Linux kernel stable tree (mirror) https://git.radix-linux.su/kernel/linux.git/atom?h=linux-4.16.y 2018-02-04T00:25:42+00:00 2018-02-04T00:25:42+00:00 Linus Torvalds torvalds@linux-foundation.org 2018-02-04T00:25:42+00:00 urn:sha1:617aebe6a97efa539cc4b8a52adccd89596e6be0 Pull hardened usercopy whitelisting from Kees Cook: "Currently, hardened usercopy performs dynamic bounds checking on slab cache objects. This is good, but still leaves a lot of kernel memory available to be copied to/from userspace in the face of bugs. To further restrict what memory is available for copying, this creates a way to whitelist specific areas of a given slab cache object for copying to/from userspace, allowing much finer granularity of access control. Slab caches that are never exposed to userspace can declare no whitelist for their objects, thereby keeping them unavailable to userspace via dynamic copy operations. (Note, an implicit form of whitelisting is the use of constant sizes in usercopy operations and get_user()/put_user(); these bypass all hardened usercopy checks since these sizes cannot change at runtime.) This new check is WARN-by-default, so any mistakes can be found over the next several releases without breaking anyone's system. The series has roughly the following sections: - remove %p and improve reporting with offset - prepare infrastructure and whitelist kmalloc - update VFS subsystem with whitelists - update SCSI subsystem with whitelists - update network subsystem with whitelists - update process memory with whitelists - update per-architecture thread_struct with whitelists - update KVM with whitelists and fix ioctl bug - mark all other allocations as not whitelisted - update lkdtm for more sensible test overage" * tag 'usercopy-v4.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux: (38 commits) lkdtm: Update usercopy tests for whitelisting usercopy: Restrict non-usercopy caches to size 0 kvm: x86: fix KVM_XEN_HVM_CONFIG ioctl kvm: whitelist struct kvm_vcpu_arch arm: Implement thread_struct whitelist for hardened usercopy arm64: Implement thread_struct whitelist for hardened usercopy x86: Implement thread_struct whitelist for hardened usercopy fork: Provide usercopy whitelisting for task_struct fork: Define usercopy region in thread_stack slab caches fork: Define usercopy region in mm_struct slab caches net: Restrict unwhitelisted proto caches to size 0 sctp: Copy struct sctp_sock.autoclose to userspace using put_user() sctp: Define usercopy region in SCTP proto slab cache caif: Define usercopy region in caif proto slab cache ip: Define usercopy region in IP proto slab cache net: Define usercopy region in struct proto slab cache scsi: Define usercopy region in scsi_sense_cache slab cache cifs: Define usercopy region in cifs_request slab cache vxfs: Define usercopy region in vxfs_inode slab cache ufs: Define usercopy region in ufs_inode_cache slab cache ... 2018-02-02T17:50:51+00:00 Linus Torvalds torvalds@linux-foundation.org 2018-02-02T17:50:51+00:00 urn:sha1:367b0df173b0ebea5d18b6971c244e260b5feb17 Pull ARM updates from Russell King: - StrongARM SA1111 updates to modernise and remove cruft - Add StrongARM gpio drivers for board GPIOs - Verify size of zImage is what we expect to avoid issues with appended DTB - nommu updates from Vladimir Murzin - page table read-write-execute checking from Jinbum Park - Broadcom Brahma-B15 cache updates from Florian Fainelli - Avoid failure with kprobes test caused by inappropriately placed kprobes - Remove __memzero optimisation (which was incorrectly being used directly by some drivers) * 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm: (32 commits) ARM: 8745/1: get rid of __memzero() ARM: 8744/1: don't discard memblock for kexec ARM: 8743/1: bL_switcher: add MODULE_LICENSE tag ARM: 8742/1: Always use REFCOUNT_FULL ARM: 8741/1: B15: fix unused label warnings ARM: 8740/1: NOMMU: Make sure we do not hold stale data in mem[] array ARM: 8739/1: NOMMU: Setup VBAR/Hivecs for secondaries cores ARM: 8738/1: Disable CONFIG_DEBUG_VIRTUAL for NOMMU ARM: 8737/1: mm: dump: add checking for writable and executable ARM: 8736/1: mm: dump: make the page table dumping seq_file ARM: 8735/1: mm: dump: make page table dumping reusable ARM: sa1100/neponset: add GPIO drivers for control and modem registers ARM: sa1100/assabet: add BCR/BSR GPIO driver ARM: 8734/1: mm: idmap: Mark variables as ro_after_init ARM: 8733/1: hw_breakpoint: Mark variables as __ro_after_init ARM: 8732/1: NOMMU: Allow userspace to access background MPU region ARM: 8727/1: MAINTAINERS: Update brcmstb entries to cover B15 code ARM: 8728/1: B15: Register reboot notifier for KEXEC ARM: 8730/1: B15: Add suspend/resume hooks ARM: 8726/1: B15: Add CPU hotplug awareness ... 2018-02-02T00:07:54+00:00 Linus Torvalds torvalds@linux-foundation.org 2018-02-02T00:07:54+00:00 urn:sha1:537433b6241e067de2d9da3bed5f4fed9c9eac58 Pull ARM SoC device tree updates from Arnd Bergmann: "We get a moderate number of new machines this time, and only one new SoC variant (Actions S700): Actions: - S700 Soc and CubieBoard7 development board - Allo.com Sparky Single-board-computer Allwinner: - Orange Pi R1 development board - Libre Computer Board ALL-H3-CC H3 single-board computer ASpeed ast2x00: - Witherspoon: OpenPower Power9 server manufactured by IBM that uses the ASPEED ast2500 - Zaius: OpenPower Power9 server manufactured by Invatech that uses the ASPEED ast2500 - Q71L: Intel Xeon server manufactured by Qanta that uses the ASPEED ast2400 AT91: - Axentia Nattis/Natte digital signage - sama5d2 PTC-ek Evaluation board Freescale/NXP i.MX: - SolidRun Humminboard2 development board - Variscite DART-MX6 SoM and Carrier-board - Technologic TS-4600 and TS-7970 development board - Toradex Colibri iMX7D SoM board - v1.5 variant of Solidrun Cubox-i and Hummingboard Freescale/NXP Layerscape: - Moxa UC-8410A Series industrial computer Gemini: - D-Link DNS-313 NAS enclosure OMAP: - LogicPD OMAP35xx SOM-LV devkit - LogicPD OMAP35xx Torpedo devkit Renesas: - r8a77970 (V3M) Starter Kit board - r8a7795 (M3-W) Salvator-XS board We finally managed to get the dtc warnings under control, with no more build-time warnings for bad device tree files. This includes fixes for the majority of platforms, including nomadik, samsung, lpc32xx, STi, spear, mediatek, freescale, qcom, realview, keystone, omap, kirkwood, renesas, hisilicon, and broadcom. Files get rearranged on a few platforms, in particular the Marvell Armada 7K/8K device tree files are changed in preparation for future SoC support, based on more than two of the same chips in one package, and some boards get renamed for oxnas for consistency. Finally, many existing SoCs gain descriptions for additional on-chip devices that we can now support with kernel drivers: - Allwinner A83t (drm, ethernet, i2c, ...), H3/H5 (USB-OTG) - Amlogic AXG family (clk, pinctrl, pwm, ...), and others (vpu, hdmi) - Aspeed clk controller support - Freescale LS1088A, LS1021A device support - Gemini Ethernet, PCI, TVE, panel - Keystone gpio, qspi, more uarts - Mediatek cpufreq, regulator, clock, reset - Marvell thermal, cpufreq, nand - Renesas SMP, thermal, timer, PWM, sound, phy, ipmmu - Rockchip Mipi, GPU, display - Samsung Exynos5433 PMU, power domain, nfc - Spreadtrum: sc9860 