)]}' { "log": [ { "commit": "ad062195731bea1624ce7160e79e0fcdaa25c1b5", "tree": "13a8b484abc78118de172195fc87d4d9e38efb80", "parents": [ "7ac63f6ba5db5e2e81e4674551d6f9ec58e70618", "f690790c9da3122dd7ee1b0d64d97973a7c34135" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:59:10 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:59:10 2019 -0700" }, "message": "Merge tag \u0027platform-drivers-x86-v5.4-1\u0027 of git://git.infradead.org/linux-platform-drivers-x86\n\nPull x86 platform-drivers updates from Andy Shevchenko:\n\n - ASUS WMI driver got a couple of updates, i.e. support of FAN is fixed\n for recent products and the charge threshold support has been added\n\n - Two uknown key events for Dell laptops are being ignored now to avoid\n spamming users with harmless messages\n\n - HP ZBook 17 G5 and ASUS Zenbook UX430UNR got accelerometer support.\n\n - Intel CherryTrail platforms had a regression with wake up. Now it\u0027s\n fixed\n\n - Intel PMC driver got fixed in order to work nicely in Xen\n environment\n\n - Intel Speed Select driver provides bucket vs core count relationship.\n Besides that the tools has been updated for better output\n\n - The PrivacyGuard is enabled on Lenovo ThinkPad laptops\n\n - Three tablets - Trekstor Primebook C11B 2-in-1, Irbis TW90 and Chuwi\n Surbook Mini - got touchscreen support\n\n* tag \u0027platform-drivers-x86-v5.4-1\u0027 of git://git.infradead.org/linux-platform-drivers-x86: (53 commits)\n MAINTAINERS: Switch PDx86 subsystem status to Odd Fixes\n platform/x86: asus-wmi: Refactor charge threshold to use the battery hooking API\n platform/x86: asus-wmi: Rename CHARGE_THRESHOLD to RSOC\n platform/x86: asus-wmi: Reorder ASUS_WMI_CHARGE_THRESHOLD\n tools/power/x86/intel-speed-select: Display core count for bucket\n platform/x86: ISST: Allow additional TRL MSRs\n tools/power/x86/intel-speed-select: Fix memory leak\n tools/power/x86/intel-speed-select: Output success/failed for command output\n tools/power/x86/intel-speed-select: Output human readable CPU list\n tools/power/x86/intel-speed-select: Change turbo ratio output to maximum turbo frequency\n tools/power/x86/intel-speed-select: Switch output to MHz\n tools/power/x86/intel-speed-select: Simplify output for turbo-freq and base-freq\n tools/power/x86/intel-speed-select: Fix cpu-count output\n tools/power/x86/intel-speed-select: Fix help option typo\n tools/power/x86/intel-speed-select: Fix package typo\n tools/power/x86/intel-speed-select: Fix a read overflow in isst_set_tdp_level_msr()\n platform/x86: intel_int0002_vgpio: Use device_init_wakeup\n platform/x86: intel_int0002_vgpio: Fix wakeups not working on Cherry Trail\n platform/x86: compal-laptop: Initialize \"value\" in ec_read_u8()\n platform/x86: touchscreen_dmi: Add info for the Trekstor Primebook C11B 2-in-1\n ...\n" }, { "commit": "7ac63f6ba5db5e2e81e4674551d6f9ec58e70618", "tree": "fae597509e6781401b6abb517dac17f663737a34", "parents": [ "e2bddc20b562ee23046ad541cf29314e4aebd934", "f7b15c74cffd760ec9959078982d8268a38456c4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:40:24 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:40:24 2019 -0700" }, "message": "Merge branch \u0027x86-vmware-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 vmware updates from Ingo Molnar:\n \"This updates the VMWARE guest driver with support for VMCALL/VMMCALL\n based hypercalls\"\n\n* \u0027x86-vmware-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n input/vmmouse: Update the backdoor call with support for new instructions\n drm/vmwgfx: Update the backdoor call with support for new instructions\n x86/vmware: Add a header file for hypercall definitions\n x86/vmware: Update platform detection code for VMCALL/VMMCALL hypercalls\n" }, { "commit": "e2bddc20b562ee23046ad541cf29314e4aebd934", "tree": "608da9a4721ec3b9b19f342f580793de6babef4e", "parents": [ "6f24671485d0d0eaeaccd910fa8148db72aac089", "83527ef7abf7c02c33a90b00f0954db35415adbd" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:39:00 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:39:00 2019 -0700" }, "message": "Merge branch \u0027x86-hyperv-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 hyperv updates from Ingo Molnar:\n \"Misc updates related to page size abstractions within the HyperV code,\n in preparation for future features\"\n\n* \u0027x86-hyperv-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n drivers: hv: vmbus: Replace page definition with Hyper-V specific one\n x86/hyperv: Add functions to allocate/deallocate page for Hyper-V\n x86/hyperv: Create and use Hyper-V page definitions\n" }, { "commit": "6f24671485d0d0eaeaccd910fa8148db72aac089", "tree": "3133b15b3e0adc92918e6b8c19c944e663993f4c", "parents": [ "ac51667b5b95f1209aa97af780cecf0cf6f4003f", "864b23f0169d5bff677e8443a7a90dfd6b090afc" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:37:44 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:37:44 2019 -0700" }, "message": "Merge branch \u0027x86-platform-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 platform update from Ingo Molnar:\n \"The biggest change is the rework of the intel/iosf_mbi locking code\n which used a few non-standard locking patterns, to make it work under\n lockdep\"\n\n* \u0027x86-platform-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/platform/uv: Fix kmalloc() NULL check routine\n x86/platform/intel/iosf_mbi Rewrite locking\n" }, { "commit": "ac51667b5b95f1209aa97af780cecf0cf6f4003f", "tree": "f670eadaf8920c82b15e48002f298ac245432289", "parents": [ "e0d60a1e68a3fbf42cdf3546004e00230d9048ba", "bc04a049f058a472695aa22905d57e2b1f4c77d9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:21:34 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:21:34 2019 -0700" }, "message": "Merge branch \u0027x86-mm-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 mm updates from Ingo Molnar:\n\n - Make cpumask_of_node() more robust against invalid node IDs\n\n - Simplify and speed up load_mm_cr4()\n\n - Unexport and remove various unused set_memory_*() APIs\n\n - Misc cleanups\n\n* \u0027x86-mm-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/mm: Fix cpumask_of_node() error condition\n x86/mm: Remove the unused set_memory_wt() function\n x86/mm: Remove set_pages_x() and set_pages_nx()\n x86/mm: Remove the unused set_memory_array_*() functions\n x86/mm: Unexport set_memory_x() and set_memory_nx()\n x86/fixmap: Cleanup outdated comments\n x86/kconfig: Remove X86_DIRECT_GBPAGES dependency on !DEBUG_PAGEALLOC\n x86/mm: Avoid redundant interrupt disable in load_mm_cr4()\n" }, { "commit": "e0d60a1e68a3fbf42cdf3546004e00230d9048ba", "tree": "e283f864ee0d53fa8df68701b463365e6b73683b", "parents": [ "22331f895298bd23ca9f99f6a237aae883c9e1c7", "6365b842aae4490ebfafadfc6bb27a6d3cc54757" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:06:29 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 19:06:29 2019 -0700" }, "message": "Merge branch \u0027x86-entry-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 entry updates from Ingo Molnar:\n \"This contains x32 and compat syscall improvements, the biggest one of\n which splits x32 syscalls into their own table, which allows new\n syscalls to share the x32 and x86-64 number - which turns the\n 512-547 special syscall numbers range into a legacy wart that won\u0027t be\n extended going forward\"\n\n* \u0027x86-entry-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/syscalls: Split the x32 syscalls into their own table\n x86/syscalls: Disallow compat entries for all types of 64-bit syscalls\n x86/syscalls: Use the compat versions of rt_sigsuspend() and rt_sigprocmask()\n x86/syscalls: Make __X32_SYSCALL_BIT be unsigned long\n" }, { "commit": "22331f895298bd23ca9f99f6a237aae883c9e1c7", "tree": "13c4dcc25d8ff39444c4158846f4c44b0d39fa04", "parents": [ "fc6fd1392a8f3d5f3d722ad9c92314477c1a2a35", "0cc5359d8fd45bc410906e009117e78e2b5b2322" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 18:47:53 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 18:47:53 2019 -0700" }, "message": "Merge branch \u0027x86-cpu-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 cpu-feature updates from Ingo Molnar:\n\n - Rework the Intel model names symbols/macros, which were decades of\n ad-hoc extensions and added random noise. It\u0027s now a coherent, easy\n to follow nomenclature.\n\n - Add new Intel CPU model IDs:\n - \"Tiger Lake\" desktop and mobile models\n - \"Elkhart Lake\" model ID\n - and the \"Lightning Mountain\" variant of Airmont, plus support code\n\n - Add the new AVX512_VP2INTERSECT instruction to cpufeatures\n\n - Remove Intel MPX user-visible APIs and the self-tests, because the\n toolchain (gcc) is not supporting it going forward. This is the\n first, lowest-risk phase of MPX removal.\n\n - Remove X86_FEATURE_MFENCE_RDTSC\n\n - Various smaller cleanups and fixes\n\n* \u0027x86-cpu-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (25 commits)\n x86/cpu: Update init data for new Airmont CPU model\n x86/cpu: Add new Airmont variant to Intel family\n x86/cpu: Add Elkhart Lake to Intel family\n x86/cpu: Add Tiger Lake to Intel family\n x86: Correct misc typos\n x86/intel: Add common OPTDIFFs\n x86/intel: Aggregate microserver naming\n x86/intel: Aggregate big core graphics naming\n x86/intel: Aggregate big core mobile naming\n x86/intel: Aggregate big core client naming\n x86/cpufeature: Explain the macro duplication\n x86/ftrace: Remove mcount() declaration\n x86/PCI: Remove superfluous returns from void functions\n x86/msr-index: Move AMD MSRs where they belong\n x86/cpu: Use constant definitions for CPU models\n lib: Remove redundant ftrace flag removal\n x86/crash: Remove unnecessary comparison\n x86/bitops: Use __builtin_constant_p() directly instead of IS_IMMEDIATE()\n x86: Remove X86_FEATURE_MFENCE_RDTSC\n x86/mpx: Remove MPX APIs\n ...\n" }, { "commit": "fc6fd1392a8f3d5f3d722ad9c92314477c1a2a35", "tree": "11497e6cba801765ad494c2ae39b783385740f49", "parents": [ "49a21e52a6baeea076301fd944268fd0d1f75be1", "701010532164eaacd415ec5683717da03f4b822d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 18:29:19 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 18:29:19 2019 -0700" }, "message": "Merge branch \u0027x86-build-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 build cleanup from Ingo Molnar:\n \"A single change that removes unnecessary asm-generic wrappers\"\n\n* \u0027x86-build-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/build: Remove unneeded uapi asm-generic wrappers\n" }, { "commit": "49a21e52a6baeea076301fd944268fd0d1f75be1", "tree": "bd929f64b63373b3d02c7aaa8f2ec46f24c0dcf9", "parents": [ "df4c0b18f2a2798f1e3ae9dcf58c024bb33e4202", "d5a1baddf1585885868cbab55989401fb97118c6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 18:27:37 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 18:27:37 2019 -0700" }, "message": "Merge branch \u0027x86-boot-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 boot code cleanup from Ingo Molnar:\n \"Clean up the BUILD_BUG_ON() definition which can cause build warnings\"\n\n* \u0027x86-boot-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/boot: Use common BUILD_BUG_ON\n" }, { "commit": "df4c0b18f2a2798f1e3ae9dcf58c024bb33e4202", "tree": "08f57b6d3966dcffba3ff056124fd80333bffc44", "parents": [ "7e67a859997aad47727aff9c5a32e160da079ce3", "e86c2c8b9380440bbe761b8e2f63ab6b04a45ac2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 18:07:08 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 18:07:08 2019 -0700" }, "message": "Merge branch \u0027x86-asm-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 asm updates from Ingo Molnar:\n\n - Add UMIP emulation/spoofing for 64-bit processes as well, because of\n Wine based gaming.\n\n - Clean up symbols/labels in low level asm code\n\n - Add an assembly optimized mul_u64_u32_div() implementation on x86-64.\n\n* \u0027x86-asm-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/umip: Add emulation (spoofing) for UMIP covered instructions in 64-bit processes as well\n x86/asm: Make some functions local labels\n x86/asm/suspend: Get rid of bogus_64_magic\n x86/math64: Provide a sane mul_u64_u32_div() implementation for x86_64\n" }, { "commit": "7e67a859997aad47727aff9c5a32e160da079ce3", "tree": "96f53425c2834de5b3276d7598782ab6412e4d5e", "parents": [ "772c1d06bd402f7ee72c61a18c2db74cd74b6758", "563c4f85f9f0d63b712081d5b4522152cdcb8b6b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 17:25:49 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 17:25:49 2019 -0700" }, "message": "Merge branch \u0027sched-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull scheduler updates from Ingo Molnar:\n\n - MAINTAINERS: Add Mark Rutland as perf submaintainer, Juri Lelli and\n Vincent Guittot as scheduler submaintainers. Add Dietmar Eggemann,\n Steven Rostedt, Ben Segall and Mel Gorman as scheduler reviewers.\n\n As perf and the scheduler is getting bigger and more complex,\n document the status quo of current responsibilities and interests,\n and spread the review pain^H^H^H^H fun via an increase in the Cc:\n linecount generated by scripts/get_maintainer.pl. :-)\n\n - Add another series of patches that brings the -rt (PREEMPT_RT) tree\n closer to mainline: split the monolithic CONFIG_PREEMPT dependencies\n into a new CONFIG_PREEMPTION category that will allow the eventual\n introduction of CONFIG_PREEMPT_RT. Still a few more hundred patches\n to go though.\n\n - Extend the CPU cgroup controller with uclamp.min and uclamp.max to\n allow the finer shaping of CPU bandwidth usage.\n\n - Micro-optimize energy-aware wake-ups from O(CPUS^2) to O(CPUS).\n\n - Improve the behavior of high CPU count, high thread count\n applications running under cpu.cfs_quota_us constraints.\n\n - Improve balancing with SCHED_IDLE (SCHED_BATCH) tasks present.\n\n - Improve CPU isolation housekeeping CPU allocation NUMA locality.\n\n - Fix deadline scheduler bandwidth calculations and logic when cpusets\n rebuilds the topology, or when it gets deadline-throttled while it\u0027s\n being offlined.\n\n - Convert the cpuset_mutex to percpu_rwsem, to allow it to be used from\n setscheduler() system calls without creating global serialization.\n Add new synchronization between cpuset topology-changing events and\n the deadline acceptance tests in setscheduler(), which were broken\n before.\n\n - Rework the active_mm state machine to be less confusing and more\n optimal.\n\n - Rework (simplify) the pick_next_task() slowpath.\n\n - Improve load-balancing on AMD EPYC systems.\n\n - ... and misc cleanups, smaller fixes and improvements - please see\n the Git log for more details.\n\n* \u0027sched-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (53 commits)\n sched/psi: Correct overly pessimistic size calculation\n sched/fair: Speed-up energy-aware wake-ups\n sched/uclamp: Always use \u0027enum uclamp_id\u0027 for clamp_id values\n sched/uclamp: Update CPU\u0027s refcount on TG\u0027s clamp changes\n sched/uclamp: Use TG\u0027s clamps to restrict TASK\u0027s clamps\n sched/uclamp: Propagate system defaults to the root group\n sched/uclamp: Propagate parent clamps\n sched/uclamp: Extend CPU\u0027s cgroup controller\n sched/topology: Improve load balancing on AMD EPYC systems\n arch, ia64: Make NUMA select SMP\n sched, perf: MAINTAINERS update, add submaintainers and reviewers\n sched/fair: Use rq_lock/unlock in online_fair_sched_group\n cpufreq: schedutil: fix equation in comment\n sched: Rework pick_next_task() slow-path\n sched: Allow put_prev_task() to drop rq-\u003elock\n sched/fair: Expose newidle_balance()\n sched: Add task_struct pointer to sched_class::set_curr_task\n sched: Rework CPU hotplug task selection\n sched/{rt,deadline}: Fix set_next_task vs pick_next_task\n sched: Fix kerneldoc comment for ia64_set_curr_task\n ...\n" }, { "commit": "772c1d06bd402f7ee72c61a18c2db74cd74b6758", "tree": "e362fc7e158b3580d810a26189ecf91ec8a4f141", "parents": [ "c7eba51cfdf9cd1ca7ed4201b30be8b2bef15ff5", "e336b4027775cb458dc713745e526fa1a1996b2a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 17:06:21 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 17:06:21 2019 -0700" }, "message": "Merge branch \u0027perf-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf updates from Ingo Molnar:\n \"Kernel side changes:\n\n - Improved kbprobes robustness\n\n - Intel PEBS support for PT hardware tracing\n\n - Other Intel PT improvements: high order pages memory footprint\n reduction and various related cleanups\n\n - Misc cleanups\n\n The perf tooling side has been very busy in this cycle, with over 300\n commits. This is an incomplete high-level summary of the many\n improvements done by over 30 developers:\n\n - Lots of updates to the following tools:\n\n \u0027perf c2c\u0027\n \u0027perf config\u0027\n \u0027perf record\u0027\n \u0027perf report\u0027\n \u0027perf script\u0027\n \u0027perf test\u0027\n \u0027perf top\u0027\n \u0027perf trace\u0027\n\n - Updates to libperf and libtraceevent, and a consolidation of the\n proliferation of x86 instruction decoder libraries.\n\n - Vendor event updates for Intel and PowerPC CPUs,\n\n - Updates to hardware tracing tooling for ARM and Intel CPUs,\n\n - ... and lots of other changes and cleanups - see the shortlog and\n Git log for details\"\n\n* \u0027perf-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (322 commits)\n kprobes: Prohibit probing on BUG() and WARN() address\n perf/x86: Make more stuff static\n x86, perf: Fix the dependency of the x86 insn decoder selftest\n objtool: Ignore intentional differences for the x86 insn decoder\n objtool: Update sync-check.sh from perf\u0027s check-headers.sh\n perf build: Ignore intentional differences for the x86 insn decoder\n perf intel-pt: Use shared x86 insn decoder\n perf intel-pt: Remove inat.c from build dependency list\n perf: Update .gitignore file\n objtool: Move x86 insn decoder to a common location\n perf metricgroup: Support multiple events for metricgroup\n perf metricgroup: Scale the metric result\n perf pmu: Change convert_scale from static to global\n perf symbols: Move mem_info and branch_info out of symbol.h\n perf auxtrace: Uninline functions that touch perf_session\n perf tools: Remove needless evlist.h include directives\n perf tools: Remove needless evlist.h include directives\n perf tools: Remove needless thread_map.h include directives\n perf tools: Remove needless thread.h include directives\n perf tools: Remove needless map.h include directives\n ...