user/sven/linux.git/arch, branch v5.4.140

arm64: fix compat syscall return truncation

2021-08-12T11:21:05Z

commit e30e8d46cf605d216a799a28c77b8a41c328613a upstream. Due to inconsistencies in the way we manipulate compat GPRs, we have a few issues today: * For audit and tracing, where error codes are handled as a (native) long, negative error codes are expected to be sign-extended to the native 64-bits, or they may fail to be matched correctly. Thus a syscall which fails with an error may erroneously be identified as failing. * For ptrace, *all* compat return values should be sign-extended for consistency with 32-bit arm, but we currently only do this for negative return codes. * As we may transiently set the upper 32 bits of some compat GPRs while in the kernel, these can be sampled by perf, which is somewhat confusing. This means that where a syscall returns a pointer above 2G, this will be sign-extended, but will not be mistaken for an error as error codes are constrained to the inclusive range [-4096, -1] where no user pointer can exist. To fix all of these, we must consistently use helpers to get/set the compat GPRs, ensuring that we never write the upper 32 bits of the return code, and always sign-extend when reading the return code. This patch does so, with the following changes: * We re-organise syscall_get_return_value() to always sign-extend for compat tasks, and reimplement syscall_get_error() atop. We update syscall_trace_exit() to use syscall_get_return_value(). * We consistently use syscall_set_return_value() to set the return value, ensureing the upper 32 bits are never set unexpectedly. * As the core audit code currently uses regs_return_value() rather than syscall_get_return_value(), we special-case this for compat_user_mode(regs) such that this will do the right thing. Going forward, we should try to move the core audit code over to syscall_get_return_value(). Cc: Reported-by: He Zhe Reported-by: weiyuchen Cc: Catalin Marinas Cc: Will Deacon Reviewed-by: Catalin Marinas Link: https://lore.kernel.org/r/20210802104200.21390-1-mark.rutland@arm.com Signed-off-by: Will Deacon [Mark: trivial conflict resolution for v5.4.y] Signed-off-by: Mark Rutland Signed-off-by: Greg Kroah-Hartman

alpha: Send stop IPI to send to online CPUs

2021-08-12T11:21:05Z

[ Upstream commit caace6ca4e06f09413fb8f8a63319594cfb7d47d ] This issue was noticed while debugging a shutdown issue where some secondary CPUs are not being shutdown correctly. A fix for that [1] requires that secondary cpus be offlined using the cpu_online_mask so that the stop operation is a no-op if CPU HOTPLUG is disabled. I, like the author in [1] looked at the architectures and found that alpha is one of two architectures that executes smp_send_stop() on all possible CPUs. On alpha, smp_send_stop() sends an IPI to all possible CPUs but only needs to send them to online CPUs. Send the stop IPI to only the online CPUs. [1] https://lkml.org/lkml/2020/1/10/250 Signed-off-by: Prarit Bhargava Cc: Richard Henderson Cc: Ivan Kokshaysky Signed-off-by: Matt Turner Signed-off-by: Sasha Levin

perf/x86/amd: Don't touch the AMD64_EVENTSEL_HOSTONLY bit inside the guest

2021-08-12T11:21:04Z

commit df51fe7ea1c1c2c3bfdb81279712fdd2e4ea6c27 upstream. If we use "perf record" in an AMD Milan guest, dmesg reports a #GP warning from an unchecked MSR access error on MSR_F15H_PERF_CTLx: [] unchecked MSR access error: WRMSR to 0xc0010200 (tried to write 0x0000020000110076) at rIP: 0xffffffff8106ddb4 (native_write_msr+0x4/0x20) [] Call Trace: [] amd_pmu_disable_event+0x22/0x90 [] x86_pmu_stop+0x4c/0xa0 [] x86_pmu_del+0x3a/0x140 The AMD64_EVENTSEL_HOSTONLY bit is defined and used on the host, while the guest perf driver should avoid such use. Fixes: 1018faa6cf23 ("perf/x86/kvm: Fix Host-Only/Guest-Only counting with SVM disabled") Signed-off-by: Like Xu Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Liam Merwick Tested-by: Kim Phillips Tested-by: Liam Merwick Link: https://lkml.kernel.org/r/20210802070850.35295-1-likexu@tencent.com Signed-off-by: Greg Kroah-Hartman

arm64: vdso: Avoid ISB after reading from cntvct_el0

2021-08-12T11:21:04Z

commit 77ec462536a13d4b428a1eead725c4818a49f0b1 upstream. We can avoid the expensive ISB instruction after reading the counter in the vDSO gettime functions by creating a fake address hazard against a dummy stack read, just like we do inside the kernel. Signed-off-by: Will Deacon Reviewed-by: Vincenzo Frascino Link: https://lore.kernel.org/r/20210318170738.7756-5-will@kernel.org Signed-off-by: Catalin Marinas Signed-off-by: Greg Kroah-Hartman Signed-off-by: Chanho Park

