-
Mon Mar 12 2018 Brian Maly <brian.maly@oracle.com> [4.1.12-112.16.4.el7uek]
- scsi: Make __scsi_remove_device go straight from BLOCKED to DEL (Bart Van Assche) [Orabug: 27339736]
- scsi: Protect SCSI device state changes with a mutex (Bart Van Assche) [Orabug: 27339736]
- scsi: Introduce scsi_start_queue() (Bart Van Assche) [Orabug: 27339736]
- scsi: avoid a permanent stop of the scsi device's request queue (Wei Fang) [Orabug: 27339736]
- x86/spectre_v2: Fix cpu offlining with IPBP. (Konrad Rzeszutek Wilk)
-
Thu Mar 08 2018 Chuck Anderson <chuck.anderson@oracle.com> [4.1.12-112.16.3.el7uek]
- retpoline: selectively disable IBRS in disable_ibrs_and_friends() (Chuck Anderson) [Orabug: 27636063]
-
Sat Mar 03 2018 Chuck Anderson <chuck.anderson@oracle.com> [4.1.12-112.16.2.el7uek]
- retpoline: move setting of sysctl_ibrs_enabled and sysctl_ibpb_enabled to where SPEC_CTRL_IBRS_INUSE and SPEC_CTRL_IBPB_INUSE are set (Chuck Anderson) [Orabug: 27556096]
- retpoline: set IBRS and IBPB in use only on the boot CPU call to init_scattered_cpuid_features() (Chuck Anderson) [Orabug: 27556096]
- retpoline: display IBPB feature status along with IBRS status (Chuck Anderson) [Orabug: 27556096]
- retpoline: move lock/unlock of spec_ctrl_mutex to check_modinfo() (Chuck Anderson) [Orabug: 27556096]
- retpoline: call clear_retpoline_fallback() with boot parm spectre_v2_heuristics=off (Chuck Anderson) [Orabug: 27556096]
- retpoline: add brackets to check_ibrs_inuse() and clear_ibpb_inuse() (Chuck Anderson) [Orabug: 27556096]
- retpoline/module: do not enable IBRS/IPBP if SPEC_CTRL_IBRS_ADMIN_DISABLED/SPEC_CTRL_IBPB_ADMIN_DISABLED is set (Chuck Anderson) [Orabug: 27547729]
- retpoline: microcode incorrectly reported as broken during early boot (Chuck Anderson) [Orabug: 27556096]
- retpoline: move lock/unlock of spec_ctrl_mutex into init_scattered_cpuid_features() (Chuck Anderson) [Orabug: 27556096]
- KVM: Disable irq while unregistering user notifier (Ignacio Alvarado)
- x86/speculation: Use IBRS if available before calling into firmware (David Woodhouse) [Orabug: 27516479]
- Revert "x86/spec_ctrl: Add 'nolfence' knob to disable fallback for spectre_v2 mitigation" (Konrad Rzeszutek Wilk) [Orabug: 27601790]
- Revert "x86/spec: Add 'lfence_enabled' in sysfs" (Konrad Rzeszutek Wilk)
- x86/spectre_v2: Fix the documentation to say the right thing. (Konrad Rzeszutek Wilk)
- x86/spectre_v2: Don't check bad microcode versions when running under hypervisors. (Konrad Rzeszutek Wilk) [Orabug: 27601735]
-
Wed Feb 28 2018 Brian Maly <brian.maly@oracle.com> [4.1.12-112.16.1.el7uek]
- firmware: dmi_scan: add SBMIOS entry and DMI tables (Ivan Khoronzhuk) [Orabug: 27045425]
- Btrfs: fix unexpected EEXIST from btrfs_get_extent (Liu Bo) [Orabug: 27033166]
- Btrfs: fix incorrect block_len in merge_extent_mapping (Liu Bo) [Orabug: 27033166]
- Btrfs: add WARN_ONCE to detect unexpected error from merge_extent_mapping (Liu Bo) [Orabug: 27033166]
- Btrfs: deal with existing encompassing extent map in btrfs_get_extent() (Omar Sandoval) [Orabug: 27033166]
- Btrfs: deal with duplciates during extent_map insertion in btrfs_get_extent (Chris Mason) [Orabug: 27033166]
- net: cdc_ether: fix divide by 0 on bad descriptors (Bjørn Mork) [Orabug: 27215191] {CVE-2017-16649}
- KVM: x86: pvclock: Handle first-time write to pvclock-page contains random junk (Liran Alon) [Orabug: 27221679]