\n" }, { "commit": "c7eba51cfdf9cd1ca7ed4201b30be8b2bef15ff5", "tree": "048bff0b92dc230f01b5af23cd1b5960615bd68a", "parents": [ "cc9b499a1f71696054a2771aae504c53eecff31d", "e57d143091f1c0b1a98140a4d2e63e113afb62c0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:49:55 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:49:55 2019 -0700" }, "message": "Merge branch \u0027locking-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull locking updates from Ingo Molnar:\n\n - improve rwsem scalability\n\n - add uninitialized rwsem debugging check\n\n - reduce lockdep\u0027s stacktrace memory usage and add diagnostics\n\n - misc cleanups, code consolidation and constification\n\n* \u0027locking-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n mutex: Fix up mutex_waiter usage\n locking/mutex: Use mutex flags macro instead of hard code\n locking/mutex: Make __mutex_owner static to mutex.c\n locking/qspinlock,x86: Clarify virt_spin_lock_key\n locking/rwsem: Check for operations on an uninitialized rwsem\n locking/rwsem: Make handoff writer optimistically spin on owner\n locking/lockdep: Report more stack trace statistics\n locking/lockdep: Reduce space occupied by stack traces\n stacktrace: Constify \u0027entries\u0027 arguments\n locking/lockdep: Make it clear that what lock_class::key points at is not modified\n" }, { "commit": "cc9b499a1f71696054a2771aae504c53eecff31d", "tree": "96b18e65c38bfcacd2f03a9012bd5384e1c54bf2", "parents": [ "98c82b4b8be60b05bc96aa4ab664ca0b0e39001f", "d3dc0168e93233ba4d4ed9a2c506c9d2b8d8cd33" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:47:38 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:47:38 2019 -0700" }, "message": "Merge branch \u0027efi-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull EFI updates from Ingo Molnar:\n\n - refactor the EFI config table handling across architectures\n\n - add support for the Dell EMC OEM config table\n\n - include AER diagnostic output to CPER handling of fatal PCIe errors\n\n* \u0027efi-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n efi: cper: print AER info of PCIe fatal error\n efi: Export Runtime Configuration Interface table to sysfs\n efi: ia64: move SAL systab handling out of generic EFI code\n efi/x86: move UV_SYSTAB handling into arch/x86\n efi: x86: move efi_is_table_address() into arch/x86\n" }, { "commit": "98c82b4b8be60b05bc96aa4ab664ca0b0e39001f", "tree": "ea7a308f3ff882f4d92c614ad11d9ac03c240beb", "parents": [ "94d18ee9340e00ee3455bb45661484093e3b2674", "ee050dc83bc326ad5ef8ee93bca344819371e7a5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:44:55 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:44:55 2019 -0700" }, "message": "Merge branch \u0027core-stacktrace-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull stacktrace fixlet from Ingo Molnar:\n \"Two comment fixes\"\n\n* \u0027core-stacktrace-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n lib/stackdepot: Fix outdated comments\n" }, { "commit": "94d18ee9340e00ee3455bb45661484093e3b2674", "tree": "6274a11e0b1201e0f1b7181649af922972d46f94", "parents": [ "d75a43c645c26ab58118bd35405666a12971350d", "4a0fa886ab79ea85e8d1be2b0df143d8249779be" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:28:19 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:28:19 2019 -0700" }, "message": "Merge branch \u0027core-rcu-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull RCU updates from Ingo Molnar:\n \"This cycle\u0027s RCU changes were:\n\n - A few more RCU flavor consolidation cleanups.\n\n - Updates to RCU\u0027s list-traversal macros improving lockdep usability.\n\n - Forward-progress improvements for no-CBs CPUs: Avoid ignoring\n incoming callbacks during grace-period waits.\n\n - Forward-progress improvements for no-CBs CPUs: Use -\u003ecblist\n structure to take advantage of others\u0027 grace periods.\n\n - Also added a small commit that avoids needlessly inflicting\n scheduler-clock ticks on callback-offloaded CPUs.\n\n - Forward-progress improvements for no-CBs CPUs: Reduce contention on\n -\u003enocb_lock guarding -\u003ecblist.\n\n - Forward-progress improvements for no-CBs CPUs: Add -\u003enocb_bypass\n list to further reduce contention on -\u003enocb_lock guarding -\u003ecblist.\n\n - Miscellaneous fixes.\n\n - Torture-test updates.\n\n - minor LKMM updates\"\n\n* \u0027core-rcu-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (86 commits)\n MAINTAINERS: Update from paulmck@linux.ibm.com to paulmck@kernel.org\n rcu: Don\u0027t include \u003clinux/ktime.h\u003e in rcutiny.h\n rcu: Allow rcu_do_batch() to dynamically adjust batch sizes\n rcu/nocb: Don\u0027t wake no-CBs GP kthread if timer posted under overload\n rcu/nocb: Reduce __call_rcu_nocb_wake() leaf rcu_node -\u003elock contention\n rcu/nocb: Reduce nocb_cb_wait() leaf rcu_node -\u003elock contention\n rcu/nocb: Advance CBs after merge in rcutree_migrate_callbacks()\n rcu/nocb: Avoid synchronous wakeup in __call_rcu_nocb_wake()\n rcu/nocb: Print no-CBs diagnostics when rcutorture writer unduly delayed\n rcu/nocb: EXP Check use and usefulness of -\u003enocb_lock_contended\n rcu/nocb: Add bypass callback queueing\n rcu/nocb: Atomic -\u003elen field in rcu_segcblist structure\n rcu/nocb: Unconditionally advance and wake for excessive CBs\n rcu/nocb: Reduce -\u003enocb_lock contention with separate -\u003enocb_gp_lock\n rcu/nocb: Reduce contention at no-CBs invocation-done time\n rcu/nocb: Reduce contention at no-CBs registry-time CB advancement\n rcu/nocb: Round down for number of no-CBs grace-period kthreads\n rcu/nocb: Avoid -\u003enocb_lock capture by corresponding CPU\n rcu/nocb: Avoid needless wakeups of no-CBs grace-period kthread\n rcu/nocb: Make __call_rcu_nocb_wake() safe for many callbacks\n ...\n" }, { "commit": "d75a43c645c26ab58118bd35405666a12971350d", "tree": "1416c3f00729d1c2088edbe16eb37652d5da27cf", "parents": [ "a480222f4c7cdafad22540f44487f009e359dfb8", "f73b3cc39c84220e6dccd463b5c8279b03514646" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:15:34 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:15:34 2019 -0700" }, "message": "Merge branch \u0027core-objtool-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull objtool build fix from Ingo Molnar:\n \"Fix objtool builds with more exotic, user-defined CFLAGS\"\n\n* \u0027core-objtool-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n objtool: Clobber user CFLAGS variable\n" }, { "commit": "a480222f4c7cdafad22540f44487f009e359dfb8", "tree": "d14c9e5852653846e32151a8f4c7299c677ea3b5", "parents": [ "cef7298262e9af841fb70d8673af45caf55300a1", "e8e4eb0fbeda570b16464208aebf5caccfb6eb95" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:11:41 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 16:11:41 2019 -0700" }, "message": "Merge branch \u0027core-headers-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull header documentation fix from Ingo Molnar:\n \"Fix the parameter description \u003casm-generic/div64.h\u003e\"\n\n* \u0027core-headers-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n asm-generic/div64: Fix documentation of do_div() parameter\n" }, { "commit": "cef7298262e9af841fb70d8673af45caf55300a1", "tree": "6953da41a16417db538a42d367c08483757a8111", "parents": [ "d47ebd684229f0048be5def6027bfcfbfe2db0d6", "44acee207844789514233341ccbb3ec5086a6f7b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:56:22 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:56:22 2019 -0700" }, "message": "Merge tag \u0027armsoc-dt\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc\n\nPull ARM DT updates from Arnd Bergmann:\n \"This is another huge branch with close to 450 changessets related to\n devicetree files, roughly half of this for 32-bit and 64-bit\n respectively. There are lots of cleanups and additional hardware\n support for platforms we already support based on SoCs from Renesas,\n ST-Microelectronics, Intel/Altera, Rockchips, Allwinner, Broadcom and\n other manufacturers.\n\n A total of 6 new SoCs and 37 new boards gets added this time, one more\n SoC will come in a follow-up branch. Most of the new boards are for\n 64-bit ARM SoCs, the others are typically for the 32-bit Cortex-A7.\n\n Going more into details for SoC platforms with new hardware support:\n\n - The Snapdragon 855 (SM8150) is Qualcomm\u0027s current high-end phone\n platform, usually paired with an external 5G modem. So far we only\n support the Qualcomm SM8150 MTP reference platform, but no actual\n products.\n\n - For the slightly older Qualcomm platforms, support for several\n interesting products is getting added: Three laptops based on\n Snapdragon 835/MSM8998 (Asus NovaGo, HP Envy X2 and Lenovo Miix\n 630), one laptop based on Snapdragon 850/sdm850 (Lenovo Yoga C630)\n and several phones based on the older Snapdragon 410/MSM8916\n (Samsung A3 and A5, Longcheer L8150 aka Android One 2nd gen \"seed\"\n aka Wileyfox Swift).\n\n - Mediatek MT7629 is a new wireless network router chip, similar to\n the older MT7623. It gets added together with the reference board\n implementation.\n\n - Allwinner V3 is a repackaged version of the existing low-end V3s\n chip, and is used in the tiny Lichee Pi Zero plus, also added here.\n There is also a new TV set-top box based on Allwinner H6, the Tanix\n TX6, and the eMMC variant of the Olimex A64-Olinuxino development\n board.\n\n - NXP i.MX8M Nano is a new member of the ever-expanding i.MX SoC\n family, similar to the i.MX8M Mini. As usual, there is a large\n number of new boards for i.MX SoCs: Einfochips i.MX8QXP AI_ML,\n SolidRun Hummingboard Pulse baseboard and System-on-Module,\n Boundary Devices i.MX8MQ Nitrogen8M, and TechNexion\n PICO-PI-IMX8M-DEV for the 64-bit i.MX8 line. For 32-bit, we get the\n Kontron i.MX6UL N6310 SoM with two baseboards, the PHYTEC\n phyBOARD-Segin SoM with three baseboards, and the Zodiac Inflight\n Innovations i.MX7 RMU2 board.\n\n - In a different NXP product line, the Layerscape LS1046A \"Freeway\"\n reference board gets added.\n\n - Amlogic SM1 (S905X3) and G12B (S922X, A311D) are updated chips from\n their set-top-box line and smart speaker with newer CPU and GPU\n cores compared to their predecessors. Both are now also supported\n by the Khadas VIM3 development board series, and the dts files for\n that get reorganized a bit to better deal with all variants.\n Another board based on SM1 that gets added is the SEI Robotics\n SEI610.\n\n - There are a handful of new x86 and Power9 server boards using\n Aspeed BMC chips that are gaining support for running Linux on the\n BMC through the OpenBMC project: Facebook\n Minipack/Wedge100/Wedge40, Lenovo Hr855xg2, and Mihawk. Notably\n these are still new machines using SoCs based on the ARM9 and ARM11\n CPU cores, as support for the new Cortex-A7 based AST2600 is still\n ramping up.\n\n - There are three new end-user products using 32-bit Rockchips SoCs:\n Mecer Xtreme Mini S6 is an Android \"mini PC\" box based on the\n low-end RK3229 chip, while the two AOpen products Chromebox Mini\n (Fievel) and Chromebase Mini (Tiger) run ChromeOS and are meant for\n commercial settings(digital signage, PoS, ...).\n\n - One more single-board computer based on the popular 64-bit RK3399\n is added: the Leez RK3399 P710\"\n\n* tag \u0027armsoc-dt\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (467 commits)\n arm64: dts: qcom: Add Lenovo Yoga C630\n ARM: dts: aspeed-g5: Fixe gpio-ranges upper limit\n ARM; dts: aspeed: mihawk: File should not be executable\n ARM: dts: aspeed: swift: Change power supplies to version 2\n ARM: dts: aspeed: vesnin: Add secondary SPI flash chip\n ARM: dts: aspeed: vesnin: Add wdt2 with alt-boot option\n ARM: dts: aspeed-g4: Add all flash chips\n ARM: dts: exynos: Enable GPU/Mali T604 on Arndale board\n ARM: dts: exynos: Enable GPU/Mali T604 on Chromebook Snow\n ARM: dts: exynos: Add GPU/Mali T604 node to Exynos5250\n ARM: dts: exynos: Fix min/max buck4 for GPU on Arndale board\n ARM: dts: exynos: Mark LDO10 as always-on on Peach Pit/Pi Chromebooks\n ARM: dts: exynos: Remove not accurate secondary ADC compatible\n arm64: dts: rockchip: limit clock rate of MMC controllers for RK3328\n arm64: dts: meson-sm1-sei610: add stdout-path property back\n arm64: dts: meson-sm1-sei610: enable DVFS\n arm64: dts: khadas-vim3: add support for the SM1 based VIM3L\n dt-bindings: arm: amlogic: add Amlogic SM1 based Khadas VIM3L bindings\n arm64: dts: khadas-vim3: move common nodes into meson-khadas-vim3.dtsi\n arm64: dts: meson: g12a: add reset to tdm formatters\n ...\n" }, { "commit": "d47ebd684229f0048be5def6027bfcfbfe2db0d6", "tree": "9b0c0c7a3084a7359f7e8cfac2e57574429a5f27", "parents": [ "399eb9b6cbf31ff6ef91a6930e2e94c703d74078", "adce1767799b6f6cf18c13e746950fc976f2e15e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:55:06 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:55:06 2019 -0700" }, "message": "Merge tag \u0027armsoc-defconfig\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc\n\nPull ARM SoC defconfig updates from Arnd Bergmann:\n \"As usual, a bunch of commits, mostly adding drivers and other options\n to defconfigs after the code was merged through another tree\"\n\n* tag \u0027armsoc-defconfig\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (32 commits)\n arm64: defconfig: Enable Qualcomm QUSB2 PHY\n arm64: defconfig: Enable the EFI Framebuffer\n arm64: defconfig: Enable Qualcomm GENI based I2C controller\n ARM: multi_v7_defconfig: Make MAX77802 regulator driver built-in\n arm64: defconfig: Enable CPU clock drivers for Qualcomm msm8916\n arm64: defconfig: Add DRM_MSM to defconfigs with ARCH_QCOM\n ARM: multi_v7_defconfig: Add DRM_MSM to defconfigs with ARCH_QCOM\n ARM: qcom_defconfig: Add DRM_MSM to defconfigs with ARCH_QCOM\n ARM: configs: aspeed_g5: Enable AST2600\n ARM: configs: multi_v7: Add ASPEED G6\n arm64: defconfig: Enable SM8150 GCC and pinctrl driver\n arm64: defconfig: Enable CONFIG_ACPI_APEI_PCIEAER\n arm64: defconfig: Enable the DesignWare watchdog\n ARM: multi_v7_defconfig: Enable SPI_STM32_QSPI support\n ARM: imx_v6_v7_defconfig: Enable the PSCI CPUidle driver\n arm64: defconfig: Enable the PSCI CPUidle driver\n arm64: defconfig: Enable Sun4i SPDIF module\n ARM: exynos_defconfig: Enable AHCI-platform SATA driver\n arm64: defconfig: CONFIG_DRM_ETNAVIV\u003dm\n ARM: imx_v6_v7_defconfig: Select the OV5645 camera driver\n ...\n" }, { "commit": "399eb9b6cbf31ff6ef91a6930e2e94c703d74078", "tree": "60c26e51c167efdfec5ab5821111df9ea90cbf7f", "parents": [ "2b97c39514a6130f38b14227a36d9cd37e650a9d", "3dc8dcb02fdba3370aec0696727e6adfe8033aa4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:52:38 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:52:38 2019 -0700" }, "message": "Merge tag \u0027armsoc-drivers\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc\n\nPull ARM SoC driver updates from Arnd Bergmann:\n \"This contains driver changes that are tightly connected to SoC\n specific code. Aside from smaller cleanups and bug fixes, here is a\n list of the notable changes.\n\n New device drivers:\n\n - The Turris Mox router has a new \"moxtet\" bus driver for its\n on-board pluggable extension bus. The same platform also gains a\n firmware driver.\n\n - The Samsung Exynos family gains a new Chipid driver exporting using\n the soc device sysfs interface\n\n - A similar socinfo driver for Qualcomm Snapdragon chips.\n\n - A firmware driver for the NXP i.MX DSP IPC protocol using shared\n memory and a mailbox\n\n Other changes:\n\n - The i.MX reset controller driver now supports the NXP i.MX8MM chip\n\n - Amlogic SoC specific drivers gain support for the S905X3 and A311D\n chips\n\n - A rework of the TI Davinci framebuffer driver to allow important\n cleanups in the platform code\n\n - A couple of device drivers for removed ARM SoC platforms are\n removed. Most of the removals were picked up by other maintainers,\n this contains whatever was left\"\n\n* tag \u0027armsoc-drivers\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (123 commits)\n bus: uniphier-system-bus: use devm_platform_ioremap_resource()\n soc: ti: ti_sci_pm_domains: Add support for exclusive and shared access\n dt-bindings: ti_sci_pm_domains: Add support for exclusive and shared access\n firmware: ti_sci: Allow for device shared and exclusive requests\n bus: imx-weim: remove incorrect __init annotations\n fbdev: remove w90x900/nuc900 platform drivers\n spi: remove w90x900 driver\n net: remove w90p910-ether driver\n net: remove ks8695 driver\n firmware: turris-mox-rwtm: Add sysfs documentation\n firmware: Add Turris Mox rWTM firmware driver\n dt-bindings: firmware: Document cznic,turris-mox-rwtm binding\n bus: moxtet: fix unsigned comparison to less than zero\n bus: moxtet: remove set but not used variable \u0027dummy\u0027\n ARM: scoop: Use the right include\n dt-bindings: power: add Amlogic Everything-Else power domains bindings\n soc: amlogic: Add support for Everything-Else power domains controller\n fbdev: da8xx: use resource management for dma\n fbdev: da8xx-fb: drop a redundant if\n fbdev: da8xx-fb: use devm_platform_ioremap_resource()\n ...\n" }, { "commit": "2b97c39514a6130f38b14227a36d9cd37e650a9d", "tree": "dfc6ae0eb6c7acd46d170bd4d2c34e2d90bcf264", "parents": [ "d0a16fe934383ecdb605ab9312d700fb9099f75e", "0366977480c43a221e4309f242d1144e85a368c3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:48:14 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:48:14 2019 -0700" }, "message": "Merge tag \u0027armsoc-soc\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc\n\nPull ARM SoC platform updates from Arnd Bergmann:\n \"The main change this time around is a cleanup of some of the oldest\n platforms based on the XScale and ARM9 CPU cores, which are between 10\n and 20 years old.\n\n The Kendin/Micrel/Microchip KS8695, Winbond/Nuvoton W90x900 and Intel\n IOP33x/IOP13xx platforms are removed after we determined that nobody\n is using them any more.\n\n The TI Davinci and NXP LPC32xx platforms on the other hand are still\n in active use and are converted to the ARCH_MULTIPLATFORM build,\n meaning that we can compile a kernel that works on these along with\n most other ARMv5 platforms. Changes toward that goal are also merged\n for IOP32x, but additional work is needed to complete this. Patches\n for the remaining ARMv5 platforms have started but need more work and\n some testing.\n\n Support for the new ASpeed AST2600 gets added, this is based on the\n Cortex-A7 ARMv7 core, and is a newer version of the existing ARMv5 and\n ARMv6 chips in the same family.\n\n Other changes include a cleanup of the ST-Ericsson ux500 platform and\n the move of the TI Davinci platform to a new clocksource driver\"\n\n[ The changes had marked INTEL_IOP_ADMA and USB_LPC32XX as being\n buildable on other platforms through COMPILE_TEST, but that causes new\n warnings that I most definitely do not want to see during the merge\n window as that could hide other issues.\n\n So the COMPILE_TEST option got disabled for them again - Linus ]\n\n* tag \u0027armsoc-soc\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (61 commits)\n ARM: multi_v5_defconfig: make DaVinci part of the ARM v5 multiplatform build\n ARM: davinci: support multiplatform build for ARM v5\n arm64: exynos: Enable exynos-chipid driver\n ARM: OMAP2+: Delete an unnecessary kfree() call in omap_hsmmc_pdata_init()\n ARM: OMAP2+: move platform-specific asm-offset.h to arch/arm/mach-omap2\n ARM: davinci: dm646x: Fix a typo in the comment\n ARM: davinci: dm646x: switch to using the clocksource driver\n ARM: davinci: dm644x: switch to using the clocksource driver\n ARM: aspeed: Enable SMP boot\n ARM: aspeed: Add ASPEED AST2600 architecture\n ARM: aspeed: Select timer in each SoC\n dt-bindings: arm: cpus: Add ASPEED SMP\n ARM: imx: stop adjusting ar8031 phy tx delay\n mailmap: map old company name to new one @microchip.com\n MAINTAINERS: at91: remove the TC entry\n MAINTAINERS: at91: Collect all pinctrl/gpio drivers in same entry\n ARM: at91: move platform-specific asm-offset.h to arch/arm/mach-at91\n MAINTAINERS: Extend patterns for Samsung SoC, Security Subsystem and clock drivers\n ARM: s3c64xx: squash samsung_usb_phy.h into setup-usb-phy.c\n ARM: debug-ll: Add support for r7s9210\n ...\n" }, { "commit": "d0a16fe934383ecdb605ab9312d700fb9099f75e", "tree": "b9763fcb1b2c7426adbffc6f0f921c79f3230609", "parents": [ "76f0f227cffb570bc5ce343b1750f14907371d80", "fcc16a9e24ba6a2bb9f3af43d892eeec2a435d18" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:38:31 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:38:31 2019 -0700" }, "message": "Merge branch \u0027parisc-5.4-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux\n\nPull parisc updates from Helge Deller:\n\n - Make the powerpc implementation to read elf files available as a\n public kexec interface so it can be re-used on other architectures\n (Sven)\n\n - Implement kexec on parisc (Sven)\n\n - Add kprobes on ftrace on parisc (Sven)\n\n - Fix kernel crash with HSC-PCI cards based on card-mode Dino\n\n - Add assembly implementations for memset, strlen, strcpy, strncpy and\n strcat\n\n - Some cleanups, documentation updates, warning fixes, ...