)]}' { "log": [ { "commit": "7928b2cbe55b2a410a0f5c1f154610059c57b1b2", "tree": "67e4a1a4fd57fbd392c3add4e1d751b9248d97b7", "parents": [ "7a163b2195cda0cddf47b5caf14a7229d4e2bea3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 15:04:29 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 15:04:29 2018 -0800" }, "message": "Linux 4.16-rc1\n" }, { "commit": "7a163b2195cda0cddf47b5caf14a7229d4e2bea3", "tree": "5867e113f0232679d5125ca7d42ae4558dd0d554", "parents": [ "a9a08845e9acbd224e4ee466f5c1275ed50054e8" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Feb 01 15:13:18 2018 -0500" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 14:37:22 2018 -0800" }, "message": "unify {de,}mangle_poll(), get rid of kernel-side POLL...\n\nexcept, again, POLLFREE and POLL_BUSY_LOOP.\n\nWith this, we finally get to the promised end result:\n\n - POLL{IN,OUT,...} are plain integers and *not* in __poll_t, so any\n stray instances of -\u003epoll() still using those will be caught by\n sparse.\n\n - eventpoll.c and select.c warning-free wrt __poll_t\n\n - no more kernel-side definitions of POLL... - userland ones are\n visible through the entire kernel (and used pretty much only for\n mangle/demangle)\n\n - same behavior as after the first series (i.e. sparc et.al. epoll(2)\n working correctly).\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a9a08845e9acbd224e4ee466f5c1275ed50054e8", "tree": "415d6e6a82e001c65e6b161539411f54ba5fe8ce", "parents": [ "ee5daa1361fceb6f482c005bcc9ba8d01b92ea5c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 14:34:03 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 14:34:03 2018 -0800" }, "message": "vfs: do bulk POLL* -\u003e EPOLL* replacement\n\nThis is the mindless scripted replacement of kernel use of POLL*\nvariables as described by Al, done by this script:\n\n for V in IN OUT PRI ERR RDNORM RDBAND WRNORM WRBAND HUP RDHUP NVAL MSG; do\n L\u003d`git grep -l -w POLL$V | grep -v \u0027^t\u0027 | grep -v /um/ | grep -v \u0027^sa\u0027 | grep -v \u0027/poll.h$\u0027|grep -v \u0027^D\u0027`\n for f in $L; do sed -i \"-es/^\\([^\\\"]*\\)\\(\\\u003cPOLL$V\\\u003e\\)/\\\\1E\\\\2/\" $f; done\n done\n\nwith de-mangling cleanups yet to come.\n\nNOTE! On almost all architectures, the EPOLL* constants have the same\nvalues as the POLL* constants do. But they keyword here is \"almost\".\nFor various bad reasons they aren\u0027t the same, and epoll() doesn\u0027t\nactually work quite correctly in some cases due to this on Sparc et al.\n\nThe next patch from Al will sort out the final differences, and we\nshould be all done.\n\nScripted-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ee5daa1361fceb6f482c005bcc9ba8d01b92ea5c", "tree": "71736f733c7efec13218756ab8635b258b8a1c53", "parents": [ "3fc928dc5308f6533d0e8004c8c63b46709a5276", "d85e2aa2e34dac793e70b900d865f48c69ecbc27" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 13:57:19 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 13:57:19 2018 -0800" }, "message": "Merge branch \u0027work.poll2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull more poll annotation updates from Al Viro:\n \"This is preparation to solving the problems you\u0027ve mentioned in the\n original poll series.\n\n After this series, the kernel is ready for running\n\n for V in IN OUT PRI ERR RDNORM RDBAND WRNORM WRBAND HUP RDHUP NVAL MSG; do\n L\u003d`git grep -l -w POLL$V | grep -v \u0027^t\u0027 | grep -v /um/ | grep -v \u0027^sa\u0027 | grep -v \u0027/poll.h$\u0027|grep -v \u0027^D\u0027`\n for f in $L; do sed -i \"-es/^\\([^\\\"]*\\)\\(\\\u003cPOLL$V\\\u003e\\)/\\\\1E\\\\2/\" $f; done\n done\n\n as a for bulk search-and-replace.\n\n After that, the kernel is ready to apply the patch to unify\n {de,}mangle_poll(), and then get rid of kernel-side POLL... uses\n entirely, and we should be all done with that stuff.\n\n Basically, that\u0027s what you suggested wrt KPOLL..., except that we can\n use EPOLL... instead - they already are arch-independent (and equal to\n what is currently kernel-side POLL...).\n\n After the preparations (in this series) switch to returning EPOLL...\n from -\u003epoll() instances is completely mechanical and kernel-side\n POLL... can go away. The last step (killing kernel-side POLL... and\n unifying {de,}mangle_poll() has to be done after the\n search-and-replace job, since we need userland-side POLL... for\n unified {de,}mangle_poll(), thus the cherry-pick at the last step.\n\n After that we will have:\n\n - POLL{IN,OUT,...} *not* in __poll_t, so any stray instances of\n -\u003epoll() still using those will be caught by sparse.\n\n - eventpoll.c and select.c warning-free wrt __poll_t\n\n - no more kernel-side definitions of POLL... - userland ones are\n visible through the entire kernel (and used pretty much only for\n mangle/demangle)\n\n - same behavior as after the first series (i.e. sparc et.al. epoll(2)\n working correctly)\"\n\n* \u0027work.poll2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:\n annotate ep_scan_ready_list()\n ep_send_events_proc(): return result via esed-\u003eres\n preparation to switching -\u003epoll() to returning EPOLL...\n add EPOLLNVAL, annotate EPOLL... and event_poll-\u003eevent\n use linux/poll.h instead of asm/poll.h\n xen: fix poll misannotation\n smc: missing poll annotations\n" }, { "commit": "3fc928dc5308f6533d0e8004c8c63b46709a5276", "tree": "97fe262892b85691899cda8d7a8aa8b67057d6d2", "parents": [ "60d7a21aedad7cd2dc9c27ca04fb8486aa6fb7da", "f8d0cbf28df14b6b6801f027a396fd593971a922" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 13:54:52 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 13:54:52 2018 -0800" }, "message": "Merge tag \u0027xtensa-20180211\u0027 of git://github.com/jcmvbkbc/linux-xtensa\n\nPull xtense fix from Max Filippov:\n \"Build fix for xtensa architecture with KASAN enabled\"\n\n* tag \u0027xtensa-20180211\u0027 of git://github.com/jcmvbkbc/linux-xtensa:\n xtensa: fix build with KASAN\n" }, { "commit": "60d7a21aedad7cd2dc9c27ca04fb8486aa6fb7da", "tree": "9da308fa03f1e60c10e815fa105415bb3dda427b", "parents": [ "d48fcbd864a008802a90c58a9ceddd9436d11a49", "e0691ebb33c12084ef11b6b59228c004e19f59c8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 13:52:32 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 11 13:52:32 2018 -0800" }, "message": "Merge tag \u0027nios2-v4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/lftan/nios2\n\nPull nios2 update from Ley Foon Tan:\n\n - clean up old Kconfig options from defconfig\n\n - remove leading 0x and 0s from bindings notation in dts files\n\n* tag \u0027nios2-v4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/lftan/nios2:\n nios2: defconfig: Cleanup from old Kconfig options\n nios2: dts: Remove leading 0x and 0s from bindings notation\n" }, { "commit": "f8d0cbf28df14b6b6801f027a396fd593971a922", "tree": "f521df51f24274f7d5ed6cff4e6553e6dbfcd766", "parents": [ "d48fcbd864a008802a90c58a9ceddd9436d11a49" ], "author": { "name": "Max Filippov", "email": "jcmvbkbc@gmail.com", "time": "Sun Feb 11 01:07:54 2018 -0800" }, "committer": { "name": "Max Filippov", "email": "jcmvbkbc@gmail.com", "time": "Sun Feb 11 09:18:12 2018 -0800" }, "message": "xtensa: fix build with KASAN\n\nThe commit 917538e212a2 (\"kasan: clean up KASAN_SHADOW_SCALE_SHIFT\nusage\") removed KASAN_SHADOW_SCALE_SHIFT definition from\ninclude/linux/kasan.h and added it to architecture-specific headers,\nexcept for xtensa. This broke the xtensa build with KASAN enabled.\nDefine KASAN_SHADOW_SCALE_SHIFT in arch/xtensa/include/asm/kasan.h\n\nReported by: kbuild test robot \u003cfengguang.wu@intel.com\u003e\nFixes: 917538e212a2 (\"kasan: clean up KASAN_SHADOW_SCALE_SHIFT usage\")\nAcked-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nSigned-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n" }, { "commit": "e0691ebb33c12084ef11b6b59228c004e19f59c8", "tree": "030c050fff51b257546092163cfc0fdd15144048", "parents": [ "5d13c73179983f8692adf397e65f2f57c613a917" ], "author": { "name": "Krzysztof Kozlowski", "email": "krzk@kernel.org", "time": "Sun Feb 11 23:01:17 2018 +0800" }, "committer": { "name": "Ley Foon Tan", "email": "ley.foon.tan@intel.com", "time": "Sun Feb 11 23:04:53 2018 +0800" }, "message": "nios2: defconfig: Cleanup from old Kconfig options\n\nRemove old, dead Kconfig option INET_LRO. It is gone since\ncommit 7bbf3cae65b6 (\"ipv4: Remove inet_lro library\").\n\nSigned-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nAcked-by: Ley Foon Tan \u003cley.foon.tan@intel.com\u003e\n" }, { "commit": "5d13c73179983f8692adf397e65f2f57c613a917", "tree": "80458c02e8798b906aa83f621a8d354ca161fe22", "parents": [ "d8a5b80568a9cb66810e75b182018e9edb68e8ff" ], "author": { "name": "Mathieu Malaterre", "email": "malat@debian.org", "time": "Sun Feb 11 22:59:18 2018 +0800" }, "committer": { "name": "Ley Foon Tan", "email": "ley.foon.tan@intel.com", "time": "Sun Feb 11 23:04:47 2018 +0800" }, "message": "nios2: dts: Remove leading 0x and 0s from bindings notation\n\nImprove the DTS files by removing all the leading \"0x\" and zeros to fix the\nfollowing dtc warnings:\n\nWarning (unit_address_format): Node /XXX unit name should not have leading \"0x\"\n\nand\n\nWarning (unit_address_format): Node /XXX unit name should not have leading 0s\n\nConverted using the following command:\n\nfind . -type f \\( -iname *.dts -o -iname *.dtsi \\) -exec sed -E -i -e \"s/@0x([0-9a-fA-F\\.]+)\\s?\\{/@\\L\\1 \\{/g\" -e \"s/@0+([0-9a-fA-F\\.]+)\\s?\\{/@\\L\\1 \\{/g\" {} +\n\nFor simplicity, two sed expressions were used to solve each warnings separately.\n\nTo make the regex expression more robust a few other issues were resolved,\nnamely setting unit-address to lower case, and adding a whitespace before the\nthe opening curly brace:\n\nhttps://elinux.org/Device_Tree_Linux#Linux_conventions\n\nThis is a follow up to commit 4c9847b7375a (\"dt-bindings: Remove leading 0x from bindings notation\")\n\nReported-by: David Daney \u003cddaney@caviumnetworks.com\u003e\nSuggested-by: Rob Herring \u003crobh@kernel.org\u003e\nSigned-off-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\nAcked-by: Ley Foon Tan \u003cley.foon.tan@intel.com\u003e\n" }, { "commit": "d48fcbd864a008802a90c58a9ceddd9436d11a49", "tree": "106829082981ab448c13094a58b5fb72d9d17da5", "parents": [ "9454473c9dccb7b9d25e5baf915a082bfd490b33", "c591c2e36ccc9a08f265841d2fd68e35327ab3c4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 14:08:26 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 14:08:26 2018 -0800" }, "message": "Merge tag \u0027pci-v4.16-fixes-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci\n\nPull PCI fix from Bjorn Helgaas:\n \"Fix a POWER9/powernv INTx regression from the merge window (Alexey\n Kardashevskiy)\"\n\n* tag \u0027pci-v4.16-fixes-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:\n powerpc/pci: Fix broken INTx configuration via OF\n" }, { "commit": "9454473c9dccb7b9d25e5baf915a082bfd490b33", "tree": "46f7f1a8886088e2f0184f1cf0e47c8ac12d4849", "parents": [ "cc5cb5af3a3363bc6f0530703895bf9c5fa2f159", "8525e5ff456592effe83640ea1702525e35b0363" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 14:05:11 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 14:05:11 2018 -0800" }, "message": "Merge tag \u0027for-linus-20180210\u0027 of git://git.kernel.dk/linux-block\n\nPull block fixes from Jens Axboe:\n \"A few fixes to round off the merge window on the block side:\n\n - a set of bcache fixes by way of Michael Lyle, from the usual bcache\n suspects.\n\n - add a simple-to-hook-into function for bpf EIO error injection.\n\n - fix blk-wbt that mischarectized flushes as reads. Improve the logic\n so that flushes and writes are accounted as writes, and only reads\n as reads. From me.\n\n - fix requeue crash in BFQ, from Paolo\"\n\n* tag \u0027for-linus-20180210\u0027 of git://git.kernel.dk/linux-block:\n block, bfq: add requeue-request hook\n bcache: fix for data collapse after re-attaching an attached device\n bcache: return attach error when no cache set exist\n bcache: set writeback_rate_update_seconds in range [1, 60] seconds\n bcache: fix for allocator and register thread race\n bcache: set error_limit correctly\n bcache: properly set task state in bch_writeback_thread()\n bcache: fix high CPU occupancy during journal\n bcache: add journal statistic\n block: Add should_fail_bio() for bpf error injection\n blk-wbt: account flush requests correctly\n" }, { "commit": "cc5cb5af3a3363bc6f0530703895bf9c5fa2f159", "tree": "24bac57b13b2bc6b5b80b4bbf22dd32966469326", "parents": [ "e9d46f74ecf5eb2c604b32bb43e632d2a5bd44a9", "1bd42d94ccab4eab5dc9dc9d303a687a61cde9bd" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 13:55:33 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 13:55:33 2018 -0800" }, "message": "Merge tag \u0027platform-drivers-x86-v4.16-3\u0027 of git://github.com/dvhart/linux-pdx86\n\nPull x86 platform driver updates from Darren Hart:\n \"Mellanox fixes and new system type support.\n\n Mostly data for new system types with a correction and an\n uninitialized variable fix\"\n\n[ Pulling from github because git.infradead.org currently seems to be\n down for some reason, but Darren had a backup location - Linus ]\n\n* tag \u0027platform-drivers-x86-v4.16-3\u0027 of git://github.com/dvhart/linux-pdx86:\n platform/x86: mlx-platform: Add support for new 200G IB and Ethernet systems\n platform/x86: mlx-platform: Add support for new msn201x system type\n platform/x86: mlx-platform: Add support for new msn274x system type\n platform/x86: mlx-platform: Fix power cable setting for msn21xx family\n platform/x86: mlx-platform: Add define for the negative bus\n platform/x86: mlx-platform: Use defines for bus assignment\n platform/mellanox: mlxreg-hotplug: Fix uninitialized variable\n" }, { "commit": "e9d46f74ecf5eb2c604b32bb43e632d2a5bd44a9", "tree": "2754cef2252b2ebe82e86136d90f0669bc24a9aa", "parents": [ "15303ba5d1cd9b28d03a980456c0978c0ea3b208", "d48b8c58c57f6edbe2965f0a5f62c5cf9593ca96" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 13:50:23 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 13:50:23 2018 -0800" }, "message": "Merge tag \u0027chrome-platform-for-linus-4.16\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/bleung/chrome-platform\n\nPull chrome platform updates from Benson Leung:\n\n - move cros_ec_dev to drivers/mfd\n\n - other small maintenance fixes\n\n[ The cros_ec_dev movement came in earlier through the MFD tree - Linus ]\n\n* tag \u0027chrome-platform-for-linus-4.16\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/bleung/chrome-platform:\n platform/chrome: Use proper protocol transfer function\n platform/chrome: cros_ec_lpc: Add support for Google Glimmer\n platform/chrome: cros_ec_lpc: Register the driver if ACPI entry is missing.\n platform/chrome: cros_ec_lpc: remove redundant pointer request\n cros_ec: fix nul-termination for firmware build info\n platform/chrome: chromeos_laptop: make chromeos_laptop const\n" }, { "commit": "15303ba5d1cd9b28d03a980456c0978c0ea3b208", "tree": "b9200d5b7474661cf36468038529a5269ee83238", "parents": [ "9a61df9e5f7471fe5be3e02bd0bed726b2761a54", "1ab03c072feb579c9fd116de25be2b211e6bff6a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 13:16:35 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 10 13:16:35 2018 -0800" }, "message": "Merge tag \u0027kvm-4.16-1\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm\n\nPull KVM updates from Radim Krčmář:\n \"ARM:\n\n - icache invalidation optimizations, improving VM startup time\n\n - support for forwarded level-triggered interrupts, improving\n performance for timers and passthrough platform devices\n\n - a small fix for power-management notifiers, and some cosmetic\n changes\n\n PPC:\n\n - add MMIO emulation for vector loads and stores\n\n - allow HPT guests to run on a radix host on POWER9 v2.2 CPUs without\n requiring the complex thread synchronization of older CPU versions\n\n - improve the handling of escalation interrupts with the XIVE\n interrupt controller\n\n - support decrement register migration\n\n - various cleanups and bugfixes.