- KVM: x86: always fill in vcpu->arch.hv_clock (Paolo Bonzini) [Orabug: 27221679]
- net/mlx4_core: allow QPs with enable_smi_admin enabled (Zhu Yanjun) [Orabug: 27256720]
- qlcnic: fix deadlock bug (Junxiao Bi) [Orabug: 27337130]
- x86/ia32/syscall: RESTORE_EXTRA_REGS when returning from syscall (Ankur Arora) [Orabug: 27534437] {CVE-2017-5715}
- x86/ia32/syscall: don't do RESTORE_EXTRA_REGS prematurely (Ankur Arora) [Orabug: 27534437] {CVE-2017-5715}
- trace: declare blk_add_trace_rq non-static on OL6 (Todd Vierling) [Orabug: 27578295]
-
Wed Feb 14 2018 Brian Maly <brian.maly@oracle.com> [4.1.12-112.15.1.el7uek]
- x86/spectre: move microcode check before kernel ibrs flags are set (Daniel Jordan) [Orabug: 27534557] {CVE-2017-5715}
- bonding: attempt to better support longer hw addresses (Jarod Wilson) [Orabug: 27535442]
- x86: make HAVE_FENTRY dependent on !SIMULATE_GCC44_KABI (Todd Vierling) [Orabug: 27533590]
- Revert "uek-rpm: configs: remove __stack_chk_fail from kABI list" (Todd Vierling) [Orabug: 27540429]
-
Mon Feb 12 2018 Chuck Anderson <chuck.anderson@oracle.com> [4.1.12-112.14.17.el7uek]
- retpoline/module: fall back to another spectre mitigation when disabling retpoline (Chuck Anderson) [Orabug: 27457402]
- retpoline/module: add bit defs for use_ibpb (Chuck Anderson) [Orabug: 27457402]
- x86/spectre_v2: Only use IBRS when ibrs_inuse tells us to (Konrad Rzeszutek Wilk)
- kernel: on OL6 only, simulate the gcc 4.4 kABI for __stack_chk_fail() (Todd Vierling) [Orabug: 27509404]
- uek-rpm: configs: Don't set HAVE_FENTRY on OL6 builds. (Todd Vierling) [Orabug: 27509404]
-
Sat Feb 10 2018 Brian Maly <brian.maly@oracle.com> [4.1.12-112.14.16.el7uek]
- x86/spectre_v2: Disable IBRS if spectre_v2=off (Konrad Rzeszutek Wilk)
- KVM/VMX: Allow direct access to MSR_IA32_SPEC_CTRL (KarimAllah Ahmed) [Orabug: 27525577]
- x86/spectre_v2: Remove 0xc2 from spectre_bad_microcodes (Darren Kenny) [Orabug: 27523395]
- x86/speculation: Use Indirect Branch Prediction Barrier in context switch (Tim Chen) [Orabug: 27524607]
- xenbus: track caller request id (Joao Martins) [Orabug: 27472583]
- uek-rpm: configs: remove __stack_chk_fail from kABI list (Todd Vierling)
- x86/entry: RESTORE_IBRS needs to be done under kernel CR3 (Ankur Arora) [Orabug: 27501741]
- Fix typo IBRS_ATT, which should be IBRS_ALL (redux) (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Add spectre_v2_heuristics= (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Do not disable IBPB when disabling IBRS (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/scattered: Fix the order. (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre: Favor IBRS on Skylake over retpoline (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/speculation: Fix typo IBRS_ATT, which should be IBRS_ALL (Darren Kenny) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre: Now that we expose 'stbibp' make sure it is correct. (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/cpufeatures: Clean up Spectre v2 related CPUID flags (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/speculation: Add basic IBPB (Indirect Branch Prediction Barrier) support (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/bugs: Drop one "mitigation" from dmesg (Borislav Petkov) [Orabug: 27478139] {CVE-2017-5715}