\n\n* \u0027parisc-5.4-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux: (25 commits)\n parisc: Have git ignore generated real2.S and firmware.c\n parisc: Disable HP HSC-PCI Cards to prevent kernel crash\n parisc: add support for kexec_file_load() syscall\n parisc: wire up kexec_file_load syscall\n parisc: add kexec syscall support\n parisc: add __pdc_cpu_rendezvous()\n kprobes/parisc: remove arch_kprobe_on_func_entry()\n kexec_elf: support 32 bit ELF files\n kexec_elf: remove unused variable in kexec_elf_load()\n kexec_elf: remove Elf_Rel macro\n kexec_elf: remove PURGATORY_STACK_SIZE\n kexec_elf: remove parsing of section headers\n kexec_elf: change order of elf_*_to_cpu() functions\n kexec: add KEXEC_ELF\n parisc: Save some bytes in dino driver\n parisc: Drop comments which are already in pci.h\n parisc: Convert eisa_enumerator to use pr_cont()\n parisc: Avoid warning when loading hppb driver\n parisc: speed up flush_tlb_all_local with qemu\n parisc: Add ALTERNATIVE_CODE() and ALT_COND_RUN_ON_QEMU\n ...\n" }, { "commit": "76f0f227cffb570bc5ce343b1750f14907371d80", "tree": "7bbbf7174a7c7cc834936a18a73fc95161b4228e", "parents": [ "58d4fafd0b4c36838077a5d7b17df537b7226f1c", "0d3d343560bad8e1b7879fe94251cfe731a2dd13" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:32:01 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:32:01 2019 -0700" }, "message": "Merge tag \u0027please-pull-ia64_for_5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux\n\nPull ia64 updates from Tony Luck:\n \"The big change here is removal of support for SGI Altix\"\n\n* tag \u0027please-pull-ia64_for_5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux: (33 commits)\n genirq: remove the is_affinity_mask_valid hook\n ia64: remove CONFIG_SWIOTLB ifdefs\n ia64: remove support for machvecs\n ia64: move the screen_info setup to common code\n ia64: move the ROOT_DEV setup to common code\n ia64: rework iommu probing\n ia64: remove the unused sn_coherency_id symbol\n ia64: remove the SGI UV simulator support\n ia64: remove the zx1 swiotlb machvec\n ia64: remove CONFIG_ACPI ifdefs\n ia64: remove CONFIG_PCI ifdefs\n ia64: remove the hpsim platform\n ia64: remove now unused machvec indirections\n ia64: remove support for the SGI SN2 platform\n drivers: remove the SGI SN2 IOC4 base support\n drivers: remove the SGI SN2 IOC3 base support\n qla2xxx: remove SGI SN2 support\n qla1280: remove SGI SN2 support\n misc/sgi-xp: remove SGI SN2 support\n char/mspec: remove SGI SN2 support\n ...\n" }, { "commit": "58d4fafd0b4c36838077a5d7b17df537b7226f1c", "tree": "1b6b824c5ca4d1a5ff72219c18ee16dd23f90c4d", "parents": [ "dbcda58ad98936079c48728c12c27a2f333fb484", "9ce06497c2722a0f9109e4cc3ce35b7a69617886" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:29:34 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:29:34 2019 -0700" }, "message": "Merge tag \u0027riscv/for-v5.4-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux\n\nPull RISC-V updates from Paul Walmsley:\n \"Add the following new features:\n\n - Generic CPU topology description support for DT-based platforms,\n including ARM64, ARM and RISC-V.\n\n - Sparsemem support\n\n - Perf callchain support\n\n - SiFive PLIC irqchip modifications, in preparation for M-mode Linux\n\n and clean up the code base:\n\n - Clean up chip-specific register (CSR) manipulation code, IPIs, TLB\n flushing, and the RISC-V CPU-local timer code\n\n - Kbuild cleanup from one of the Kbuild maintainers\"\n\n[ The CPU topology parts came in through the arm64 tree with a shared\n branch - Linus ]\n\n* tag \u0027riscv/for-v5.4-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:\n irqchip/sifive-plic: set max threshold for ignored handlers\n riscv: move the TLB flush logic out of line\n riscv: don\u0027t use the rdtime(h) pseudo-instructions\n riscv: cleanup riscv_cpuid_to_hartid_mask\n riscv: optimize send_ipi_single\n riscv: cleanup send_ipi_mask\n riscv: refactor the IPI code\n riscv: Add support for libdw\n riscv: Add support for perf registers sampling\n riscv: Add perf callchain support\n riscv: add arch/riscv/Kbuild\n RISC-V: Implement sparsemem\n riscv: Using CSR numbers to access CSRs\n" }, { "commit": "dbcda58ad98936079c48728c12c27a2f333fb484", "tree": "6da524b6fee55b28ab47cb8753aeca2ecbd35f1e", "parents": [ "e77fafe9afb53b7f4d8176c5cd5c10c43a905bc8", "0f1979b402df5f0dd86425830ddaa191d70f3655" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:28:12 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 15:28:12 2019 -0700" }, "message": "Merge tag \u0027m68k-for-v5.4-tag1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k\n\nPull m68k updates from Geert Uytterhoeven:\n\n - ioremap() cleanups\n\n - defconfig updates\n\n - small fixes and cleanups\n\n* tag \u0027m68k-for-v5.4-tag1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k:\n m68k: Remove ioremap_fullcache()\n m68k: Simplify ioremap_nocache()\n m68k: defconfig: Update defconfigs for v5.3-rc2\n m68k: atari: Rename shifter to shifter_st to avoid conflict\n m68k: Prevent some compiler warnings in Coldfire builds\n m68k: mac: Revisit floppy disc controller base addresses\n" }, { "commit": "e77fafe9afb53b7f4d8176c5cd5c10c43a905bc8", "tree": "828ad771a2951f7ac06111c3a9a30e0f368a9b5e", "parents": [ "52a5525214d0d612160154d902956eca0558b7c0", "e376897f424a1c807779a2635f62eb02d7e382f9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:31:40 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:31:40 2019 -0700" }, "message": "Merge tag \u0027arm64-upstream\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux\n\nPull arm64 updates from Will Deacon:\n \"Although there isn\u0027t tonnes of code in terms of line count, there are\n a fair few headline features which I\u0027ve noted both in the tag and also\n in the merge commits when I pulled everything together.\n\n The part I\u0027m most pleased with is that we had 35 contributors this\n time around, which feels like a big jump from the usual small group of\n core arm64 arch developers. Hopefully they all enjoyed it so much that\n they\u0027ll continue to contribute, but we\u0027ll see.\n\n It\u0027s probably worth highlighting that we\u0027ve pulled in a branch from\n the risc-v folks which moves our CPU topology code out to where it can\n be shared with others.\n\n Summary:\n\n - 52-bit virtual addressing in the kernel\n\n - New ABI to allow tagged user pointers to be dereferenced by\n syscalls\n\n - Early RNG seeding by the bootloader\n\n - Improve robustness of SMP boot\n\n - Fix TLB invalidation in light of recent architectural\n clarifications\n\n - Support for i.MX8 DDR PMU\n\n - Remove direct LSE instruction patching in favour of static keys\n\n - Function error injection using kprobes\n\n - Support for the PPTT \"thread\" flag introduced by ACPI 6.3\n\n - Move PSCI idle code into proper cpuidle driver\n\n - Relaxation of implicit I/O memory barriers\n\n - Build with RELR relocations when toolchain supports them\n\n - Numerous cleanups and non-critical fixes\"\n\n* tag \u0027arm64-upstream\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (114 commits)\n arm64: remove __iounmap\n arm64: atomics: Use K constraint when toolchain appears to support it\n arm64: atomics: Undefine internal macros after use\n arm64: lse: Make ARM64_LSE_ATOMICS depend on JUMP_LABEL\n arm64: asm: Kill \u0027asm/atomic_arch.h\u0027\n arm64: lse: Remove unused \u0027alt_lse\u0027 assembly macro\n arm64: atomics: Remove atomic_ll_sc compilation unit\n arm64: avoid using hard-coded registers for LSE atomics\n arm64: atomics: avoid out-of-line ll/sc atomics\n arm64: Use correct ll/sc atomic constraints\n jump_label: Don\u0027t warn on __exit jump entries\n docs/perf: Add documentation for the i.MX8 DDR PMU\n perf/imx_ddr: Add support for AXI ID filtering\n arm64: kpti: ensure patched kernel text is fetched from PoU\n arm64: fix fixmap copy for 16K pages and 48-bit VA\n perf/smmuv3: Validate groups for global filtering\n perf/smmuv3: Validate group size\n arm64: Relax Documentation/arm64/tagged-pointers.rst\n arm64: kvm: Replace hardcoded \u00271\u0027 with SYS_PAR_EL1_F\n arm64: mm: Ignore spurious translation faults taken from the kernel\n ...\n" }, { "commit": "52a5525214d0d612160154d902956eca0558b7c0", "tree": "a459e3fc3977531d9f1c389eda7b49aaa0f416d0", "parents": [ "bbfe0d6b8b730af4954a0e0e741217eb3e1c58bc", "e95adb9add75affb98570a518c902f50e5fcce1b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:14:40 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:14:40 2019 -0700" }, "message": "Merge tag \u0027iommu-updates-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu\n\nPull iommu updates from Joerg Roedel:\n\n - batched unmap support for the IOMMU-API\n\n - support for unlocked command queueing in the ARM-SMMU driver\n\n - rework the ATS support in the ARM-SMMU driver\n\n - more refactoring in the ARM-SMMU driver to support hardware\n implemention specific quirks and errata\n\n - bounce buffering DMA-API implementatation in the Intel VT-d driver\n for untrusted devices (like Thunderbolt devices)\n\n - fixes for runtime PM support in the OMAP iommu driver\n\n - MT8183 IOMMU support in the Mediatek IOMMU driver\n\n - rework of the way the IOMMU core sets the default domain type for\n groups. Changing the default domain type on x86 does not require two\n kernel parameters anymore.\n\n - more smaller fixes and cleanups\n\n* tag \u0027iommu-updates-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (113 commits)\n iommu/vt-d: Declare Broadwell igfx dmar support snafu\n iommu/vt-d: Add Scalable Mode fault information\n iommu/vt-d: Use bounce buffer for untrusted devices\n iommu/vt-d: Add trace events for device dma map/unmap\n iommu/vt-d: Don\u0027t switch off swiotlb if bounce page is used\n iommu/vt-d: Check whether device requires bounce buffer\n swiotlb: Split size parameter to map/unmap APIs\n iommu/omap: Mark pm functions __maybe_unused\n iommu/ipmmu-vmsa: Disable cache snoop transactions on R-Car Gen3\n iommu/ipmmu-vmsa: Move IMTTBCR_SL0_TWOBIT_* to restore sort order\n iommu: Don\u0027t use sme_active() in generic code\n iommu/arm-smmu-v3: Fix build error without CONFIG_PCI_ATS\n iommu/qcom: Use struct_size() helper\n iommu: Remove wrong default domain comments\n iommu/dma: Fix for dereferencing before null checking\n iommu/mediatek: Clean up struct mtk_smi_iommu\n memory: mtk-smi: Get rid of need_larbid\n iommu/mediatek: Fix VLD_PA_RNG register backup when suspend\n memory: mtk-smi: Add bus_sel for mt8183\n memory: mtk-smi: Invoke pm runtime_callback to enable clocks\n ...\n" }, { "commit": "bbfe0d6b8b730af4954a0e0e741217eb3e1c58bc", "tree": "bd6a9cb698c0002e49db2919fb4323addc34a8b2", "parents": [ "31dda85e49da949fdea5fd4e1fdeeabc44e59625", "11c43bb022b373d2fdb84950ebf1911362f3f010" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:06:50 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:06:50 2019 -0700" }, "message": "Merge tag \u0027gpio-v5.4-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio\n\nPull GPIO updates from Linus Walleij:\n \"This is the bulk of changes in the GPIO subsystem for the v5.4 kernel\n cycle.\n\n Core changes:\n\n - Support hierarchical GPIO irqchips.\n\n We now have three consumers that can use this: Intel IXP4xx,\n ThunderX and Qualcomm SPMI GPIO (in the pinctrl subsystem).\n\n The support code has been long in the making and hashed out so it\n should be easily adaptable for all hierarchical irqchip parents.\n The code only gets compiled in if hierarchical irqchip is used at\n the topmost irq controller at least, as the hierarchical irqchip\n requires strict hierarchy all the way up in the system.\n\n - Determine the need for a \"valid_mask\" for GPIO lines on the\n gpio_chip and conversely for the \"valid_mask\" for the GPIO\n interrupt chip interrupt lines by looking for a .init_valid_mask()\n callback in the main chip or GPIO interrupt chip respectively.\n Allocate it with bitmap_alloc().\n\n - Isolate the device tree/open firmware GPIO description code out in\n its own file properly.\n\n - Isolate the ACPI GPIO description code out in its own file\n properly.\n\n - Drop a whole lot of #ifdef:s in the main includes: it does not hurt\n to keep the include items around, and we get quicker and clearer\n compile failures if the appropriate kernel symbols are not selected\n for drivers.\n\n New/deleted drivers:\n\n - New driver for Aspeed SGPIO.\n\n - The KS8695 driver is deleted as the platform gets deleted from\n arch/arm in this kernel cycle.\n\n - The Cirrus Logic Madera driver now supports CS47L92 and CS47L15.\n\n - The Freescale MPC8xxx now supports LS1028A and LS1088A.\n\n Driver improvements:\n\n - We pass the GPIO irqchip intialization by directly filling in the\n struct instead of using set-up functions (the new way) for Intel\n MID, Lynxpoint, Merrifield, XLP, HLWD, Aspeed, ZX, VF610, TQMX86,\n MT7621, Zynq and EP93xx.\n\n Out-of-band changes:\n\n - Fix a GPIO header inclusion in Unicore - no response from\n maintainer.\n\n - Drop FMC subsystem from MAINTAINERS - was deleted in the GPIO tree\n last cycle so let\u0027s mop up the shards\"\n\n* tag \u0027gpio-v5.4-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio: (82 commits)\n gpiolib: of: add a fallback for wlf,reset GPIO name\n gpio: htc-egpio: Remove unused exported htc_egpio_get_wakeup_irq()\n gpio: remove explicit comparison with 0\n gpio: creg-snps: use devm_platform_ioremap_resource() to simplify code\n gpio: devres: Switch to EXPORT_SYMBOL_GPL()\n gpio: of: Switch to EXPORT_SYMBOL_GPL()\n gpio: of: Make of_gpio_simple_xlate() private\n gpio: of: Make of_get_named_gpiod_flags() private\n gpio: aspeed: Add in ast2600 details to Aspeed driver\n gpio: aspeed: Use ngpio property from device tree if available\n gpio: aspeed: Setup irqchip dynamically\n gpio/aspeed: Fix incorrect number of banks\n gpio: aspeed: Update documentation with ast2600 controllers\n gpio: Initialize the irqchip valid_mask with a callback\n gpiolib: acpi: make acpi_can_fallback_to_crs() static\n gpio: Fix further merge errors\n gpio: Fix up merge collision in include file\n gpio: of: Normalize return code variable name\n gpio: gpiolib: Normalize return code variable name\n gpio: ep93xx: Pass irqchip when adding gpiochip\n ...\n" }, { "commit": "31dda85e49da949fdea5fd4e1fdeeabc44e59625", "tree": "ed2f9ada1e77e9712257edca798868eaecf1cb34", "parents": [ "aa62325dc38de2be8b1c27eb180ad3751b3f58ec", "6030f42d20cedc04df019891851320f3e257038b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:04:46 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:04:46 2019 -0700" }, "message": "Merge tag \u0027i3c/for-5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux\n\nPull i3c updates from Boris Brezillon:\n \"Core changes:\n\n - Export i3c_device_match_id() so driver can get per-device data\n\n - Add addr and lvr fields to i2c_dev_desc so we can attach I2C\n devices that are not described in the DT\n\n - Add a missing of_node_put()\n\n - Fix a memory leak\n\n - Use dev_to_i3cmaster() instead of open-coding it\n\n Driver changes:\n\n - Use for_each_set_bit() in the Cadence driver\"\n\n* tag \u0027i3c/for-5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux:\n i3c: master: Use dev_to_i3cmaster()\n i3c: master: fix a memory leak bug\n i3c: add addr and lvr to i2c_dev_desc structure\n i3c: master: cdns: Use for_each_set_bit()\n i3c: master: Add of_node_put() before return\n i3c: move i3c_device_match_id to device.c and export it\n" }, { "commit": "aa62325dc38de2be8b1c27eb180ad3751b3f58ec", "tree": "a95691be94ab37128f6aaddcf526d5c9ad2336ff", "parents": [ "c4d11ccb2b5cec6cdef7aeeb0017473d23031d96", "b769c5ba8aedc395ed04abe6db84a556d28beec1" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:02:43 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 14:02:43 2019 -0700" }, "message": "Merge tag \u0027spi-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi\n\nPull spi updates from Mark Brown:\n \"The big theme for this release has been performance, we\u0027ve had a\n series of unrelated overhauls of a few drivers all with a big\n peformance component.\n\n Otherwise it\u0027s been relatively quiet, highlights include:\n\n - A big overhaul of the spi-fsl-dspi driver improving the code\n quality, performance and stability from Vladimir Oltean.\n\n - A big performance enhancement for the bc2835 (Raspberry Pi) driver\n for unidirectional transfers from Lukas Wunner.\n\n - Improved performance on small transfers for the uniphier driver\n from Keiji Hayashibara.\n\n - Lots of coccinelle generated cleanups from Yue Haibing.\n\n - New device support for Freescale ls2080a and Nuvoton NPCM FIU\"\n\n* tag \u0027spi-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi: (102 commits)\n spi: mediatek: support large PA\n spi: mediatek: add spi support for mt6765 IC\n dt-bindings: spi: update bindings for MT6765 SoC\n spi: bcm2835: Speed up RX-only DMA transfers by zero-filling TX FIFO\n spi: bcm2835: Speed up TX-only DMA transfers by clearing RX FIFO\n dmaengine: bcm2835: Avoid accessing memory when copying zeroes\n spi: bcm2835: Cache CS register value for -\u003eprepare_message()\n dmaengine: bcm2835: Document struct bcm2835_dmadev\n spi: Guarantee cacheline alignment of driver-private data\n dmaengine: bcm2835: Allow reusable descriptors\n dmaengine: bcm2835: Allow cyclic transactions without interrupt\n spi: bcm2835: Drop dma_pending flag\n spi: bcm2835: Work around DONE bit erratum\n spi-gpio: Use PTR_ERR_OR_ZERO() in spi_gpio_request()\n spi: Use an abbreviated pointer to ctlr-\u003ecur_msg in __spi_pump_messages\n spi: npcm-fiu: remove set but not used variable \u0027retlen\u0027\n spi: fsl-spi: use devm_platform_ioremap_resource() to simplify code\n spi: zynq-qspi: use devm_platform_ioremap_resource() to simplify code\n spi: zynqmp: use devm_platform_ioremap_resource() to simplify code\n spi: xlp: use devm_platform_ioremap_resource() to simplify code\n ...\n" }, { "commit": "c4d11ccb2b5cec6cdef7aeeb0017473d23031d96", "tree": "2cae7e62c8852d634b6e79eba865b3bfc7fae0c1", "parents": [ "0372fd1a70c4bc0731486851abe2048993f94a8d", "c4ad85026d4dd5a3f65c04b4564fe273e37e5b88" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:58:43 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:58:43 2019 -0700" }, "message": "Merge tag \u0027regulator-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator\n\nPull regulator updates from Mark Brown:\n \"A small update for the regualtor API for this cycle, some small fixes\n and a bunch of new devices but none of them very big.\n\n The most stand out thing is the regulator-fixed-clock driver which is\n for regulators where the enable control is done by using a clock\n instead of a GPIO or register write, a novel hardware design that had\n not previously come up.\n\n Summary:\n\n - Added a keyword pattern for regulator_get_optional() since usage of\n that API generally needs extra review.\n\n - Operating mode and suspend state support for act8865.\n\n - New device support for Active Semiconductor ACT8600 chargers,\n Mediatek MT6358, Qualcomm SM8150, regulator-fixed-clock, and\n Synoptics SY20276, SY20278 and SY8824E\"\n\n* tag \u0027regulator-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator: (52 commits)\n regulator: core: Fix error return for /sys access\n regulator: da9211: fix obtaining \"enable\" GPIO\n regulator: max77686: fix obtaining \"maxim,ena\" GPIO\n regulator: uniphier: Add Pro5 USB3 VBUS support\n dt-bindings: regulator: add regulator-fixed-clock binding\n regulator: fixed: add possibility to enable by clock\n regulator: s2mps11: Consistently use local variable\n regulator: lp87565: Simplify lp87565_buck_set_ramp_delay\n regulator: slg51000: use devm_gpiod_get_optional() in probe\n regulator: lp8788-ldo: make array en_mask static const, makes object smaller\n regulator: tps65132: Stop parsing DT when gpio is not found\n regulator: Defer init completion for a while after late_initcall\n regulator: add missing \u0027static inline\u0027 to a helper\u0027s stub\n regulator: provide regulator_bulk_set_supply_names()\n MAINTAINERS: Add keyword pattern on regulator_get_optional()\n regulator: sy8824x: add prefixes to BUCK_EN and MODE macros\n regulator: sy8824x: use c++style for the comment block near SPDX\n regulator: mt6358: Add BROKEN dependency while waiting for MFD to merge\n regulator: mt6358: Add support for MT6358 regulator\n regulator: Add document for MT6358 regulator\n ...