\n\n s390:\n\n - Cornelia Huck passed maintainership to Janosch Frank\n\n - exitless interrupts for emulated devices\n\n - cleanup of cpuflag handling\n\n - kvm_stat counter improvements\n\n - VSIE improvements\n\n - mm cleanup\n\n x86:\n\n - hypervisor part of SEV\n\n - UMIP, RDPID, and MSR_SMI_COUNT emulation\n\n - paravirtualized TLB shootdown using the new KVM_VCPU_PREEMPTED bit\n\n - allow guests to see TOPOEXT, GFNI, VAES, VPCLMULQDQ, and more\n AVX512 features\n\n - show vcpu id in its anonymous inode name\n\n - many fixes and cleanups\n\n - per-VCPU MSR bitmaps (already merged through x86/pti branch)\n\n - stable KVM clock when nesting on Hyper-V (merged through\n x86/hyperv)\"\n\n* tag \u0027kvm-4.16-1\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm: (197 commits)\n KVM: PPC: Book3S: Add MMIO emulation for VMX instructions\n KVM: PPC: Book3S HV: Branch inside feature section\n KVM: PPC: Book3S HV: Make HPT resizing work on POWER9\n KVM: PPC: Book3S HV: Fix handling of secondary HPTEG in HPT resizing code\n KVM: PPC: Book3S PR: Fix broken select due to misspelling\n KVM: x86: don\u0027t forget vcpu_put() in kvm_arch_vcpu_ioctl_set_sregs()\n KVM: PPC: Book3S PR: Fix svcpu copying with preemption enabled\n KVM: PPC: Book3S HV: Drop locks before reading guest memory\n kvm: x86: remove efer_reload entry in kvm_vcpu_stat\n KVM: x86: AMD Processor Topology Information\n x86/kvm/vmx: do not use vm-exit instruction length for fast MMIO when running nested\n kvm: embed vcpu id to dentry of vcpu anon inode\n kvm: Map PFN-type memory regions as writable (if possible)\n x86/kvm: Make it compile on 32bit and with HYPYERVISOR_GUEST\u003dn\n KVM: arm/arm64: Fixup userspace irqchip static key optimization\n KVM: arm/arm64: Fix userspace_irqchip_in_use counting\n KVM: arm/arm64: Fix incorrect timer_is_pending logic\n MAINTAINERS: update KVM/s390 maintainers\n MAINTAINERS: add Halil as additional vfio-ccw maintainer\n MAINTAINERS: add David as a reviewer for KVM/s390\n ...\n" }, { "commit": "c591c2e36ccc9a08f265841d2fd68e35327ab3c4", "tree": "9e08bfbbbd4c88bdb011602607a5ae6275299616", "parents": [ "ab8c609356fbe8dbcd44df11e884ce8cddf3739e" ], "author": { "name": "Alexey Kardashevskiy", "email": "aik@ozlabs.ru", "time": "Fri Feb 09 17:23:58 2018 +1100" }, "committer": { "name": "Bjorn Helgaas", "email": "bhelgaas@google.com", "time": "Sat Feb 10 11:49:56 2018 -0600" }, "message": "powerpc/pci: Fix broken INTx configuration via OF\n\n59f47eff03a0 (\"powerpc/pci: Use of_irq_parse_and_map_pci() helper\")\nreplaced of_irq_parse_pci() + irq_create_of_mapping() with\nof_irq_parse_and_map_pci(), but neglected to capture the virq\nreturned by irq_create_of_mapping(), so virq remained zero, which\ncaused INTx configuration to fail.\n\nSave the virq value returned by of_irq_parse_and_map_pci() and correct\nthe virq declaration to match the of_irq_parse_and_map_pci() signature.\n\nFixes: 59f47eff03a0 \"powerpc/pci: Use of_irq_parse_and_map_pci() helper\"\nSigned-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n[bhelgaas: changelog]\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e" }, { "commit": "9a61df9e5f7471fe5be3e02bd0bed726b2761a54", "tree": "d10ace20372c87b37919fa704d14cecdde3a35ce", "parents": [ "7a501609c2cb73381e925827c504a4c2c2cb0817", "523ca58b7db2e30e3c185a7927dd80a30c1bc743" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 19:32:41 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 19:32:41 2018 -0800" }, "message": "Merge tag \u0027kbuild-v4.16-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild\n\nPull more Kbuild updates from Masahiro Yamada:\n \"Makefile changes:\n - enable unused-variable warning that was wrongly disabled for clang\n\n Kconfig changes:\n - warn about blank \u0027help\u0027 and fix existing instances\n - fix \u0027choice\u0027 behavior to not write out invisible symbols\n - fix misc weirdness\n\n Coccinell changes:\n - fix false positive of free after managed memory alloc detection\n - improve performance of NULL dereference detection\"\n\n* tag \u0027kbuild-v4.16-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild: (21 commits)\n kconfig: remove const qualifier from sym_expand_string_value()\n kconfig: add xrealloc() helper\n kconfig: send error messages to stderr\n kconfig: echo stdin to stdout if either is redirected\n kconfig: remove check_stdin()\n kconfig: remove \u0027config*\u0027 pattern from .gitignnore\n kconfig: show \u0027?\u0027 prompt even if no help text is available\n kconfig: do not write choice values when their dependency becomes n\n coccinelle: deref_null: avoid useless computation\n coccinelle: devm_free: reduce false positives\n kbuild: clang: disable unused variable warnings only when constant\n kconfig: Warn if help text is blank\n nios2: kconfig: Remove blank help text\n arm: vt8500: kconfig: Remove blank help text\n MIPS: kconfig: Remove blank help text\n MIPS: BCM63XX: kconfig: Remove blank help text\n lib/Kconfig.debug: Remove blank help text\n Staging: rtl8192e: kconfig: Remove blank help text\n Staging: rtl8192u: kconfig: Remove blank help text\n mmc: kconfig: Remove blank help text\n ...\n" }, { "commit": "7a501609c2cb73381e925827c504a4c2c2cb0817", "tree": "e16eb57e0b245a7082e4ee1827fd2f9587e6e97e", "parents": [ "878e66d06fd098a744b9caafdd0e99bba415ce37" ], "author": { "name": "Al Viro", "email": "viro@ZenIV.linux.org.uk", "time": "Sat Feb 10 01:35:16 2018 +0000" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 19:28:01 2018 -0800" }, "message": "mconsole_proc(): don\u0027t mess with file-\u003ef_pos\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "878e66d06fd098a744b9caafdd0e99bba415ce37", "tree": "71a5cceab7713eb01b1a15a544124dbe2f6e07fc", "parents": [ "c839682c719f0e3dc851951c9e2eeb8a41cd9609", "cf5eebae2cd28d37581507668605f4d23cd7218d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 19:22:17 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 19:22:17 2018 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull misc vfs fixes from Al Viro.\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:\n seq_file: fix incomplete reset on read from zero offset\n kernfs: fix regression in kernfs_fop_write caused by wrong type\n" }, { "commit": "523ca58b7db2e30e3c185a7927dd80a30c1bc743", "tree": "59c135f8bde7d4ec24f1270c1d8615b8b142f296", "parents": [ "d717f24d8c68081caae2374cf5ea6c4e62c490fc" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Fri Feb 09 01:19:08 2018 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Sat Feb 10 11:31:49 2018 +0900" }, "message": "kconfig: remove const qualifier from sym_expand_string_value()\n\nThis function returns realloc\u0027ed memory, so the returned pointer\nmust be passed to free() when done. So, \u0027const\u0027 qualifier is odd.\nIt is allowed to modify the expanded string.\n\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n" }, { "commit": "d717f24d8c68081caae2374cf5ea6c4e62c490fc", "tree": "325008cf6d2fea2dfa15e4d49ae4d4d79fb1ff2b", "parents": [ "9e3e10c725360b9d07018cfcd5b7b6b7d325fae5" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Fri Feb 09 01:19:07 2018 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Sat Feb 10 11:26:04 2018 +0900" }, "message": "kconfig: add xrealloc() helper\n\nWe already have xmalloc(), xcalloc(). Add xrealloc() as well\nto save tedious error handling.\n\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n" }, { "commit": "1bd42d94ccab4eab5dc9dc9d303a687a61cde9bd", "tree": "5e0db902885b81bd9ba821b69586453356dd40a6", "parents": [ "a49a41482f61a48ff00f63f809ac0d802cb75424" ], "author": { "name": "Vadim Pasternak", "email": "vadimp@mellanox.com", "time": "Fri Feb 09 23:59:32 2018 +0000" }, "committer": { "name": "Darren Hart (VMware)", "email": "dvhart@infradead.org", "time": "Fri Feb 09 17:23:20 2018 -0800" }, "message": "platform/x86: mlx-platform: Add support for new 200G IB and Ethernet systems\n\nIt adds support for new Mellanox system types of basic classes qmb7, sn34,\nsn37, containing systems QMB700 (40x200GbE InfiniBand switch), SN3700\n(32x200GbE and 16x400GbE Ethernet switch) and SN3410 (6x400GbE plus\n48x50GbE Ethernet switch). These are the Top of the Rack systems, equipped\nwith Mellanox COM-Express carrier board and switch board with Mellanox\nQuantum device, which supports InfiniBand switching with 40X200G ports and\nline rate of up to HDR speed or with Mellanox Spectrum-2 device, which\nsupports Ethernet switching with 32X200G ports line rate of up to HDR\nspeed.\n\nSigned-off-by: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\nSigned-off-by: Darren Hart (VMware) \u003cdvhart@infradead.org\u003e\n" }, { "commit": "a49a41482f61a48ff00f63f809ac0d802cb75424", "tree": "4de18cdc3250f26129bb83f6a3ef96f8293aeab0", "parents": [ "ef08e14a3832c88bb8d5ccbb88eab48642fc6aa9" ], "author": { "name": "Vadim Pasternak", "email": "vadimp@mellanox.com", "time": "Fri Feb 09 23:59:31 2018 +0000" }, "committer": { "name": "Darren Hart (VMware)", "email": "dvhart@infradead.org", "time": "Fri Feb 09 17:23:16 2018 -0800" }, "message": "platform/x86: mlx-platform: Add support for new msn201x system type\n\nIt adds support for new Mellanox system types of basic half unit size\nclass msn201x, containing system MSN2010 (18x10GbE plus 4x4x25GbE) half\nand its derivatives. This is the Top of the Rack system, equipped with\nMellanox Small Form Factor carrier board and switch board with Mellanox\nSpectrum device, which supports Ethernet switching with 32X100G ports line\nrate of up to EDR speed.\n\nSigned-off-by: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\nSigned-off-by: Darren Hart (VMware) \u003cdvhart@infradead.org\u003e\n" }, { "commit": "ef08e14a3832c88bb8d5ccbb88eab48642fc6aa9", "tree": "55fac70f63a892010ba394e4469c57a2325353dd", "parents": [ "6016f7d54bc49e3c570d5c7adbb463fe87e19dd7" ], "author": { "name": "Vadim Pasternak", "email": "vadimp@mellanox.com", "time": "Fri Feb 09 23:59:30 2018 +0000" }, "committer": { "name": "Darren Hart (VMware)", "email": "dvhart@infradead.org", "time": "Fri Feb 09 17:23:07 2018 -0800" }, "message": "platform/x86: mlx-platform: Add support for new msn274x system type\n\nIt adds support for new Mellanox system types of basic class msn274x,\ncontaining system MSN2740 (32x100GbE Ethernet switch with cost reduction)\nand its derivatives. These are the Top of the Rack system, equipped with\nMellanox Small Form Factor carrier board and switch board with Mellanox\nSpectrum device, which supports Ethernet switching with 32X100G ports line\nrate of up to EDR speed.\n\nSigned-off-by: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\nSigned-off-by: Darren Hart (VMware) \u003cdvhart@infradead.org\u003e\n" }, { "commit": "c839682c719f0e3dc851951c9e2eeb8a41cd9609", "tree": "ec10374627a2f2446e14506cb545b713ab2909fe", "parents": [ "82f0a41e1980318ea4cdae20cdce7b33cb9c8946", "2fa56a494484f19e06bf4f3464b2155a92beafac" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 15:34:18 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 15:34:18 2018 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net\n\nPull networking fixes from David Miller:\n\n 1) Make allocations less aggressive in x_tables, from Minchal Hocko.\n\n 2) Fix netfilter flowtable Kconfig deps, from Pablo Neira Ayuso.\n\n 3) Fix connection loss problems in rtlwifi, from Larry Finger.\n\n 4) Correct DRAM dump length for some chips in ath10k driver, from Yu\n Wang.\n\n 5) Fix ABORT handling in rxrpc, from David Howells.\n\n 6) Add SPDX tags to Sun networking drivers, from Shannon Nelson.\n\n 7) Some ipv6 onlink handling fixes, from David Ahern.\n\n 8) Netem packet scheduler interval calcualtion fix from Md. Islam.\n\n 9) Don\u0027t put crypto buffers on-stack in rxrpc, from David Howells.\n\n10) Fix handling of error non-delivery status in netlink multicast\n delivery over multiple namespaces, from Nicolas Dichtel.\n\n11) Missing xdp flush in tuntap driver, from Jason Wang.\n\n12) Synchonize RDS protocol netns/module teardown with rds object\n management, from Sowini Varadhan.\n\n13) Add nospec annotations to mpls, from Dan Williams.\n\n14) Fix SKB truesize handling in TIPC, from Hoang Le.\n\n15) Interrupt masking fixes in stammc from Niklas Cassel.\n\n16) Don\u0027t allow ptr_ring objects to be sized outside of kmalloc\u0027s\n limits, from Jason Wang.\n\n17) Don\u0027t allow SCTP chunks to be built which will have a length\n exceeding the chunk header\u0027s 16-bit length field, from Alexey\n Kodanev.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (82 commits)\n ibmvnic: Remove skb-\u003eprotocol checks in ibmvnic_xmit\n bpf: fix rlimit in reuseport net selftest\n sctp: verify size of a new chunk in _sctp_make_chunk()\n s390/qeth: fix SETIP command handling\n s390/qeth: fix underestimated count of buffer elements\n ptr_ring: try vmalloc() when kmalloc() fails\n ptr_ring: fail early if queue occupies more than KMALLOC_MAX_SIZE\n net: stmmac: remove redundant enable of PMT irq\n net: stmmac: rename GMAC_INT_DEFAULT_MASK for dwmac4\n net: stmmac: discard disabled flags in interrupt status register\n ibmvnic: Reset long term map ID counter\n tools/libbpf: handle issues with bpf ELF objects containing .eh_frames\n selftests/bpf: add selftest that use test_libbpf_open\n selftests/bpf: add test program for loading BPF ELF files\n tools/libbpf: improve the pr_debug statements to contain section numbers\n bpf: Sync kernel ABI header with tooling header for bpf_common.h\n net: phy: fix phy_start to consider PHY_IGNORE_INTERRUPT\n net: thunder: change q_len\u0027s type to handle max ring size\n tipc: fix skb truesize/datasize ratio control\n net/sched: cls_u32: fix cls_u32 on filter replace\n ...\n" }, { "commit": "82f0a41e1980318ea4cdae20cdce7b33cb9c8946", "tree": "633222659a5b180f546d63af36f2db533b86455d", "parents": [ "858f45bff3b8be61d91e87ef90dddd68433cbffa", "0afa6b4412988019db14c6bfb8c6cbdf120ca9ad" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:55:30 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:55:30 2018 -0800" }, "message": "Merge tag \u0027nfs-for-4.16-2\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs\n\nPull more NFS client updates from Trond Myklebust:\n \"A few bugfixes and some small sunrpc latency/performance improvements\n before the merge window closes:\n\n Stable fixes:\n\n - fix an incorrect calculation of the RDMA send scatter gather\n element limit\n\n - fix an Oops when attempting to free resources after RDMA device\n removal\n\n Bugfixes:\n\n - SUNRPC: Ensure we always release the TCP socket in a timely fashion\n when the connection is shut down.\n\n - SUNRPC: Don\u0027t call __UDPX_INC_STATS() from a preemptible context\n\n Latency/Performance:\n\n - SUNRPC: Queue latency sensitive socket tasks to the less contended\n xprtiod queue\n\n - SUNRPC: Make the xprtiod workqueue unbounded.\n\n - SUNRPC: Make the rpciod workqueue unbounded\"\n\n* tag \u0027nfs-for-4.16-2\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs:\n SUNRPC: Don\u0027t call __UDPX_INC_STATS() from a preemptible context\n fix parallelism for rpc tasks\n Make the xprtiod workqueue unbounded.\n SUNRPC: Queue latency-sensitive socket tasks to xprtiod\n SUNRPC: Ensure we always close the socket after a connection shuts down\n xprtrdma: Fix BUG after a device removal\n xprtrdma: Fix calculation of ri_max_send_sges\n" }, { "commit": "858f45bff3b8be61d91e87ef90dddd68433cbffa", "tree": "5ccfb594195abf08913beb68676a1450a1036ceb", "parents": [ "8158c2ffa4f48d1e4ed3d0e198651f007b598137", "1c130ae00b769a2e2df41bad3d6051ee8234b636" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:49:46 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:49:46 2018 -0800" }, "message": "Merge branch \u0027for-next\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending\n\nPull SCSI target updates from Nicholas Bellinger:\n \"The highlights include:\n\n - numerous target-core-user improvements related to queue full and\n timeout handling. (MNC)\n\n - prevent target-core-user corruption when invalid data page is\n requested. (MNC)\n\n - add target-core device action configfs attributes to allow\n user-space to trigger events separate from existing attributes\n exposed to end-users. (MNC)\n\n - fix iscsi-target NULL pointer dereference 4.6+ regression in CHAP\n error path. (David Disseldorp)\n\n - avoid target-core backend UNMAP callbacks if range is zero. (Andrei\n Vagin)\n\n - fix a iscsi-target 4.14+ regression related multiple PDU logins,\n that was exposed due to removal of TCP prequeue support. (Florian\n Westphal + MNC)\n\n Also, there is a iser-target bug still being worked on for post -rc1\n code to address a long standing issue resulting in persistent\n ib_post_send() failures, for RNICs with small max_send_sge\"\n\n* \u0027for-next\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (36 commits)\n iscsi-target: make sure to wake up sleeping login worker\n tcmu: Fix trailing semicolon\n tcmu: fix cmd user after free\n target: fix destroy device in target_configure_device\n tcmu: allow userspace to reset ring\n target core: add device action configfs files\n tcmu: fix error return code in tcmu_configure_device()\n target_core_user: add cmd id to broken ring message\n target: add SAM_STAT_BUSY sense reason\n tcmu: prevent corruption when invalid data page requested\n target: don\u0027t call an unmap callback if a range length is zero\n target/iscsi: avoid NULL dereference in CHAP auth error path\n cxgbit: call neigh_event_send() to update MAC address\n target: tcm_loop: Use seq_puts() in tcm_loop_show_info()\n target: tcm_loop: Delete an unnecessary return statement in tcm_loop_submission_work()\n target: tcm_loop: Delete two unnecessary variable initialisations in tcm_loop_issue_tmr()\n target: tcm_loop: Combine substrings for 26 messages\n target: tcm_loop: Improve a size determination in two functions\n target: tcm_loop: Delete an error message for a failed memory allocation in four functions\n sbp-target: Delete an error message for a failed memory allocation in three functions\n ...