- x86/nospec: Fix header guards names (Borislav Petkov) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Don't spam the console with these: (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/cpufeature: Blacklist SPEC_CTRL/PRED_CMD on early Spectre v2 microcodes (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/cpu: Keep model defines sorted by model number (Andy Shevchenko) [Orabug: 27478139] {CVE-2017-5715}
- x86/pti: Do not enable PTI on CPUs which are not vulnerable to Meltdown (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/msr: Add definitions for new speculation control MSRs (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/cpufeatures: Add AMD feature bits for Speculation Control (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Print what options are available. (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Add VMEXIT_FILL_RSB instead of RETPOLINE (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre: If IBRS is enabled disable "Filling RSB on context switch" (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- KVM: VMX: Allow direct access to MSR_IA32_SPEC_CTRL (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Don't allow {ibrs,ipbp,lfence}_enabled to be toggled if retpoline (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre: Fix retpoline_enabled (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre: Update sysctl values if toggled only by set_{ibrs,ibpb}_disabled (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- retpoline/module: Taint kernel for missing retpoline in module (Andi Kleen) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline: Fill RSB on context switch for affected CPUs (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline: Optimize inline assembler for vmexit_fill_RSB (Andi Kleen) [Orabug: 27478139] {CVE-2017-5715}
- kprobes/x86: Disable optimizing on the function jumps to indirect thunk (Masami Hiramatsu) [Orabug: 27478139] {CVE-2017-5715}
- kprobes/x86: Blacklist indirect thunk functions for kprobes (Masami Hiramatsu) [Orabug: 27478139] {CVE-2017-5715}
- retpoline: Introduce start/end markers of indirect thunk (Masami Hiramatsu) [Orabug: 27478139] {CVE-2017-5715}
- x86/mce: Make machine check speculation protected (Thomas Gleixner) [Orabug: 27478139] {CVE-2017-5715}
- kbuild: modversions for EXPORT_SYMBOL() for asm (Nicholas Piggin) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline: Add LFENCE to the retpoline/RSB filling RSB macros (Tom Lendacky) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline: Remove compile time warning (Thomas Gleixner) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline: Fill return stack buffer on vmexit (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline/irq32: Convert assembler indirect jumps (Andi Kleen) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline/checksum32: Convert assembler indirect jumps (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline/xen: Convert Xen hypercall indirect jumps (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline/hyperv: Convert assembler indirect jumps (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline/ftrace: Convert ftrace assembler indirect jumps (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline/entry: Convert entry assembler indirect jumps (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline/crypto: Convert crypto assembler indirect jumps (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Add disable_ibrs_and_friends (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Figure out if STUFF_RSB macro needs to be used. (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre_v2: Figure out when to use IBRS. (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre: Add IBRS option. (Konrad Rzeszutek Wilk) [Orabug: 27478139] {CVE-2017-5715}