\n" }, { "commit": "0372fd1a70c4bc0731486851abe2048993f94a8d", "tree": "9e86c9f2d54a187569008546778ada4eeec047ed", "parents": [ "6729fb666a3b5a9a5ffa1bb6589127f7e4d35c38", "1bd4584626a9715634d2cb91ae2ed0364c070b01" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:57:02 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:57:02 2019 -0700" }, "message": "Merge tag \u0027regmap-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap\n\nPull regmap updates from Mark Brown:\n \"Only two changes for this release, one fix for error handling with\n runtime PM and a change from Greg removing error handling from debugfs\n API calls now that they implement user visible error reporting\"\n\n* tag \u0027regmap-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap:\n regmap-irq: Correct error paths in regmap_irq_thread for pm_runtime\n regmap: no need to check return value of debugfs_create functions\n" }, { "commit": "6729fb666a3b5a9a5ffa1bb6589127f7e4d35c38", "tree": "e4cb26eff52f6ed9a0dd65fb6aaa7ecbc402f2af", "parents": [ "8e97be2acd59e07539c5002e12ba6a9c0bb5e83b", "4e19e72f45d360975c59df8272f98bff59f6b748" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:44:16 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:44:16 2019 -0700" }, "message": "Merge tag \u0027hwmon-for-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging\n\nPull hwmon updates from Guenter Roeck:\n \"New drivers:\n - Inspur Power System power supply driver\n - Synaptics AS370 PVT sensor driver\n\n Chip support:\n - support SHTC3 in shtc1 driver\n - support NCT6116 in nct6775 driver\n - support AMD family 17h, model 70h CPUs in k10temp driver\n - support PCT2075 in lm75 driver\n\n Removed drivers:\n - ads1015 driver (now supported in iio)\n\n Other changes:\n - Convert drivers to use devm_i2c_new_dummy_device\n - Substantial structural improvements in lm75 driver adding support\n for writing sample interval for supported chips\n - Add support for PSU version 2 to ibm-cffps driver\n - Add support for power attribute to iio_hwmon bridge\n - Add support for additional fan, voltage and temperature attributes\n to nct7904 driver\n - Convert adt7475 driver to use hwmon_device_register_with_groups()\n - Convert k8temp driver to use hwmon_device_register_with_info()\n - Various other improvements and minor fixes\"\n\n* tag \u0027hwmon-for-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging: (48 commits)\n hwmon: submitting-patches: Add note on comment style\n hwmon: submitting-patches: Point to with_info API\n hwmon: (nct7904) Fix incorrect SMI status register setting of LTD temperature and fan.\n hwmon: (shtc1) add support for the SHTC3 sensor\n hwmon: (shtc1) fix shtc1 and shtw1 id mask\n hwmon: (lm75) Aproximate sample times to data-sheet values\n hwmon: (w83793d) convert to use devm_i2c_new_dummy_device\n hwmon: (w83792d) convert to use devm_i2c_new_dummy_device\n hwmon: (w83791d) convert to use devm_i2c_new_dummy_device\n hwmon: (as370-hwmon) fix devm_platform_ioremap_resource.cocci warnings\n hwmon: (lm75) Add support for writing sampling period on PCT2075\n hwmon: (lm75) Add support for writing conversion time for TMP112\n hwmon: (lm75) Move updating the sample interval to its own function\n hwmon: (lm75) Support configuring the sample time for various chips\n hwmon: (nct7904) Fix incorrect temperature limitation register setting of LTD.\n hwmon: (as370-hwmon) Add DT bindings for Synaptics AS370 PVT\n hwmon: Add Synaptics AS370 PVT sensor driver\n pmbus: (ibm-cffps) Add support for version 2 of the PSU\n dt-bindings: hwmon: Document ibm,cffps2 compatible string\n hwmon: (iio_hwmon) Enable power exporting from IIO\n ...\n" }, { "commit": "8e97be2acd59e07539c5002e12ba6a9c0bb5e83b", "tree": "5e1658dde240c462c0db0d9ee7ba2fe4d059153c", "parents": [ "ff881842e12563f25d69a4e2e373f1ad392a7027", "b6ff24f7b5101101ff897dfdde3f37924e676bc2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:42:25 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:42:25 2019 -0700" }, "message": "Merge branch \u0027ras-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull RAS updates from Borislav Petkov:\n \"The latest meager RAS updates:\n\n - Enable processing of action-optional MCEs which have the Overflow\n bit set (Tony Luck)\n\n - -Wmissing-prototypes warning fix and a build fix (Valdis\n Klētnieks)\"\n\n* \u0027ras-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n RAS: Build debugfs.o only when enabled in Kconfig\n RAS: Fix prototype warnings\n x86/mce: Don\u0027t check for the overflow bit on action optional machine checks\n" }, { "commit": "ff881842e12563f25d69a4e2e373f1ad392a7027", "tree": "8dc850f99c3d42b32fb953d72ae9551cf6741ecb", "parents": [ "a7bd4bcf138e7ec95c00d55fee158f6be378029b", "3e443eb353eda6f4b4796e07f2599683fa752f1d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:38:45 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:38:45 2019 -0700" }, "message": "Merge tag \u0027edac_for_5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras\n\nPull EDAC updates from Borislav Petkov:\n \"The new thing this time around is that we have three maintainers now\n and a new, old repo. New because it is new for the EDAC tree which is\n hosted there from now on and old because it is Tony\u0027s and mine\u0027s old\n RAS repo which we still use occasionally when the stuff isn\u0027t in tip.\n\n Summary:\n\n - EDAC tree has three maintainers and one new designated reviewer\n now, so that the work can scale better.\n\n - New driver for Mellanox\u0027 BlueField SoC DDR controller (Shravan\n Kumar Ramani)\n\n - AMD Rome support in amd64_edac (Yazen Ghannam and Isaac Vaughn)\n\n - Misc fixes, cleanups and code improvements\"\n\n* tag \u0027edac_for_5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:\n EDAC/amd64: Add PCI device IDs for family 17h, model 70h\n MAINTAINERS: Add Robert as a EDAC reviewer\n EDAC/mc_sysfs: Make debug messages consistent\n EDAC/mc_sysfs: Remove pointless gotos\n EDAC: Prefer \u0027unsigned int\u0027 to bare use of \u0027unsigned\u0027\n EDAC/amd64: Support asymmetric dual-rank DIMMs\n EDAC/amd64: Cache secondary Chip Select registers\n EDAC/amd64: Decode syndrome before translating address\n EDAC/amd64: Find Chip Select memory size using Address Mask\n EDAC/amd64: Initialize DIMM info for systems with more than two channels\n EDAC/amd64: Recognize DRAM device type ECC capability\n EDAC/amd64: Support more than two controllers for chip selects handling\n EDAC/mc: Cleanup _edac_mc_free() code\n EDAC, pnd2: Fix ioremap() size in dnv_rd_reg()\n EDAC, mellanox: Add ECC support for BlueField DDR4\n EDAC/altera: Use the proper type for the IRQ status bits\n EDAC/mc: Fix grain_bits calculation\n edac: altera: Move Stratix10 SDRAM ECC to peripheral\n MAINTAINERS: update EDAC entry to reflect current tree and maintainers\n" }, { "commit": "a7bd4bcf138e7ec95c00d55fee158f6be378029b", "tree": "e7fabdb4357243ebe96e757a3ea9a0532bc66b36", "parents": [ "c17112a5c413f20188da276c138484e7127cdc84", "e8bd417aab0c72bfb54465596b16085702ba0405" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:34:04 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 13:34:04 2019 -0700" }, "message": "Merge tag \u0027tpmdd-next-20190902\u0027 of git://git.infradead.org/users/jjs/linux-tpmdd\n\nPull tpm updates from Jarkko Sakkinen:\n \"A new driver for fTPM living inside ARM TEE was added this round.\n\n In addition to that, there are three bug fixes and one clean up\"\n\n* tag \u0027tpmdd-next-20190902\u0027 of git://git.infradead.org/users/jjs/linux-tpmdd:\n tpm/tpm_ftpm_tee: Document fTPM TEE driver\n tpm/tpm_ftpm_tee: A driver for firmware TPM running inside TEE\n tpm: Remove a deprecated comments about implicit sysfs locking\n tpm_tis_core: Set TPM_CHIP_FLAG_IRQ before probing for interrupts\n tpm_tis_core: Turn on the TPM before probing IRQ\u0027s\n MAINTAINERS: fix style in KEYS-TRUSTED entry\n" }, { "commit": "c17112a5c413f20188da276c138484e7127cdc84", "tree": "9fb0607a64e4dd49bbf405ed456221640fa972d8", "parents": [ "4d856f72c10ecb060868ed10ff1b1453943fc6c8", "821cc7b0b205c0df64cce59aacc330af251fa8f7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 09:28:19 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Sep 16 09:28:19 2019 -0700" }, "message": "Merge tag \u0027core-process-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux\n\nPull pidfd/waitid updates from Christian Brauner:\n \"This contains two features and various tests.\n\n First, it adds support for waiting on process through pidfds by adding\n the P_PIDFD type to the waitid() syscall. This completes the basic\n functionality of the pidfd api (cf. [1]). In the meantime we also have\n a new adition to the userspace projects that make use of the pidfd\n api. The qt project was nice enough to send a mail pointing out that\n they have a pr up to switch to the pidfd api (cf. [2]).\n\n Second, this tag contains an extension to the waitid() syscall to make\n it possible to wait on the current process group in a race free manner\n (even though the actual problem is very unlikely) by specifing 0\n together with the P_PGID type. This extension traces back to a\n discussion on the glibc development mailing list.\n\n There are also a range of tests for the features above. Additionally,\n the test-suite which detected the pidfd-polling race we fixed in [3]\n is included in this tag\"\n\n[1] https://lwn.net/Articles/794707/\n[2] https://codereview.qt-project.org/c/qt/qtbase/+/108456\n[3] commit b191d6491be6 (\"pidfd: fix a poll race when setting exit_state\")\n\n* tag \u0027core-process-v5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux:\n waitid: Add support for waiting for the current process group\n tests: add pidfd poll tests\n tests: move common definitions and functions into pidfd.h\n pidfd: add pidfd_wait tests\n pidfd: add P_PIDFD to waitid()\n" }, { "commit": "563c4f85f9f0d63b712081d5b4522152cdcb8b6b", "tree": "92351ab2d60d5fd7ce5745d21c9a60ce6d46c5dd", "parents": [ "4adcdcea717cb2d8436bef00dd689aa5bc76f11b", "09c7e8b21d67c3c78ab9701dbc0fb1e9f14a0ba5" ], "author": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Mon Sep 16 14:04:28 2019 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Mon Sep 16 14:05:04 2019 +0200" }, "message": "Merge branch \u0027sched/rt\u0027 into sched/core, to pick up -rt changes\n\nPick up the first couple of patches working towards PREEMPT_RT.\n\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "4d856f72c10ecb060868ed10ff1b1453943fc6c8", "tree": "f4734bac3f77ce0c1192d6772792a19b9e23c5c9", "parents": [ "72dbcf72156641fde4d8ea401e977341bfd35a05" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Sep 15 14:19:32 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Sep 15 14:19:32 2019 -0700" }, "message": "Linux 5.3\n" }, { "commit": "72dbcf72156641fde4d8ea401e977341bfd35a05", "tree": "48a69287e39099681405f1981e8280cf2d780835", "parents": [ "1609d7604b847a9820e63393d1a3b6cac7286d40" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Sep 15 12:32:03 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Sep 15 12:32:03 2019 -0700" }, "message": "Revert \"ext4: make __ext4_get_inode_loc plug\"\n\nThis reverts commit b03755ad6f33b7b8cd7312a3596a2dbf496de6e7.\n\nThis is sad, and done for all the wrong reasons. Because that commit is\ngood, and does exactly what it says: avoids a lot of small disk requests\nfor the inode table read-ahead.\n\nHowever, it turns out that it causes an entirely unrelated problem: the\ngetrandom() system call was introduced back in 2014 by commit\nc6e9d6f38894 (\"random: introduce getrandom(2) system call\"), and people\nuse it as a convenient source of good random numbers.\n\nBut part of the current semantics for getrandom() is that it waits for\nthe entropy pool to fill at least partially (unlike /dev/urandom). And\nat least ArchLinux apparently has a systemd that uses getrandom() at\nboot time, and the improvements in IO patterns means that existing\ninstallations suddenly start hanging, waiting for entropy that will\nnever happen.\n\nIt seems to be an unlucky combination of not _quite_ enough entropy,\ntogether with a particular systemd version and configuration. Lennart\nsays that the systemd-random-seed process (which is what does this early\naccess) is supposed to not block any other boot activity, but sadly that\ndoesn\u0027t actually seem to be the case (possibly due bogus dependencies on\ncryptsetup for encrypted swapspace).\n\nThe correct fix is to fix getrandom() to not block when it\u0027s not\nappropriate, but that fix is going to take a lot more discussion. Do we\njust make it act like /dev/urandom by default, and add a new flag for\n\"wait for entropy\"? Do we add a boot-time option? Or do we just limit\nthe amount of time it will wait for entropy?\n\nSo in the meantime, we do the revert to give us time to discuss the\neventual fix for the fundamental problem, at which point we can re-apply\nthe ext4 inode table access optimization.\n\nReported-by: Ahmed S. Darwish \u003cdarwish.07@gmail.com\u003e\nCc: Ted Ts\u0027o \u003ctytso@mit.edu\u003e\nCc: Willy Tarreau \u003cw@1wt.eu\u003e\nCc: Alexander E. Patrakov \u003cpatrakov@gmail.com\u003e\nCc: Lennart Poettering \u003cmzxreary@0pointer.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b769c5ba8aedc395ed04abe6db84a556d28beec1", "tree": "dd00c955f899f65785d60c2b4673785c28054b30", "parents": [ "262a2f33454fcecdc2032ca84d6fecdb08233468", "fdeae8f5a2e5eb3fcc9c295bfb28503c3abd4d6e" ], "author": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Sun Sep 15 10:32:06 2019 +0100" }, "committer": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Sun Sep 15 10:32:06 2019 +0100" }, "message": "Merge branch \u0027spi-5.4\u0027 into spi-next\n" }, { "commit": "262a2f33454fcecdc2032ca84d6fecdb08233468", "tree": "0689eaf3ef298f1ebd62db89a6c19680e5eb479c", "parents": [ "f74c2bb98776e2de508f4d607cd519873065118e", "4c524191c0a21d758b519087c64f84348095e940" ], "author": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Sun Sep 15 10:32:04 2019 +0100" }, "committer": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Sun Sep 15 10:32:04 2019 +0100" }, "message": "Merge branch \u0027spi-5.3\u0027 into spi-linus\n" }, { "commit": "1609d7604b847a9820e63393d1a3b6cac7286d40", "tree": "993914907707ceff9eb965f8c519f0a91f5ab192", "parents": [ "1f9c632cde0c3d781463a88ce430a8dd4a7c1a0e", "a9c20bb0206ae9384bd470a6832dd8913730add9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 16:07:40 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 16:07:40 2019 -0700" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm\n\nPull kvm fixes from Paolo Bonzini:\n \"The main change here is a revert of reverts. We recently simplified\n some code that was thought unnecessary; however, since then KVM has\n grown quite a few cond_resched()s and for that reason the simplified\n code is prone to livelocks---one CPUs tries to empty a list of guest\n page tables while the others keep adding to them. This adds back the\n generation-based zapping of guest page tables, which was not\n unnecessary after all.\n\n On top of this, there is a fix for a kernel memory leak and a couple\n of s390 fixlets as well\"\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm:\n KVM: x86/mmu: Reintroduce fast invalidate/zap for flushing memslot\n KVM: x86: work around leak of uninitialized stack contents\n KVM: nVMX: handle page fault in vmread\n KVM: s390: Do not leak kernel stack data in the KVM_S390_INTERRUPT ioctl\n KVM: s390: kvm_s390_vm_start_migration: check dirty_bitmap before using it as target for memset()\n" }, { "commit": "1f9c632cde0c3d781463a88ce430a8dd4a7c1a0e", "tree": "0ec9848980252d49870d3e09fde699be97138dd1", "parents": [ "b03c036e6f96340dd311817c7b964dad183c4141", "0d4a3f2abbef73b9e5bb5f12213c275565473588" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 16:02:49 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 16:02:49 2019 -0700" }, "message": "Merge tag \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost\n\nPull virtio fix from Michael Tsirkin:\n \"A last minute revert\n\n The 32-bit build got broken by the latest defence in depth patch.\n Revert and we\u0027ll try again in the next cycle\"\n\n* tag \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:\n Revert \"vhost: block speculation of translated descriptors\"\n" }, { "commit": "b03c036e6f96340dd311817c7b964dad183c4141", "tree": "ad58fa152e995a03d63958ea98f344721ce88e8b", "parents": [ "36024fcf8d28999f270908e75675d43b099ff7b3", "474efecb65dceb15f793b6e2f2b226e952f0f8e9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 15:58:02 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 15:58:02 2019 -0700" }, "message": "Merge tag \u0027riscv/for-v5.3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux\n\nPull RISC-V fix from Paul Walmsley:\n \"Last week, Palmer and I learned that there was an error in the RISC-V\n kernel image header format that could make it less compatible with the\n ARM64 kernel image header format. I had missed this error during my\n original reviews of the patch.\n\n The kernel image header format is an interface that impacts\n bootloaders, QEMU, and other user tools. Those packages must be\n updated to align with whatever is merged in the kernel. We would like\n to avoid proliferating these image formats by keeping the RISC-V\n header as close as possible to the existing ARM64 header. Since the\n arch/riscv patch that adds support for the image header was merged\n with our v5.3-rc1 pull request as commit 0f327f2aaad6a (\"RISC-V: Add\n an Image header that boot loader can parse.\"), we think it wise to try\n to fix this error before v5.3 is released.\n\n The fix itself should be backwards-compatible with any project that\n has already merged support for premature versions of this interface.\n It primarily involves ensuring that the RISC-V image header has\n something useful in the same field as the ARM64 image header\"\n\n* tag \u0027riscv/for-v5.3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:\n riscv: modify the Image header to improve compatibility with the ARM64 header\n" }, { "commit": "0d4a3f2abbef73b9e5bb5f12213c275565473588", "tree": "5aa5d51eb73baf137252c43fd26aa33f4e897446", "parents": [ "060423bfdee3f8bc6e2c1bac97de24d5415e2bc4" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@redhat.com", "time": "Sat Sep 14 15:21:51 2019 -0400" }, "committer": { "name": "Michael S. Tsirkin", "email": "mst@redhat.com", "time": "Sat Sep 14 15:21:51 2019 -0400" }, "message": "Revert \"vhost: block speculation of translated descriptors\"\n\nThis reverts commit a89db445fbd7f1f8457b03759aa7343fa530ef6b.\n\nI was hasty to include this patch, and it breaks the build on 32 bit.\nDefence in depth is good but let\u0027s do it properly.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\n" }, { "commit": "36024fcf8d28999f270908e75675d43b099ff7b3", "tree": "547c289f2ee7c24b4314b072abafb67780378792", "parents": [ "1c4c5e2528af0c803fb1171632074f4070229a75", "4d7ffcf3bf1be98d876c570cab8fc31d9fa92725" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 12:20:38 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 12:20:38 2019 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from David Miller:\n\n 1) Don\u0027t corrupt xfrm_interface parms before validation, from Nicolas\n Dichtel.\n\n 2) Revert use of usb-wakeup in btusb, from Mario Limonciello.\n\n 3) Block ipv6 packets in bridge netfilter if ipv6 is disabled, from\n Leonardo Bras.\n\n 4) IPS_OFFLOAD not honored in ctnetlink, from Pablo Neira Ayuso.\n\n 5) Missing ULP check in sock_map, from John Fastabend.\n\n 6) Fix receive statistic handling in forcedeth, from Zhu Yanjun.\n\n 7) Fix length of SKB allocated in 6pack driver, from Christophe\n JAILLET.\n\n 8) ip6_route_info_create() returns an error pointer, not NULL. From\n Maciej Żenczykowski.\n\n 9) Only add RDS sock to the hashes after rs_transport is set, from\n Ka-Cheong Poon.\n\n10) Don\u0027t double clean TX descriptors in ixgbe, from Ilya Maximets.