\n" }, { "commit": "8158c2ffa4f48d1e4ed3d0e198651f007b598137", "tree": "382ac8db301f927ec4cdedb745d79fde9ba62a9a", "parents": [ "a28348322f79799d34867138806aa3cd15993d59", "878cb3fb06c67e8f1a452346e0bc6bb85f29b0a0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:47:09 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:47:09 2018 -0800" }, "message": "Merge tag \u0027trace-v4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing fixes from Steven Rostedt:\n \"Al Viro discovered some breakage with the parsing of the\n set_ftrace_filter as well as the removing of function probes.\n\n This fixes the code with Al\u0027s suggestions. I also added a few\n selftests to test the broken cases such that they wont happen\n again\"\n\n* tag \u0027trace-v4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n selftests/ftrace: Add more tests for removing of function probes\n selftests/ftrace: Add some missing glob checks\n selftests/ftrace: Have reset_ftrace_filter handle multiple instances\n selftests/ftrace: Have reset_ftrace_filter handle modules\n tracing: Fix parsing of globs with a wildcard at the beginning\n ftrace: Remove incorrect setting of glob search field\n" }, { "commit": "a28348322f79799d34867138806aa3cd15993d59", "tree": "41521d484e29c27f928cd99b42b9a6bb5cffec96", "parents": [ "58fcba61dea4a238f907bf7eb8b9c95427c27f28", "5f60a56494ea5518376b274dd93b3ceee9a783fb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:42:57 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:42:57 2018 -0800" }, "message": "Merge tag \u00274.16-minor-rc-SMB3-fixes\u0027 of git://git.samba.org/sfrench/cifs-2.6\n\nPull cifs fixes from Steve French:\n \"There are a couple additional security fixes that are still being\n tested that are not in this set.\"\n\n* tag \u00274.16-minor-rc-SMB3-fixes\u0027 of git://git.samba.org/sfrench/cifs-2.6:\n Add missing structs and defines from recent SMB3.1.1 documentation\n address lock imbalance warnings in smbdirect.c\n cifs: silence compiler warnings showing up with gcc-8.0.0\n Add some missing debug fields in server and tcon structs\n" }, { "commit": "58fcba61dea4a238f907bf7eb8b9c95427c27f28", "tree": "0ec863b5e9440d8eb474c6a510a552048f7f5fd1", "parents": [ "f9f1e414128ea58d8e848a0275db0f644c9e9f45", "b9058afcd6c725a8d293bb0a1b30bad68113235e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:40:16 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 14:40:16 2018 -0800" }, "message": "Merge tag \u0027fbdev-v4.16-fix\u0027 of git://github.com/bzolnier/linux\n\nPull fbdev fix from Bartlomiej Zolnierkiewicz:\n \"Fix building of the omapfb driver (Tomi Valkeinen)\"\n\n* tag \u0027fbdev-v4.16-fix\u0027 of git://github.com/bzolnier/linux:\n video: omapfb: fix missing #includes\n" }, { "commit": "1ab03c072feb579c9fd116de25be2b211e6bff6a", "tree": "c207a69c7943a464a83c97c9cc02a45c30c42bfa", "parents": [ "80132f4c0cf9e9966216ef4368bce530055dbb1d", "09f984961c137c4b252c368adab7e1c9f035fa59" ], "author": { "name": "Radim Krčmář", "email": "rkrcmar@redhat.com", "time": "Fri Feb 09 21:36:57 2018 +0100" }, "committer": { "name": "Radim Krčmář", "email": "rkrcmar@redhat.com", "time": "Fri Feb 09 22:03:06 2018 +0100" }, "message": "Merge tag \u0027kvm-ppc-next-4.16-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc\n\nSecond PPC KVM update for 4.16\n\nSeven fixes that are either trivial or that address bugs that people\nare actually hitting. The main ones are:\n\n- Drop spinlocks before reading guest memory\n\n- Fix a bug causing corruption of VCPU state in PR KVM with preemption\n enabled\n\n- Make HPT resizing work on POWER9\n\n- Add MMIO emulation for vector loads and stores, because guests now\n use these instructions in memcpy and similar routines.\n" }, { "commit": "80132f4c0cf9e9966216ef4368bce530055dbb1d", "tree": "bded8f3e2f88f8e073e78e6663e6f2e33aa61421", "parents": [ "8dbfb2bf1bb3848a8069164e205635b2675c24fe", "904e14fb7cb96401a7dc803ca2863fd5ba32ffe6" ], "author": { "name": "Radim Krčmář", "email": "rkrcmar@redhat.com", "time": "Fri Feb 02 18:26:58 2018 +0100" }, "committer": { "name": "Radim Krčmář", "email": "rkrcmar@redhat.com", "time": "Fri Feb 09 21:35:35 2018 +0100" }, "message": "Merge branch \u0027msr-bitmaps\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm\n\nThis topic branch allocates separate MSR bitmaps for each VCPU.\nThis is required for the IBRS enablement to choose, on a per-VM\nbasis, whether to intercept the SPEC_CTRL and PRED_CMD MSRs;\nthe IBRS enablement comes in through the tip tree.\n" }, { "commit": "2fa56a494484f19e06bf4f3464b2155a92beafac", "tree": "38c548bd82b9846c8ef039ce2569b69c07ffcb8c", "parents": [ "941ff6f11c020913f5cddf543a9ec63475d7c082" ], "author": { "name": "John Allen", "email": "jallen@linux.vnet.ibm.com", "time": "Fri Feb 09 13:19:46 2018 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:47:00 2018 -0500" }, "message": "ibmvnic: Remove skb-\u003eprotocol checks in ibmvnic_xmit\n\nHaving these checks in ibmvnic_xmit causes problems with VLAN\ntagging and balance-alb/tlb bonding modes. The restriction they\nimposed can be removed.\n\nSigned-off-by: John Allen \u003cjallen@linux.vnet.ibm.com\u003e\nSigned-off-by: Nathan Fontenot \u003cnfont@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "941ff6f11c020913f5cddf543a9ec63475d7c082", "tree": "d99421a9a776126da53568d094c34a3e6ec96924", "parents": [ "07f2c7ab6f8d0a7e7c5764c4e6cc9c52951b9d9c" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 09 14:49:44 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:46:37 2018 -0500" }, "message": "bpf: fix rlimit in reuseport net selftest\n\nFix two issues in the reuseport_bpf selftests that were\nreported by Linaro CI:\n\n [...]\n + ./reuseport_bpf\n ---- IPv4 UDP ----\n Testing EBPF mod 10...\n Reprograming, testing mod 5...\n ./reuseport_bpf: ebpf error. log:\n 0: (bf) r6 \u003d r1\n 1: (20) r0 \u003d *(u32 *)skb[0]\n 2: (97) r0 %\u003d 10\n 3: (95) exit\n processed 4 insns\n : Operation not permitted\n + echo FAIL\n [...]\n ---- IPv4 TCP ----\n Testing EBPF mod 10...\n ./reuseport_bpf: failed to bind send socket: Address already in use\n + echo FAIL\n [...]\n\nFor the former adjust rlimit since this was the cause of\nfailure for loading the BPF prog, and for the latter add\nSO_REUSEADDR.\n\nReported-by: Naresh Kamboju \u003cnaresh.kamboju@linaro.org\u003e\nLink: https://bugs.linaro.org/show_bug.cgi?id\u003d3502\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "07f2c7ab6f8d0a7e7c5764c4e6cc9c52951b9d9c", "tree": "fa4ac76cfca94b8848007559e04ab1d837ffb055", "parents": [ "7b30d51a8f830bd1c1103d31d0314ea6f1106707" ], "author": { "name": "Alexey Kodanev", "email": "alexey.kodanev@oracle.com", "time": "Fri Feb 09 17:35:23 2018 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:31:38 2018 -0500" }, "message": "sctp: verify size of a new chunk in _sctp_make_chunk()\n\nWhen SCTP makes INIT or INIT_ACK packet the total chunk length\ncan exceed SCTP_MAX_CHUNK_LEN which leads to kernel panic when\ntransmitting these packets, e.g. the crash on sending INIT_ACK:\n\n[ 597.804948] skbuff: skb_over_panic: text:00000000ffae06e4 len:120168\n put:120156 head:000000007aa47635 data:00000000d991c2de\n tail:0x1d640 end:0xfec0 dev:\u003cNULL\u003e\n...\n[ 597.976970] ------------[ cut here ]------------\n[ 598.033408] kernel BUG at net/core/skbuff.c:104!\n[ 600.314841] Call Trace:\n[ 600.345829] \u003cIRQ\u003e\n[ 600.371639] ? sctp_packet_transmit+0x2095/0x26d0 [sctp]\n[ 600.436934] skb_put+0x16c/0x200\n[ 600.477295] sctp_packet_transmit+0x2095/0x26d0 [sctp]\n[ 600.540630] ? sctp_packet_config+0x890/0x890 [sctp]\n[ 600.601781] ? __sctp_packet_append_chunk+0x3b4/0xd00 [sctp]\n[ 600.671356] ? sctp_cmp_addr_exact+0x3f/0x90 [sctp]\n[ 600.731482] sctp_outq_flush+0x663/0x30d0 [sctp]\n[ 600.788565] ? sctp_make_init+0xbf0/0xbf0 [sctp]\n[ 600.845555] ? sctp_check_transmitted+0x18f0/0x18f0 [sctp]\n[ 600.912945] ? sctp_outq_tail+0x631/0x9d0 [sctp]\n[ 600.969936] sctp_cmd_interpreter.isra.22+0x3be1/0x5cb0 [sctp]\n[ 601.041593] ? sctp_sf_do_5_1B_init+0x85f/0xc30 [sctp]\n[ 601.104837] ? sctp_generate_t1_cookie_event+0x20/0x20 [sctp]\n[ 601.175436] ? sctp_eat_data+0x1710/0x1710 [sctp]\n[ 601.233575] sctp_do_sm+0x182/0x560 [sctp]\n[ 601.284328] ? sctp_has_association+0x70/0x70 [sctp]\n[ 601.345586] ? sctp_rcv+0xef4/0x32f0 [sctp]\n[ 601.397478] ? sctp6_rcv+0xa/0x20 [sctp]\n...\n\nHere the chunk size for INIT_ACK packet becomes too big, mostly\nbecause of the state cookie (INIT packet has large size with\nmany address parameters), plus additional server parameters.\n\nLater this chunk causes the panic in skb_put_data():\n\n skb_packet_transmit()\n sctp_packet_pack()\n skb_put_data(nskb, chunk-\u003eskb-\u003edata, chunk-\u003eskb-\u003elen);\n\n\u0027nskb\u0027 (head skb) was previously allocated with packet-\u003esize\nfrom u16 \u0027chunk-\u003echunk_hdr-\u003elength\u0027.\n\nAs suggested by Marcelo we should check the chunk\u0027s length in\n_sctp_make_chunk() before trying to allocate skb for it and\ndiscard a chunk if its size bigger than SCTP_MAX_CHUNK_LEN.\n\nSigned-off-by: Alexey Kodanev \u003calexey.kodanev@oracle.com\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leinter@gmail.com\u003e\nAcked-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7b30d51a8f830bd1c1103d31d0314ea6f1106707", "tree": "577be160ca8f5a410d2d10eeb911c0c7286eb36a", "parents": [ "0bf7800f1799b5b1fd7d4f024e9ece53ac489011", "1c5b2216fbb973a9410e0b06389740b5c1289171" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:30:23 2018 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:30:23 2018 -0500" }, "message": "Merge branch \u0027s390-qeth-fixes\u0027\n\nJulian Wiedmann says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\ns390/qeth: fixes 2018-02-09\n\nplease apply the following two qeth patches for 4.16 and stable.\n\nOne restricts a command quirk to the intended commandd type,\nwhile the other fixes an off-by-one during data transmission\nthat can cause qeth to build malformed buffer descriptors.\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": "1c5b2216fbb973a9410e0b06389740b5c1289171", "tree": "577be160ca8f5a410d2d10eeb911c0c7286eb36a", "parents": [ "89271c65edd599207dd982007900506283c90ae3" ], "author": { "name": "Julian Wiedmann", "email": "jwi@linux.vnet.ibm.com", "time": "Fri Feb 09 11:03:50 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:30:23 2018 -0500" }, "message": "s390/qeth: fix SETIP command handling\n\nsend_control_data() applies some special handling to SETIP v4 IPA\ncommands. But current code parses *all* command types for the SETIP\ncommand code. Limit the command code check to IPA commands.\n\nFixes: 5b54e16f1a54 (\"qeth: do not spin for SETIP ip assist command\")\nSigned-off-by: Julian Wiedmann \u003cjwi@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "89271c65edd599207dd982007900506283c90ae3", "tree": "8914755732ad1d2d5eca6c0be6d64d05806eb23d", "parents": [ "0bf7800f1799b5b1fd7d4f024e9ece53ac489011" ], "author": { "name": "Ursula Braun", "email": "ubraun@linux.vnet.ibm.com", "time": "Fri Feb 09 11:03:49 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:30:23 2018 -0500" }, "message": "s390/qeth: fix underestimated count of buffer elements\n\nFor a memory range/skb where the last byte falls onto a page boundary\n(ie. \u0027end\u0027 is of the form xxx...xxx001), the PFN_UP() part of the\ncalculation currently doesn\u0027t round up to the next PFN due to an\noff-by-one error.\nThus qeth believes that the skb occupies one page less than it\nactually does, and may select a IO buffer that doesn\u0027t have enough spare\nbuffer elements to fit all of the skb\u0027s data.\nHW detects this as a malformed buffer descriptor, and raises an\nexception which then triggers device recovery.\n\nFixes: 2863c61334aa (\"qeth: refactor calculation of SBALE count\")\nSigned-off-by: Ursula Braun \u003cubraun@linux.vnet.ibm.com\u003e\nSigned-off-by: Julian Wiedmann \u003cjwi@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0bf7800f1799b5b1fd7d4f024e9ece53ac489011", "tree": "0570b566eede6848db5042f3af8c500a45496532", "parents": [ "6e6e41c3112276288ccaf80c70916779b84bb276" ], "author": { "name": "Jason Wang", "email": "jasowang@redhat.com", "time": "Fri Feb 09 17:45:50 2018 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:28:57 2018 -0500" }, "message": "ptr_ring: try vmalloc() when kmalloc() fails\n\nThis patch switch to use kvmalloc_array() for using a vmalloc()\nfallback to help in case kmalloc() fails.\n\nReported-by: syzbot+e4d4f9ddd4295539735d@syzkaller.appspotmail.com\nFixes: 2e0ab8ca83c12 (\"ptr_ring: array based FIFO for pointers\")\nSigned-off-by: Jason Wang \u003cjasowang@redhat.com\u003e\nAcked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6e6e41c3112276288ccaf80c70916779b84bb276", "tree": "336e9b2806125740656b3fcb6f0dc895ccd69acf", "parents": [ "909ebd5840d158670eb88b2f73e9d38a894f28e2" ], "author": { "name": "Jason Wang", "email": "jasowang@redhat.com", "time": "Fri Feb 09 17:45:49 2018 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:28:57 2018 -0500" }, "message": "ptr_ring: fail early if queue occupies more than KMALLOC_MAX_SIZE\n\nTo avoid slab to warn about exceeded size, fail early if queue\noccupies more than KMALLOC_MAX_SIZE.\n\nReported-by: syzbot+e4d4f9ddd4295539735d@syzkaller.appspotmail.com\nFixes: 2e0ab8ca83c12 (\"ptr_ring: array based FIFO for pointers\")\nSigned-off-by: Jason Wang \u003cjasowang@redhat.com\u003e\nAcked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "909ebd5840d158670eb88b2f73e9d38a894f28e2", "tree": "ebec54574b541a88c0465c9a2b59ce45ace5faa1", "parents": [ "faefaa97215a0c05105d7ae180fe1a3b5979ad1f", "1029117127540fef4edcf4f0887dc3e1f7d5adb2" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:23:04 2018 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:23:04 2018 -0500" }, "message": "Merge branch \u0027stmmac-irq-fixes-cleanups\u0027\n\nNiklas Cassel says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nstmmac irq fixes/cleanups\n\nA couple of small stmmac irq fixes/cleanups.\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": "1029117127540fef4edcf4f0887dc3e1f7d5adb2", "tree": "ebec54574b541a88c0465c9a2b59ce45ace5faa1", "parents": [ "e879b7ab3739d8f9990961fc7df2f63861bd780b" ], "author": { "name": "Niklas Cassel", "email": "niklas.cassel@axis.com", "time": "Fri Feb 09 17:22:47 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:23:04 2018 -0500" }, "message": "net: stmmac: remove redundant enable of PMT irq\n\nFor dwmac4, GMAC_INT_DEFAULT_ENABLE already includes\nGMAC_INT_PMT_EN, so it is redundant to check if hw-\u003epmt\nis set, and if so, setting the bit again.\n\nFor dwmac1000, GMAC_INT_DEFAULT_MASK does not include\nGMAC_INT_DISABLE_PMT, so it is redundant to check if\nhw-\u003epmt is set, and if so, clearing an already cleared bit.\n\nImprove code readability by removing this redundant code.\n\nSigned-off-by: Niklas Cassel \u003cniklas.cassel@axis.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e879b7ab3739d8f9990961fc7df2f63861bd780b", "tree": "18bfd32e7c76e1f415d480b0dddcfc920e24627c", "parents": [ "1b84ca187510f60f00f4e15255043ce19bb30410" ], "author": { "name": "Niklas Cassel", "email": "niklas.cassel@axis.com", "time": "Fri Feb 09 17:22:46 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:23:04 2018 -0500" }, "message": "net: stmmac: rename GMAC_INT_DEFAULT_MASK for dwmac4\n\nGMAC_INT_DEFAULT_MASK is written to the interrupt enable register.