- x86/spectre: Add boot time option to select Spectre v2 mitigation (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/retpoline: Add initial retpoline support (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- kconfig.h: use __is_defined() to check if MODULE is defined (Masahiro Yamada) [Orabug: 27478139] {CVE-2017-5715}
- EXPORT_SYMBOL() for asm (Al Viro) [Orabug: 27478139] {CVE-2017-5715}
- x86/asm: Make asm/alternative.h safe from assembly (Andy Lutomirski) [Orabug: 27478139] {CVE-2017-5715}
- x86/kbuild: enable modversions for symbols exported from asm (Adam Borowski) [Orabug: 27478139] {CVE-2017-5715}
- x86/asm: Use register variable to get stack pointer value (Andrey Ryabinin) [Orabug: 27478139] {CVE-2017-5715}
- x86/mm/32: Move setup_clear_cpu_cap(X86_FEATURE_PCID) earlier (Andy Lutomirski) [Orabug: 27478139] {CVE-2017-5715}
- x86/alternatives: Add missing '
' at end of ALTERNATIVE inline asm (David Woodhouse) [Orabug: 27478139] {CVE-2017-5715}
- x86/alternatives: Fix optimize_nops() checking (Borislav Petkov) [Orabug: 27478139] {CVE-2017-5715}
- x86/microcode/intel: Extend BDW late-loading with a revision check (Jia Zhang) [Orabug: 27343573]
- x86/microcode/intel: Disable late loading on model 79 (Borislav Petkov) [Orabug: 27343573]
- x86/spec: Fix spectre_v1 bug and mitigation indicators (John Haxby) [Orabug: 27470707]
- x86: Fix compile issues if CONFIG_XEN not defined (Konrad Rzeszutek Wilk)
- x86/mitigation/spectre_v2: Add reporting of 'lfence' (Konrad Rzeszutek Wilk)
- x86/spec: Add 'lfence_enabled' in sysfs (Konrad Rzeszutek Wilk)
- x86/spec_ctrl: Add 'nolfence' knob to disable fallback for spectre_v2 mitigation (Konrad Rzeszutek Wilk)
- x86/IBRS/IBPB: Remove procfs interface to ibrs/ibpb_enable (Boris Ostrovsky) [Orabug: 27415916]
- x86: Use PRED_CMD MSR when ibpb is enabled (Konrad Rzeszutek Wilk)
-
Thu Feb 08 2018 Brian Maly <brian.maly@oracle.com> [4.1.12-112.14.15.el7uek]
- nvme-pci: Remove nvme_setup_prps BUG_ON (Keith Busch) [Orabug: 27448367]
- block: Check for gaps on front and back merges (Jens Axboe) [Orabug: 27448367]
- block: Copy a user iovec if it includes gaps (Sagi Grimberg) [Orabug: 27448367]
- block: Replace SG_GAPS with new queue limits mask (Keith Busch) [Orabug: 27448367]
-
Thu Feb 01 2018 Brian Maly <brian.maly@oracle.com> [4.1.12-112.14.14.el7uek]
- drivers/char/mem.c: deny access in open operation when securelevel is set (Ethan Zhao) [Orabug: 27234850] [Orabug: 27234850]
- hugetlb: fix nr_pmds accounting with shared page tables (Kirill A. Shutemov) [Orabug: 26988581]
- x86/IBRS: Drop unnecessary WRITE_ONCE (Boris Ostrovsky) [Orabug: 27416198]
- x86/IBRS: Don't try to change IBRS mode if IBRS is not available (Boris Ostrovsky) [Orabug: 27416198]
- x86/IBRS: Remove support for IBRS_ENABLED_USER mode (Boris Ostrovsky) [Orabug: 27416198]
- x86: Include linux/device.h in bugs_64.c (Boris Ostrovsky) [Orabug: 27418896]