\n\n11) Presence of transmit IPSEC offload in an SKB is not tested for\n correctly in ixgbe and ixgbevf. From Steffen Klassert and Jeff\n Kirsher.\n\n12) Need rcu_barrier() when register_netdevice() takes one of the\n notifier based failure paths, from Subash Abhinov Kasiviswanathan.\n\n13) Fix leak in sctp_do_bind(), from Mao Wenan.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (72 commits)\n cdc_ether: fix rndis support for Mediatek based smartphones\n sctp: destroy bucket if failed to bind addr\n sctp: remove redundant assignment when call sctp_get_port_local\n sctp: change return type of sctp_get_port_local\n ixgbevf: Fix secpath usage for IPsec Tx offload\n sctp: Fix the link time qualifier of \u0027sctp_ctrlsock_exit()\u0027\n ixgbe: Fix secpath usage for IPsec TX offload.\n net: qrtr: fix memort leak in qrtr_tun_write_iter\n net: Fix null de-reference of device refcount\n ipv6: Fix the link time qualifier of \u0027ping_v6_proc_exit_net()\u0027\n tun: fix use-after-free when register netdev failed\n tcp: fix tcp_ecn_withdraw_cwr() to clear TCP_ECN_QUEUE_CWR\n ixgbe: fix double clean of Tx descriptors with xdp\n ixgbe: Prevent u8 wrapping of ITR value to something less than 10us\n mlx4: fix spelling mistake \"veify\" -\u003e \"verify\"\n net: hns3: fix spelling mistake \"undeflow\" -\u003e \"underflow\"\n net: lmc: fix spelling mistake \"runnin\" -\u003e \"running\"\n NFC: st95hf: fix spelling mistake \"receieve\" -\u003e \"receive\"\n net/rds: An rds_sock is added too early to the hash table\n mac80211: Do not send Layer 2 Update frame before authorization\n ...\n" }, { "commit": "1c4c5e2528af0c803fb1171632074f4070229a75", "tree": "73638cf03626fec0a815080dd384b47a15a212da", "parents": [ "592b8d8759ceb7086e1683e1796c7110e6c2ae8f", "87b5d602a1cc76169b8d81ec2c74c8d95d9350dc" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 12:08:19 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 12:08:19 2019 -0700" }, "message": "Merge tag \u0027mmc-v5.3-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc\n\nPull MMC fixes from Ulf Hansson:\n\n - tmio: Fixup runtime PM management during probe and remove\n\n - sdhci-pci-o2micro: Fix eMMC initialization for an AMD SoC\n\n - bcm2835: Prevent lockups when terminating work\n\n* tag \u0027mmc-v5.3-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:\n mmc: tmio: Fixup runtime PM management during remove\n mmc: tmio: Fixup runtime PM management during probe\n Revert \"mmc: tmio: move runtime PM enablement to the driver implementations\"\n Revert \"mmc: sdhci: Remove unneeded quirk2 flag of O2 SD host controller\"\n Revert \"mmc: bcm2835: Terminate timeout work synchronously\"\n" }, { "commit": "592b8d8759ceb7086e1683e1796c7110e6c2ae8f", "tree": "734b938984201039c1e35517d6d9a1c67b8bd051", "parents": [ "a7f89616b7376495424f682b6086e0c391a89a1d", "e6bb711600db23eef2fa0c16a2d361e17b45bb28" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 11:54:57 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Sep 14 11:54:57 2019 -0700" }, "message": "Merge tag \u0027drm-fixes-2019-09-13\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"From the maintainer summit, just some last minute fixes for final:\n\n lima:\n - fix gem_wait ioctl\n\n core:\n - constify modes list\n\n i915:\n - DP MST high color depth regression\n - GPU hangs on vulkan compute workloads\"\n\n* tag \u0027drm-fixes-2019-09-13\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/lima: fix lima_gem_wait() return value\n drm/i915: Restore relaxed padding (OCL_OOB_SUPPRES_ENABLE) for skl+\n drm/i915: Limit MST to \u003c\u003d 8bpc once again\n drm/modes: Make the whitelist more const\n" }, { "commit": "a9c20bb0206ae9384bd470a6832dd8913730add9", "tree": "2a6bf3c6d9bc73c1c44c972143f6656927f50111", "parents": [ "002c5f73c508f7df5681bda339831c27f3c1aef4", "53936b5bf35e140ae27e4bbf0447a61063f400da" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Sat Sep 14 09:25:30 2019 +0200" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Sat Sep 14 09:25:30 2019 +0200" }, "message": "Merge tag \u0027kvm-s390-master-5.3-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into kvm-master\n\nKVM: s390: Fixes for 5.3\n\n- prevent a user triggerable oops in the migration code\n- do not leak kernel stack content\n" }, { "commit": "002c5f73c508f7df5681bda339831c27f3c1aef4", "tree": "2fd5017a8aea7c525cdc3592935b5b452a93e116", "parents": [ "541ab2aeb28251bf7135c7961f3a6080eebcc705" ], "author": { "name": "Sean Christopherson", "email": "sean.j.christopherson@intel.com", "time": "Thu Sep 12 19:46:02 2019 -0700" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Sat Sep 14 09:25:11 2019 +0200" }, "message": "KVM: x86/mmu: Reintroduce fast invalidate/zap for flushing memslot\n\nJames Harvey reported a livelock that was introduced by commit\nd012a06ab1d23 (\"Revert \"KVM: x86/mmu: Zap only the relevant pages when\nremoving a memslot\"\").\n\nThe livelock occurs because kvm_mmu_zap_all() as it exists today will\nvoluntarily reschedule and drop KVM\u0027s mmu_lock, which allows other vCPUs\nto add shadow pages. With enough vCPUs, kvm_mmu_zap_all() can get stuck\nin an infinite loop as it can never zap all pages before observing lock\ncontention or the need to reschedule. The equivalent of kvm_mmu_zap_all()\nthat was in use at the time of the reverted commit (4e103134b8623, \"KVM:\nx86/mmu: Zap only the relevant pages when removing a memslot\") employed\na fast invalidate mechanism and was not susceptible to the above livelock.\n\nThere are three ways to fix the livelock:\n\n- Reverting the revert (commit d012a06ab1d23) is not a viable option as\n the revert is needed to fix a regression that occurs when the guest has\n one or more assigned devices. It\u0027s unlikely we\u0027ll root cause the device\n assignment regression soon enough to fix the regression timely.\n\n- Remove the conditional reschedule from kvm_mmu_zap_all(). However, although\n removing the reschedule would be a smaller code change, it\u0027s less safe\n in the sense that the resulting kvm_mmu_zap_all() hasn\u0027t been used in\n the wild for flushing memslots since the fast invalidate mechanism was\n introduced by commit 6ca18b6950f8d (\"KVM: x86: use the fast way to\n invalidate all pages\"), back in 2013.\n\n- Reintroduce the fast invalidate mechanism and use it when zapping shadow\n pages in response to a memslot being deleted/moved, which is what this\n patch does.\n\nFor all intents and purposes, this is a revert of commit ea145aacf4ae8\n(\"Revert \"KVM: MMU: fast invalidate all pages\"\") and a partial revert of\ncommit 7390de1e99a70 (\"Revert \"KVM: x86: use the fast way to invalidate\nall pages\"\"), i.e. restores the behavior of commit 5304b8d37c2a5 (\"KVM:\nMMU: fast invalidate all pages\") and commit 6ca18b6950f8d (\"KVM: x86:\nuse the fast way to invalidate all pages\") respectively.\n\nFixes: d012a06ab1d23 (\"Revert \"KVM: x86/mmu: Zap only the relevant pages when removing a memslot\"\")\nReported-by: James Harvey \u003cjamespharvey20@gmail.com\u003e\nCc: Alex Willamson \u003calex.williamson@redhat.com\u003e\nCc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "541ab2aeb28251bf7135c7961f3a6080eebcc705", "tree": "98b9ca06198ed91ff5f188d7b4c94c00de82c254", "parents": [ "f7eea636c3d505fe6f1d1066234f1aaf7171b681" ], "author": { "name": "Fuqian Huang", "email": "huangfq.daxian@gmail.com", "time": "Thu Sep 12 12:18:17 2019 +0800" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Sat Sep 14 09:25:11 2019 +0200" }, "message": "KVM: x86: work around leak of uninitialized stack contents\n\nEmulation of VMPTRST can incorrectly inject a page fault\nwhen passed an operand that points to an MMIO address.\nThe page fault will use uninitialized kernel stack memory\nas the CR2 and error code.\n\nThe right behavior would be to abort the VM with a KVM_EXIT_INTERNAL_ERROR\nexit to userspace; however, it is not an easy fix, so for now just ensure\nthat the error code and CR2 are zero.\n\nSigned-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nCc: stable@vger.kernel.org\n[add comment]\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "f7eea636c3d505fe6f1d1066234f1aaf7171b681", "tree": "9130f21fdcaf7da6057af6f683e6ef23df255e50", "parents": [ "a7f89616b7376495424f682b6086e0c391a89a1d" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Sat Sep 14 00:26:27 2019 +0200" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Sat Sep 14 09:25:02 2019 +0200" }, "message": "KVM: nVMX: handle page fault in vmread\n\nThe implementation of vmread to memory is still incomplete, as it\nlacks the ability to do vmread to I/O memory just like vmptrst.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "474efecb65dceb15f793b6e2f2b226e952f0f8e9", "tree": "823295a3eb9436fe5e9277c553666caf3aa04adf", "parents": [ "f74c2bb98776e2de508f4d607cd519873065118e" ], "author": { "name": "Paul Walmsley", "email": "paul.walmsley@sifive.com", "time": "Fri Sep 13 18:35:50 2019 -0700" }, "committer": { "name": "Paul Walmsley", "email": "paul.walmsley@sifive.com", "time": "Fri Sep 13 19:03:52 2019 -0700" }, "message": "riscv: modify the Image header to improve compatibility with the ARM64 header\n\nPart of the intention during the definition of the RISC-V kernel image\nheader was to lay the groundwork for a future merge with the ARM64\nimage header. One error during my original review was not noticing\nthat the RISC-V header\u0027s \"magic\" field was at a different size and\nposition than the ARM64\u0027s \"magic\" field. If the existing ARM64 Image\nheader parsing code were to attempt to parse an existing RISC-V kernel\nimage header format, it would see a magic number 0. This is\nundesirable, since it\u0027s our intention to align as closely as possible\nwith the ARM64 header format. Another problem was that the original\n\"res3\" field was not being initialized correctly to zero.\n\nAddress these issues by creating a 32-bit \"magic2\" field in the RISC-V\nheader which matches the ARM64 \"magic\" field. RISC-V binaries will\nstore \"RSC\\x05\" in this field. The intention is that the use of the\nexisting 64-bit \"magic\" field in the RISC-V header will be deprecated\nover time. Increment the minor version number of the file format to\nindicate this change, and update the documentation accordingly. Fix\nthe assembler directives in head.S to ensure that reserved fields are\nproperly zero-initialized.\n\nSigned-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\nReported-by: Palmer Dabbelt \u003cpalmer@sifive.com\u003e\nReviewed-by: Palmer Dabbelt \u003cpalmer@sifive.com\u003e\nCc: Atish Patra \u003catish.patra@wdc.com\u003e\nCc: Karsten Merker \u003cmerker@debian.org\u003e\nLink: https://lore.kernel.org/linux-riscv/194c2f10c9806720623430dbf0cc59a965e50448.camel@wdc.com/T/#u\nLink: https://lore.kernel.org/linux-riscv/mhng-755b14c4-8f35-4079-a7ff-e421fd1b02bc@palmer-si-x1e/T/#t" }, { "commit": "4d7ffcf3bf1be98d876c570cab8fc31d9fa92725", "tree": "f40866eb676676df6994561b89ad364da3ad13e6", "parents": [ "ae3b06ed55b1554e9a91bf959c6b0b5e212e7f4d" ], "author": { "name": "Bjørn Mork", "email": "bjorn@mork.no", "time": "Thu Sep 12 10:42:00 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Sep 13 22:08:13 2019 +0200" }, "message": "cdc_ether: fix rndis support for Mediatek based smartphones\n\nA Mediatek based smartphone owner reports problems with USB\ntethering in Linux. The verbose USB listing shows a rndis_host\ninterface pair (e0/01/03 + 10/00/00), but the driver fails to\nbind with\n\n[ 355.960428] usb 1-4: bad CDC descriptors\n\nThe problem is a failsafe test intended to filter out ACM serial\nfunctions using the same 02/02/ff class/subclass/protocol as RNDIS.\nThe serial functions are recognized by their non-zero bmCapabilities.\n\nNo RNDIS function with non-zero bmCapabilities were known at the time\nthis failsafe was added. But it turns out that some Wireless class\nRNDIS functions are using the bmCapabilities field. These functions\nare uniquely identified as RNDIS by their class/subclass/protocol, so\nthe failing test can safely be disabled. The same applies to the two\ntypes of Misc class RNDIS functions.\n\nApplying the failsafe to Communication class functions only retains\nthe original functionality, and fixes the problem for the Mediatek based\nsmartphone.\n\nTow examples of CDC functional descriptors with non-zero bmCapabilities\nfrom Wireless class RNDIS functions are:\n\n0e8d:000a Mediatek Crosscall Spider X5 3G Phone\n\n CDC Header:\n bcdCDC 1.10\n CDC ACM:\n bmCapabilities 0x0f\n connection notifications\n sends break\n line coding and serial state\n get/set/clear comm features\n CDC Union:\n bMasterInterface 0\n bSlaveInterface 1\n CDC Call Management:\n bmCapabilities 0x03\n call management\n use DataInterface\n bDataInterface 1\n\nand\n\n19d2:1023 ZTE K4201-z\n\n CDC Header:\n bcdCDC 1.10\n CDC ACM:\n bmCapabilities 0x02\n line coding and serial state\n CDC Call Management:\n bmCapabilities 0x03\n call management\n use DataInterface\n bDataInterface 1\n CDC Union:\n bMasterInterface 0\n bSlaveInterface 1\n\nThe Mediatek example is believed to apply to most smartphones with\nMediatek firmware. The ZTE example is most likely also part of a larger\nfamily of devices/firmwares.\n\nSuggested-by: Lars Melin \u003clarsm17@gmail.com\u003e\nSigned-off-by: Bjørn Mork \u003cbjorn@mork.no\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ae3b06ed55b1554e9a91bf959c6b0b5e212e7f4d", "tree": "0716b44421ec6eae6739a33321ec4d3eecc3b8df", "parents": [ "8f6617badcc96a582678ea36ea96490c5ff26eb4", "29b99f54a8e63f871cf4060406d09fd8b0d744d8" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Sep 13 22:06:20 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Sep 13 22:06:20 2019 +0200" }, "message": "Merge branch \u0027sctp_do_bind-leak\u0027\n\nMao Wenan says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nfix memory leak for sctp_do_bind\n\nFirst two patches are to do cleanup, remove redundant assignment,\nand change return type of sctp_get_port_local.\nThird patch is to fix memory leak for sctp_do_bind if failed\nto bind address.\n\nv2: add one patch to change return type of sctp_get_port_local.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "29b99f54a8e63f871cf4060406d09fd8b0d744d8", "tree": "0716b44421ec6eae6739a33321ec4d3eecc3b8df", "parents": [ "e0e4b8de10ad3affe83cc91a1cf6d1e48cca333e" ], "author": { "name": "Mao Wenan", "email": "maowenan@huawei.com", "time": "Thu Sep 12 12:02:19 2019 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Sep 13 22:06:20 2019 +0200" }, "message": "sctp: destroy bucket if failed to bind addr\n\nThere is one memory leak bug report:\nBUG: memory leak\nunreferenced object 0xffff8881dc4c5ec0 (size 40):\n comm \"syz-executor.0\", pid 5673, jiffies 4298198457 (age 27.578s)\n hex dump (first 32 bytes):\n 02 00 00 00 81 88 ff ff 00 00 00 00 00 00 00 00 ................\n f8 63 3d c1 81 88 ff ff 00 00 00 00 00 00 00 00 .c\u003d.............\n backtrace:\n [\u003c0000000072006339\u003e] sctp_get_port_local+0x2a1/0xa00 [sctp]\n [\u003c00000000c7b379ec\u003e] sctp_do_bind+0x176/0x2c0 [sctp]\n [\u003c000000005be274a2\u003e] sctp_bind+0x5a/0x80 [sctp]\n [\u003c00000000b66b4044\u003e] inet6_bind+0x59/0xd0 [ipv6]\n [\u003c00000000c68c7f42\u003e] __sys_bind+0x120/0x1f0 net/socket.c:1647\n [\u003c000000004513635b\u003e] __do_sys_bind net/socket.c:1658 [inline]\n [\u003c000000004513635b\u003e] __se_sys_bind net/socket.c:1656 [inline]\n [\u003c000000004513635b\u003e] __x64_sys_bind+0x3e/0x50 net/socket.c:1656\n [\u003c0000000061f2501e\u003e] do_syscall_64+0x72/0x2e0 arch/x86/entry/common.c:296\n [\u003c0000000003d1e05e\u003e] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n\nThis is because in sctp_do_bind, if sctp_get_port_local is to\ncreate hash bucket successfully, and sctp_add_bind_addr failed\nto bind address, e.g return -ENOMEM, so memory leak found, it\nneeds to destroy allocated bucket.\n\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nSigned-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\nAcked-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e0e4b8de10ad3affe83cc91a1cf6d1e48cca333e", "tree": "5921d1bbc3bac880b1ba0d175b8fa3d7d01493f0", "parents": [ "8e2ef6abd4653b1cfeefe262017ecdd5eaf40fef" ], "author": { "name": "Mao Wenan", "email": "maowenan@huawei.com", "time": "Thu Sep 12 12:02:18 2019 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Sep 13 22:06:20 2019 +0200" }, "message": "sctp: remove redundant assignment when call sctp_get_port_local\n\nThere are more parentheses in if clause when call sctp_get_port_local\nin sctp_do_bind, and redundant assignment to \u0027ret\u0027. This patch is to\ndo cleanup.\n\nSigned-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\nAcked-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8e2ef6abd4653b1cfeefe262017ecdd5eaf40fef", "tree": "32c066b1f02081a3f3194c51342ba2a278bd3349", "parents": [ "8f6617badcc96a582678ea36ea96490c5ff26eb4" ], "author": { "name": "Mao Wenan", "email": "maowenan@huawei.com", "time": "Thu Sep 12 12:02:17 2019 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Sep 13 22:06:20 2019 +0200" }, "message": "sctp: change return type of sctp_get_port_local\n\nCurrently sctp_get_port_local() returns a long\nwhich is either 0,1 or a pointer casted to long.\nIt\u0027s neither of the callers use the return value since\ncommit 62208f12451f (\"net: sctp: simplify sctp_get_port\").\nNow two callers are sctp_get_port and sctp_do_bind,\nthey actually assumend a casted to an int was the same as\na pointer casted to a long, and they don\u0027t save the return\nvalue just check whether it is zero or non-zero, so\nit would better change return type from long to int for\nsctp_get_port_local.\n\nSigned-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8f6617badcc96a582678ea36ea96490c5ff26eb4", "tree": "8d254820d274adb5a7c9f03abd63eae24f08a584", "parents": [ "b456d72412ca8797234449c25815e82f4e1426c0" ], "author": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Thu Sep 12 12:07:34 2019 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Sep 13 15:52:10 2019 +0200" }, "message": "ixgbevf: Fix secpath usage for IPsec Tx offload\n\nPort the same fix for ixgbe to ixgbevf.\n\nThe ixgbevf driver currently does IPsec Tx offloading\nbased on an existing secpath. However, the secpath\ncan also come from the Rx side, in this case it is\nmisinterpreted for Tx offload and the packets are\ndropped with a \"bad sa_idx\" error. Fix this by using\nthe xfrm_offload() function to test for Tx offload.\n\nCC: Shannon Nelson \u003csnelson@pensando.io\u003e\nFixes: 7f68d4306701 (\"ixgbevf: enable VF IPsec offload operations\")\nReported-by: Jonathan Tooker \u003cjonathan@reliablehosting.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\nAcked-by: Shannon Nelson \u003csnelson@pensando.io\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4e19e72f45d360975c59df8272f98bff59f6b748", "tree": "6407298626146714ca567b0055767860004682f8", "parents": [ "9b0cffa6c46239b64c60e8c4484a41310708bcec" ], "author": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Fri Sep 13 05:43:06 2019 -0700" }, "committer": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Fri Sep 13 05:43:06 2019 -0700" }, "message": "hwmon: submitting-patches: Add note on comment style\n\nAsk for standard multi-line comments, and ask for consistent\ncomment style.\n\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n" }, { "commit": "9b0cffa6c46239b64c60e8c4484a41310708bcec", "tree": "4e6e9d279963ce3c37dd358eb61c9b90b56014ad", "parents": [ "3b710d7ae5d5138a0d4d876a4ea12bfbb8055efb" ], "author": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Fri Sep 13 05:06:45 2019 -0700" }, "committer": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Fri Sep 13 05:06:45 2019 -0700" }, "message": "hwmon: submitting-patches: Point to with_info API\n\nNew driver should use devm_hwmon_device_register_with_info() or\nhwmon_device_register_with_info() to register with the hwmon subsystem.