\nIn previous versions of the IP (e.g. dwmac1000), this register was\ninstead an interrupt mask register.\nTo improve clarity and reflect reality, rename GMAC_INT_DEFAULT_MASK\nto GMAC_INT_DEFAULT_ENABLE.\n\nSigned-off-by: Niklas Cassel \u003cniklas.cassel@axis.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1b84ca187510f60f00f4e15255043ce19bb30410", "tree": "f59b834d18e15e91ca89f56728389e00a4af8fa5", "parents": [ "faefaa97215a0c05105d7ae180fe1a3b5979ad1f" ], "author": { "name": "Niklas Cassel", "email": "niklas.cassel@axis.com", "time": "Fri Feb 09 17:22:45 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:23:04 2018 -0500" }, "message": "net: stmmac: discard disabled flags in interrupt status register\n\nThe interrupt status register in both dwmac1000 and dwmac4 ignores\ninterrupt enable (for dwmac4) / interrupt mask (for dwmac1000).\nTherefore, if we want to check only the bits that can actually trigger\nan irq, we have to filter the interrupt status register manually.\n\nCommit 0a764db10337 (\"stmmac: Discard masked flags in interrupt status\nregister\") fixed this for dwmac1000. Fix the same issue for dwmac4.\n\nJust like commit 0a764db10337 (\"stmmac: Discard masked flags in\ninterrupt status register\"), this makes sure that we do not get\nspurious link up/link down prints.\n\nSigned-off-by: Niklas Cassel \u003cniklas.cassel@axis.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "faefaa97215a0c05105d7ae180fe1a3b5979ad1f", "tree": "05ec5c9be1529f7b919e0ddb50c38160062dca7c", "parents": [ "437a4db66df3bc9f139cfbfc66913ea207d9162a" ], "author": { "name": "Thomas Falcon", "email": "tlfalcon@linux.vnet.ibm.com", "time": "Fri Feb 09 11:41:09 2018 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:20:39 2018 -0500" }, "message": "ibmvnic: Reset long term map ID counter\n\nWhen allocating RX or TX buffer pools, the driver needs to provide a\nunique mapping ID to firmware for each pool. This value is assigned\nusing a counter which is incremented after a new pool is created. The\nID can be an integer ranging from 1-255. When migrating to a device\nthat requests a different number of queues, this value was not being\nreset properly. As a result, after enough migrations, the counter\nexceeded the upper bound and pool creation failed. This is fixed by\nresetting the counter to one in this case.\n\nSigned-off-by: Thomas Falcon \u003ctlfalcon@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "437a4db66df3bc9f139cfbfc66913ea207d9162a", "tree": "24c796fd31a3298da7d4ea75652afc5e22dc28f8", "parents": [ "08f5138512180a479ce6b9d23b825c9f4cd3be77", "d977ae593b2d3f9ef0df795eda93f4e6bc92b323" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:05:10 2018 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Feb 09 14:05:10 2018 -0500" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf\n\nDaniel Borkmann says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\npull-request: bpf 2018-02-09\n\nThe following pull-request contains BPF updates for your *net* tree.\n\nThe main changes are:\n\n1) Two fixes for BPF sockmap in order to break up circular map references\n from programs attached to sockmap, and detaching related sockets in\n case of socket close() event. For the latter we get rid of the\n smap_state_change() and plug into ULP infrastructure, which will later\n also be used for additional features anyway such as TX hooks. For the\n second issue, dependency chain is broken up via map release callback\n to free parse/verdict programs, all from John.\n\n2) Fix a libbpf relocation issue that was found while implementing XDP\n support for Suricata project. Issue was that when clang was invoked\n with default target instead of bpf target, then various other e.g.\n debugging relevant sections are added to the ELF file that contained\n relocation entries pointing to non-BPF related sections which libbpf\n trips over instead of skipping them. Test cases for libbpf are added\n as well, from Jesper.\n\n3) Various misc fixes for bpftool and one for libbpf: a small addition\n to libbpf to make sure it recognizes all standard section prefixes.\n Then, the Makefile in bpftool/Documentation is improved to explicitly\n check for rst2man being installed on the system as we otherwise risk\n installing empty man pages; the man page for bpftool-map is corrected\n and a set of missing bash completions added in order to avoid shipping\n bpftool where the completions are only partially working, from Quentin.\n\n4) Fix applying the relocation to immediate load instructions in the\n nfp JIT which were missing a shift, from Jakub.\n\n5) Two fixes for the BPF kernel selftests: handle CONFIG_BPF_JIT_ALWAYS_ON\u003dy\n gracefully in test_bpf.ko module and mark them as FLAG_EXPECTED_FAIL\n in this case; and explicitly delete the veth devices in the two tests\n test_xdp_{meta,redirect}.sh before dismantling the netnses as when\n selftests are run in batch mode, then workqueue to handle destruction\n might not have finished yet and thus veth creation in next test under\n same dev name would fail, from Yonghong.\n\n6) Fix test_kmod.sh to check the test_bpf.ko module path before performing\n an insmod, and fallback to modprobe. Especially the latter is useful\n when having a device under test that has the modules installed instead,\n from Naresh.\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": "f9f1e414128ea58d8e848a0275db0f644c9e9f45", "tree": "9fe49804e165527116c49ca207247de1defbcf0a", "parents": [ "dff839f27dc8d70e191562c8e78b0a9a88028362", "781198f1f373c3e350dbeb3af04a7d4c81c1b8d7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 10:07:39 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 10:07:39 2018 -0800" }, "message": "Merge tag \u0027for-linus-4.16-rc1-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip\n\nPull xen fixes from Juergen Gross:\n \"Only five small fixes for issues when running under Xen\"\n\n* tag \u0027for-linus-4.16-rc1-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:\n xen: Fix {set,clear}_foreign_p2m_mapping on autotranslating guests\n pvcalls-back: do not return error on inet_accept EAGAIN\n xen-netfront: Fix race between device setup and open\n xen/grant-table: Use put_page instead of free_page\n x86/xen: init %gs very early to avoid page faults with stack protector\n" }, { "commit": "dff839f27dc8d70e191562c8e78b0a9a88028362", "tree": "58173c5913ee68e0d8acc8ef31808bbe416d3758", "parents": [ "54ce685cae30c106f062d714c11e644ab1b93b51", "48973df8c9c51612acc870ad1a885b0cf27c3356" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 09:58:37 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 09:58:37 2018 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\nPull s390 updates from Heiko Carstens:\n \"The main thing in this merge is the defense for the Spectre\n vulnerabilities. But there are other updates as well, the changes in\n more detail:\n\n - An s390 specific implementation of the array_index_mask_nospec\n function to the defense against spectre v1\n\n - Two patches to utilize the new PPA-12/PPA-13 instructions to run\n the kernel and/or user space with reduced branch predicton.\n\n - The s390 variant of the \u0027retpoline\u0027 spectre v2 defense called\n \u0027expoline\u0027. There is no return instruction for s390, instead an\n indirect branch is used for function return\n\n The s390 defense mechanism for indirect branches works by using an\n execute-type instruction with the indirect branch as the target of\n the execute. In effect that turns off the prediction for the\n indirect branch.\n\n - Scrub registers in entry.S that contain user controlled values to\n prevent the speculative use of these values.\n\n - Re-add the second parameter for the s390 specific runtime\n instrumentation system call and move the header file to uapi. The\n second parameter will continue to do nothing but older kernel\n versions only accepted valid real-time signal numbers. The details\n will be documented in the man-page for the system call.\n\n - Corrections and improvements for the s390 specific documentation\n\n - Add a line to /proc/sysinfo to display the CPU model dependent\n license-internal-code identifier\n\n - A header file include fix for eadm.\n\n - An error message fix in the kprobes code.\n\n - The removal of an outdated ARCH_xxx select statement\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:\n s390/kconfig: Remove ARCH_WANTS_PROT_NUMA_PROT_NONE select\n s390: introduce execute-trampolines for branches\n s390: run user space and KVM guests with modified branch prediction\n s390: add options to change branch prediction behaviour for the kernel\n s390/alternative: use a copy of the facility bit mask\n s390: add optimized array_index_mask_nospec\n s390: scrub registers on kernel entry and KVM exit\n s390/cio: fix kernel-doc usage\n s390/runtime_instrumentation: re-add signum system call parameter\n s390/cpum_cf: correct counter number of LAST_HOST_TRANSLATIONS\n s390/kprobes: Fix %p uses in error messages\n s390/runtime instrumentation: provide uapi header file\n s390/sysinfo: add and display licensed internal code identifier\n s390/docs: reword airq section\n s390/docs: mention subchannel types\n s390/cmf: fix kerneldoc\n s390/eadm: fix CONFIG_BLOCK include dependency\n" }, { "commit": "54ce685cae30c106f062d714c11e644ab1b93b51", "tree": "40869cc4a3aabfb6d9ef1cb4f68835cca2e8558c", "parents": [ "a051c14b8db35cb269e9d91e11fc3573b6f7475d", "d4abd46b7e7279a61a1aa76d1d1b16a990178e75" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 09:44:25 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 09:44:25 2018 -0800" }, "message": "Merge tag \u0027acpi-part2-4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull more ACPI updates from Rafael Wysocki:\n \"These are mostly fixes and cleanups, a few new quirks, a couple of\n updates related to the handling of ACPI tables and ACPICA copyrights\n refreshment.\n\n Specifics:\n\n - Update the ACPICA kernel code to upstream revision 20180105\n including:\n * Assorted fixes (Jung-uk Kim)\n * Support for X32 ABI compilation (Anuj Mittal)\n * Update of ACPICA copyrights to 2018 (Bob Moore)\n\n - Prepare for future modifications to avoid executing the _STA\n control method too early (Hans de Goede)\n\n - Make the processor performance control library code ignore _PPC\n notifications if they cannot be handled and fix up the C1 idle\n state definition when it is used as a fallback state (Chen Yu,\n Yazen Ghannam)\n\n - Make it possible to use the SPCR table on x86 and to replace the\n original IORT table with a new one from initrd (Prarit Bhargava,\n Shunyong Yang)\n\n - Add battery-related quirks for Asus UX360UA and UX410UAK and add\n quirks for table parsing on Dell XPS 9570 and Precision M5530 (Kai\n Heng Feng)\n\n - Address static checker warnings in the CPPC code (Gustavo Silva)\n\n - Avoid printing a raw pointer to the kernel log in the smart battery\n driver (Greg Kroah-Hartman)\"\n\n* tag \u0027acpi-part2-4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n ACPI: sbshc: remove raw pointer from printk() message\n ACPI: SPCR: Make SPCR available to x86\n ACPI / CPPC: Use 64-bit arithmetic instead of 32-bit\n ACPI / tables: Add IORT to injectable table list\n ACPI / bus: Parse tables as term_list for Dell XPS 9570 and Precision M5530\n ACPICA: Update version to 20180105\n ACPICA: All acpica: Update copyrights to 2018\n ACPI / processor: Set default C1 idle state description\n ACPI / battery: Add quirk for Asus UX360UA and UX410UAK\n ACPI: processor_perflib: Do not send _PPC change notification if not ready\n ACPI / scan: Use acpi_bus_get_status() to initialize ACPI_TYPE_DEVICE devs\n ACPI / bus: Do not call _STA on battery devices with unmet dependencies\n PCI: acpiphp_ibm: prepare for acpi_get_object_info() no longer returning status\n ACPI: export acpi_bus_get_status_handle()\n ACPICA: Add a missing pair of parentheses\n ACPICA: Prefer ACPI_TO_POINTER() over ACPI_ADD_PTR()\n ACPICA: Avoid NULL pointer arithmetic\n ACPICA: Linux: add support for X32 ABI compilation\n ACPI / video: Use true for boolean value\n" }, { "commit": "a051c14b8db35cb269e9d91e11fc3573b6f7475d", "tree": "35d603e7619b2a64d55fd9529519eac629b24173", "parents": [ "f1517df8701c9f12dae9ce7f43a5d300a6917619", "616f16093d852215f6a1188de1bafdc1f8892924" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 09:40:33 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 09 09:40:33 2018 -0800" }, "message": "Merge tag \u0027pm-part2-4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull more power management updates from Rafael Wysocki:\n \"These are mostly fixes and cleanups and removal of the no longer\n needed at32ap-cpufreq driver.\n\n Specifics:\n\n - Drop the at32ap-cpufreq driver which is useless after the removal\n of the corresponding arch (Corentin LABBE).\n\n - Fix a regression from the 4.14 cycle in the APM idle driver by\n making it initialize the polling state properly (Rafael Wysocki).\n\n - Fix a crash on failing system suspend due to a missing check in the\n cpufreq core (Bo Yan).\n\n - Make the intel_pstate driver initialize the hardware-managed\n P-state control (HWP) feature on CPU0 upon resume from system\n suspend if HWP had been enabled before the system was suspended\n (Chen Yu).\n\n - Fix up the SCPI cpufreq driver after recent changes (Sudeep Holla,\n Wei Yongjun).\n\n - Avoid pointer subtractions during frequency table walks in cpufreq\n (Dominik Brodowski).\n\n - Avoid the check for ProcFeedback in ST/CZ in the cpufreq driver for\n AMD processors and add a MODULE_ALIAS for cpufreq on ARM IMX (Akshu\n Agrawal, Nicolas Chauvet).\n\n - Fix the prototype of swsusp_arch_resume() on x86 (Arnd Bergmann).\n\n - Fix up the parsing of power domains DT data (Ulf Hansson)\"\n\n* tag \u0027pm-part2-4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n arm: imx: Add MODULE_ALIAS for cpufreq\n cpufreq: Add and use cpufreq_for_each_{valid_,}entry_idx()\n cpufreq: intel_pstate: Enable HWP during system resume on CPU0\n cpufreq: scpi: fix error return code in scpi_cpufreq_init()\n x86: hibernate: fix swsusp_arch_resume() prototype\n PM / domains: Fix up domain-idle-states OF parsing\n cpufreq: scpi: fix static checker warning cdev isn\u0027t an ERR_PTR\n cpufreq: remove at32ap-cpufreq\n cpufreq: AMD: Ignore the check for ProcFeedback in ST/CZ\n x86: PM: Make APM idle driver initialize polling state\n cpufreq: Skip cpufreq resume if it\u0027s not suspended\n" }, { "commit": "0afa6b4412988019db14c6bfb8c6cbdf120ca9ad", "tree": "1df23fb04c2fc0fbee7783369ead6c5ec0941e86", "parents": [ "f515f86b34b2e7d4b24cc9b7375c9e749895088e" ], "author": { "name": "Trond Myklebust", "email": "trond.myklebust@primarydata.com", "time": "Fri Feb 09 09:39:42 2018 -0500" }, "committer": { "name": "Trond Myklebust", "email": "trond.myklebust@primarydata.com", "time": "Fri Feb 09 09:39:42 2018 -0500" }, "message": "SUNRPC: Don\u0027t call __UDPX_INC_STATS() from a preemptible context\n\nCalling __UDPX_INC_STATS() from a preemptible context leads to a\nwarning of the form:\n\n BUG: using __this_cpu_add() in preemptible [00000000] code: kworker/u5:0/31\n caller is xs_udp_data_receive_workfn+0x194/0x270\n CPU: 1 PID: 31 Comm: kworker/u5:0 Not tainted 4.15.0-rc8-00076-g90ea9f1 #2\n Workqueue: xprtiod xs_udp_data_receive_workfn\n Call Trace:\n dump_stack+0x85/0xc1\n check_preemption_disabled+0xce/0xe0\n xs_udp_data_receive_workfn+0x194/0x270\n process_one_work+0x318/0x620\n worker_thread+0x20a/0x390\n ? process_one_work+0x620/0x620\n kthread+0x120/0x130\n ? __kthread_bind_mask+0x60/0x60\n ret_from_fork+0x24/0x30\n\nSince we\u0027re taking a spinlock in those functions anyway, let\u0027s fix the\nissue by moving the call so that it occurs under the spinlock.\n\nReported-by: kernel test robot \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n" }, { "commit": "b9058afcd6c725a8d293bb0a1b30bad68113235e", "tree": "e9bdbe12f03ecdfa3f801782d10e63622a86de46", "parents": [ "f1517df8701c9f12dae9ce7f43a5d300a6917619" ], "author": { "name": "Tomi Valkeinen", "email": "tomi.valkeinen@ti.com", "time": "Fri Feb 09 14:43:49 2018 +0100" }, "committer": { "name": "Bartlomiej Zolnierkiewicz", "email": "b.zolnierkie@samsung.com", "time": "Fri Feb 09 14:43:49 2018 +0100" }, "message": "video: omapfb: fix missing #includes\n\nThe omapfb driver fails to build after commit 23c35f48f5fb\n(\"pinctrl: remove include file from \u003clinux/device.h\u003e\") because it\nrelies on the \u003clinux/pinctrl/consumer.h\u003e and \u003clinux/seq_file.h\u003e\nbeing pulled in by the \u003clinux/device.h\u003e header implicitly.