KVM: x86/mmu: Fix per-cpu counter corruption on 32-bit builds

2021-08-12T11:21:03Z

commit d5aaad6f83420efb8357ac8e11c868708b22d0a9 upstream. Take a signed 'long' instead of an 'unsigned long' for the number of pages to add/subtract to the total number of pages used by the MMU. This fixes a zero-extension bug on 32-bit kernels that effectively corrupts the per-cpu counter used by the shrinker. Per-cpu counters take a signed 64-bit value on both 32-bit and 64-bit kernels, whereas kvm_mod_used_mmu_pages() takes an unsigned long and thus an unsigned 32-bit value on 32-bit kernels. As a result, the value used to adjust the per-cpu counter is zero-extended (unsigned -> signed), not sign-extended (signed -> signed), and so KVM's intended -1 gets morphed to 4294967295 and effectively corrupts the counter. This was found by a staggering amount of sheer dumb luck when running kvm-unit-tests on a 32-bit KVM build. The shrinker just happened to kick in while running tests and do_shrink_slab() logged an error about trying to free a negative number of objects. The truly lucky part is that the kernel just happened to be a slightly stale build, as the shrinker no longer yells about negative objects as of commit 18bb473e5031 ("mm: vmscan: shrink deferred objects proportional to priority"). vmscan: shrink_slab: mmu_shrink_scan+0x0/0x210 [kvm] negative objects to delete nr=-858993460 Fixes: bc8a3d8925a8 ("kvm: mmu: Fix overflow on kvm mmu page limit calculation") Cc: stable@vger.kernel.org Cc: Ben Gardon Signed-off-by: Sean Christopherson Message-Id: <20210804214609.1096003-1-seanjc@google.com> Reviewed-by: Jim Mattson Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman

KVM: x86: accept userspace interrupt only if no event is injected

2021-08-12T11:21:03Z

commit fa7a549d321a4189677b0cea86e58d9db7977f7b upstream. Once an exception has been injected, any side effects related to the exception (such as setting CR2 or DR6) have been taked place. Therefore, once KVM sets the VM-entry interruption information field or the AMD EVENTINJ field, the next VM-entry must deliver that exception. Pending interrupts are processed after injected exceptions, so in theory it would not be a problem to use KVM_INTERRUPT when an injected exception is present. However, DOSEMU is using run->ready_for_interrupt_injection to detect interrupt windows and then using KVM_SET_SREGS/KVM_SET_REGS to inject the interrupt manually. For this to work, the interrupt window must be delayed after the completion of the previous event injection. Cc: stable@vger.kernel.org Reported-by: Stas Sergeev Tested-by: Stas Sergeev Fixes: 71cc849b7093 ("KVM: x86: Fix split-irqchip vs interrupt injection window request") Reviewed-by: Sean Christopherson Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman

MIPS: Malta: Do not byte-swap accesses to the CBUS UART

2021-08-12T11:21:02Z

commit 9a936d6c3d3d6c33ecbadf72dccdb567b5cd3c72 upstream. Correct big-endian accesses to the CBUS UART, a Malta on-board discrete TI16C550C part wired directly to the system controller's device bus, and do not use byte swapping with the 32-bit accesses to the device. The CBUS is used for devices such as the boot flash memory needed early on in system bootstrap even before PCI has been initialised. Therefore it uses the system controller's device bus, which follows the endianness set with the CPU, which means no byte-swapping is ever required for data accesses to CBUS, unlike with PCI. The CBUS UART uses the UPIO_MEM32 access method, that is the `readl' and `writel' MMIO accessors, which on the MIPS platform imply byte-swapping with PCI systems. Consequently the wrong byte lane is accessed with the big-endian configuration and the UART is not correctly accessed. As it happens the UPIO_MEM32BE access method makes use of the `ioread32' and `iowrite32' MMIO accessors, which still use `readl' and `writel' respectively, however they byte-swap data passed, effectively cancelling swapping done with the accessors themselves and making it suitable for the CBUS UART. Make the CBUS UART switch between UPIO_MEM32 and UPIO_MEM32BE then, based on the endianness selected. With this change in place the device is correctly recognised with big-endian Malta at boot, along with the Super I/O devices behind PCI: Serial: 8250/16550 driver, 5 ports, IRQ sharing enabled printk: console [ttyS0] disabled serial8250.0: ttyS0 at I/O 0x3f8 (irq = 4, base_baud = 115200) is a 16550A printk: console [ttyS0] enabled printk: bootconsole [uart8250] disabled serial8250.0: ttyS1 at I/O 0x2f8 (irq = 3, base_baud = 115200) is a 16550A serial8250.0: ttyS2 at MMIO 0x1f000900 (irq = 20, base_baud = 230400) is a 16550A Fixes: e7c4782f92fc ("[MIPS] Put an end to 's long and annyoing existence") Cc: stable@vger.kernel.org # v2.6.23+ Reviewed-by: Philippe Mathieu-Daudé Signed-off-by: Maciej W. Rozycki Link: https://lore.kernel.org/r/alpine.DEB.2.21.2106260524430.37803@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman

mips: Fix non-POSIX regexp

2021-08-12T11:20:57Z

[ Upstream commit 28bbbb9875a35975904e46f9b06fa689d051b290 ] When cross compiling a MIPS kernel on a BSD based HOSTCC leads to errors like SYNC include/config/auto.conf.cmd - due to: .config egrep: empty (sub)expression UPD include/config/kernel.release HOSTCC scripts/dtc/dtc.o - due to target missing It turns out that egrep uses this egrep pattern: (|MINOR_|PATCHLEVEL_) This is not valid syntax or gives undefined results according to POSIX 9.5.3 ERE Grammar https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap09.html It seems to be silently accepted by the Linux egrep implementation while a BSD host complains. Such patterns can be replaced by a transformation like "(|p1|p2)" -> "(p1|p2)?" Fixes: 48c35b2d245f ("[MIPS] There is no __GNUC_MAJOR__") Signed-off-by: H. Nikolaus Schaller Signed-off-by: Masahiro Yamada Signed-off-by: Sasha Levin

omap5-board-common: remove not physically existing vdds_1v8_main fixed-regulator

2021-08-12T11:20:55Z

[ Upstream commit c68ef4ad180e09805fa46965d15e1dfadf09ffa5 ] This device tree include file describes a fixed-regulator connecting smps7_reg output (1.8V) to some 1.8V rail and consumers (vdds_1v8_main). This regulator does not physically exist. I assume it was introduced as a wrapper around smps7_reg to provide a speaking signal name "vdds_1v8_main" as label. This fixed-regulator without real function was not an issue in driver code until Commit 98e48cd9283d ("regulator: core: resolve supply for boot-on/always-on regulators") introduced a new check for regulator initialization which makes Palmas regulator registration fail: [ 5.407712] ldo1: supplied by vsys_cobra [ 5.412748] ldo2: supplied by vsys_cobra [ 5.417603] palmas-pmic 48070000.i2c:palmas@48:palmas_pmic: failed to register 48070000.i2c:palmas@48:palmas_pmic regulator The reason is that the supply-chain of regulators is too long and goes from ldo3 through the virtual vdds_1v8_main regulator and then back to smps7. This adds a cross-dependency of probing Palmas regulators and the fixed-regulator which leads to probe deferral by the new check and is no longer resolved. Since we do not control what device tree files including this one reference (either &vdds_1v8_main or &smps7_reg or both) we keep both labels for smps7 for compatibility. Fixes: 98e48cd9283d ("regulator: core: resolve supply for boot-on/always-on regulators") Signed-off-by: H. Nikolaus Schaller Signed-off-by: Tony Lindgren Signed-off-by: Sasha Levin

ARM: dts: am437x-l4: fix typo in can@0 node

2021-08-12T11:20:55Z

[ Upstream commit 0162a9964365fd26e34575e121b17d021204c481 ] Replace clock-name with clock-names. Fixes: 2a4117df9b43 ("ARM: dts: Fix dcan driver probe failed on am437x platform") Signed-off-by: Dario Binacchi Signed-off-by: Tony Lindgren Signed-off-by: Sasha Levin