\n\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n" }, { "commit": "87b5d602a1cc76169b8d81ec2c74c8d95d9350dc", "tree": "e9ff9b5cab17d2e5a86311ddfd0f718c7f23e0cc", "parents": [ "aa86f1a3887523d78bfadd1c4e4df8f919336511" ], "author": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Fri Sep 13 11:20:22 2019 +0200" }, "committer": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Fri Sep 13 13:49:09 2019 +0200" }, "message": "mmc: tmio: Fixup runtime PM management during remove\n\nAccessing the device when it may be runtime suspended is a bug, which is\nthe case in tmio_mmc_host_remove(). Let\u0027s fix the behaviour.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\nTested-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n" }, { "commit": "aa86f1a3887523d78bfadd1c4e4df8f919336511", "tree": "d5cf7ea5e5a9f6b05dbf7ae0499791da730b672c", "parents": [ "8861474a105c91e812d38cd65d3be795c7f32c4f" ], "author": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Fri Sep 13 11:19:26 2019 +0200" }, "committer": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Fri Sep 13 13:49:04 2019 +0200" }, "message": "mmc: tmio: Fixup runtime PM management during probe\n\nThe tmio_mmc_host_probe() calls pm_runtime_set_active() to update the\nruntime PM status of the device, as to make it reflect the current status\nof the HW. This works fine for most cases, but unfortunate not for all.\nEspecially, there is a generic problem when the device has a genpd attached\nand that genpd have the -\u003estart|stop() callbacks assigned.\n\nMore precisely, if the driver calls pm_runtime_set_active() during\n-\u003eprobe(), genpd does not get to invoke the -\u003estart() callback for it,\nwhich means the HW isn\u0027t really fully powered on. Furthermore, in the next\nphase, when the device becomes runtime suspended, genpd will invoke the\n-\u003estop() callback for it, potentially leading to usage count imbalance\nproblems, depending on what\u0027s implemented behind the callbacks of course.\n\nTo fix this problem, convert to call pm_runtime_get_sync() from\ntmio_mmc_host_probe() rather than pm_runtime_set_active(). Additionally, to\navoid bumping usage counters and unnecessary re-initializing the HW the\nfirst time the tmio driver\u0027s -\u003eruntime_resume() callback is called,\nintroduce a state flag to keeping track of this.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\nTested-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n" }, { "commit": "8861474a105c91e812d38cd65d3be795c7f32c4f", "tree": "bc2999c029207a93fb1f32ac2153316ea46f97c5", "parents": [ "49baa01c8b99ef92958e18fb58ebeb5dfdcde8af" ], "author": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Fri Sep 13 10:03:15 2019 +0200" }, "committer": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Fri Sep 13 13:48:35 2019 +0200" }, "message": "Revert \"mmc: tmio: move runtime PM enablement to the driver implementations\"\n\nThis reverts commit 7ff213193310ef8d0ee5f04f79d791210787ac2c.\n\nIt turns out that the above commit introduces other problems. For example,\ncalling pm_runtime_set_active() must not be done prior calling\npm_runtime_enable() as that makes it fail. This leads to additional\nproblems, such as clock enables being wrongly balanced.\n\nRather than fixing the problem on top, let\u0027s start over by doing a revert.\n\nFixes: 7ff213193310 (\"mmc: tmio: move runtime PM enablement to the driver implementations\")\nSigned-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\nTested-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n" }, { "commit": "fdeae8f5a2e5eb3fcc9c295bfb28503c3abd4d6e", "tree": "ae88c754807581f67075a4b54d761a120a0f97bb", "parents": [ "2c231e0ab638519759c9a7203ecced0aa457c6c3" ], "author": { "name": "luhua.xu", "email": "luhua.xu@mediatek.com", "time": "Wed Sep 11 05:55:31 2019 -0400" }, "committer": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Fri Sep 13 10:41:10 2019 +0100" }, "message": "spi: mediatek: support large PA\n\nAdd spi large PA(max\u003d64G) support for DMA transfer.\n\nSigned-off-by: luhua.xu \u003cluhua.xu@mediatek.com\u003e\nLink: https://lore.kernel.org/r/1568195731-3239-4-git-send-email-luhua.xu@mediatek.com\nSigned-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n" }, { "commit": "2c231e0ab638519759c9a7203ecced0aa457c6c3", "tree": "f76923ed29ac28c89c2c61c679b7d8ec9c9ee511", "parents": [ "7359d108d44b823d5da049ee12302125a08f2216" ], "author": { "name": "luhua.xu", "email": "luhua.xu@mediatek.com", "time": "Wed Sep 11 05:55:30 2019 -0400" }, "committer": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Fri Sep 13 10:40:37 2019 +0100" }, "message": "spi: mediatek: add spi support for mt6765 IC\n\nThis patch add spi support for mt6765 IC.\n\nSigned-off-by: luhua.xu \u003cluhua.xu@mediatek.com\u003e\nLink: https://lore.kernel.org/r/1568195731-3239-3-git-send-email-luhua.xu@mediatek.com\nSigned-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n" }, { "commit": "7359d108d44b823d5da049ee12302125a08f2216", "tree": "afa0ea0bc8d709835dde2df40e6a8619142393b1", "parents": [ "2b8279aec1829da4dc645e8731c28d2f5458d652" ], "author": { "name": "luhua.xu", "email": "luhua.xu@mediatek.com", "time": "Wed Sep 11 05:55:29 2019 -0400" }, "committer": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Fri Sep 13 10:40:08 2019 +0100" }, "message": "dt-bindings: spi: update bindings for MT6765 SoC\n\nAdd a DT binding documentation for the MT6765 soc.\n\nSigned-off-by: luhua.xu \u003cluhua.xu@mediatek.com\u003e\nLink: https://lore.kernel.org/r/1568195731-3239-2-git-send-email-luhua.xu@mediatek.com\nSigned-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n" }, { "commit": "a7f89616b7376495424f682b6086e0c391a89a1d", "tree": "55cf3b4ccb41880d6d0282f91370bffddf33c320", "parents": [ "1b304a1ae45de4df7d773f0a39d1100aabca615b", "97a61369830ab085df5aed0ff9256f35b07d425a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 13 09:52:01 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 13 09:52:01 2019 +0100" }, "message": "Merge branch \u0027for-5.3-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup\n\nPull cgroup fix from Tejun Heo:\n \"Roman found and fixed a bug in the cgroup2 freezer which allows new\n child cgroup to escape frozen state\"\n\n* \u0027for-5.3-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:\n cgroup: freezer: fix frozen state inheritance\n kselftests: cgroup: add freezer mkdir test\n" }, { "commit": "1b304a1ae45de4df7d773f0a39d1100aabca615b", "tree": "f8727159a28159faaf07df87ec1d0175d6d9582a", "parents": [ "505a8ec7e11ae5236c4a154a1e24ef49a8349600", "18dfa7117a3f379862dcd3f67cadd678013bb9dd" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 13 09:48:47 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 13 09:48:47 2019 +0100" }, "message": "Merge tag \u0027for-5.3-rc8-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux\n\nPull btrfs fixes from David Sterba:\n \"Here are two fixes, one of them urgent fixing a bug introduced in 5.2\n and reported by many users. It took time to identify the root cause,\n catching the 5.3 release is higly desired also to push the fix to 5.2\n stable tree.\n\n The bug is a mess up of return values after adding proper error\n handling and honestly the kind of bug that can cause sleeping\n disorders until it\u0027s caught. My appologies to everybody who was\n affected.\n\n Summary of what could happen:\n\n 1) either a hang when committing a transaction, if this happens\n there\u0027s no risk of corruption, still the hang is very inconvenient\n and can\u0027t be resolved without a reboot\n\n 2) writeback for some btree nodes may never be started and we end up\n committing a transaction without noticing that, this is really\n serious and that will lead to the \"parent transid verify failed\"\n messages\"\n\n* tag \u0027for-5.3-rc8-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:\n Btrfs: fix unwritten extent buffers and hangs on future writeback attempts\n Btrfs: fix assertion failure during fsync and use of stale transaction\n" }, { "commit": "4adcdcea717cb2d8436bef00dd689aa5bc76f11b", "tree": "4c009a87ab0a147025002f30e70a9bea8490a6c9", "parents": [ "eb92692b2544d3f415887dbbc98499843dfe568b" ], "author": { "name": "Miles Chen", "email": "miles.chen@mediatek.com", "time": "Thu Sep 12 18:34:52 2019 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Fri Sep 13 07:49:28 2019 +0200" }, "message": "sched/psi: Correct overly pessimistic size calculation\n\nWhen passing a equal or more then 32 bytes long string to psi_write(),\npsi_write() copies 31 bytes to its buf and overwrites buf[30]\nwith \u0027\\0\u0027. Which makes the input string 1 byte shorter than\nit should be.\n\nFix it by copying sizeof(buf) bytes when nbytes \u003e\u003d sizeof(buf).\n\nThis does not cause problems in normal use case like:\n\"some 500000 10000000\" or \"full 500000 10000000\" because they\nare less than 32 bytes in length.\n\n\t/* assuming nbytes \u003d\u003d 35 */\n\tchar buf[32];\n\n\tbuf_size \u003d min(nbytes, (sizeof(buf) - 1)); /* buf_size \u003d 31 */\n\tif (copy_from_user(buf, user_buf, buf_size))\n\t\treturn -EFAULT;\n\n\tbuf[buf_size - 1] \u003d \u0027\\0\u0027; /* buf[30] \u003d \u0027\\0\u0027 */\n\nBefore:\n\n %cd /proc/pressure/\n %echo \"123456789|123456789|123456789|1234\" \u003e memory\n [ 22.473497] nbytes\u003d35,buf_size\u003d31\n [ 22.473775] 123456789|123456789|123456789| (print 30 chars)\n %sh: write error: Invalid argument\n\n %echo \"123456789|123456789|123456789|1\" \u003e memory\n [ 64.916162] nbytes\u003d32,buf_size\u003d31\n [ 64.916331] 123456789|123456789|123456789| (print 30 chars)\n %sh: write error: Invalid argument\n\nAfter:\n\n %cd /proc/pressure/\n %echo \"123456789|123456789|123456789|1234\" \u003e memory\n [ 254.837863] nbytes\u003d35,buf_size\u003d32\n [ 254.838541] 123456789|123456789|123456789|1 (print 31 chars)\n %sh: write error: Invalid argument\n\n %echo \"123456789|123456789|123456789|1\" \u003e memory\n [ 9965.714935] nbytes\u003d32,buf_size\u003d32\n [ 9965.715096] 123456789|123456789|123456789|1 (print 31 chars)\n %sh: write error: Invalid argument\n\nAlso remove the superfluous parentheses.\n\nSigned-off-by: Miles Chen \u003cmiles.chen@mediatek.com\u003e\nCc: \u003clinux-mediatek@lists.infradead.org\u003e\nCc: \u003cwsd_upstream@mediatek.com\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nLink: https://lkml.kernel.org/r/20190912103452.13281-1-miles.chen@mediatek.com\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "eb92692b2544d3f415887dbbc98499843dfe568b", "tree": "d1542b308184900dad99eb6545e23207421cc61d", "parents": [ "0413d7f33e60751570fd6c179546bde2f7d82dcb" ], "author": { "name": "Quentin Perret", "email": "quentin.perret@arm.com", "time": "Thu Sep 12 11:44:04 2019 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Fri Sep 13 07:45:17 2019 +0200" }, "message": "sched/fair: Speed-up energy-aware wake-ups\n\nEAS computes the energy impact of migrating a waking task when deciding\non which CPU it should run. However, the current approach is known to\nhave a high algorithmic complexity, which can result in prohibitively\nhigh wake-up latencies on systems with complex energy models, such as\nsystems with per-CPU DVFS. On such systems, the algorithm complexity is\nin O(n^2) (ignoring the cost of searching for performance states in the\nEM) with \u0027n\u0027 the number of CPUs.\n\nTo address this, re-factor the EAS wake-up path to compute the energy\n\u0027delta\u0027 (with and without the task) on a per-performance domain basis,\nrather than system-wide, which brings the complexity down to O(n).\n\nNo functional changes intended.\n\nTest results\n~~~~~~~~~~~~\n\n* Setup: Tested on a Google Pixel 3, with a Snapdragon 845 (4+4 CPUs,\n A55/A75). Base kernel is 5.3-rc5 + Pixel3 specific patches. Android\n userspace, no graphics.\n\n* Test case: Run a periodic rt-app task, with 16ms period, ramping down\n from 70% to 10%, in 5% steps of 500 ms each (json avail. at [1]).\n Frequencies of all CPUs are pinned to max (using scaling_min_freq\n CPUFreq sysfs entries) to reduce variability. The time to run\n select_task_rq_fair() is measured using the function profiler\n (/sys/kernel/debug/tracing/trace_stat/function*). See the test script\n for more details [2].\n\nTest 1:\n\nI hacked the DT to \u0027fake\u0027 per-CPU DVFS. That is, we end up with one\nCPUFreq policy per CPU (8 policies in total). Since all frequencies are\npinned to max for the test, this should have no impact on the actual\nfrequency selection, but it does in the EAS calculation.\n\n +---------------------------+----------------------------------+\n | Without patch | With patch |\n+-----+-----+----------+----------+-----+-----------------+----------+\n| CPU | Hit | Avg (us) | s^2 (us) | Hit | Avg (us) | s^2 (us) |\n|-----+-----+----------+----------+-----+-----------------+----------+\n| 0 | 274 | 38.303 | 1750.239 | 401 | 14.126 (-63.1%) | 146.625 |\n| 1 | 197 | 49.529 | 1695.852 | 314 | 16.135 (-67.4%) | 167.525 |\n| 2 | 142 | 34.296 | 1758.665 | 302 | 14.133 (-58.8%) | 130.071 |\n| 3 | 172 | 31.734 | 1490.975 | 641 | 14.637 (-53.9%) | 139.189 |\n| 4 | 316 | 7.834 | 178.217 | 425 | 5.413 (-30.9%) | 20.803 |\n| 5 | 447 | 8.424 | 144.638 | 556 | 5.929 (-29.6%) | 27.301 |\n| 6 | 581 | 14.886 | 346.793 | 456 | 5.711 (-61.6%) | 23.124 |\n| 7 | 456 | 10.005 | 211.187 | 997 | 4.708 (-52.9%) | 21.144 |\n+-----+-----+----------+----------+-----+-----------------+----------+\n * Hit, Avg and s^2 are as reported by the function profiler\n\nTest 2:\nI also ran the same test with a normal DT, with 2 CPUFreq policies, to\nsee if this causes regressions in the most common case.\n\n +---------------------------+----------------------------------+\n | Without patch | With patch |\n+-----+-----+----------+----------+-----+-----------------+----------+\n| CPU | Hit | Avg (us) | s^2 (us) | Hit | Avg (us) | s^2 (us) |\n|-----+-----+----------+----------+-----+-----------------+----------+\n| 0 | 345 | 22.184 | 215.321 | 580 | 18.635 (-16.0%) | 146.892 |\n| 1 | 358 | 18.597 | 200.596 | 438 | 12.934 (-30.5%) | 104.604 |\n| 2 | 359 | 25.566 | 200.217 | 397 | 10.826 (-57.7%) | 74.021 |\n| 3 | 362 | 16.881 | 200.291 | 718 | 11.455 (-32.1%) | 102.280 |\n| 4 | 457 | 3.822 | 9.895 | 757 | 4.616 (+20.8%) | 13.369 |\n| 5 | 344 | 4.301 | 7.121 | 594 | 5.320 (+23.7%) | 18.798 |\n| 6 | 472 | 4.326 | 7.849 | 464 | 5.648 (+30.6%) | 22.022 |\n| 7 | 331 | 4.630 | 13.937 | 408 | 5.299 (+14.4%) | 18.273 |\n+-----+-----+----------+----------+-----+-----------------+----------+\n * Hit, Avg and s^2 are as reported by the function profiler\n\nIn addition to these two tests, I also ran 50 iterations of the Lisa\nEAS functional test suite [3] with this patch applied on Arm Juno r0,\nArm Juno r2, Arm TC2 and Hikey960, and could not see any regressions\n(all EAS functional tests are passing).\n\n [1] https://paste.debian.net/1100055/\n [2] https://paste.debian.net/1100057/\n [3] https://github.com/ARM-software/lisa/blob/master/lisa/tests/scheduler/eas_behaviour.py\n\nSigned-off-by: Quentin Perret \u003cquentin.perret@arm.com\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: dietmar.eggemann@arm.com\nCc: juri.lelli@redhat.com\nCc: morten.rasmussen@arm.com\nCc: qais.yousef@arm.com\nCc: qperret@qperret.net\nCc: rjw@rjwysocki.net\nCc: tkjos@google.com\nCc: valentin.schneider@arm.com\nCc: vincent.guittot@linaro.org\nLink: https://lkml.kernel.org/r/20190912094404.13802-1-qperret@qperret.net\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "97a61369830ab085df5aed0ff9256f35b07d425a", "tree": "6030eff72cde967285d9ee8c8cdc330399e2f899", "parents": [ "44e9d308a51fbf6d2e5a913f7452a5d5f1902249" ], "author": { "name": "Roman Gushchin", "email": "guro@fb.com", "time": "Thu Sep 12 10:56:45 2019 -0700" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Thu Sep 12 14:04:45 2019 -0700" }, "message": "cgroup: freezer: fix frozen state inheritance\n\nIf a new child cgroup is created in the frozen cgroup hierarchy\n(one or more of ancestor cgroups is frozen), the CGRP_FREEZE cgroup\nflag should be set. Otherwise if a process will be attached to the\nchild cgroup, it won\u0027t become frozen.\n\nThe problem can be reproduced with the test_cgfreezer_mkdir test.\n\nThis is the output before this patch:\n ~/test_freezer\n ok 1 test_cgfreezer_simple\n ok 2 test_cgfreezer_tree\n ok 3 test_cgfreezer_forkbomb\n Cgroup /sys/fs/cgroup/cg_test_mkdir_A/cg_test_mkdir_B isn\u0027t frozen\n not ok 4 test_cgfreezer_mkdir\n ok 5 test_cgfreezer_rmdir\n ok 6 test_cgfreezer_migrate\n ok 7 test_cgfreezer_ptrace\n ok 8 test_cgfreezer_stopped\n ok 9 test_cgfreezer_ptraced\n ok 10 test_cgfreezer_vfork\n\nAnd with this patch:\n ~/test_freezer\n ok 1 test_cgfreezer_simple\n ok 2 test_cgfreezer_tree\n ok 3 test_cgfreezer_forkbomb\n ok 4 test_cgfreezer_mkdir\n ok 5 test_cgfreezer_rmdir\n ok 6 test_cgfreezer_migrate\n ok 7 test_cgfreezer_ptrace\n ok 8 test_cgfreezer_stopped\n ok 9 test_cgfreezer_ptraced\n ok 10 test_cgfreezer_vfork\n\nReported-by: Mark Crossen \u003cmcrossen@fb.com\u003e\nSigned-off-by: Roman Gushchin \u003cguro@fb.com\u003e\nFixes: 76f969e8948d (\"cgroup: cgroup v2 freezer\")\nCc: Tejun Heo \u003ctj@kernel.org\u003e\nCc: stable@vger.kernel.org # v5.2+\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n" }, { "commit": "44e9d308a51fbf6d2e5a913f7452a5d5f1902249", "tree": "2af3fa472bb1fba03698ef3367073149d409336f", "parents": [ "505a8ec7e11ae5236c4a154a1e24ef49a8349600" ], "author": { "name": "Roman Gushchin", "email": "guro@fb.com", "time": "Thu Sep 12 10:56:44 2019 -0700" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Thu Sep 12 14:04:40 2019 -0700" }, "message": "kselftests: cgroup: add freezer mkdir test\n\nAdd a new cgroup freezer selftest, which checks that if a cgroup is\nfrozen, their new child cgroups will properly inherit the frozen\nstate.\n\nIt creates a parent cgroup, freezes it, creates a child cgroup\nand populates it with a dummy process. Then it checks that both\nparent and child cgroup are frozen.\n\nSigned-off-by: Roman Gushchin \u003cguro@fb.com\u003e\nCc: Tejun Heo \u003ctj@kernel.org\u003e\nCc: Shuah Khan \u003cshuah@kernel.org\u003e\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n" }, { "commit": "3b710d7ae5d5138a0d4d876a4ea12bfbb8055efb", "tree": "1a9a735f34b25274006b66d6c55465a3f8900aa4", "parents": [ "ffd96868ac5d4c22bba1ba2175d124cb2775f3f7" ], "author": { "name": "amy.shih", "email": "amy.shih@advantech.com.tw", "time": "Thu Sep 12 11:33:00 2019 +0000" }, "committer": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Thu Sep 12 11:52:08 2019 -0700" }, "message": "hwmon: (nct7904) Fix incorrect SMI status register setting of LTD temperature and fan.\n\nAccording to datasheet, the SMI status register setting of LTD\ntemperature is SMI_STS3, and the SMI status register setting\nof fan is SMI_STS5 and SMI_STS6.\n\nSigned-off-by: amy.shih \u003camy.shih@advantech.com.tw\u003e\nLink: https://lore.kernel.org/r/20190912113300.4714-1-Amy.Shih@advantech.com.tw\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n" }, { "commit": "f690790c9da3122dd7ee1b0d64d97973a7c34135", "tree": "aac52bb7cc694451bf24a8bec27df96b2242efea", "parents": [ "7973353e92ee1e7ca3b2eb361a4b7cb66c92abee" ], "author": { "name": "Andy Shevchenko", "email": "andriy.shevchenko@linux.intel.com", "time": "Thu Sep 12 16:10:49 2019 +0300" }, "committer": { "name": "Andy Shevchenko", "email": "andriy.shevchenko@linux.intel.com", "time": "Thu Sep 12 17:36:42 2019 +0300" }, "message": "MAINTAINERS: Switch PDx86 subsystem status to Odd Fixes\n\nDue to shift of priorities the actual status of the subsystem is Odd Fixes.\n\nSigned-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n" }, { "commit": "505a8ec7e11ae5236c4a154a1e24ef49a8349600", "tree": "9d1fac93a7f0a94710763bc2ef19ff4e623ca262", "parents": [ "98dcb386e5c3d53da2ed0b14b5930a01c90ad36a" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Thu Sep 12 13:56:34 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 14:55:03 2019 +0100" }, "message": "Revert \"drm/i915/userptr: Acquire the page lock around set_page_dirty()\"\n\nThe userptr put_pages can be called from inside try_to_unmap, and so\nenters with the page lock held on one of the object\u0027s backing pages. We\ncannot take the page lock ourselves for fear of recursion.