\n\nInclude these headers explicitly to avoid the build failures.\n\nFixes: 23c35f48f5fb (\"pinctrl: remove include file from \u003clinux/device.h\u003e\")\nSigned-off-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\nTested-by: Tony Lindgren \u003ctony@atomide.com\u003e\n[b.zolnierkie: fix include order and patch description]\nSigned-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n" }, { "commit": "09f984961c137c4b252c368adab7e1c9f035fa59", "tree": "e1f509042ec0fce63894916fc43a40096d44507d", "parents": [ "d20fe50a7b3c8f936f7347e9cab2e9dd89c1199d" ], "author": { "name": "Jose Ricardo Ziviani", "email": "joserz@linux.vnet.ibm.com", "time": "Sat Feb 03 18:24:26 2018 -0200" }, "committer": { "name": "Paul Mackerras", "email": "paulus@ozlabs.org", "time": "Fri Feb 09 16:51:51 2018 +1100" }, "message": "KVM: PPC: Book3S: Add MMIO emulation for VMX instructions\n\nThis patch provides the MMIO load/store vector indexed\nX-Form emulation.\n\nInstructions implemented:\nlvx: the quadword in storage addressed by the result of EA \u0026\n0xffff_ffff_ffff_fff0 is loaded into VRT.\n\nstvx: the contents of VRS are stored into the quadword in storage\naddressed by the result of EA \u0026 0xffff_ffff_ffff_fff0.\n\nReported-by: Gopesh Kumar Chaudhary \u003cgopchaud@in.ibm.com\u003e\nReported-by: Balamuruhan S \u003cbala24@linux.vnet.ibm.com\u003e\nSigned-off-by: Jose Ricardo Ziviani \u003cjoserz@linux.vnet.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n" }, { "commit": "d20fe50a7b3c8f936f7347e9cab2e9dd89c1199d", "tree": "006f51196b43e606bfa12c958bb005b2ab5a5158", "parents": [ "790a9df5fbef982f2a6992194fe497dd2b794a3d" ], "author": { "name": "Alexander Graf", "email": "agraf@suse.de", "time": "Thu Feb 08 18:38:53 2018 +0100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@ozlabs.org", "time": "Fri Feb 09 15:39:41 2018 +1100" }, "message": "KVM: PPC: Book3S HV: Branch inside feature section\n\nWe ended up with code that did a conditional branch inside a feature\nsection to code outside of the feature section. Depending on how the\nobject file gets organized, that might mean we exceed the 14bit\nrelocation limit for conditional branches:\n\n arch/powerpc/kvm/built-in.o:arch/powerpc/kvm/book3s_hv_rmhandlers.S:416:(__ftr_alt_97+0x8): relocation truncated to fit: R_PPC64_REL14 against `.text\u0027+1ca4\n\nSo instead of doing a conditional branch outside of the feature section,\nlet\u0027s just jump at the end of the same, making the branch very short.\n\nSigned-off-by: Alexander Graf \u003cagraf@suse.de\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n" }, { "commit": "790a9df5fbef982f2a6992194fe497dd2b794a3d", "tree": "d88cbd77fafe3909e242b77be4c768284498a4b7", "parents": [ "05f2bb0313a2855e491dadfc8319b7da261d7074" ], "author": { "name": "David Gibson", "email": "david@gibson.dropbear.id.au", "time": "Fri Feb 02 14:29:08 2018 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@ozlabs.org", "time": "Fri Feb 09 15:36:36 2018 +1100" }, "message": "KVM: PPC: Book3S HV: Make HPT resizing work on POWER9\n\nThis adds code to enable the HPT resizing code to work on POWER9,\nwhich uses a slightly modified HPT entry format compared to POWER8.\nOn POWER9, we convert HPTEs read from the HPT from the new format to\nthe old format so that the rest of the HPT resizing code can work as\nbefore. HPTEs written to the new HPT are converted to the new format\nas the last step before writing them into the new HPT.\n\nThis takes out the checks added by commit bcd3bb63dbc8 (\"KVM: PPC:\nBook3S HV: Disable HPT resizing on POWER9 for now\", 2017-02-18),\nnow that HPT resizing works on POWER9.\n\nOn POWER9, when we pivot to the new HPT, we now call\nkvmppc_setup_partition_table() to update the partition table in order\nto make the hardware use the new HPT.\n\n[paulus@ozlabs.org - added kvmppc_setup_partition_table() call,\n wrote commit message.]\n\nTested-by: Laurent Vivier \u003clvivier@redhat.com\u003e\nSigned-off-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n" }, { "commit": "05f2bb0313a2855e491dadfc8319b7da261d7074", "tree": "5e75fbfcc885ae5235eaaef4ef2ef310160ab56d", "parents": [ "57ea5f161a7de5b1913c212d04f57a175b159fdf" ], "author": { "name": "Paul Mackerras", "email": "paulus@ozlabs.org", "time": "Wed Feb 07 19:49:54 2018 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@ozlabs.org", "time": "Fri Feb 09 15:35:24 2018 +1100" }, "message": "KVM: PPC: Book3S HV: Fix handling of secondary HPTEG in HPT resizing code\n\nThis fixes the computation of the HPTE index to use when the HPT\nresizing code encounters a bolted HPTE which is stored in its\nsecondary HPTE group. The code inverts the HPTE group number, which\nis correct, but doesn\u0027t then mask it with new_hash_mask. As a result,\nnew_pteg will be effectively negative, resulting in new_hptep\npointing before the new HPT, which will corrupt memory.\n\nIn addition, this removes two BUG_ON statements. The condition that\nthe BUG_ONs were testing -- that we have computed the hash value\nincorrectly -- has never been observed in testing, and if it did\noccur, would only affect the guest, not the host. Given that\nBUG_ON should only be used in conditions where the kernel (i.e.\nthe host kernel, in this case) can\u0027t possibly continue execution,\nit is not appropriate here.\n\nReviewed-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n" }, { "commit": "6016f7d54bc49e3c570d5c7adbb463fe87e19dd7", "tree": "f9799e11aa602edd4715fa9dbd66adeead039fbd", "parents": [ "1778567a2063c3027f3ab03748fadab5840eba33" ], "author": { "name": "Vadim Pasternak", "email": "vadimp@mellanox.com", "time": "Fri Feb 02 08:45:47 2018 +0000" }, "committer": { "name": "Darren Hart (VMware)", "email": "dvhart@infradead.org", "time": "Thu Feb 08 17:47:51 2018 -0800" }, "message": "platform/x86: mlx-platform: Fix power cable setting for msn21xx family\n\nAdd dedicated structure with power cable setting for Mellanox msn21xx\nfamily. These systems do not have a physical device for the power unit\ncontroller. When the power cable is inserted or removed, the relevant\ninterrupt signal is handled, the status is updated, but no device is\nassociated with the signal.\n\nAdd definition for interrupt low aggregation signal. On system from\nmsn21xx family, low aggregation mask should be removed in order to allow\nsignal to hit CPU.\n\nFixes: 6613d18e9038 (\"platform/x86: mlx-platform: Move module from arch/x86\")\nSigned-off-by: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\nSigned-off-by: Darren Hart (VMware) \u003cdvhart@infradead.org\u003e\n" }, { "commit": "1778567a2063c3027f3ab03748fadab5840eba33", "tree": "e1f866e6cb40bec7971ba05f9d7c99f7e7f6f885", "parents": [ "ba814fdd0eac312be9cf204066760ef8d0f57221" ], "author": { "name": "Vadim Pasternak", "email": "vadimp@mellanox.com", "time": "Fri Feb 02 08:45:46 2018 +0000" }, "committer": { "name": "Darren Hart (VMware)", "email": "dvhart@infradead.org", "time": "Thu Feb 08 17:45:52 2018 -0800" }, "message": "platform/x86: mlx-platform: Add define for the negative bus\n\nAdd define for the negative bus ID in order to use it in case no hotplug\ndevice is associated with the hotplug interrupt signal. In this case,\nthe signal will be handled by the mlxreg-hotplug driver, but no device\nwill be associated with the signal.\n\nSigned-off-by: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\nSigned-off-by: Darren Hart (VMware) \u003cdvhart@infradead.org\u003e\n" }, { "commit": "ba814fdd0eac312be9cf204066760ef8d0f57221", "tree": "0ae14f1a6d5ff503b2f55730bc543928fa04be9c", "parents": [ "b81e830c9ad041aba81b495d9386dc31aed3f1e6" ], "author": { "name": "Vadim Pasternak", "email": "vadimp@mellanox.com", "time": "Fri Feb 02 08:45:45 2018 +0000" }, "committer": { "name": "Darren Hart (VMware)", "email": "dvhart@infradead.org", "time": "Thu Feb 08 17:45:02 2018 -0800" }, "message": "platform/x86: mlx-platform: Use defines for bus assignment\n\nAdd defines for the bus IDs, used for hotplug device topology to improve\ncode readability. Defines added for FAN and power units.\n\nSigned-off-by: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\nSigned-off-by: Darren Hart (VMware) \u003cdvhart@infradead.org\u003e\n" }, { "commit": "b81e830c9ad041aba81b495d9386dc31aed3f1e6", "tree": "3ff5c94d1d2b283f87fcbfb83ffa545464c96472", "parents": [ "9a7c551ba7aae3648473179f234e960cfd46eaf4" ], "author": { "name": "Geert Uytterhoeven", "email": "geert@linux-m68k.org", "time": "Wed Feb 07 10:12:04 2018 +0100" }, "committer": { "name": "Darren Hart (VMware)", "email": "dvhart@infradead.org", "time": "Thu Feb 08 17:43:37 2018 -0800" }, "message": "platform/mellanox: mlxreg-hotplug: Fix uninitialized variable\n\nWith gcc-4.1.2:\n\n drivers/platform/mellanox/mlxreg-hotplug.c: In function ‘mlxreg_hotplug_health_work_helper’:\n drivers/platform/mellanox/mlxreg-hotplug.c:347: warning: ‘ret’ is used uninitialized in this function\n\nIndeed, if mlxreg_core_item.count is zero, ret is used uninitialized.\n\nWhile this is unlikely to happen (it is set to ARRAY_SIZE(...) in x86\nboard files), this is done in another source file, so fix this by\npreinitializing ret to zero.\n\nFixes: c6acad68eb2dbffd (\"platform/mellanox: mlxreg-hotplug: Modify to use a regmap interface\")\nSigned-off-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nAcked-by: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\nSigned-off-by: Darren Hart (VMware) \u003cdvhart@infradead.org\u003e\n" }, { "commit": "d977ae593b2d3f9ef0df795eda93f4e6bc92b323", "tree": "0e3c09034cb8bc79fea586681d4de51689a23325", "parents": [ "69fe98edee48b4c3c7d90cb7eea1f1662202137d", "e3d91b0ca523d53158f435a3e13df7f0cb360ea2" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 09 00:26:18 2018 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 09 00:26:28 2018 +0100" }, "message": "Merge branch \u0027bpf-libbpf-relo-fix-and-tests\u0027\n\nJesper Dangaard Brouer says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nWhile playing with using libbpf for the Suricata project, we had\nissues LLVM \u003e\u003d 4.0.1 generating ELF files that could not be loaded\nwith libbpf (tools/lib/bpf/).\n\nDuring the troubleshooting phase, I wrote a test program and improved\nthe debugging output in libbpf. I turned this into a selftests\nprogram, and it also serves as a code example for libbpf in itself.\n\nI discovered that there are at least three ELF load issues with\nlibbpf. I left them as TODO comments in (tools/testing/selftests/bpf)\ntest_libbpf.sh. I\u0027ve only fixed the load issue with eh_frames, and\nother types of relo-section that does not have exec flags. We can\nwork on the other issues later.\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: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n" }, { "commit": "e3d91b0ca523d53158f435a3e13df7f0cb360ea2", "tree": "0e3c09034cb8bc79fea586681d4de51689a23325", "parents": [ "f09b2e382e9a7053e3ae6f2fb6535efd5760cf5d" ], "author": { "name": "Jesper Dangaard Brouer", "email": "brouer@redhat.com", "time": "Thu Feb 08 12:48:32 2018 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 09 00:25:12 2018 +0100" }, "message": "tools/libbpf: handle issues with bpf ELF objects containing .eh_frames\n\nV3: More generic skipping of relo-section (suggested by Daniel)\n\nIf clang \u003e\u003d 4.0.1 is missing the option \u0027-target bpf\u0027, it will cause\nllc/llvm to create two ELF sections for \"Exception Frames\", with\nsection names \u0027.eh_frame\u0027 and \u0027.rel.eh_frame\u0027.\n\nThe BPF ELF loader library libbpf fails when loading files with these\nsections. The other in-kernel BPF ELF loader in samples/bpf/bpf_load.c,\nhandle this gracefully. And iproute2 loader also seems to work with these\n\"eh\" sections.\n\nThe issue in libbpf is caused by bpf_object__elf_collect() skipping\nsome sections, and later when performing relocation it will be\npointing to a skipped section, as these sections cannot be found by\nbpf_object__find_prog_by_idx() in bpf_object__collect_reloc().\n\nThis is a general issue that also occurs for other sections, like\ndebug sections which are also skipped and can have relo section.\n\nAs suggested by Daniel. To avoid keeping state about all skipped\nsections, instead perform a direct qlookup in the ELF object. Lookup\nthe section that the relo-section points to and check if it contains\nexecutable machine instructions (denoted by the sh_flags\nSHF_EXECINSTR). Use this check to also skip irrelevant relo-sections.\n\nNote, for samples/bpf/ the \u0027-target bpf\u0027 parameter to clang cannot be used\ndue to incompatibility with asm embedded headers, that some of the samples\ninclude. This is explained in more details by Yonghong Song in bpf_devel_QA.\n\nSigned-off-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n" }, { "commit": "f09b2e382e9a7053e3ae6f2fb6535efd5760cf5d", "tree": "91b16d7227e33007319adfed253b945425a1dbec", "parents": [ "864db336c677c288d81524c30a656804785902f9" ], "author": { "name": "Jesper Dangaard Brouer", "email": "brouer@redhat.com", "time": "Thu Feb 08 12:48:27 2018 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 09 00:25:12 2018 +0100" }, "message": "selftests/bpf: add selftest that use test_libbpf_open\n\nThis script test_libbpf.sh will be part of the \u0027make run_tests\u0027\ninvocation, but can also be invoked manually in this directory,\nand a verbose mode can be enabled via setting the environment\nvariable $VERBOSE like:\n\n $ VERBOSE\u003dyes ./test_libbpf.sh\n\nThe script contains some tests that are commented out, as they\ncurrently fail. They are reminders about what we need to improve\nfor the libbpf loader library.\n\nSigned-off-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n" }, { "commit": "864db336c677c288d81524c30a656804785902f9", "tree": "84c4458cafdc4048669a3c47c698e2849fbb933f", "parents": [ "077c066a6c5445423147496e6c9b9a2c2d2ee762" ], "author": { "name": "Jesper Dangaard Brouer", "email": "brouer@redhat.com", "time": "Thu Feb 08 12:48:22 2018 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 09 00:24:38 2018 +0100" }, "message": "selftests/bpf: add test program for loading BPF ELF files\n\nV2: Moved program into selftests/bpf from tools/libbpf\n\nThis program can be used on its own for testing/debugging if a\nBPF ELF-object file can be loaded with libbpf (from tools/lib/bpf).\n\nIf something is wrong with the ELF object, the program have\na --debug mode that will display the ELF sections and especially\nthe skipped sections. This allows for quickly identifying the\nproblematic ELF section number, which can be corrolated with the\nreadelf tool.\n\nThe program signal error via return codes, and also have\na --quiet mode, which is practical for use in scripts like\nselftests/bpf.\n\nSigned-off-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n" }, { "commit": "077c066a6c5445423147496e6c9b9a2c2d2ee762", "tree": "9e1fbf72eea030e467249b7d4ff385b8bf476f2c", "parents": [ "8c88181ed452707f3815827225517b1964463b95" ], "author": { "name": "Jesper Dangaard Brouer", "email": "brouer@redhat.com", "time": "Thu Feb 08 12:48:17 2018 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 09 00:24:38 2018 +0100" }, "message": "tools/libbpf: improve the pr_debug statements to contain section numbers\n\nWhile debugging a bpf ELF loading issue, I needed to correlate the\nELF section number with the failed relocation section reference.\nThus, add section numbers/index to the pr_debug.\n\nIn debug mode, also print section that were skipped. This helped\nme identify that a section (.eh_frame) was skipped, and this was\nthe reason the relocation section (.rel.eh_frame) could not find\nthat section number.\n\nThe section numbers corresponds to the readelf tools Section Headers [Nr].\n\nSigned-off-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n" }, { "commit": "8c88181ed452707f3815827225517b1964463b95", "tree": "6608f0e4c485b76e2f121afc3e13945880f3dc1a", "parents": [ "69fe98edee48b4c3c7d90cb7eea1f1662202137d" ], "author": { "name": "Jesper Dangaard Brouer", "email": "brouer@redhat.com", "time": "Thu Feb 08 12:48:12 2018 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 09 00:24:38 2018 +0100" }, "message": "bpf: Sync kernel ABI header with tooling header for bpf_common.h\n\nI recently fixed up a lot of commits that forgot to keep the tooling\nheaders in sync. And then I forgot to do the same thing in commit\ncb5f7334d479 (\"bpf: add comments to BPF ld/ldx sizes\"). Let correct\nthat before people notice ;-).