11-08rtc: stmp3xxx: unify register access macrosHarald Geyer1-14/+13
Use STMP_OFFSET_REG_(SET|CLR) instead of defining _SET and _CLR for STMP3XXX_RTC_CTRL and STMP3XXX_RTC_PERSISTENT0 - no functional changes. Signed-off-by: Harald Geyer <[email protected]> Signed-off-by: Alexandre Belloni <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.