\n\nReported-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nReported-by: Martin Wilck \u003cMartin.Wilck@suse.com\u003e\nReported-by: Leo Kraav \u003cleho@kraav.com\u003e\nFixes: aa56a292ce62 (\"drm/i915/userptr: Acquire the page lock around set_page_dirty()\")\nReferences: https://bugzilla.kernel.org/show_bug.cgi?id\u003d203317\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nCc: Jani Nikula \u003cjani.nikula@intel.com\u003e\nCc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "98dcb386e5c3d53da2ed0b14b5930a01c90ad36a", "tree": "761717c9fdfc022d64ca9e556a5e35b1dd3e669a", "parents": [ "95217783b7f6f331e7a6675e0a31fb9a5a1b9a36", "a0eb9abd8af92d1aa34bc1e24dfbd1ba0bd6a56c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 14:50:14 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 14:50:14 2019 +0100" }, "message": "Merge tag \u0027for-linus-20190912\u0027 of gitolite.kernel.org:pub/scm/linux/kernel/git/brauner/linux\n\nPull clone3 fix from Christian Brauner:\n \"This is a last-minute bugfix for clone3() that should go in before we\n release 5.3 with clone3().\n\n clone3() did not verify that the exit_signal argument was set to a\n valid signal. This can be used to cause a crash by specifying a signal\n greater than NSIG. e.g. -1.\n\n The commit from Eugene adds a check to copy_clone_args_from_user() to\n verify that the exit signal is limited by CSIGNAL as with legacy\n clone() and that the signal is valid. With this we don\u0027t get the\n legacy clone behavior were an invalid signal could be handed down and\n would only be detected and then ignored in do_notify_parent(). Users\n of clone3() will now get a proper error right when they pass an\n invalid exit signal. Note, that this is not a change in user-visible\n behavior since no kernel with clone3() has been released yet\"\n\n* tag \u0027for-linus-20190912\u0027 of gitolite.kernel.org:pub/scm/linux/kernel/git/brauner/linux:\n fork: block invalid exit signals with clone3()\n" }, { "commit": "fcc16a9e24ba6a2bb9f3af43d892eeec2a435d18", "tree": "e1ba26d5c0c39bb398f8013d0486d05fa0ca8ae3", "parents": [ "5fa1659105fac63e0f3c199b476025c2e04111ce" ], "author": { "name": "Jeroen Roovers", "email": "jer@gentoo.org", "time": "Tue Sep 10 11:45:14 2019 +0200" }, "committer": { "name": "Helge Deller", "email": "deller@gmx.de", "time": "Thu Sep 12 15:50:07 2019 +0200" }, "message": "parisc: Have git ignore generated real2.S and firmware.c\n\nThese files are not covered in globs from any other .gitignore files.\n\nSigned-off-by: Jeroen Roovers \u003cjer@gentoo.org\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n" }, { "commit": "95217783b7f6f331e7a6675e0a31fb9a5a1b9a36", "tree": "b6b96e107fc5e9d1306f33b129da37d3c6166301", "parents": [ "ad32b4800c2b59a303e337ec1f6c2f122418cebe", "afa8b475c1aec185a8e106c48b3832e0b88bc2de" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 14:47:35 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 14:47:35 2019 +0100" }, "message": "Merge branch \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Ingo Molnar:\n \"A KVM guest fix, and a kdump kernel relocation errors fix\"\n\n* \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/timer: Force PIT initialization when !X86_FEATURE_ARAT\n x86/purgatory: Change compiler flags from -mcmodel\u003dkernel to -mcmodel\u003dlarge to fix kexec relocation errors\n" }, { "commit": "e6bb711600db23eef2fa0c16a2d361e17b45bb28", "tree": "259acfe777f792eb021aab0b1a1893a80c457873", "parents": [ "911ad0b6110d385474cb1dd91b1ef51ca6aeff93", "21670bd78a25001cf8ef2679b378c73fb73b904f" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Sep 12 23:14:29 2019 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Sep 12 23:14:35 2019 +1000" }, "message": "Merge tag \u0027drm-misc-fixes-2019-09-12\u0027 of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes\n\ndrm-misc-fixes for v5.3 final:\n- Constify modes whitelist harder.\n- Fix lima driver gem_wait ioctl.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/99e52e7a-d4ce-6a2c-0501-bc559a710955@linux.intel.com\n" }, { "commit": "911ad0b6110d385474cb1dd91b1ef51ca6aeff93", "tree": "5215836a9ed3780c45e3ce852e6d0e9ef6aec461", "parents": [ "f74c2bb98776e2de508f4d607cd519873065118e", "2eb0964eec5f1d99f9eaf4963eee267acc72b615" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Sep 12 23:11:36 2019 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Sep 12 23:11:36 2019 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2019-09-11\u0027 of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes\n\nFinal drm/i915 fixes for v5.3:\n- Fox DP MST high color depth regression\n- Fix GPU hangs on Vulkan compute workloads\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Jani Nikula \u003cjani.nikula@intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/877e6e27qm.fsf@intel.com\n" }, { "commit": "a0eb9abd8af92d1aa34bc1e24dfbd1ba0bd6a56c", "tree": "b4e8d8523ac5dd6bd19a603ebb2b35a9852114be", "parents": [ "f74c2bb98776e2de508f4d607cd519873065118e" ], "author": { "name": "Eugene Syromiatnikov", "email": "esyr@redhat.com", "time": "Wed Sep 11 18:45:40 2019 +0100" }, "committer": { "name": "Christian Brauner", "email": "christian.brauner@ubuntu.com", "time": "Thu Sep 12 14:56:33 2019 +0200" }, "message": "fork: block invalid exit signals with clone3()\n\nPreviously, higher 32 bits of exit_signal fields were lost when copied\nto the kernel args structure (that uses int as a type for the respective\nfield). Moreover, as Oleg has noted, exit_signal is used unchecked, so\nit has to be checked for sanity before use; for the legacy syscalls,\napplying CSIGNAL mask guarantees that it is at least non-negative;\nhowever, there\u0027s no such thing is done in clone3() code path, and that\ncan break at least thread_group_leader.\n\nThis commit adds a check to copy_clone_args_from_user() to verify that\nthe exit signal is limited by CSIGNAL as with legacy clone() and that\nthe signal is valid. With this we don\u0027t get the legacy clone behavior\nwere an invalid signal could be handed down and would only be detected\nand ignored in do_notify_parent(). Users of clone3() will now get a\nproper error when they pass an invalid exit signal. Note, that this is\nnot user-visible behavior since no kernel with clone3() has been\nreleased yet.\n\nThe following program will cause a splat on a non-fixed clone3() version\nand will fail correctly on a fixed version:\n\n #define _GNU_SOURCE\n #include \u003clinux/sched.h\u003e\n #include \u003clinux/types.h\u003e\n #include \u003csched.h\u003e\n #include \u003cstdio.h\u003e\n #include \u003cstdlib.h\u003e\n #include \u003csys/syscall.h\u003e\n #include \u003csys/wait.h\u003e\n #include \u003cunistd.h\u003e\n\n int main(int argc, char *argv[])\n {\n pid_t pid \u003d -1;\n struct clone_args args \u003d {0};\n args.exit_signal \u003d -1;\n\n pid \u003d syscall(__NR_clone3, \u0026args, sizeof(struct clone_args));\n if (pid \u003c 0)\n exit(EXIT_FAILURE);\n\n if (pid \u003d\u003d 0)\n exit(EXIT_SUCCESS);\n\n wait(NULL);\n\n exit(EXIT_SUCCESS);\n }\n\nFixes: 7f192e3cd316 (\"fork: add clone3\")\nReported-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nSuggested-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nSuggested-by: Dmitry V. Levin \u003cldv@altlinux.org\u003e\nSigned-off-by: Eugene Syromiatnikov \u003cesyr@redhat.com\u003e\nLink: https://lore.kernel.org/r/4b38fa4ce420b119a4c6345f42fe3cec2de9b0b5.1568223594.git.esyr@redhat.com\n[christian.brauner@ubuntu.com: simplify check and rework commit message]\nSigned-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n" }, { "commit": "53936b5bf35e140ae27e4bbf0447a61063f400da", "tree": "2fdf79317c34f75861183dfa54b340bd91b6cdd2", "parents": [ "13a17cc0526f08d1df9507f7484176371cd263a0" ], "author": { "name": "Thomas Huth", "email": "thuth@redhat.com", "time": "Thu Sep 12 13:54:38 2019 +0200" }, "committer": { "name": "Christian Borntraeger", "email": "borntraeger@de.ibm.com", "time": "Thu Sep 12 14:12:21 2019 +0200" }, "message": "KVM: s390: Do not leak kernel stack data in the KVM_S390_INTERRUPT ioctl\n\nWhen the userspace program runs the KVM_S390_INTERRUPT ioctl to inject\nan interrupt, we convert them from the legacy struct kvm_s390_interrupt\nto the new struct kvm_s390_irq via the s390int_to_s390irq() function.\nHowever, this function does not take care of all types of interrupts\nthat we can inject into the guest later (see do_inject_vcpu()). Since we\ndo not clear out the s390irq values before calling s390int_to_s390irq(),\nthere is a chance that we copy random data from the kernel stack which\ncould be leaked to the userspace later.\n\nSpecifically, the problem exists with the KVM_S390_INT_PFAULT_INIT\ninterrupt: s390int_to_s390irq() does not handle it, and the function\n__inject_pfault_init() later copies irq-\u003eu.ext which contains the\nrandom kernel stack data. This data can then be leaked either to\nthe guest memory in __deliver_pfault_init(), or the userspace might\nretrieve it directly with the KVM_S390_GET_IRQ_STATE ioctl.\n\nFix it by handling that interrupt type in s390int_to_s390irq(), too,\nand by making sure that the s390irq struct is properly pre-initialized.\nAnd while we\u0027re at it, make sure that s390int_to_s390irq() now\ndirectly returns -EINVAL for unknown interrupt types, so that we\nimmediately get a proper error code in case we add more interrupt\ntypes to do_inject_vcpu() without updating s390int_to_s390irq()\nsometime in the future.\n\nCc: stable@vger.kernel.org\nReviewed-by: David Hildenbrand \u003cdavid@redhat.com\u003e\nReviewed-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\nReviewed-by: Janosch Frank \u003cfrankja@linux.ibm.com\u003e\nSigned-off-by: Thomas Huth \u003cthuth@redhat.com\u003e\nLink: https://lore.kernel.org/kvm/20190912115438.25761-1-thuth@redhat.com\nSigned-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\n" }, { "commit": "b456d72412ca8797234449c25815e82f4e1426c0", "tree": "5b2d0bf2ee4e55abc6cc232772729ee3fcfb3ba1", "parents": [ "f39b683d35dfa93a58f1b400a8ec0ff81296b37c" ], "author": { "name": "Christophe JAILLET", "email": "christophe.jaillet@wanadoo.fr", "time": "Wed Sep 11 18:02:39 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 12 12:55:28 2019 +0100" }, "message": "sctp: Fix the link time qualifier of \u0027sctp_ctrlsock_exit()\u0027\n\nThe \u0027.exit\u0027 functions from \u0027pernet_operations\u0027 structure should be marked\nas __net_exit, not __net_init.\n\nFixes: 8e2d61e0aed2 (\"sctp: fix race on protocol/netns initialization\")\nSigned-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3dc8dcb02fdba3370aec0696727e6adfe8033aa4", "tree": "554677270e03bbae10bfeff9347d85057f9e1fcf", "parents": [ "ef92bfdaa7123bc8bdaed3f0f6e008500dd555da", "1709510221c57fd566479c228434ff9edd6435be" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Thu Sep 12 13:46:20 2019 +0200" }, "committer": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Thu Sep 12 13:46:20 2019 +0200" }, "message": "Merge tag \u0027qcom-drivers-for-5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux into arm/drivers\n\nQualcomm ARM Based Driver Updates for v5.4\n\n* Add AOSS QMP support\n* Various fixups for Qualcomm SCM\n* Add socinfo driver\n* Add SoC serial number attribute and associated APIs\n* Add SM8150 and SC7180 support in Qualcomm SCM\n* Fixup max processor count in SMEM\n\n* tag \u0027qcom-drivers-for-5.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux:\n soc: qcom: aoss: Add AOSS QMP support\n dt-bindings: soc: qcom: aoss: Add SM8150 and SC7180 support\n dt-bindings: firmware: scm: Add SM8150 and SC7180 support\n dt-bindings: firmware: scm: re-order compatible list\n soc: qcom: smem: Update max processor count\n soc: qcom: socinfo: Annotate switch cases with fall through\n soc: qcom: Extend AOSS QMP driver to support resources that are used to wake up the SoC.\n soc: qcom: socinfo: Expose image information\n soc: qcom: socinfo: Expose custom attributes\n soc: qcom: Add socinfo driver\n base: soc: Export soc_device_register/unregister APIs\n base: soc: Add serial_number attribute to soc\n firmware: qcom_scm: Cleanup code in qcom_scm_assign_mem()\n firmware: qcom_scm: Fix some typos in docs and printks\n firmware: qcom_scm: Use proper types for dma mappings\n" }, { "commit": "f39b683d35dfa93a58f1b400a8ec0ff81296b37c", "tree": "df54cf71c691c9941ad3b29cda254a9ebf7ad273", "parents": [ "a21b7f0cff1906a93a0130b74713b15a0b36481d" ], "author": { "name": "Steffen Klassert", "email": "steffen.klassert@secunet.com", "time": "Thu Sep 12 13:01:44 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 12 12:43:14 2019 +0100" }, "message": "ixgbe: Fix secpath usage for IPsec TX offload.\n\nThe ixgbe driver currently does IPsec TX offloading\nbased on an existing secpath. However, the secpath\ncan also come from the RX side, in this case it is\nmisinterpreted for TX offload and the packets are\ndropped with a \"bad sa_idx\" error. Fix this by using\nthe xfrm_offload() function to test for TX offload.\n\nFixes: 592594704761 (\"ixgbe: process the Tx ipsec offload\")\nReported-by: Michael Marley \u003cmichael@michaelmarley.com\u003e\nSigned-off-by: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "18dfa7117a3f379862dcd3f67cadd678013bb9dd", "tree": "bc89eb54b42ee5e462a70a3d55d9cfce20c5d3e2", "parents": [ "410f954cb1d1c79ae485dd83a175f21954fd87cd" ], "author": { "name": "Filipe Manana", "email": "fdmanana@suse.com", "time": "Wed Sep 11 17:42:00 2019 +0100" }, "committer": { "name": "David Sterba", "email": "dsterba@suse.com", "time": "Thu Sep 12 13:37:25 2019 +0200" }, "message": "Btrfs: fix unwritten extent buffers and hangs on future writeback attempts\n\nThe lock_extent_buffer_io() returns 1 to the caller to tell it everything\nwent fine and the callers needs to start writeback for the extent buffer\n(submit a bio, etc), 0 to tell the caller everything went fine but it does\nnot need to start writeback for the extent buffer, and a negative value if\nsome error happened.\n\nWhen it\u0027s about to return 1 it tries to lock all pages, and if a try lock\non a page fails, and we didn\u0027t flush any existing bio in our \"epd\", it\ncalls flush_write_bio(epd) and overwrites the return value of 1 to 0 or\nan error. The page might have been locked elsewhere, not with the goal\nof starting writeback of the extent buffer, and even by some code other\nthan btrfs, like page migration for example, so it does not mean the\nwriteback of the extent buffer was already started by some other task,\nso returning a 0 tells the caller (btree_write_cache_pages()) to not\nstart writeback for the extent buffer. Note that epd might currently have\neither no bio, so flush_write_bio() returns 0 (success) or it might have\na bio for another extent buffer with a lower index (logical address).\n\nSince we return 0 with the EXTENT_BUFFER_WRITEBACK bit set on the\nextent buffer and writeback is never started for the extent buffer,\nfuture attempts to writeback the extent buffer will hang forever waiting\non that bit to be cleared, since it can only be cleared after writeback\ncompletes. Such hang is reported with a trace like the following:\n\n [49887.347053] INFO: task btrfs-transacti:1752 blocked for more than 122 seconds.\n [49887.347059] Not tainted 5.2.13-gentoo #2\n [49887.347060] \"echo 0 \u003e /proc/sys/kernel/hung_task_timeout_secs\" disables this message.\n [49887.347062] btrfs-transacti D 0 1752 2 0x80004000\n [49887.347064] Call Trace:\n [49887.347069] ? __schedule+0x265/0x830\n [49887.347071] ? bit_wait+0x50/0x50\n [49887.347072] ? bit_wait+0x50/0x50\n [49887.347074] schedule+0x24/0x90\n [49887.347075] io_schedule+0x3c/0x60\n [49887.347077] bit_wait_io+0x8/0x50\n [49887.347079] __wait_on_bit+0x6c/0x80\n [49887.347081] ? __lock_release.isra.29+0x155/0x2d0\n [49887.347083] out_of_line_wait_on_bit+0x7b/0x80\n [49887.347084] ? var_wake_function+0x20/0x20\n [49887.347087] lock_extent_buffer_for_io+0x28c/0x390\n [49887.347089] btree_write_cache_pages+0x18e/0x340\n [49887.347091] do_writepages+0x29/0xb0\n [49887.347093] ? kmem_cache_free+0x132/0x160\n [49887.347095] ? convert_extent_bit+0x544/0x680\n [49887.347097] filemap_fdatawrite_range+0x70/0x90\n [49887.347099] btrfs_write_marked_extents+0x53/0x120\n [49887.347100] btrfs_write_and_wait_transaction.isra.4+0x38/0xa0\n [49887.347102] btrfs_commit_transaction+0x6bb/0x990\n [49887.347103] ? start_transaction+0x33e/0x500\n [49887.347105] transaction_kthread+0x139/0x15c\n\nSo fix this by not overwriting the return value (ret) with the result\nfrom flush_write_bio(). We also need to clear the EXTENT_BUFFER_WRITEBACK\nbit in case flush_write_bio() returns an error, otherwise it will hang\nany future attempts to writeback the extent buffer, and undo all work\ndone before (set back EXTENT_BUFFER_DIRTY, etc).\n\nThis is a regression introduced in the 5.2 kernel.\n\nFixes: 2e3c25136adfb (\"btrfs: extent_io: add proper error handling to lock_extent_buffer_for_io()\")\nFixes: f4340622e0226 (\"btrfs: extent_io: Move the BUG_ON() in flush_write_bio() one level up\")\nReported-by: Zdenek Sojka \u003czsojka@seznam.cz\u003e\nLink: https://lore.kernel.org/linux-btrfs/GpO.2yos.3WGDOLpx6t%7D.1TUDYM@seznam.cz/T/#u\nReported-by: Stefan Priebe - Profihost AG \u003cs.priebe@profihost.ag\u003e\nLink: https://lore.kernel.org/linux-btrfs/5c4688ac-10a7-fb07-70e8-c5d31a3fbb38@profihost.ag/T/#t\nReported-by: Drazen Kacar \u003cdrazen.kacar@oradian.com\u003e\nLink: https://lore.kernel.org/linux-btrfs/DB8PR03MB562876ECE2319B3E579590F799C80@DB8PR03MB5628.eurprd03.prod.outlook.com/\nBugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204377\nSigned-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\nSigned-off-by: David Sterba \u003cdsterba@suse.com\u003e\n" }, { "commit": "410f954cb1d1c79ae485dd83a175f21954fd87cd", "tree": "3bc8a3210514d677064ee3c8c0522a7565ca6194", "parents": [ "07301df7d2fc220d3de5f7ad804dcb941400cb00" ], "author": { "name": "Filipe Manana", "email": "fdmanana@suse.com", "time": "Tue Sep 10 15:26:49 2019 +0100" }, "committer": { "name": "David Sterba", "email": "dsterba@suse.com", "time": "Thu Sep 12 13:37:19 2019 +0200" }, "message": "Btrfs: fix assertion failure during fsync and use of stale transaction\n\nSometimes when fsync\u0027ing a file we need to log that other inodes exist and\nwhen we need to do that we acquire a reference on the inodes and then drop\nthat reference using iput() after logging them.\n\nThat generally is not a problem except if we end up doing the final iput()\n(dropping the last reference) on the inode and that inode has a link count\nof 0, which can happen in a very short time window if the logging path\ngets a reference on the inode while it\u0027s being unlinked.\n\nIn that case we end up getting the eviction callback, btrfs_evict_inode(),\ninvoked through the iput() call chain which needs to drop all of the\ninode\u0027s items from its subvolume btree, and in order to do that, it needs\nto join a transaction at the helper function evict_refill_and_join().\nHowever because the task previously started a transaction at the fsync\nhandler, btrfs_sync_file(), it has current-\u003ejournal_info already pointing\nto a transaction handle and therefore evict_refill_and_join() will get\nthat transaction handle from btrfs_join_transaction(). From this point on,\ntwo different problems can happen:\n\n1) evict_refill_and_join() will often change the transaction handle\u0027s\n block reserve (-\u003eblock_rsv) and set its -\u003ebytes_reserved field to a\n value greater than 0. If evict_refill_and_join() never commits the\n transaction, the eviction handler ends up decreasing the reference\n count (-\u003euse_count) of the transaction handle through the call to\n btrfs_end_transaction(), and after that point we have a transaction\n handle with a NULL -\u003eblock_rsv (which is the value prior to the\n transaction join from evict_refill_and_join()) and a -\u003ebytes_reserved\n value greater than 0. If after the eviction/iput completes the inode\n logging path hits an error or it decides that it must fallback to a\n transaction commit, the btrfs fsync handle, btrfs_sync_file(), gets a\n non-zero value from btrfs_log_dentry_safe(), and because of that\n non-zero value it tries to commit the transaction using a handle with\n a NULL -\u003eblock_rsv and a non-zero -\u003ebytes_reserved value. This makes\n the transaction commit hit an assertion failure at\n btrfs_trans_release_metadata() because -\u003ebytes_reserved is not zero but\n the -\u003eblock_rsv is NULL. The produced stack trace for that is like the\n following:\n\n [192922.917158] assertion failed: !trans-\u003ebytes_reserved, file: fs/btrfs/transaction.c, line: 816\n [192922.917553] ------------[ cut here ]------------\n [192922.917922] kernel BUG at fs/btrfs/ctree.h:3532!