\n\nLawrence did partly fix/sync this for bpf.h in commit d6d4f60c3a09\n(\"bpf: add selftest for tcpbpf\").\n\nFixes: cb5f7334d479 (\"bpf: add comments to BPF ld/ldx sizes\")\nSigned-off-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n" }, { "commit": "f1517df8701c9f12dae9ce7f43a5d300a6917619", "tree": "b7ba72e70fcdf17306d2a488b5b245e578ad52aa", "parents": [ "9d21874da8ec0e0043c85cde8dda173e74ffc24d", "175e03101d36c3034f3c80038d4c28838351a7f2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 15:18:32 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 15:18:32 2018 -0800" }, "message": "Merge tag \u0027nfsd-4.16\u0027 of git://linux-nfs.org/~bfields/linux\n\nPull nfsd update from Bruce Fields:\n \"A fairly small update this time around. Some cleanup, RDMA fixes,\n overlayfs fixes, and a fix for an NFSv4 state bug.\n\n The bigger deal for nfsd this time around was Jeff Layton\u0027s\n already-merged i_version patches\"\n\n* tag \u0027nfsd-4.16\u0027 of git://linux-nfs.org/~bfields/linux:\n svcrdma: Fix Read chunk round-up\n NFSD: hide unused svcxdr_dupstr()\n nfsd: store stat times in fill_pre_wcc() instead of inode times\n nfsd: encode stat-\u003emtime for getattr instead of inode-\u003ei_mtime\n nfsd: return RESOURCE not GARBAGE_ARGS on too many ops\n nfsd4: don\u0027t set lock stateid\u0027s sc_type to CLOSED\n nfsd: Detect unhashed stids in nfsd4_verify_open_stid()\n sunrpc: remove dead code in svc_sock_setbufsize\n svcrdma: Post Receives in the Receive completion handler\n nfsd4: permit layoutget of executable-only files\n lockd: convert nlm_rqst.a_count from atomic_t to refcount_t\n lockd: convert nlm_lockowner.count from atomic_t to refcount_t\n lockd: convert nsm_handle.sm_count from atomic_t to refcount_t\n" }, { "commit": "9d21874da8ec0e0043c85cde8dda173e74ffc24d", "tree": "8c02dce2b04ff5d70cafbd82a8bff383f2a49604", "parents": [ "4ed8244ef8847c8ad7414e1a12ba45fef7998721", "ac665d9423474e64e64b34b0e2cea43601b50d7d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 14:39:29 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 14:39:29 2018 -0800" }, "message": "Merge branch \u0027idr-2018-02-06\u0027 of git://git.infradead.org/users/willy/linux-dax\n\nPull idr updates from Matthew Wilcox:\n\n - test-suite improvements\n\n - replace the extended API by improving the normal API\n\n - performance improvement for IDRs which are 1-based rather than\n 0-based\n\n - add documentation\n\n* \u0027idr-2018-02-06\u0027 of git://git.infradead.org/users/willy/linux-dax:\n idr: Add documentation\n idr: Make 1-based IDRs more efficient\n idr: Warn if old iterators see large IDs\n idr: Rename idr_for_each_entry_ext\n idr: Remove idr_alloc_ext\n cls_u32: Convert to idr_alloc_u32\n cls_u32: Reinstate cyclic allocation\n cls_flower: Convert to idr_alloc_u32\n cls_bpf: Convert to use idr_alloc_u32\n cls_basic: Convert to use idr_alloc_u32\n cls_api: Convert to idr_alloc_u32\n net sched actions: Convert to use idr_alloc_u32\n idr: Add idr_alloc_u32 helper\n idr: Delete idr_find_ext function\n idr: Delete idr_replace_ext function\n idr: Delete idr_remove_ext function\n IDR test suite: Check handling negative end correctly\n idr test suite: Fix ida_test_random()\n radix tree test suite: Remove ARRAY_SIZE\n" }, { "commit": "4ed8244ef8847c8ad7414e1a12ba45fef7998721", "tree": "3edcbbb546bf5d8c00c51053c0a344ae5e81942d", "parents": [ "a0f79386a4968b4925da6db2d1daffd0605a4402", "b86729109c5fd0a480300f40608aac68764b5adf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 14:37:32 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 14:37:32 2018 -0800" }, "message": "Merge tag \u0027gcc-plugins-v4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux\n\nPull gcc plugins updates from Kees Cook:\n\n - update includes for gcc 8 (Valdis Kletnieks)\n\n - update initializers for gcc 8\n\n* tag \u0027gcc-plugins-v4.16-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:\n gcc-plugins: Use dynamic initializers\n gcc-plugins: Add include required by GCC release 8\n" }, { "commit": "f515f86b34b2e7d4b24cc9b7375c9e749895088e", "tree": "8f39d79722093381684830f6f9a1b13f702abfe9", "parents": [ "90ea9f1b60c679049619a79d9fc1557bc41c4973" ], "author": { "name": "Olga Kornievskaia", "email": "aglo@umich.edu", "time": "Thu Jun 29 09:25:36 2017 -0400" }, "committer": { "name": "Trond Myklebust", "email": "trond.myklebust@primarydata.com", "time": "Thu Feb 08 16:24:35 2018 -0500" }, "message": "fix parallelism for rpc tasks\n\nHi folks,\n\nOn a multi-core machine, is it expected that we can have parallel RPCs\nhandled by each of the per-core workqueue?\n\nIn testing a read workload, observing via \"top\" command that a single\n\"kworker\" thread is running servicing the requests (no parallelism).\nIt\u0027s more prominent while doing these operations over krb5p mount.\n\nWhat has been suggested by Bruce is to try this and in my testing I\nsee then the read workload spread among all the kworker threads.\n\nSigned-off-by: Olga Kornievskaia \u003ckolga@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n" }, { "commit": "08f5138512180a479ce6b9d23b825c9f4cd3be77", "tree": "a38fc27693c17bbe239cc6876c39043a92ba558c", "parents": [ "88c991a91729b402bbfbf247fdba16ac21c369ab" ], "author": { "name": "Heiner Kallweit", "email": "hkallweit1@gmail.com", "time": "Thu Feb 08 21:01:48 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:35:51 2018 -0500" }, "message": "net: phy: fix phy_start to consider PHY_IGNORE_INTERRUPT\n\nThis condition wasn\u0027t adjusted when PHY_IGNORE_INTERRUPT (-2) was added\nlong ago. In case of PHY_IGNORE_INTERRUPT the MAC interrupt indicates\nalso PHY state changes and we should do what the symbol says.\n\nFixes: 84a527a41f38 (\"net: phylib: fix interrupts re-enablement in phy_start\")\nSigned-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nReviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "88c991a91729b402bbfbf247fdba16ac21c369ab", "tree": "dc9d25a3d8290f40c5cf5669b69a7584a53c066d", "parents": [ "e0c42c8e3e94f6c478f8c96814d4a2d19d2204b2" ], "author": { "name": "Dean Nelson", "email": "dnelson@redhat.com", "time": "Thu Feb 08 14:21:05 2018 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:34:23 2018 -0500" }, "message": "net: thunder: change q_len\u0027s type to handle max ring size\n\nThe Cavium thunder nicvf driver supports rx/tx rings of up to 65536 entries per.\nThe number of entires are stored in the q_len member of struct q_desc_mem. The\nproblem is that q_len being a u16, results in 65536 becoming 0.\n\nIn getting pointers to descriptors in the rings, the driver uses q_len minus 1\nas a mask after incrementing the pointer, in order to go back to the beginning\nand not go past the end of the ring.\n\nWith the q_len set to 0 the mask is no longer correct and the driver does go\nbeyond the end of the ring, causing various ills. Usually the first thing that\nshows up is a \"NETDEV WATCHDOG: enP2p1s0f1 (nicvf): transmit queue 7 timed out\"\nwarning.\n\nThis patch remedies the problem by changing q_len to a u32.\n\nSigned-off-by: Dean Nelson \u003cdnelson@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e0c42c8e3e94f6c478f8c96814d4a2d19d2204b2", "tree": "52478ce91a12eb79c3f0c3d4753d187d80886520", "parents": [ "55b3280d1e471795c08dbbe17325720a843e104c", "99ffd198f07f46f3a8e64399249a8333c09063df" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:32:25 2018 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:32:25 2018 -0500" }, "message": "Merge tag \u0027wireless-drivers-next-for-davem-2018-02-08\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers-next\n\nKalle Valo says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nwireless-drivers-next patches for 4.16\n\nThe most important here is the ssb fix, it has been reported by the\nusers frequently and the fix just missed the final v4.15. Also\nnumerous other fixes, mt76 had multiple problems with aggregation and\na long standing unaligned access bug in rtlwifi is finally fixed.\n\nMajor changes:\n\nath10k\n\n* correct firmware RAM dump length for QCA6174/QCA9377\n\n* add new QCA988X device id\n\n* fix a kernel panic during pci probe\n\n* revert a recent commit which broke ath10k firmware metadata parsing\n\nath9k\n\n* fix a noise floor regression introduced during the merge window\n\n* add new device id\n\nrtlwifi\n\n* fix unaligned access seen on ARM architecture\n\nmt76\n\n* various aggregation fixes which fix connection stalls\n\nssb\n\n* fix b43 and b44 on non-MIPS which broke in v4.15-rc9\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": "55b3280d1e471795c08dbbe17325720a843e104c", "tree": "540f61caa564a39a197d6980c0a244f1cd93a807", "parents": [ "eb53f7af6f15285e2f6ada97285395343ce9f433" ], "author": { "name": "Hoang Le", "email": "hoang.h.le@dektek.com.au", "time": "Thu Feb 08 17:16:25 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:30:40 2018 -0500" }, "message": "tipc: fix skb truesize/datasize ratio control\n\nIn commit d618d09a68e4 (\"tipc: enforce valid ratio between skb truesize\nand contents\") we introduced a test for ensuring that the condition\ntruesize/datasize \u003c\u003d 4 is true for a received buffer. Unfortunately this\ntest has two problems.\n\n- Because of the integer arithmetics the test\n if (skb-\u003etruesize / buf_roundup_len(skb) \u003e 4) will miss all\n ratios [4 \u003c ratio \u003c 5], which was not the intention.\n- The buffer returned by skb_copy() inherits skb-\u003etruesize of the\n original buffer, which doesn\u0027t help the situation at all.\n\nIn this commit, we change the ratio condition and replace skb_copy()\nwith a call to skb_copy_expand() to finally get this right.\n\nAcked-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\nSigned-off-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "eb53f7af6f15285e2f6ada97285395343ce9f433", "tree": "a0d8288be4929ac39f3b115e1f159f8979b6b21e", "parents": [ "3968523f855050b8195134da951b87c20bd66130" ], "author": { "name": "Ivan Vecera", "email": "ivecera@redhat.com", "time": "Thu Feb 08 16:10:39 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:27:58 2018 -0500" }, "message": "net/sched: cls_u32: fix cls_u32 on filter replace\n\nThe following sequence is currently broken:\n\n # tc qdisc add dev foo ingress\n # tc filter replace dev foo protocol all ingress \\\n u32 match u8 0 0 action mirred egress mirror dev bar1\n # tc filter replace dev foo protocol all ingress \\\n handle 800::800 pref 49152 \\\n u32 match u8 0 0 action mirred egress mirror dev bar2\n Error: cls_u32: Key node flags do not match passed flags.\n We have an error talking to the kernel, -1\n\nThe error comes from u32_change() when comparing new and\nexisting flags. The existing ones always contains one of\nTCA_CLS_FLAGS_{,NOT}_IN_HW flag depending on offloading state.\nThese flags cannot be passed from userspace so the condition\n(n-\u003eflags !\u003d flags) in u32_change() always fails.\n\nFix the condition so the flags TCA_CLS_FLAGS_NOT_IN_HW and\nTCA_CLS_FLAGS_IN_HW are not taken into account.\n\nFixes: 24d3dc6d27ea (\"net/sched: cls_u32: Reflect HW offload status\")\nSigned-off-by: Ivan Vecera \u003civecera@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3968523f855050b8195134da951b87c20bd66130", "tree": "1bf66a780a1e70748dcbd036281015a1d510c564", "parents": [ "ebeeb1ad9b8adcc37c2ec21a96f39e9d35199b46" ], "author": { "name": "Dan Williams", "email": "dan.j.williams@intel.com", "time": "Wed Feb 07 22:34:24 2018 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:24:12 2018 -0500" }, "message": "mpls, nospec: Sanitize array index in mpls_label_ok()\n\nmpls_label_ok() validates that the \u0027platform_label\u0027 array index from a\nuserspace netlink message payload is valid. Under speculation the\nmpls_label_ok() result may not resolve in the CPU pipeline until after\nthe index is used to access an array element. Sanitize the index to zero\nto prevent userspace-controlled arbitrary out-of-bounds speculation, a\nprecursor for a speculative execution side channel vulnerability.\n\nCc: \u003cstable@vger.kernel.org\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ebeeb1ad9b8adcc37c2ec21a96f39e9d35199b46", "tree": "25c1f3a044ab1c8f8e15031f606b00ae35db2ed0", "parents": [ "79a8a642bf05cd0dced20621f6fef9d884124abd" ], "author": { "name": "Sowmini Varadhan", "email": "sowmini.varadhan@oracle.com", "time": "Sat Feb 03 04:26:51 2018 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:23:52 2018 -0500" }, "message": "rds: tcp: use rds_destroy_pending() to synchronize netns/module teardown and rds connection/workq management\n\nAn rds_connection can get added during netns deletion between lines 528\nand 529 of\n\n 506 static void rds_tcp_kill_sock(struct net *net)\n :\n /* code to pull out all the rds_connections that should be destroyed */\n :\n 528 spin_unlock_irq(\u0026rds_tcp_conn_lock);\n 529 list_for_each_entry_safe(tc, _tc, \u0026tmp_list, t_tcp_node)\n 530 rds_conn_destroy(tc-\u003et_cpath-\u003ecp_conn);\n\nSuch an rds_connection would miss out the rds_conn_destroy()\nloop (that cancels all pending work) and (if it was scheduled\nafter netns deletion) could trigger the use-after-free.\n\nA similar race-window exists for the module unload path\nin rds_tcp_exit -\u003e rds_tcp_destroy_conns\n\nConcurrency with netns deletion (rds_tcp_kill_sock()) must be handled\nby checking check_net() before enqueuing new work or adding new\nconnections.\n\nConcurrency with module-unload is handled by maintaining a module\nspecific flag that is set at the start of the module exit function,\nand must be checked before enqueuing new work or adding new connections.\n\nThis commit refactors existing RDS_DESTROY_PENDING checks added by\ncommit 3db6e0d172c9 (\"rds: use RCU to synchronize work-enqueue with\nconnection teardown\") and consolidates all the concurrency checks\nlisted above into the function rds_destroy_pending().\n\nSigned-off-by: Sowmini Varadhan \u003csowmini.varadhan@oracle.com\u003e\nAcked-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a0f79386a4968b4925da6db2d1daffd0605a4402", "tree": "096fb4be74f080900bba38afefcc7de0148f34e3", "parents": [ "81153336eb76b253ba7852f3f1de525bb98f8c4d", "74e938c22705c7b80d4422e84eea927fc78e60a8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 12:20:41 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 12:20:41 2018 -0800" }, "message": "Merge tag \u0027for-linus-4.16\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hubcap/linux\n\nPull orangefs updates from Mike Marshall:\n \"Mostly cleanups, but three bug fixes:\n\n - don\u0027t pass garbage return codes back up the call chain (Mike\n Marshall)\n\n - fix stale inode test (Martin Brandenburg)\n\n - fix off-by-one errors (Xiongfeng Wang)\n\n Also add Martin as a reviewer in the Maintainers file\"\n\n* tag \u0027for-linus-4.16\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hubcap/linux:\n orangefs: reverse sense of is-inode-stale test in d_revalidate\n orangefs: simplify orangefs_inode_is_stale\n Orangefs: don\u0027t propogate whacky error codes\n orangefs: use correct string length\n orangefs: make orangefs_make_bad_inode static\n orangefs: remove ORANGEFS_KERNEL_DEBUG\n orangefs: remove gossip_ldebug and gossip_lerr\n orangefs: make orangefs_client_debug_init static\n MAINTAINERS: update orangefs list and add myself as reviewer\n" }, { "commit": "79a8a642bf05cd0dced20621f6fef9d884124abd", "tree": "71e480e6404ec409f6f9fa6dfac7372a1c14d559", "parents": [ "e728789c52afccc1275cba1dd812f03abe16ea3c" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Feb 07 17:44:38 2018 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:15:48 2018 -0500" }, "message": "net: Whitelist the skbuff_head_cache \"cb\" field\n\nMost callers of put_cmsg() use a \"sizeof(foo)\" for the length argument.\nWithin put_cmsg(), a copy_to_user() call is made with a dynamic size, as a\nresult of the cmsg header calculations. This means that hardened usercopy\nwill examine the copy, even though it was technically a fixed size and\nshould be implicitly whitelisted. All the put_cmsg() calls being built\nfrom values in skbuff_head_cache are coming out of the protocol-defined\n\"cb\" field, so whitelist this field entirely instead of creating per-use\nbounce buffers, for which there are concerns about performance.\n\nOriginal report was:\n\nBad or missing usercopy whitelist? Kernel memory exposure attempt detected from SLAB object \u0027skbuff_head_cache\u0027 (offset 64, size 16)!