clocks - Tegra TX2 PSDI, HDMI, I2C,SMMU, display, fuse, ..." * tag 'armsoc-dt' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (690 commits) arm64: dts: stratix10: fix SPI settings ARM: dts: socfpga: add i2c reset signals arm64: dts: stratix10: add USB ECC reset bit arm64: dts: stratix10: enable USB on the devkit ARM: dts: socfpga: disable over-current for Arria10 USB devkit ARM: dts: Nokia N9: add support for up/down keys in the dts ARM: dts: nomadik: add interrupt-parent for clcd ARM: dts: Add ethernet to a bunch of platforms ARM: dts: Add ethernet to the Gemini SoC ARM: dts: rename oxnas dts files ARM: dts: s5pv210: add interrupt-parent for ohci ARM: lpc3250: fix uda1380 gpio numbers ARM: dts: STi: Add gpio polarity for "hdmi,hpd-gpio" property ARM: dts: dra7: Reduce shut down temperature of non-cpu thermal zones ARM: dts: n900: Add aliases for lcd and tvout displays ARM: dts: Update ti-sysc data for existing users ARM: dts: Fix smartreflex compatible for omap3 shared mpu-iva instance arm64: dts: marvell: armada-80x0: Fix pinctrl compatible string arm: spear13xx: Fix spics gpio controller's warning arm: spear13xx: Fix dmas cells ... 2018-01-21T15:36:50+00:00 Arnd Bergmann arnd@arndb.de 2018-01-16T13:48:14+00:00 urn:sha1:ec80eb467171b511635b9e3086fec357f79afe3b Discarding the memblock arrays usually works, but causes problems with kexec, as pointed out by this kbuild warning: WARNING: vmlinux.o(.text+0x7c60): Section mismatch in reference from the function machine_kexec_prepare() to the function .meminit.text:memblock_is_region_memory() This lets us keep the memblock structures around whenever kexec is enabled, but otherwise still drops them. Fixes: cf1b09908a23 ("ARM: 8693/1: discard memblock arrays when possible") Acked-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> 2018-01-21T15:32:26+00:00 Jinbum Park jinb.park7@gmail.com 2018-01-09T23:54:37+00:00 urn:sha1:b26d07a0f5568bec782072006f25bbaaed49fb2e refcount_t overflow detection is implemented as two way. 1. REFCOUNT_FULL - It means the full refcount_t implementation which has validation but is slightly slower. - (fd25d19f6b8d ("locking/refcount: Create unchecked atomic_t implementation")) 2. ARCH_HAS_REFCOUNT - refcount_t overflow detection can be optimized via an arch-dependent way. - It is based on atomic_t infrastructure with some instruction added for detection. - It is faster than REFCOUNT_FULL, as fast as unprotected atomic_t infrastructure. - (7a46ec0e2f48 ("locking/refcounts, x86/asm: Implement fast refcount overflow protection")) ARCH_HAS_REFCOUNT has implemented for x86, not implemented for others. In the case of arm64, Will Deacon said he didn't want the specialized "fast but technically incomplete" refcounting as seen with x86's. But rather to set REFCOUNT_FULL by default because no one could point to real-world performance impacts with REFCOUNT_FULL vs unprotected atomic_t infrastructure. This is the reason arm64 ended up enabling REFCOUNT_FULL. (4adcec1164de ("arm64: Always use REFCOUNT_FULL")) As with the decision of arm64, arm can set REFCOUNT_FULL by default. Acked-by: Kees Cook <keescook@chromium.org> Signed-off-by: Jinbum Park <jinb.park7@gmail.com> Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> 2018-01-21T15:32:21+00:00 Vladimir Murzin vladimir.murzin@arm.com 2017-12-18T10:48:42+00:00 urn:sha1:c7780ab56c091a9ba95a3278e6e1d9c73afb5052 While running MPS2 platform (NOMMU) with DTB placed below PHYS_OFFSET following warning poped up: ------------[ cut here ]------------ WARNING: CPU: 0 PID: 0 at arch/arm/mm/physaddr.c:42 __virt_to_phys+0x2f/0x40 virt_to_phys used for non-linear address: 00004000 (0x4000) CPU: 0 PID: 0 Comm: swapper Not tainted 4.15.0-rc1-5a31bf2-clean+ #2767 Hardware name: MPS2 (Device Tree Support) [<2100bf39>] (unwind_backtrace) from [<2100b3ff>] (show_stack+0xb/0xc) [<2100b3ff>] (show_stack) from [<2100e697>] (__warn+0x87/0xac) [<2100e697>] (__warn) from [<2100e6db>] (warn_slowpath_fmt+0x1f/0x28) [<2100e6db>] (warn_slowpath_fmt) from [<2100c603>] (__virt_to_phys+0x2f/0x40) [<2100c603>] (__virt_to_phys) from [<2116a499>] (early_init_fdt_reserve_self+0xd/0x24) [<2116a499>] (early_init_fdt_reserve_self) from [<2116222d>] (arm_memblock_init+0xb5/0xf8) [<2116222d>] (arm_memblock_init) from [<21161cad>] (setup_arch+0x38b/0x50e) [<21161cad>] (setup_arch) from [<21160455>] (start_kernel+0x31/0x280) [<21160455>] (start_kernel) from [<00000000>] ( (null)) random: get_random_bytes called from init_oops_id+0x17/0x2c with crng_init=0 ---[ end trace 0000000000000000 ]--- Platforms without MMU support run with 1:1 (i.e. linear) memory mapping, so disable CONFIG_DEBUG_VIRTUAL. Fixes: e377cd8221eb ("ARM: 8640/1: Add support for CONFIG_DEBUG_VIRTUAL") Signed-off-by: Vladimir Murzin <vladimir.murzin@arm.com> Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> 2018-01-15T20:08:06+00:00 Kees Cook keescook@chromium.org 2017-08-16T21:09:13+00:00 urn:sha1:08626a6056aad824c43d34ce587ab2b01f49d1a4 While ARM32 carries FPU state in the thread structure that is saved and restored during signal handling, it doesn't need to declare a usercopy whitelist, since existing accessors are all either using a bounce buffer (for which whitelisting isn't checking the slab), are statically sized (which will bypass the hardened usercopy check), or both. Cc: Russell King <linux@armlinux.org.uk> Cc: Ingo Molnar <mingo@kernel.org> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: "Peter Zijlstra (Intel)" <peterz@infradead.org> Cc: linux-arm-kernel@lists.infradead.org Signed-off-by: Kees Cook <keescook@chromium.org> 2018-01-15T08:35:06+00:00 Christoph Hellwig hch@lst.de 2018-01-09T15:30:23+00:00 urn:sha1:002e67454f61bb67d8071ac4d0cacb86a01d18e0 The trivial direct mapping implementation already does a virtual to physical translation which isn't strictly a noop, and will soon learn to do non-direct but linear physical to dma translations through the device offset and a few small tricks. Rename it to a better fitting name. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com> 2018-01-10T15:40:54+00:00 Christoph Hellwig hch@lst.de 2018-01-10T15:21:13+00:00 urn:sha1:ea8c64ace86647260ec4255f483e5844d62af2df phys_to_dma, dma_to_phys and dma_capable are helpers published by architecture code for use of swiotlb and xen-swiotlb only. Drivers are not supposed to use these directly, but use the DMA API instead. Move these to a new asm/dma-direct.h helper, included by a linux/dma-direct.h wrapper that provides the default linear mapping unless the architecture wants to override it. In the MIPS case the existing dma-coherent.h is reused for now as untangling it will take a bit of work. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Robin Murphy <robin.murphy@arm.com> 2017-12-21T15:11:12+00:00 Masahiro Yamada yamada.masahiro@socionext.com 2017-11-27T02:19:23+00:00 urn:sha1:ea04d6b456b2b7a7163680fd1bd19567203faf7a ARCH_S3C24XX is a very exceptional platform that some DT files in arch/arm/boot/dts/, but does not select USE_OF. All the other platforms with DT files correctly select USE_OF directly or indirectly (Most of them are either ARCH_MULTIPLATFORM or ARM_SINGLE_ARMV7M). With ARCH_S3C24XX fixed, "ifeq ($(CONFIG_OF),y)" in DT Makefile can be deleted. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de>