\n [192922.918310] invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC PTI\n [192922.918666] CPU: 2 PID: 883 Comm: fsstress Tainted: G W 5.1.4-btrfs-next-47 #1\n [192922.919035] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.2-0-gf9626ccb91-prebuilt.qemu-project.org 04/01/2014\n [192922.919801] RIP: 0010:assfail.constprop.25+0x18/0x1a [btrfs]\n (...)\n [192922.920925] RSP: 0018:ffffaebdc8a27da8 EFLAGS: 00010286\n [192922.921315] RAX: 0000000000000051 RBX: ffff95c9c16a41c0 RCX: 0000000000000000\n [192922.921692] RDX: 0000000000000000 RSI: ffff95cab6b16838 RDI: ffff95cab6b16838\n [192922.922066] RBP: ffff95c9c16a41c0 R08: 0000000000000000 R09: 0000000000000000\n [192922.922442] R10: ffffaebdc8a27e70 R11: 0000000000000000 R12: ffff95ca731a0980\n [192922.922820] R13: 0000000000000000 R14: ffff95ca84c73338 R15: ffff95ca731a0ea8\n [192922.923200] FS: 00007f337eda4e80(0000) GS:ffff95cab6b00000(0000) knlGS:0000000000000000\n [192922.923579] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [192922.923948] CR2: 00007f337edad000 CR3: 00000001e00f6002 CR4: 00000000003606e0\n [192922.924329] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [192922.924711] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [192922.925105] Call Trace:\n [192922.925505] btrfs_trans_release_metadata+0x10c/0x170 [btrfs]\n [192922.925911] btrfs_commit_transaction+0x3e/0xaf0 [btrfs]\n [192922.926324] btrfs_sync_file+0x44c/0x490 [btrfs]\n [192922.926731] do_fsync+0x38/0x60\n [192922.927138] __x64_sys_fdatasync+0x13/0x20\n [192922.927543] do_syscall_64+0x60/0x1c0\n [192922.927939] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n (...)\n [192922.934077] ---[ end trace f00808b12068168f ]---\n\n2) If evict_refill_and_join() decides to commit the transaction, it will\n be able to do it, since the nested transaction join only increments the\n transaction handle\u0027s -\u003euse_count reference counter and it does not\n prevent the transaction from getting committed. This means that after\n eviction completes, the fsync logging path will be using a transaction\n handle that refers to an already committed transaction. What happens\n when using such a stale transaction can be unpredictable, we are at\n least having a use-after-free on the transaction handle itself, since\n the transaction commit will call kmem_cache_free() against the handle\n regardless of its -\u003euse_count value, or we can end up silently losing\n all the updates to the log tree after that iput() in the logging path,\n or using a transaction handle that in the meanwhile was allocated to\n another task for a new transaction, etc, pretty much unpredictable\n what can happen.\n\nIn order to fix both of them, instead of using iput() during logging, use\nbtrfs_add_delayed_iput(), so that the logging path of fsync never drops\nthe last reference on an inode, that step is offloaded to a safe context\n(usually the cleaner kthread).\n\nThe assertion failure issue was sporadically triggered by the test case\ngeneric/475 from fstests, which loads the dm error target while fsstress\nis running, which lead to fsync failing while logging inodes with -EIO\nerrors and then trying later to commit the transaction, triggering the\nassertion failure.\n\nCC: stable@vger.kernel.org # 4.4+\nReviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nSigned-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\nSigned-off-by: David Sterba \u003cdsterba@suse.com\u003e\n" }, { "commit": "13a17cc0526f08d1df9507f7484176371cd263a0", "tree": "c3d7bf2eda08fb3a614ae05c4731f4ac444641d8", "parents": [ "609488bc979f99f805f34e9a32c1e3b71179d10b" ], "author": { "name": "Igor Mammedov", "email": "imammedo@redhat.com", "time": "Wed Sep 11 03:52:18 2019 -0400" }, "committer": { "name": "Christian Borntraeger", "email": "borntraeger@de.ibm.com", "time": "Thu Sep 12 13:09:17 2019 +0200" }, "message": "KVM: s390: kvm_s390_vm_start_migration: check dirty_bitmap before using it as target for memset()\n\nIf userspace doesn\u0027t set KVM_MEM_LOG_DIRTY_PAGES on memslot before calling\nkvm_s390_vm_start_migration(), kernel will oops with:\n\n Unable to handle kernel pointer dereference in virtual kernel address space\n Failing address: 0000000000000000 TEID: 0000000000000483\n Fault in home space mode while using kernel ASCE.\n AS:0000000002a2000b R2:00000001bff8c00b R3:00000001bff88007 S:00000001bff91000 P:000000000000003d\n Oops: 0004 ilc:2 [#1] SMP\n ...\n Call Trace:\n ([\u003c001fffff804ec552\u003e] kvm_s390_vm_set_attr+0x347a/0x3828 [kvm])\n [\u003c001fffff804ecfc0\u003e] kvm_arch_vm_ioctl+0x6c0/0x1998 [kvm]\n [\u003c001fffff804b67e4\u003e] kvm_vm_ioctl+0x51c/0x11a8 [kvm]\n [\u003c00000000008ba572\u003e] do_vfs_ioctl+0x1d2/0xe58\n [\u003c00000000008bb284\u003e] ksys_ioctl+0x8c/0xb8\n [\u003c00000000008bb2e2\u003e] sys_ioctl+0x32/0x40\n [\u003c000000000175552c\u003e] system_call+0x2b8/0x2d8\n INFO: lockdep is turned off.\n Last Breaking-Event-Address:\n [\u003c0000000000dbaf60\u003e] __memset+0xc/0xa0\n\ndue to ms-\u003edirty_bitmap being NULL, which might crash the host.\n\nMake sure that ms-\u003edirty_bitmap is set before using it or\nreturn -EINVAL otherwise.\n\nCc: \u003cstable@vger.kernel.org\u003e\nFixes: afdad61615cc (\"KVM: s390: Fix storage attributes migration with memory slots\")\nSigned-off-by: Igor Mammedov \u003cimammedo@redhat.com\u003e\nLink: https://lore.kernel.org/kvm/20190911075218.29153-1-imammedo@redhat.com/\nReviewed-by: David Hildenbrand \u003cdavid@redhat.com\u003e\nReviewed-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\nReviewed-by: Claudio Imbrenda \u003cimbrenda@linux.ibm.com\u003e\nReviewed-by: Cornelia Huck \u003ccohuck@redhat.com\u003e\nReviewed-by: Janosch Frank \u003cfrankja@linux.ibm.com\u003e\nSigned-off-by: Janosch Frank \u003cfrankja@linux.ibm.com\u003e\nSigned-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\n" }, { "commit": "a21b7f0cff1906a93a0130b74713b15a0b36481d", "tree": "4a1f5603ba981ff7d39988495841a8a2b1be91be", "parents": [ "10cc514f451a0f239aa34f91bc9dc954a9397840" ], "author": { "name": "Navid Emamdoost", "email": "navid.emamdoost@gmail.com", "time": "Wed Sep 11 10:09:02 2019 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 12 11:58:44 2019 +0100" }, "message": "net: qrtr: fix memort leak in qrtr_tun_write_iter\n\nIn qrtr_tun_write_iter the allocated kbuf should be release in case of\nerror or success return.\n\nv2 Update: Thanks to David Miller for pointing out the release on success\npath as well.\n\nSigned-off-by: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "10cc514f451a0f239aa34f91bc9dc954a9397840", "tree": "d0f714afc565abac165983c4870e62b6beda2f3d", "parents": [ "d23dbc479a8e813db4161a695d67da0e36557846" ], "author": { "name": "Subash Abhinov Kasiviswanathan", "email": "subashab@codeaurora.org", "time": "Tue Sep 10 14:02:57 2019 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 12 11:55:34 2019 +0100" }, "message": "net: Fix null de-reference of device refcount\n\nIn event of failure during register_netdevice, free_netdev is\ninvoked immediately. free_netdev assumes that all the netdevice\nrefcounts have been dropped prior to it being called and as a\nresult frees and clears out the refcount pointer.\n\nHowever, this is not necessarily true as some of the operations\nin the NETDEV_UNREGISTER notifier handlers queue RCU callbacks for\ninvocation after a grace period. The IPv4 callback in_dev_rcu_put\ntries to access the refcount after free_netdev is called which\nleads to a null de-reference-\n\n44837.761523: \u003c6\u003e Unable to handle kernel paging request at\n virtual address 0000004a88287000\n44837.761651: \u003c2\u003e pc : in_dev_finish_destroy+0x4c/0xc8\n44837.761654: \u003c2\u003e lr : in_dev_finish_destroy+0x2c/0xc8\n44837.762393: \u003c2\u003e Call trace:\n44837.762398: \u003c2\u003e in_dev_finish_destroy+0x4c/0xc8\n44837.762404: \u003c2\u003e in_dev_rcu_put+0x24/0x30\n44837.762412: \u003c2\u003e rcu_nocb_kthread+0x43c/0x468\n44837.762418: \u003c2\u003e kthread+0x118/0x128\n44837.762424: \u003c2\u003e ret_from_fork+0x10/0x1c\n\nFix this by waiting for the completion of the call_rcu() in\ncase of register_netdevice errors.\n\nFixes: 93ee31f14f6f (\"[NET]: Fix free_netdev on register_netdev failure.\")\nCc: Sean Tranchetti \u003cstranche@codeaurora.org\u003e\nSigned-off-by: Subash Abhinov Kasiviswanathan \u003csubashab@codeaurora.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d23dbc479a8e813db4161a695d67da0e36557846", "tree": "874b6982de49a551bb6bc0a591ce754c4b3d2bc1", "parents": [ "77f22f92dff8e7b45c7786a430626d38071d4670" ], "author": { "name": "Christophe JAILLET", "email": "christophe.jaillet@wanadoo.fr", "time": "Tue Sep 10 13:29:59 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 12 11:20:33 2019 +0100" }, "message": "ipv6: Fix the link time qualifier of \u0027ping_v6_proc_exit_net()\u0027\n\nThe \u0027.exit\u0027 functions from \u0027pernet_operations\u0027 structure should be marked\nas __net_exit, not __net_init.\n\nFixes: d862e5461423 (\"net: ipv6: Implement /proc/net/icmp6.\")\nSigned-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "77f22f92dff8e7b45c7786a430626d38071d4670", "tree": "52300e28bf190304d83ee1a53dcf439a04c106af", "parents": [ "13d5231cc062fd037b2464d3595351d328d1cdf7" ], "author": { "name": "Yang Yingliang", "email": "yangyingliang@huawei.com", "time": "Tue Sep 10 18:56:57 2019 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 12 11:17:26 2019 +0100" }, "message": "tun: fix use-after-free when register netdev failed\n\nI got a UAF repport in tun driver when doing fuzzy test:\n\n[ 466.269490] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n[ 466.271792] BUG: KASAN: use-after-free in tun_chr_read_iter+0x2ca/0x2d0\n[ 466.271806] Read of size 8 at addr ffff888372139250 by task tun-test/2699\n[ 466.271810]\n[ 466.271824] CPU: 1 PID: 2699 Comm: tun-test Not tainted 5.3.0-rc1-00001-g5a9433db2614-dirty #427\n[ 466.271833] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014\n[ 466.271838] Call Trace:\n[ 466.271858] dump_stack+0xca/0x13e\n[ 466.271871] ? tun_chr_read_iter+0x2ca/0x2d0\n[ 466.271890] print_address_description+0x79/0x440\n[ 466.271906] ? vprintk_func+0x5e/0xf0\n[ 466.271920] ? tun_chr_read_iter+0x2ca/0x2d0\n[ 466.271935] __kasan_report+0x15c/0x1df\n[ 466.271958] ? tun_chr_read_iter+0x2ca/0x2d0\n[ 466.271976] kasan_report+0xe/0x20\n[ 466.271987] tun_chr_read_iter+0x2ca/0x2d0\n[ 466.272013] do_iter_readv_writev+0x4b7/0x740\n[ 466.272032] ? default_llseek+0x2d0/0x2d0\n[ 466.272072] do_iter_read+0x1c5/0x5e0\n[ 466.272110] vfs_readv+0x108/0x180\n[ 466.299007] ? compat_rw_copy_check_uvector+0x440/0x440\n[ 466.299020] ? fsnotify+0x888/0xd50\n[ 466.299040] ? __fsnotify_parent+0xd0/0x350\n[ 466.299064] ? fsnotify_first_mark+0x1e0/0x1e0\n[ 466.304548] ? vfs_write+0x264/0x510\n[ 466.304569] ? ksys_write+0x101/0x210\n[ 466.304591] ? do_preadv+0x116/0x1a0\n[ 466.304609] do_preadv+0x116/0x1a0\n[ 466.309829] do_syscall_64+0xc8/0x600\n[ 466.309849] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n[ 466.309861] RIP: 0033:0x4560f9\n[ 466.309875] Code: 00 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 \u003c48\u003e 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48\n[ 466.309889] RSP: 002b:00007ffffa5166e8 EFLAGS: 00000206 ORIG_RAX: 0000000000000127\n[ 466.322992] RAX: ffffffffffffffda RBX: 0000000000400460 RCX: 00000000004560f9\n[ 466.322999] RDX: 0000000000000003 RSI: 00000000200008c0 RDI: 0000000000000003\n[ 466.323007] RBP: 00007ffffa516700 R08: 0000000000000004 R09: 0000000000000000\n[ 466.323014] R10: 0000000000000000 R11: 0000000000000206 R12: 000000000040cb10\n[ 466.323021] R13: 0000000000000000 R14: 00000000006d7018 R15: 0000000000000000\n[ 466.323057]\n[ 466.323064] Allocated by task 2605:\n[ 466.335165] save_stack+0x19/0x80\n[ 466.336240] __kasan_kmalloc.constprop.8+0xa0/0xd0\n[ 466.337755] kmem_cache_alloc+0xe8/0x320\n[ 466.339050] getname_flags+0xca/0x560\n[ 466.340229] user_path_at_empty+0x2c/0x50\n[ 466.341508] vfs_statx+0xe6/0x190\n[ 466.342619] __do_sys_newstat+0x81/0x100\n[ 466.343908] do_syscall_64+0xc8/0x600\n[ 466.345303] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n[ 466.347034]\n[ 466.347517] Freed by task 2605:\n[ 466.348471] save_stack+0x19/0x80\n[ 466.349476] __kasan_slab_free+0x12e/0x180\n[ 466.350726] kmem_cache_free+0xc8/0x430\n[ 466.351874] putname+0xe2/0x120\n[ 466.352921] filename_lookup+0x257/0x3e0\n[ 466.354319] vfs_statx+0xe6/0x190\n[ 466.355498] __do_sys_newstat+0x81/0x100\n[ 466.356889] do_syscall_64+0xc8/0x600\n[ 466.358037] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n[ 466.359567]\n[ 466.360050] The buggy address belongs to the object at ffff888372139100\n[ 466.360050] which belongs to the cache names_cache of size 4096\n[ 466.363735] The buggy address is located 336 bytes inside of\n[ 466.363735] 4096-byte region [ffff888372139100, ffff88837213a100)\n[ 466.367179] The buggy address belongs to the page:\n[ 466.368604] page:ffffea000dc84e00 refcount:1 mapcount:0 mapping:ffff8883df1b4f00 index:0x0 compound_mapcount: 0\n[ 466.371582] flags: 0x2fffff80010200(slab|head)\n[ 466.372910] raw: 002fffff80010200 dead000000000100 dead000000000122 ffff8883df1b4f00\n[ 466.375209] raw: 0000000000000000 0000000000070007 00000001ffffffff 0000000000000000\n[ 466.377778] page dumped because: kasan: bad access detected\n[ 466.379730]\n[ 466.380288] Memory state around the buggy address:\n[ 466.381844] ffff888372139100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n[ 466.384009] ffff888372139180: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n[ 466.386131] \u003effff888372139200: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n[ 466.388257] ^\n[ 466.390234] ffff888372139280: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n[ 466.392512] ffff888372139300: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n[ 466.394667] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\ntun_chr_read_iter() accessed the memory which freed by free_netdev()\ncalled by tun_set_iff():\n\n CPUA CPUB\n tun_set_iff()\n alloc_netdev_mqs()\n tun_attach()\n tun_chr_read_iter()\n tun_get()\n tun_do_read()\n tun_ring_recv()\n register_netdevice() \u003c-- inject error\n goto err_detach\n tun_detach_all() \u003c-- set RCV_SHUTDOWN\n free_netdev() \u003c-- called from\n err_free_dev path\n netdev_freemem() \u003c-- free the memory\n without check refcount\n (In this path, the refcount cannot prevent\n freeing the memory of dev, and the memory\n will be used by dev_put() called by\n tun_chr_read_iter() on CPUB.)\n (Break from tun_ring_recv(),\n because RCV_SHUTDOWN is set)\n tun_put()\n dev_put() \u003c-- use the memory\n freed by netdev_freemem()\n\nPut the publishing of tfile-\u003etun after register_netdevice(),\nso tun_get() won\u0027t get the tun pointer that freed by\nerr_detach path if register_netdevice() failed.\n\nFixes: eb0fb363f920 (\"tuntap: attach queue 0 before registering netdevice\")\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nSuggested-by: Jason Wang \u003cjasowang@redhat.com\u003e\nSigned-off-by: Yang Yingliang \u003cyangyingliang@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ad32b4800c2b59a303e337ec1f6c2f122418cebe", "tree": "1f46d5c2a8a2ee947646e86e4917f239e84f311b", "parents": [ "6dcf6a4eb95a8a78c181cd27132f41aca36aeb94", "060423bfdee3f8bc6e2c1bac97de24d5415e2bc4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 11:07:31 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 11:07:31 2019 +0100" }, "message": "Merge tag \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost\n\nPull virtio fixes from Michael Tsirkin:\n \"Last minute bugfixes.\n\n A couple of security things.\n\n And an error handling bugfix that is never encountered by most people,\n but that also makes it kind of safe to push at the last minute, and it\n helps push the fix to stable a bit sooner\"\n\n* tag \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:\n vhost: make sure log_num \u003c in_num\n vhost: block speculation of translated descriptors\n virtio_ring: fix unmap of indirect descriptors\n" }, { "commit": "6dcf6a4eb95a8a78c181cd27132f41aca36aeb94", "tree": "6f9837d576eb010e3aac64c95946669b87cabea5", "parents": [ "95779fe8506d4f750f1e66e958b6b3c405182d7a", "310aa0a25b338b3100c94880c9a69bec8ce8c3ae" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 11:04:50 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 11:04:50 2019 +0100" }, "message": "Merge branch \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf fix from Ingo Molnar:\n \"Fix an initialization bug in the hw-breakpoints, which triggered on\n the ARM platform\"\n\n* \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n perf/hw_breakpoint: Fix arch_hw_breakpoint use-before-initialization\n" }, { "commit": "95779fe8506d4f750f1e66e958b6b3c405182d7a", "tree": "db674fe9706bb1f54d0147509d17923bc634b7b1", "parents": [ "840ce8f8073edb3ff3d2c2c7a6ef211f4176961c", "eddf3e9c7c7e4d0707c68d1bb22cc6ec8aef7d4a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 11:02:00 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 11:02:00 2019 +0100" }, "message": "Merge branch \u0027irq-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull irq fix from Ingo Molnar:\n \"Fix a race in the IRQ resend mechanism, which can result in a NULL\n dereference crash\"\n\n* \u0027irq-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n genirq: Prevent NULL pointer dereference in resend_irqs()\n" }, { "commit": "840ce8f8073edb3ff3d2c2c7a6ef211f4176961c", "tree": "fe95baa99a3ea93fe2c073fee333d28fb03d1c13", "parents": [ "9c09f623487178867d8000af337585b84b1f53d9", "c1432423a16825ef94e019808ed7330ee33ffc69" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 10:58:47 2019 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 12 10:58:47 2019 +0100" }, "message": "Merge tag \u0027pinctrl-v5.3-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl\n\nPull pin control fix from Linus Walleij:\n \"Hopefully last pin control fix: a single patch for some Aspeed\n problems. The BMCs are much happier now\"\n\n* tag \u0027pinctrl-v5.3-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:\n pinctrl: aspeed: Fix spurious mux failures on the AST2500\n" }, { "commit": "11c43bb022b373d2fdb84950ebf1911362f3f010", "tree": "d46e79d87bb01d1c2f2f565c10a468c850058c92", "parents": [ "c34a024e4ee087b6d9e534160990082c3d9e7103" ], "author": { "name": "Dmitry Torokhov", "email": "dmitry.torokhov@gmail.com", "time": "Wed Sep 11 00:52:05 2019 -0700" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Thu Sep 12 10:29:17 2019 +0100" }, "message": "gpiolib: of: add a fallback for wlf,reset GPIO name\n\nThe old Arizona binding did not use -gpio or -gpios suffix, so\ndevm_gpiod_get() does not work for it. As it is the one of a few users\nof devm_gpiod_get_from_of_node() API that I want to remove, I\u0027d rather\nhave a small quirk in the gpiolib OF handler, and switch Arizona\ndriver to devm_gpiod_get().\n\nSigned-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nLink: https://lore.kernel.org/r/20190911075215.78047-2-dmitry.torokhov@gmail.com\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n" } ], "next": "c34a024e4ee087b6d9e534160990082c3d9e7103" }