\nWARNING: CPU: 0 PID: 3663 at mm/usercopy.c:81 usercopy_warn+0xdb/0x100 mm/usercopy.c:76\n...\n __check_heap_object+0x89/0xc0 mm/slab.c:4426\n check_heap_object mm/usercopy.c:236 [inline]\n __check_object_size+0x272/0x530 mm/usercopy.c:259\n check_object_size include/linux/thread_info.h:112 [inline]\n check_copy_size include/linux/thread_info.h:143 [inline]\n copy_to_user include/linux/uaccess.h:154 [inline]\n put_cmsg+0x233/0x3f0 net/core/scm.c:242\n sock_recv_errqueue+0x200/0x3e0 net/core/sock.c:2913\n packet_recvmsg+0xb2e/0x17a0 net/packet/af_packet.c:3296\n sock_recvmsg_nosec net/socket.c:803 [inline]\n sock_recvmsg+0xc9/0x110 net/socket.c:810\n ___sys_recvmsg+0x2a4/0x640 net/socket.c:2179\n __sys_recvmmsg+0x2a9/0xaf0 net/socket.c:2287\n SYSC_recvmmsg net/socket.c:2368 [inline]\n SyS_recvmmsg+0xc4/0x160 net/socket.c:2352\n entry_SYSCALL_64_fastpath+0x29/0xa0\n\nReported-by: syzbot+e2d6cfb305e9f3911dea@syzkaller.appspotmail.com\nFixes: 6d07d1cd300f (\"usercopy: Restrict non-usercopy caches to size 0\")\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e728789c52afccc1275cba1dd812f03abe16ea3c", "tree": "aa3bade1d059d3eaeaceb4d884dcbd75fe8f7084", "parents": [ "ec95dffa408f0c24c0b358f3723c6ba262190965" ], "author": { "name": "Mathieu Malaterre", "email": "malat@debian.org", "time": "Wed Feb 07 20:35:00 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:13:30 2018 -0500" }, "message": "net: Extra \u0027_get\u0027 in declaration of arch_get_platform_mac_address\n\nIn commit c7f5d105495a (\"net: Add eth_platform_get_mac_address() helper.\"),\ntwo declarations were added:\n\n int eth_platform_get_mac_address(struct device *dev, u8 *mac_addr);\n unsigned char *arch_get_platform_get_mac_address(void);\n\nAn extra \u0027_get\u0027 was introduced in arch_get_platform_get_mac_address, remove\nit. Fix compile warning using W\u003d1:\n\n CC net/ethernet/eth.o\nnet/ethernet/eth.c:523:24: warning: no previous prototype for ‘arch_get_platform_mac_address’ [-Wmissing-prototypes]\n unsigned char * __weak arch_get_platform_mac_address(void)\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n AR net/ethernet/built-in.o\n\nSigned-off-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "81153336eb76b253ba7852f3f1de525bb98f8c4d", "tree": "635fe0495b762841eb72e5ee40d9d88563674691", "parents": [ "ef9417e8a903d3a68a83ea2da32f1db030341c37", "4d673da14533b32fe8d3125b5b7be4fea14e39a8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 12:12:04 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 12:12:04 2018 -0800" }, "message": "Merge tag \u0027afs-next-20180208\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs\n\nPull afs updates from David Howells:\n \"Four fixes:\n\n - add a missing put\n\n - two fixes to reset the address iteration cursor correctly\n\n - fix setting up the fileserver iteration cursor.\n\n Two cleanups:\n\n - remove some dead code\n\n - rearrange a function to be more logically laid out\n\n And one new feature:\n\n - Support AFS dynamic root.\n\n With this one should be able to do, say:\n\n mkdir /afs\n mount -t afs none /afs -o dyn\n\n to create a dynamic root and then, provided you have keyutils\n installed, do:\n\n ls /afs/grand.central.org\n\n and:\n\n ls /afs/umich.edu\n\n to list the root volumes of both those organisations\u0027 AFS cells\n without requiring any other setup (the kernel upcall to a program\n in the keyutils package to do DNS access as does NFS)\"\n\n* tag \u0027afs-next-20180208\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs:\n afs: Support the AFS dynamic root\n afs: Rearrange afs_select_fileserver() a little\n afs: Remove unused code\n afs: Fix server list handling\n afs: Need to clear responded flag in addr cursor\n afs: Fix missing cursor clearance\n afs: Add missing afs_put_cell()\n" }, { "commit": "ec95dffa408f0c24c0b358f3723c6ba262190965", "tree": "4e99e52a2abb59aebb15c3d2bb57ad23fe95f2b6", "parents": [ "583133b35ecaea84590e92e2732d1a472747ca7d" ], "author": { "name": "Nathan Fontenot", "email": "nfont@linux.vnet.ibm.com", "time": "Wed Feb 07 13:00:24 2018 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:11:15 2018 -0500" }, "message": "ibmvnic: queue reset when CRQ gets closed during reset\n\nWhile handling a driver reset we get a H_CLOSED return trying\nto send a CRQ event. When this occurs we need to queue up another\nreset attempt. Without doing this we see instances where the driver\nis left in a closed state because the reset failed and there is no\nfurther attempts to reset the driver.\n\nSigned-off-by: Nathan Fontenot \u003cnfont@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "583133b35ecaea84590e92e2732d1a472747ca7d", "tree": "2f9ec72915048e04e8cc6ee0bfa3e423b3025635", "parents": [ "4ff66cae7f10b65b028dc3bdaaad9cc2989ef6ae" ], "author": { "name": "Gustavo A. R. Silva", "email": "gustavo@embeddedor.com", "time": "Wed Feb 07 10:17:29 2018 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 15:05:16 2018 -0500" }, "message": "atm: he: use 64-bit arithmetic instead of 32-bit\n\nAdd suffix ULL to constants 272, 204, 136 and 68 in order to give the\ncompiler complete information about the proper arithmetic to use.\nNotice that these constants are used in contexts that expect\nexpressions of type unsigned long long (64 bits, unsigned).\n\nThe following expressions are currently being evaluated using 32-bit\narithmetic:\n\n272 * mult\n204 * mult\n136 * mult\n68 * mult\n\nAddresses-Coverity-ID: 201058\nSigned-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ef9417e8a903d3a68a83ea2da32f1db030341c37", "tree": "731162b8b3b26707ae98ab40830433951df52ca1", "parents": [ "605dc7761d2701f73c17183649de0e3044609817", "fedbd940d1c2e37e6b11cf84f67b30dfcbf6fba5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 12:03:54 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 12:03:54 2018 -0800" }, "message": "Merge tag \u0027iommu-updates-v4.16\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu\n\nPull IOMMU updates from Joerg Roedel:\n \"This time there are not a lot of changes coming from the IOMMU side.\n\n That is partly because I returned from my parental leave late in the\n development process and probably partly because everyone was busy with\n Spectre and Meltdown mitigation work and didn\u0027t find the time for\n IOMMU work. So here are the few changes that queued up for this merge\n window:\n\n - 5-level page-table support for the Intel IOMMU.\n\n - error reporting improvements for the AMD IOMMU driver\n\n - additional DT bindings for ipmmu-vmsa (Renesas)\n\n - small fixes and cleanups\"\n\n* tag \u0027iommu-updates-v4.16\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:\n iommu: Clean up of_iommu_init_fn\n iommu/ipmmu-vmsa: Remove redundant of_iommu_init_fn hook\n iommu/msm: Claim bus ops on probe\n iommu/vt-d: Enable 5-level paging mode in the PASID entry\n iommu/vt-d: Add a check for 5-level paging support\n iommu/vt-d: Add a check for 1GB page support\n iommu/vt-d: Enable upto 57 bits of domain address width\n iommu/vt-d: Use domain instead of cache fetching\n iommu/exynos: Don\u0027t unconditionally steal bus ops\n iommu/omap: Fix debugfs_create_*() usage\n iommu/vt-d: clean up pr_irq if request_threaded_irq fails\n iommu: Check the result of iommu_group_get() for NULL\n iommu/ipmmu-vmsa: Add r8a779(70|95) DT bindings\n iommu/ipmmu-vmsa: Add r8a7796 DT binding\n iommu/amd: Set the device table entry PPR bit for IOMMU V2 devices\n iommu/amd - Record more information about unknown events\n" }, { "commit": "605dc7761d2701f73c17183649de0e3044609817", "tree": "2fc9ed48d50e0552f5f48d23c725930bb950973f", "parents": [ "fe26adf431a58d620819618e52a10bf9b5cfde52", "d3fdd701248290b511683ff8d348073295f07833" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 11:48:49 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 11:48:49 2018 -0800" }, "message": "Merge branch \u0027pcmcia\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/brodo/pcmcia\n\nPull pcmcia updates from Dominik Brodowski:\n \"The linux-pcmcia mailing list was shut down, so offer an alternative\n path for patches in MAINTAINERS.\n\n Also, throw in two odd fixes for the pcmcia subsystem\"\n\n* \u0027pcmcia\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/brodo/pcmcia:\n pcmcia: soc_common: Handle return value of clk_prepare_enable\n pcmcia: use proper printk format for resource\n pcmcia: remove mailing list, update MAINTAINERS\n" }, { "commit": "fe26adf431a58d620819618e52a10bf9b5cfde52", "tree": "7e5fd39db303f0291f24c6054f51f8a1121bd587", "parents": [ "9e95dae76b53e67b64bb8e8468d2285b1dc34720", "94fc27ac487a80daf42f97b1a0503d029f3c1325" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 11:42:05 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 11:42:05 2018 -0800" }, "message": "Merge tag \u0027drm-for-v4.16-part2-fixes\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull more drm updates from Dave Airlie:\n \"Ben missed sending his nouveau tree, but he really didn\u0027t have much\n stuff in it:\n\n - GP108 acceleration support is enabled by \"secure boot\" support\n\n - some clockgating work on Kepler, and bunch of fixes\n\n - the bulk of the diff is regenerated firmware files, the change to\n them really isn\u0027t that large.\n\n Otherwise this contains regular Intel and AMDGPU fixes\"\n\n* tag \u0027drm-for-v4.16-part2-fixes\u0027 of git://people.freedesktop.org/~airlied/linux: (59 commits)\n drm/i915/bios: add DP max link rate to VBT child device struct\n drm/i915/cnp: Properly handle VBT ddc pin out of bounds.\n drm/i915/cnp: Ignore VBT request for know invalid DDC pin.\n drm/i915/cmdparser: Do not check past the cmd length.\n drm/i915/cmdparser: Check reg_table_count before derefencing.\n drm/i915/bxt, glk: Increase PCODE timeouts during CDCLK freq changing\n drm/i915/gvt: Use KVM r/w to access guest opregion\n drm/i915/gvt: Fix aperture read/write emulation when enable x-no-mmap\u003don\n drm/i915/gvt: only reset execlist state of one engine during VM engine reset\n drm/i915/gvt: refine intel_vgpu_submission_ops as per engine ops\n drm/amdgpu: re-enable CGCG on CZ and disable on ST\n drm/nouveau/clk: fix gcc-7 -Wint-in-bool-context warning\n drm/nouveau/mmu: Fix trailing semicolon\n drm/nouveau: Introduce NvPmEnableGating option\n drm/nouveau: Add support for SLCG for Kepler2\n drm/nouveau: Add support for BLCG on Kepler2\n drm/nouveau: Add support for BLCG on Kepler1\n drm/nouveau: Add support for basic clockgating on Kepler1\n drm/nouveau/kms/nv50: fix handling of gamma since atomic conversion\n drm/nouveau/kms/nv50: use INTERPOLATE_257_UNITY_RANGE LUT on newer chipsets\n ...\n" }, { "commit": "9e95dae76b53e67b64bb8e8468d2285b1dc34720", "tree": "482551f8e837b34486d20b3c2273bf5f89b2000b", "parents": [ "a8c6db00bff2bdb466d80b7ec7ff25d327071f9b", "16515a6d54183349b858b9c05e483afc55fa7948" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 11:38:59 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 11:38:59 2018 -0800" }, "message": "Merge tag \u0027ceph-for-4.16-rc1\u0027 of git://github.com/ceph/ceph-client\n\nPull ceph updates from Ilya Dryomov:\n \"Things have been very quiet on the rbd side, as work continues on the\n big ticket items slated for the next merge window.\n\n On the CephFS side we have a large number of cap handling\n improvements, a fix for our long-standing abuse of -\u003ejournal_info in\n ceph_readpages() and yet another dentry pointer management patch\"\n\n* tag \u0027ceph-for-4.16-rc1\u0027 of git://github.com/ceph/ceph-client:\n ceph: improving efficiency of syncfs\n libceph: check kstrndup() return value\n ceph: try to allocate enough memory for reserved caps\n ceph: fix race of queuing delayed caps\n ceph: delete unreachable code in ceph_check_caps()\n ceph: limit rate of cap import/export error messages\n ceph: fix incorrect snaprealm when adding caps\n ceph: fix un-balanced fsc-\u003ewriteback_count update\n ceph: track read contexts in ceph_file_info\n ceph: avoid dereferencing invalid pointer during cached readdir\n ceph: use atomic_t for ceph_inode_info::i_shared_gen\n ceph: cleanup traceless reply handling for rename\n ceph: voluntarily drop Fx cap for readdir request\n ceph: properly drop caps for setattr request\n ceph: voluntarily drop Lx cap for link/rename requests\n ceph: voluntarily drop Ax cap for requests that create new inode\n rbd: whitelist RBD_FEATURE_OPERATIONS feature bit\n rbd: don\u0027t NULL out -\u003eobj_request in rbd_img_obj_parent_read_full()\n rbd: use kmem_cache_zalloc() in rbd_img_request_create()\n rbd: obj_request-\u003ecompletion is unused\n" }, { "commit": "a8c6db00bff2bdb466d80b7ec7ff25d327071f9b", "tree": "04704a2f851249f8ddfd0b8fe95fdd8efaf6d202", "parents": [ "c0136321924dd338bb8fc5661c4b0e27441a8d04" ], "author": { "name": "Nicolas Pitre", "email": "nicolas.pitre@linaro.org", "time": "Wed Feb 07 22:08:08 2018 -0500" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 11:37:31 2018 -0800" }, "message": "cramfs: better MTD dependency expression\n\nCommit b9f5fb1800d8 (\"cramfs: fix MTD dependency\") did what it says.\n\nSince commit 9059a3493efe (\"kconfig: fix relational operators for bool\nand tristate symbols\") it is possible to do it slightly better though.\n\nSigned-off-by: Nicolas Pitre \u003cnico@linaro.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "4ff66cae7f10b65b028dc3bdaaad9cc2989ef6ae", "tree": "18b63a85d9a6bbcfcb3674e0c57e684f8a671104", "parents": [ "762c330d670e3d4b795cf7a8d761866fdd1eef49" ], "author": { "name": "Christian Brauner", "email": "christian.brauner@ubuntu.com", "time": "Wed Feb 07 13:53:20 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 14:33:20 2018 -0500" }, "message": "rtnetlink: require unique netns identifier\n\nSince we\u0027ve added support for IFLA_IF_NETNSID for RTM_{DEL,GET,SET,NEW}LINK\nit is possible for userspace to send us requests with three different\nproperties to identify a target network namespace. This affects at least\nRTM_{NEW,SET}LINK. Each of them could potentially refer to a different\nnetwork namespace which is confusing. For legacy reasons the kernel will\npick the IFLA_NET_NS_PID property first and then look for the\nIFLA_NET_NS_FD property but there is no reason to extend this type of\nbehavior to network namespace ids. The regression potential is quite\nminimal since the rtnetlink requests in question either won\u0027t allow\nIFLA_IF_NETNSID requests before 4.16 is out (RTM_{NEW,SET}LINK) or don\u0027t\nsupport IFLA_NET_NS_{PID,FD} (RTM_{DEL,GET}LINK) in the first place.\n\nSigned-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\nAcked-by: Jiri Benc \u003cjbenc@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "762c330d670e3d4b795cf7a8d761866fdd1eef49", "tree": "249dbef044fc3ecf0b3f2d449e63080e9f802359", "parents": [ "cb9f7a9a5c96a773bbc9c70660dc600cfff82f82" ], "author": { "name": "Jason Wang", "email": "jasowang@redhat.com", "time": "Wed Feb 07 17:14:46 2018 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 14:10:30 2018 -0500" }, "message": "tuntap: add missing xdp flush\n\nWhen using devmap to redirect packets between interfaces,\nxdp_do_flush() is usually a must to flush any batched\npackets. Unfortunately this is missed in current tuntap\nimplementation.\n\nUnlike most hardware driver which did XDP inside NAPI loop and call\nxdp_do_flush() at then end of each round of poll. TAP did it in the\ncontext of process e.g tun_get_user(). So fix this by count the\npending redirected packets and flush when it exceeds NAPI_POLL_WEIGHT\nor MSG_MORE was cleared by sendmsg() caller.\n\nWith this fix, xdp_redirect_map works again between two TAPs.\n\nFixes: 761876c857cb (\"tap: XDP support\")\nSigned-off-by: Jason Wang \u003cjasowang@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9e3e10c725360b9d07018cfcd5b7b6b7d325fae5", "tree": "c3de9ca1974443fb36319e74ac0b7b893a3a39eb", "parents": [ "f3ff6fb5db68bcd460e9880d5fb4902520dd645b" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Tue Feb 06 09:34:41 2018 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Fri Feb 09 04:10:10 2018 +0900" }, "message": "kconfig: send error messages to stderr\n\nThese messages should be directed to stderr.\n\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nReviewed-by: Ulf Magnusson \u003culfalizer@gmail.com\u003e\n" }, { "commit": "f3ff6fb5db68bcd460e9880d5fb4902520dd645b", "tree": "40c06df6cc38a07f8e4daebeebf0d35bd8d32e27", "parents": [ "d2a04648a5dbc3d1d043b35257364f0197d4d868" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Thu Feb 08 14:56:40 2018 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Fri Feb 09 04:10:10 2018 +0900" }, "message": "kconfig: echo stdin to stdout if either is redirected\n\nIf stdio is not tty, conf_askvalue() puts additional new line to\nprevent prompts from being concatenated into a single line. This\ncare is missing in conf_choice(), so a \u0027choice\u0027 prompt and the next\nprompt are shown in the same line.