- x86/spectre: Drop the warning about ibrs being obsolete. (Konrad Rzeszutek Wilk)
- x86/spec: Don't print the Missing arguments for option spectre_v2. (Konrad Rzeszutek Wilk)
- x86/spec: Also print IBRS if IBPB is disabled. (Konrad Rzeszutek Wilk)
- x86/IBPB: Provide debugfs interface for changing IBPB mode (Boris Ostrovsky) [Orabug: 27449065]
- xen: Make PV Dom0 Linux kernel NUMA aware (Elena Ufimtseva)
- net/rds: Fix incorrect error handling (Håkon Bugge) [Orabug: 26848729]
- net/rds: use multiple sge than buddy allocation in congestion code (Wei Lin Guay) [Orabug: 26848729]
- Revert "RDS: fix the sg allocation based on actual message size" (Wei Lin Guay) [Orabug: 26848729]
- Revert "RDS: avoid large pages for sg allocation for TCP transport" (Wei Lin Guay) [Orabug: 26848729]
- Revert "net/rds: Reduce memory footprint in rds_sendmsg" (Wei Lin Guay) [Orabug: 26848729]
- net/rds: reduce memory footprint during ib_post_recv in IB transport (Wei Lin Guay) [Orabug: 26848729]
- net/rds: reduce memory footprint during rds_sendmsg with IB transport (Wei Lin Guay) [Orabug: 26848729]
- net/rds: set the rds_ib_init_frag based on supported sge (Wei Lin Guay) [Orabug: 26848729]
- bnxt_en: Fix possible corrupted NVRAM parameters from firmware response. (Michael Chan) [Orabug: 27199588]
- x86, kasan: Fix build failure on KASAN=y && KMEMCHECK=y kernels (Andrey Ryabinin) [Orabug: 27255122]
- x86, efi, kasan: Fix build failure on !KASAN && KMEMCHECK=y kernels (Andrey Ryabinin) [Orabug: 27255122]
- x86, efi, kasan: #undef memset/memcpy/memmove per arch (Andrey Ryabinin) [Orabug: 27255122]
- Revert "Makefile: Build with -Werror=date-time if the compiler supports it" (Gayatri Vasudevan) [Orabug: 27255122]
- dccp: CVE-2017-8824: use-after-free in DCCP code (Mohamed Ghannam) [Orabug: 27290300] {CVE-2017-8824}
- x86/efi: Initialize and display UEFI secure boot state a bit later during init (Daniel Kiper) [Orabug: 27309477]
- x86/espfix: Init espfix on the boot CPU side (Zhu Guihua) [Orabug: 27344552]
- x86/espfix: Add 'cpu' parameter to init_espfix_ap() (Zhu Guihua) [Orabug: 27344552]
- ALSA: pcm: prevent UAF in snd_pcm_info (Robb Glasser) [Orabug: 27344841] {CVE-2017-0861} {CVE-2017-0861}
- fs/ocfs2: remove page cache for converted direct write (Wengang Wang)
- Revert "ocfs2: code clean up for direct io" (Wengang Wang)
- assoc_array: Fix a buggy node-splitting case (David Howells) [Orabug: 27364592] {CVE-2017-12193} {CVE-2017-12193}
- Sanitize 'move_pages()' permission checks (Linus Torvalds) [Orabug: 27364690] {CVE-2017-14140}
- pti: compile fix for when PTI is disabled (Pavel Tatashin) [Orabug: 27383147] {CVE-2017-5754}
- sctp: do not peel off an assoc from one netns to another one (Xin Long) [Orabug: 27386999] {CVE-2017-15115}
- net: ipv4: fix for a race condition in raw_sendmsg (Mohamed Ghannam) [Orabug: 27390682] {CVE-2017-17712}
- mlx4: add mstflint secure boot access kernel support (Qing Huang) [Orabug: 27404202]
- x86: Move STUFF_RSB in to the idt macro (Konrad Rzeszutek Wilk)
- x86/spec: STUFF_RSB _before_ ENABLE_IBRS (Konrad Rzeszutek Wilk)
- x86: Move ENABLE_IBRS in the interrupt macro. (Konrad Rzeszutek Wilk) [Orabug: 27449045]
-
Thu Jan 18 2018 Brian Maly <brian.maly@oracle.com> [4.1.12-112.14.13.el7uek]
- Revert "kernel.spec: Require the new microcode_ctl." (Brian Maly)