\n\nMove the code into xfgets() to cater to all cases. To improve this\nmore, let\u0027s echo stdin to stdout. This clarifies what keys were\ninput from stdio and the stdout looks like as if it were from tty.\n\nI removed the isatty(2) check since stderr is unrelated here.\n\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nReviewed-by: Ulf Magnusson \u003culfalizer@gmail.com\u003e\n" }, { "commit": "d2a04648a5dbc3d1d043b35257364f0197d4d868", "tree": "ce691395a58e1a965143d0207407f5da39fdb848", "parents": [ "cd58a91def2accba1a07960889e89a63d372119a" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Thu Feb 08 14:56:39 2018 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Fri Feb 09 04:10:10 2018 +0900" }, "message": "kconfig: remove check_stdin()\n\nExcept silentoldconfig, valid_stdin is 1, so check_stdin() is no-op.\n\noldconfig and silentoldconfig work almost in the same way except that\nthe latter generates additional files under include/. Both ask users\nfor input for new symbols.\n\nI do not know why only silentoldconfig requires stdio be tty.\n\n $ rm -f .config; touch .config\n $ yes \"\" | make oldconfig \u003e stdout\n $ rm -f .config; touch .config\n $ yes \"\" | make silentoldconfig \u003e stdout\n make[1]: *** [silentoldconfig] Error 1\n make: *** [silentoldconfig] Error 2\n $ tail -n 4 stdout\n Console input/output is redirected. Run \u0027make oldconfig\u0027 to update configuration.\n\n scripts/kconfig/Makefile:40: recipe for target \u0027silentoldconfig\u0027 failed\n Makefile:507: recipe for target \u0027silentoldconfig\u0027 failed\n\nRedirection is useful, for example, for testing where we want to give\nparticular key inputs from a test file, then check the result.\n\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nReviewed-by: Ulf Magnusson \u003culfalizer@gmail.com\u003e\n" }, { "commit": "cd58a91def2accba1a07960889e89a63d372119a", "tree": "2ef42064705ddb95e1b5948f02a955ec69c5a897", "parents": [ "4f208f392103e8d5bcc7558dba69348b27d8ce42" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Tue Feb 06 09:34:45 2018 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Fri Feb 09 04:10:09 2018 +0900" }, "message": "kconfig: remove \u0027config*\u0027 pattern from .gitignnore\n\nI could not figure out why this pattern should be ignored.\nChecking commit 1e65174a3378 (\"Add some basic .gitignore files\")\ndid not help.\n\nLet\u0027s remove this pattern, then see if it is really needed.\n\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nReviewed-by: Ulf Magnusson \u003culfalizer@gmail.com\u003e\n" }, { "commit": "4f208f392103e8d5bcc7558dba69348b27d8ce42", "tree": "829ca771d530086527ecb69abfe03d2c63200ba7", "parents": [ "cb67ab2cd2b8abd9650292c986c79901e3073a59" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Tue Feb 06 09:34:43 2018 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Fri Feb 09 04:10:09 2018 +0900" }, "message": "kconfig: show \u0027?\u0027 prompt even if no help text is available\n\n\u0027make config\u0027, \u0027make oldconfig\u0027, etc. always receive \u0027?\u0027 as a valid\ninput and show useful information even if no help text is available.\n\n------------------------\u003e8------------------------\nfoo (FOO) [N/y] (NEW) ?\n\nThere is no help available for this option.\nSymbol: FOO [\u003dn]\nType : bool\nPrompt: foo\n Defined at Kconfig:1\n------------------------\u003e8------------------------\n\nHowever, \u0027?\u0027 is not shown in the prompt if its help text is missing.\nLet\u0027s show \u0027?\u0027 all the time so that the prompt and the behavior match.\n\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nReviewed-by: Ulf Magnusson \u003culfalizer@gmail.com\u003e\n" }, { "commit": "cb67ab2cd2b8abd9650292c986c79901e3073a59", "tree": "6f0f2968c93697b981b195a08fd0c830a13c89d5", "parents": [ "a2b0fe7435faee6f6fbb27409878013bc4727e98" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Tue Feb 06 09:34:42 2018 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Fri Feb 09 04:08:05 2018 +0900" }, "message": "kconfig: do not write choice values when their dependency becomes n\n\n\"# CONFIG_... is not set\" for choice values are wrongly written into\nthe .config file if they are once visible, then become invisible later.\n\n Test case\n ---------\n\n---------------------------(Kconfig)----------------------------\nconfig A\n\tbool \"A\"\n\nchoice\n\tprompt \"Choice ?\"\n\tdepends on A\n\nconfig CHOICE_B\n\tbool \"Choice B\"\n\nconfig CHOICE_C\n\tbool \"Choice C\"\n\nendchoice\n----------------------------------------------------------------\n\n---------------------------(.config)----------------------------\nCONFIG_A\u003dy\n----------------------------------------------------------------\n\nWith the Kconfig and .config above,\n\n $ make config\n scripts/kconfig/conf --oldaskconfig Kconfig\n *\n * Linux Kernel Configuration\n *\n A (A) [Y/n] n\n #\n # configuration written to .config\n #\n $ cat .config\n #\n # Automatically generated file; DO NOT EDIT.\n # Linux Kernel Configuration\n #\n # CONFIG_A is not set\n # CONFIG_CHOICE_B is not set\n # CONFIG_CHOICE_C is not set\n\nHere,\n\n # CONFIG_CHOICE_B is not set\n # CONFIG_CHOICE_C is not set\n\nshould not be written into the .config file because their dependency\n\"depends on A\" is unmet.\n\nCurrently, there is no code that clears SYMBOL_WRITE of choice values.\n\nClear SYMBOL_WRITE for all symbols in sym_calc_value(), then set it\nagain after calculating visibility. To simplify the logic, set the\nflag if they have non-n visibility, regardless of types, and regardless\nof whether they are choice values or not.\n\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nReviewed-by: Ulf Magnusson \u003culfalizer@gmail.com\u003e\n" }, { "commit": "cb9f7a9a5c96a773bbc9c70660dc600cfff82f82", "tree": "bbd5a31b8a13d94e94321dcf56960e6f91f88000", "parents": [ "8c2f826dc36314059ac146c78d3bf8056b626446" ], "author": { "name": "Nicolas Dichtel", "email": "nicolas.dichtel@6wind.com", "time": "Tue Feb 06 14:48:32 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 14:03:18 2018 -0500" }, "message": "netlink: ensure to loop over all netns in genlmsg_multicast_allns()\n\nNowadays, nlmsg_multicast() returns only 0 or -ESRCH but this was not the\ncase when commit 134e63756d5f was pushed.\nHowever, there was no reason to stop the loop if a netns does not have\nlisteners.\nReturns -ESRCH only if there was no listeners in all netns.\n\nTo avoid having the same problem in the future, I didn\u0027t take the\nassumption that nlmsg_multicast() returns only 0 or -ESRCH.\n\nFixes: 134e63756d5f (\"genetlink: make netns aware\")\nCC: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nSigned-off-by: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8c2f826dc36314059ac146c78d3bf8056b626446", "tree": "21719e202e2e3f0ada53fb30562a72cdec238842", "parents": [ "c70255868148a498ba418bc6c2f9df212d30d393" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Thu Feb 08 15:59:07 2018 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 08 13:48:29 2018 -0500" }, "message": "rxrpc: Don\u0027t put crypto buffers on the stack\n\nDon\u0027t put buffers of data to be handed to crypto on the stack as this may\ncause an assertion failure in the kernel (see below). Fix this by using an\nkmalloc\u0027d buffer instead.\n\nkernel BUG at ./include/linux/scatterlist.h:147!\n...\nRIP: 0010:rxkad_encrypt_response.isra.6+0x191/0x1b0 [rxrpc]\nRSP: 0018:ffffbe2fc06cfca8 EFLAGS: 00010246\nRAX: 0000000000000000 RBX: ffff989277d59900 RCX: 0000000000000028\nRDX: 0000259dc06cfd88 RSI: 0000000000000025 RDI: ffffbe30406cfd88\nRBP: ffffbe2fc06cfd60 R08: ffffbe2fc06cfd08 R09: ffffbe2fc06cfd08\nR10: 0000000000000000 R11: 0000000000000000 R12: 1ffff7c5f80d9f95\nR13: ffffbe2fc06cfd88 R14: ffff98927a3f7aa0 R15: ffffbe2fc06cfd08\nFS: 0000000000000000(0000) GS:ffff98927fc00000(0000) knlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nCR2: 000055b1ff28f0f8 CR3: 000000001b412003 CR4: 00000000003606f0\nDR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\nDR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\nCall Trace:\n rxkad_respond_to_challenge+0x297/0x330 [rxrpc]\n rxrpc_process_connection+0xd1/0x690 [rxrpc]\n ? process_one_work+0x1c3/0x680\n ? __lock_is_held+0x59/0xa0\n process_one_work+0x249/0x680\n worker_thread+0x3a/0x390\n ? process_one_work+0x680/0x680\n kthread+0x121/0x140\n ? kthread_create_worker_on_cpu+0x70/0x70\n ret_from_fork+0x3a/0x50\n\nReported-by: Jonathan Billings \u003cjsbillings@jsbillings.org\u003e\nReported-by: Marc Dionne \u003cmarc.dionne@auristor.com\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nTested-by: Jonathan Billings \u003cjsbillings@jsbillings.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c0136321924dd338bb8fc5661c4b0e27441a8d04", "tree": "099cc60dfe6793309b21474a00e4d10087a7cae8", "parents": [ "846ade7dd2e630a309a8c57302046e8c4037b8df", "3a0a397ff5ff8b56ca9f7908b75dee6bf0b5fabb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 10:44:25 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 10:44:25 2018 -0800" }, "message": "Merge tag \u0027arm64-upstream\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux\n\nPull more arm64 updates from Catalin Marinas:\n \"As I mentioned in the last pull request, there\u0027s a second batch of\n security updates for arm64 with mitigations for Spectre/v1 and an\n improved one for Spectre/v2 (via a newly defined firmware interface\n API).\n\n Spectre v1 mitigation:\n\n - back-end version of array_index_mask_nospec()\n\n - masking of the syscall number to restrict speculation through the\n syscall table\n\n - masking of __user pointers prior to deference in uaccess routines\n\n Spectre v2 mitigation update:\n\n - using the new firmware SMC calling convention specification update\n\n - removing the current PSCI GET_VERSION firmware call mitigation as\n vendors are deploying new SMCCC-capable firmware\n\n - additional branch predictor hardening for synchronous exceptions\n and interrupts while in user mode\n\n Meltdown v3 mitigation update:\n\n - Cavium Thunder X is unaffected but a hardware erratum gets in the\n way. The kernel now starts with the page tables mapped as global\n and switches to non-global if kpti needs to be enabled.\n\n Other:\n\n - Theoretical trylock bug fixed\"\n\n* tag \u0027arm64-upstream\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (38 commits)\n arm64: Kill PSCI_GET_VERSION as a variant-2 workaround\n arm64: Add ARM_SMCCC_ARCH_WORKAROUND_1 BP hardening support\n arm/arm64: smccc: Implement SMCCC v1.1 inline primitive\n arm/arm64: smccc: Make function identifiers an unsigned quantity\n firmware/psci: Expose SMCCC version through psci_ops\n firmware/psci: Expose PSCI conduit\n arm64: KVM: Add SMCCC_ARCH_WORKAROUND_1 fast handling\n arm64: KVM: Report SMCCC_ARCH_WORKAROUND_1 BP hardening support\n arm/arm64: KVM: Turn kvm_psci_version into a static inline\n arm/arm64: KVM: Advertise SMCCC v1.1\n arm/arm64: KVM: Implement PSCI 1.0 support\n arm/arm64: KVM: Add smccc accessors to PSCI code\n arm/arm64: KVM: Add PSCI_VERSION helper\n arm/arm64: KVM: Consolidate the PSCI include files\n arm64: KVM: Increment PC after handling an SMC trap\n arm: KVM: Fix SMCCC handling of unimplemented SMC/HVC calls\n arm64: KVM: Fix SMCCC handling of unimplemented SMC/HVC calls\n arm64: entry: Apply BP hardening for suspicious interrupts from EL0\n arm64: entry: Apply BP hardening for high-priority synchronous exceptions\n arm64: futex: Mask __user pointers prior to dereference\n ...\n" }, { "commit": "846ade7dd2e630a309a8c57302046e8c4037b8df", "tree": "021892148643db7cd0c898ad5939ecc7836a7330", "parents": [ "977e41524dae8fed9c82e3dd298f3b48282fc0b8", "d25cc43c6775bff6b8e3dad97c747954b805e421" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 10:41:00 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 08 10:41:00 2018 -0800" }, "message": "Merge tag \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost\n\nPull virtio/vhost updates from Michael Tsirkin:\n \"virtio, vhost: fixes, cleanups, features\n\n This includes the disk/cache memory stats for for the virtio balloon,\n as well as multiple fixes and cleanups\"\n\n* tag \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:\n vhost: don\u0027t hold onto file pointer for VHOST_SET_LOG_FD\n vhost: don\u0027t hold onto file pointer for VHOST_SET_VRING_ERR\n vhost: don\u0027t hold onto file pointer for VHOST_SET_VRING_CALL\n ringtest: ring.c malloc \u0026 memset to calloc\n virtio_vop: don\u0027t kfree device on register failure\n virtio_pci: don\u0027t kfree device on register failure\n virtio: split device_register into device_initialize and device_add\n vhost: remove unused lock check flag in vhost_dev_cleanup()\n vhost: Remove the unused variable.\n virtio_blk: print capacity at probe time\n virtio: make VIRTIO a menuconfig to ease disabling it all\n virtio/ringtest: virtio_ring: fix up need_event math\n virtio/ringtest: fix up need_event math\n virtio: virtio_mmio: make of_device_ids const.\n firmware: Use PTR_ERR_OR_ZERO()\n virtio-mmio: Use PTR_ERR_OR_ZERO()\n vhost/scsi: Improve a size determination in four functions\n virtio_balloon: include disk/file caches memory statistics\n" }, { "commit": "175e03101d36c3034f3c80038d4c28838351a7f2", "tree": "a75cafad97a96181e570cb5914e06e62aa5b98ba", "parents": [ "2285ae760de2cbe4b91a55c0a8033ec332fb8049" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Fri Feb 02 14:28:59 2018 -0500" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@redhat.com", "time": "Thu Feb 08 13:40:17 2018 -0500" }, "message": "svcrdma: Fix Read chunk round-up\n\nA single NFSv4 WRITE compound can often have three operations:\nPUTFH, WRITE, then GETATTR.\n\nWhen the WRITE payload is sent in a Read chunk, the client places\nthe GETATTR in the inline part of the RPC/RDMA message, just after\nthe WRITE operation (sans payload). The position value in the Read\nchunk enables the receiver to insert the Read chunk at the correct\nplace in the received XDR stream; that is between the WRITE and\nGETATTR.\n\nAccording to RFC 8166, an NFS/RDMA client does not have to add XDR\nround-up to the Read chunk that carries the WRITE payload. The\nreceiver adds XDR round-up padding if it is absent and the\nreceiver\u0027s XDR decoder requires it to be present.\n\nCommit 193bcb7b3719 (\"svcrdma: Populate tail iovec when receiving\")\nattempted to add support for receiving such a compound so that just\nthe WRITE payload appears in rq_arg\u0027s page list, and the trailing\nGETATTR is placed in rq_arg\u0027s tail iovec. (TCP just strings the\nwhole compound into the head iovec and page list, without regard\nto the alignment of the WRITE payload).\n\nThe server transport logic also had to accommodate the optional XDR\nround-up of the Read chunk, which it did simply by lengthening the\ntail iovec when round-up was needed. This approach is adequate for\nthe NFSv2 and NFSv3 WRITE decoders.\n\nUnfortunately it is not sufficient for nfsd4_decode_write. When the\nRead chunk length is a couple of bytes less than PAGE_SIZE, the\ncomputation at the end of nfsd4_decode_write allows argp-\u003epagelen to\ngo negative, which breaks the logic in read_buf that looks for the\ntail iovec.\n\nThe result is that a WRITE operation whose payload length is just\nless than a multiple of a page succeeds, but the subsequent GETATTR\nin the same compound fails with NFS4ERR_OP_ILLEGAL because the XDR\ndecoder can\u0027t find it. Clients ignore the error, but they must\nupdate their attribute cache via a separate round trip.\n\nAs nfsd4_decode_write appears to expect the payload itself to always\nhave appropriate XDR round-up, have svc_rdma_build_normal_read_chunk\nadd the Read chunk XDR round-up to the page_len rather than\nlengthening the tail iovec.\n\nReported-by: Olga Kornievskaia \u003ckolga@netapp.com\u003e\nFixes: 193bcb7b3719 (\"svcrdma: Populate tail iovec when receiving\")\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nTested-by: Olga Kornievskaia \u003ckolga@netapp.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n" } ], "next": "2285ae760de2cbe4b91a55c0a8033ec332fb8049" }